From: Luke Kanies Date: 23:22 on 12 Oct 2005 Subject: Blogging sucks How is it that people can be so geeky and so geek-illiterate at the same time? Who the hell writes HTML any more? Don't we have, like, 100 simple markup languages that make it 10x easier and better? Really. I've been writing a software product (deserving of much hate, I am sure) recently, and have had little luck finding people with whom to discuss the trials and trevails of this process. So, I decided to begin recording the conversations I have with myself, at the least so that I have them to look back over, and maybe even to draw some more interest onto the software. Unfortunately, blog software seems to largely suck. (I know, I hear you gasping with surprise -- who knew?) First, it seems to be a pretty simple setup -- I write something, it gets put online. Ok, there are some other features -- comments (can get quite complicated), categories, pinging (don't ask), and tags. Ok, I understand. But what's the most important thing, the one thing you figure they'd get right? That's right, text editing. I write a lot of code snippets in my entries (because, you know, I'm writing code and developing a language) and I can't seem to get code snippets to show up well at all. Really. They look like crap, in addition to looking nothing like what I want. There's lots of hate here, and my options appear to be 1) don't blog, 2) do the writing but do it in a vacuum with no connectedness to the rest of the blog world, 3) use shitty software, 4) host something myself and hack it till it works. All of these suck. I understand the blog software is written for people who want to post pictures, not code snippets, and want to write using WYSIWYG not Vim, but please, this isn't that hard.
From: Steve Peters Date: 00:06 on 13 Oct 2005 Subject: Re: Blogging sucks ------=_Part_32080_30509310.1129158401918 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Content-Disposition: inline On 10/12/05, Luke Kanies <luke@xxxxxxx.xxx> wrote: > > First, it seems to be a pretty simple setup -- I write something, it > gets put online. Ok, there are some other features -- comments (can get > quite complicated), categories, pinging (don't ask), and tags. Ok, I > understand. > > Pinging is one of the most hateful parts of writing blog software. Mainly because it involves web services. Maybe it was just hateful because I was using Soap::Lite with Perl. Pure hate! Maybe it was because I was trying to interface with .NET-based web services. Ass tearingly painful hate! WSDL...what shithead designed this ludicrousness? My experiences with web services of all kinds have left me to realize that it is such a painful waste of time that I'd much rather clean the litter box or shave my dog's ass than try to debug a web service. Steve ------=_Part_32080_30509310.1129158401918 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Content-Disposition: inline On 10/12/05, <b class=3D"gmail_sendername">Luke Kanies</b> <<a href=3D"m= ailto:luke@xxxxxxx.xxx">luke@xxxxxxx.xxx</a>> wrote:<div><span class=3D"= gmail_quote"></span><blockquote class=3D"gmail_quote" style=3D"border-left:= 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex= ;"> First, it seems to be a pretty simple setup -- I write something, it<br>get= s put online. Ok, there are some other features -- comments (can= get<br>quite complicated), categories, pinging (don't ask), and tags. = ; Ok, I<br> understand.<br><br> </blockquote></div><br> Pinging is one of the most hateful parts of writing blog software. Mainly because it involves web services. Maybe it was just hateful because I was using Soap::Lite with Perl. Pure hate! Maybe it was because I was trying to interface with .NET-based web services. Ass tearingly painful hate! WSDL...what shithead designed this ludicrousness? My experiences with web services of all kinds have left me to realize that it is such a painful waste of time that I'd much rather clean the litter box or shave my dog's ass than try to debug a web service.<br> <br> Steve<br> <br> ------=_Part_32080_30509310.1129158401918--
From: A. Pagaltzis Date: 01:05 on 13 Oct 2005 Subject: Re: Blogging sucks * Luke Kanies <luke@xxxxxxx.xxx> [2005-10-13 00:25]: > There's lots of hate here, and my options appear to be 1) don't > blog, 2) do the writing but do it in a vacuum with no > connectedness to the rest of the blog world, 3) use shitty > software, 4) host something myself and hack it till it works. I have set myself up so that I can filter the contents of my clipboard through Markdown[1] at the touch of a key (intercepted by the window manager) – write stuff, select all, copy, Win-Shift-X, paste. Makes writing HTML in text edit boxes on web pages much less hateful, because the Markdown syntax as such is almost non-hateful. [1]: http://daringfireball.net/projects/markdown/ The converter has its own set of hate, like weird bugs in some corner cases or the syntax, but that really goes without saying, and at least it’s a bunch better than suffering through heaps of angle-bracketed noise. Regards,
From: Chris Devers
Date: 02:01 on 13 Oct 2005
Subject: Re: Blogging sucks
On Wed, 12 Oct 2005, Luke Kanies wrote:
> Who the hell writes HTML any more? Don't we have, like, 100 simple
> markup languages that make it 10x easier and better? Really.
<p> Oh, I dunno, a minimal, useful set of HTML isn't <i>that</i>
painful. I've seen people, with a straight face, say that it's
reasonable to put more effort into <tt>LaTeX</tt> or such things, so
sprinkling in a little bit of HTML doesn't seem that bad to me. </p>
<pre>
if ( you_have_code_to_embed() ) {
wrap_it_in_a_pre_block_and_get_on_with_your_life();
}
</pre>
<p> It could be better, sure, but for minimal HTML like this, a lot of
the "solutions" I've heard of seem more cumbersome than the original
problem they were supposed to solve. </p>
<p> YMMV. </p>
<p>
-- <br />
Chris Devers
</p>
From: peter (Peter da Silva)
Date: 03:31 on 13 Oct 2005
Subject: Re: Blogging sucks
> <pre>
>
> if ( you_have_code_to_embed() ) {
> wrap_it_in_a_pre_block_and_get_on_with_your_life();
> }
>
> </pre>
<pre>
for(i = 0; i < 10; i++) {
puts("Hey! Where'd my code go?");
if(!you_can_see_this())
break;
}
if(i>=10)
puts("Oh, there it is!");
</pre>
From: Chris Devers
Date: 03:36 on 13 Oct 2005
Subject: Re: Blogging sucks
On Wed, 12 Oct 2005, Peter da Silva wrote:
>> <pre>
>>
>> if ( you_have_code_to_embed() ) {
>> wrap_it_in_a_pre_block_and_get_on_with_your_life();
>> }
>>
>> </pre>
>
> <pre>
> for(i = 0; i < 10; i++) {
> puts("Hey! Where'd my code go?");
> if(!you_can_see_this())
> break;
> }
> if(i>=10)
> puts("Oh, there it is!");
> </pre>
Well, okay, there is that... :-)
This way may lie madness, but what about an upload filter that
auto-escapes angle brackets for anything wrapped in <pre> or <code>
tags? Surely some of the blogging toolkits have such abilities, as
plugins if not as core functionality, no?
From: Chris Nandor Date: 05:01 on 13 Oct 2005 Subject: Re: Blogging sucks At 22:36 -0400 2005.10.12, Chris Devers wrote: >This way may lie madness, but what about an upload filter that >auto-escapes angle brackets for anything wrapped in <pre> or <code> >tags? Surely some of the blogging toolkits have such abilities, as >plugins if not as core functionality, no? Slash (Slashdot, use.perl.org) have a fake tag called <ecode> that does a decent job of autoformatting your code, but of course it has problems with whitespace, because, well, it's HTML. And as <pre> is harmful, we don't allow it. But <ecode> does a decent enough job.
From: Peter da Silva Date: 05:27 on 13 Oct 2005 Subject: Re: Blogging sucks > This way may lie madness, but what about an upload filter that > auto-escapes angle brackets for anything wrapped in <pre> or <code> > tags? Surely some of the blogging toolkits have such abilities, as > plugins if not as core functionality, no? What if you've got </pre> or </code> as part of the code you're entering? Yo may be writing about HTML. I have a better idea. Don't use HTML as the input format. Don't use XML either. Use some format that isn't similar to what you're going to be working on. I recommend basing it on .em RUNOFF markup, with paired delimiters for literal sections. .lit .end Anything up to a .end is literal. Really literal. .end No, I'm .em not kidding.
From: Chris Nandor Date: 06:21 on 13 Oct 2005 Subject: Re: Blogging sucks At 23:27 -0500 2005.10.12, Peter da Silva wrote: >Use some format that isn't similar to what you're >going to be working on. > >I recommend basing it on >.em RUNOFF >markup, with paired delimiters for literal sections. >.lit .end >Anything up to a .end is literal. Which is great until you want to have an ".end" in your code ...
From: A. Pagaltzis Date: 07:47 on 13 Oct 2005 Subject: Re: Blogging sucks * Peter da Silva <peter@xxxxxxx.xxx> [2005-10-13 06:30]: > I recommend basing it on > .em RUNOFF > markup, with paired delimiters for literal sections. Markdown (yes, yes, I’m cheerleading) has a simple solution: anything indentend as a code block is literal. Period. Even Markdown syntax itself. It doesn’t matter. You indent it by 4 spaces, it’s literal. (This works in conjunction with other block formatting constructs, ie if you have a quote char and 4 spaces at the start of the line you’ll get a blockquoted code block.) Regards,
From: Peter da Silva Date: 16:20 on 13 Oct 2005 Subject: Re: Blogging sucks On Oct 13, 2005, at 8:48 AM, A. Pagaltzis wrote: > * Peter da Silva <peter@xxxxxxx.xxx> [2005-10-13 13:10]: >> On Oct 13, 2005, at 1:47 AM, A. Pagaltzis wrote: >>> Markdown (yes, yes, I=92m cheerleading) has a simple solution: >>> anything indentend as a code block is literal. Period. >> [insert tremendous hate for Python here] > Uhm, okay, but what does that have to do with anything? Reserving indentation for stuff like this is hateful. Period.=
From: Rafael Garcia-Suarez Date: 16:26 on 13 Oct 2005 Subject: Re: Blogging sucks On 10/13/05, Peter da Silva <peter@xxxxxxx.xxx> wrote: > On Oct 13, 2005, at 8:48 AM, A. Pagaltzis wrote: > > * Peter da Silva <peter@xxxxxxx.xxx> [2005-10-13 13:10]: > >> On Oct 13, 2005, at 1:47 AM, A. Pagaltzis wrote: > >>> Markdown (yes, yes, I'm cheerleading) has a simple solution: > >>> anything indentend as a code block is literal. Period. > > >> [insert tremendous hate for Python here] > > > Uhm, okay, but what does that have to do with anything? > > Reserving indentation for stuff like this is hateful. Period. On the other hand POD doesn't do differently. And I don't think I hate POD. I mostly love it (not all the tools that have grown around it, though)
From: Aaron J. Grier Date: 05:25 on 14 Oct 2005 Subject: Re: Blogging sucks On Thu, Oct 13, 2005 at 05:26:28PM +0200, Rafael Garcia-Suarez wrote: > On the other hand POD doesn't do differently. And I don't think I hate > POD. I mostly love it (not all the tools that have grown around it, > though) why the hell can't I "perldoc spamassassin" and have it do what I want? perldoc Mail::SpamAssassin::Conf ? fuck that. it's not how I invoke it. where's my "man 5 spamassassin" ?
From: Abigail
Date: 08:55 on 14 Oct 2005
Subject: Re: Blogging sucks
--ULyIDA2m8JTe+TiX
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable
On Thu, Oct 13, 2005 at 09:25:07PM -0700, Aaron J. Grier wrote:
> On Thu, Oct 13, 2005 at 05:26:28PM +0200, Rafael Garcia-Suarez wrote:
> > On the other hand POD doesn't do differently. And I don't think I hate
> > POD. I mostly love it (not all the tools that have grown around it,
> > though)
>=20
> why the hell can't I "perldoc spamassassin" and have it do what I want?
>=20
> perldoc Mail::SpamAssassin::Conf ? fuck that. it's not how I invoke it.
> where's my "man 5 spamassassin" ?
No idea. Why would it be in section 5? It's a command after all.
$ man 5 spamassassin > /dev/null 2>&1; echo $?
16
$ man 1 spamassassin > /dev/null 2>&1; echo $?
0
$ which spamassassin
/opt/perl/bin/spamassassin
Abigail
--ULyIDA2m8JTe+TiX
Content-Type: application/pgp-signature
Content-Disposition: inline
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.0 (GNU/Linux)
iD8DBQFDT2RwBOh7Ggo6rasRAq9kAJ9UfSIFgMFSwVLBzJpWkqiKgwKJcQCgqYIy
6VqOS5RYGsJ3KRjXEhsevhg=
=Txeg
-----END PGP SIGNATURE-----
--ULyIDA2m8JTe+TiX--
From: Aaron J. Grier Date: 19:01 on 14 Oct 2005 Subject: Re: Blogging sucks On Fri, Oct 14, 2005 at 09:55:28AM +0200, Abigail wrote: > > perldoc Mail::SpamAssassin::Conf ? fuck that. it's not how I invoke it. > > where's my "man 5 spamassassin" ? > > No idea. Why would it be in section 5? It's a command after all. > > $ man 5 spamassassin > /dev/null 2>&1; echo $? > 16 > $ man 1 spamassassin > /dev/null 2>&1; echo $? > 0 > $ which spamassassin > /opt/perl/bin/spamassassin I'm specifically referring to the configuration file (Mail::SpamAssassin::Conf), hence section 5. I invariably end up checking the spamassassin man page to remind myself what the magic perldoc variable (Mail::SpamAssassin::Conf) is. of course this just a specific instance of the tower of babel which is on-line documentation. (roff/texinfo/perldoc/sgml/html/etc...)
From: peter (Peter da Silva) Date: 19:45 on 14 Oct 2005 Subject: Re: Blogging sucks > of course this just a specific instance of the tower of babel which is > on-line documentation. (roff/texinfo/perldoc/sgml/html/etc...) Oh, if only more had come from The Committee for the Common Man than some tweaks to DocBook.
From: Abigail
Date: 20:41 on 14 Oct 2005
Subject: Re: Blogging sucks
--WRT3RXLOp/bBMgTI
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable
On Fri, Oct 14, 2005 at 11:01:12AM -0700, Aaron J. Grier wrote:
> On Fri, Oct 14, 2005 at 09:55:28AM +0200, Abigail wrote:
> > > perldoc Mail::SpamAssassin::Conf ? fuck that. it's not how I invoke =
it.
> > > where's my "man 5 spamassassin" ?
> >=20
> > No idea. Why would it be in section 5? It's a command after all.
> >=20
> > $ man 5 spamassassin > /dev/null 2>&1; echo $?
> > 16
> > $ man 1 spamassassin > /dev/null 2>&1; echo $?
> > 0
> > $ which spamassassin
> > /opt/perl/bin/spamassassin
>=20
> I'm specifically referring to the configuration file
> (Mail::SpamAssassin::Conf), hence section 5.
Well, if I want to find the manual page of 'Mail::SpamAssassin::Conf', I
would tell man that I want 'Mail::SpamAssassin::Conf', and not 'spamassassi=
n'.
$ man Mail::SpamAssassin::Conf > /dev/null 2>&1; echo $?
0
It's not installed in section 5 though - it's in section 3. But I seldomly
use the section numbers.
> I invariably end up checking the spamassassin man page to remind myself
> what the magic perldoc variable (Mail::SpamAssassin::Conf) is.
The only time I use perldoc is if I want the -f functionality. For complete
manual pages, 'man' will do fine. 'perldoc' is mainly there for Windows
weenies, who aren't used to manual pages.=20
Abigail
--WRT3RXLOp/bBMgTI
Content-Type: application/pgp-signature
Content-Disposition: inline
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.0 (GNU/Linux)
iD8DBQFDUAnmBOh7Ggo6rasRAmePAKC3Zz+ti/UEJrDEHmvGDb1I+UzYqwCfWiKi
JieZM08foypgBxkL1T1mDsg=
=p+eX
-----END PGP SIGNATURE-----
--WRT3RXLOp/bBMgTI--
From: A. Pagaltzis Date: 21:28 on 14 Oct 2005 Subject: Re: Blogging sucks * Abigail <abigail@xxxxxxx.xx> [2005-10-14 21:45]: > Well, if I want to find the manual page of > 'Mail::SpamAssassin::Conf', I would tell man that I want > 'Mail::SpamAssassin::Conf', and not 'spamassassin'. > > $ man Mail::SpamAssassin::Conf > /dev/null 2>&1; echo $? > 0 > > It's not installed in section 5 though - it's in section 3. But > I seldomly use the section numbers. I think what he is saying is that Mail::SpamAssassin::Conf should be called spamassassin(5). Regards,
From: Martin Ebourne Date: 23:25 on 14 Oct 2005 Subject: Re: Blogging sucks On Fri, 2005-10-14 at 21:41 +0200, Abigail wrote: > The only time I use perldoc is if I want the -f functionality. For complete > manual pages, 'man' will do fine. 'perldoc' is mainly there for Windows > weenies, who aren't used to manual pages. And on badly set up unix systems which don't have the man pages. Especially slowaris it seems. Martin.
From: A. Pagaltzis Date: 17:49 on 13 Oct 2005 Subject: Re: Blogging sucks * Peter da Silva <peter@xxxxxxx.xxx> [2005-10-13 17:25]: > Reserving indentation for stuff like this is hateful. Period. It isn’t anything like Python at all. There is only one level of indentation. Once a block is indented far enough to be considered a code block, all further indentation is literal. Not to mention that *all* alternative solutions are *really* hateful. Regards,
From: Luke Kanies Date: 17:52 on 13 Oct 2005 Subject: Re: Blogging sucks On Thu, 13 Oct 2005, A. Pagaltzis wrote: > It isn't anything like Python at all. There is only one level of > indentation. Once a block is indented far enough to be considered > a code block, all further indentation is literal. > > Not to mention that *all* alternative solutions are *really* > hateful. Strangely, I don't find Restructured Text's use of indentation to determine markup nearly as distasteful as I find python's. This is probably because text markup is 10x easier than code markup (e.g., I almost never have to worry about things like line continuations).
From: Abigail
Date: 20:41 on 13 Oct 2005
Subject: Significant whitespace (was Re: Blogging sucks)
--iFRdW5/EC4oqxDHL
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable
On Thu, Oct 13, 2005 at 11:52:16AM -0500, Luke Kanies wrote:
> On Thu, 13 Oct 2005, A. Pagaltzis wrote:
>=20
> > It isn't anything like Python at all. There is only one level of
> > indentation. Once a block is indented far enough to be considered
> > a code block, all further indentation is literal.
> >
> > Not to mention that *all* alternative solutions are *really*
> > hateful.
>=20
> Strangely, I don't find Restructured Text's use of indentation to
> determine markup nearly as distasteful as I find python's. This is
> probably because text markup is 10x easier than code markup (e.g.,
> I almost never have to worry about things like line continuations).
When it comes to hating significant whitespace, nothing comes close to Perl=
6:
sub square {my $x =3D shift; return $x * $x}
print square(1) * 2 # Prints 2.
print square (1) * 2 # Prints 4.
my %hash; # Empty hash.
if %hash{1} {print "Found it"} # No output.
if %hash {1} {print "Found it"} # Prints 'Found it'.
I rather take all Pythons significant whitespace rules (which aren't
that bad, and wouldn't require me to change my coding style) repeatedly
than Perl6 whitespace rules (which conflict with almost every line I
write) once.
Evil rules. Pure evil. Hates doesn't come close to describe it.
Abigail (not touching Perl6 with a 10-feet pole).
--iFRdW5/EC4oqxDHL
Content-Type: application/pgp-signature
Content-Disposition: inline
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.0 (GNU/Linux)
iD8DBQFDTrhXBOh7Ggo6rasRAhX2AJ9VHcAL9+FjOQGJJHta93OZeOMnhwCgt2xQ
PfKLH6JPBZjBj+UAB9XBCHY=
=KH+n
-----END PGP SIGNATURE-----
--iFRdW5/EC4oqxDHL--
From: Darrell Fuhriman Date: 20:55 on 13 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) > When it comes to hating significant whitespace, nothing comes close to Perl6: *boggle* I'd say that the lunatics have taken over the asylum, except that they were always in charge. However, they do seem to have gone off their meds. > I rather take all Pythons significant whitespace rules (which aren't > that bad, and wouldn't require me to change my coding style) repeatedly Yah, I think Python's whitespace rules are really not worth getting your panties in a wad over. It doesn't bother me at all -- but maybe that's because that's how I write code anyway. It would certainly prevent abominations like the procmail code from happening, so it can't be all bad. Darrell
From: Aaron J. Grier Date: 05:57 on 14 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) On Thu, Oct 13, 2005 at 12:55:51PM -0700, Darrell Fuhriman wrote: > It would certainly prevent abominations like the procmail code from > happening, so it can't be all bad. that has become one of my "difficult-to-read-C" references. at least it's consistent in its awkwardness...
From: peter (Peter da Silva)
Date: 21:17 on 13 Oct 2005
Subject: Re: Significant whitespace (was Re: Blogging sucks)
> When it comes to hating significant whitespace, nothing comes close to Perl6:
> sub square {my $x = shift; return $x * $x}
> print square(1) * 2 # Prints 2.
> print square (1) * 2 # Prints 4.
Holy screaming sentient inkwells full of boiling blood and maggots.
I don't care much for Perl5, but at least it was mostly an improvement
over perl4. But this...
From: Luke Kanies
Date: 21:22 on 13 Oct 2005
Subject: Re: Significant whitespace (was Re: Blogging sucks)
On Thu, 13 Oct 2005, Abigail wrote:
> When it comes to hating significant whitespace, nothing comes close to Perl6:
>
> sub square {my $x = shift; return $x * $x}
> print square(1) * 2 # Prints 2.
> print square (1) * 2 # Prints 4.
>
>
> my %hash; # Empty hash.
> if %hash{1} {print "Found it"} # No output.
> if %hash {1} {print "Found it"} # Prints 'Found it'.
I'm not sure who wins, this crap from perl, or the fact that a trailing
comma affects how 'print' functions in python:
print "this is a test"
print "another test"
print "a third test", # notice the trailing comma
print "a fourth test"'
Yields:
this is a test
another test
a third test a fourth test
Yes, a trailing comma determines whether 'print' adds a carriage return for
you. I can't even imagine how painfully difficult it must have been to
write a parser that can somehow tell functions whether they were called with
a trailing comma, considering that a comma (at least in this context) is
usually entirely for the parser -- they wouldn't even be there if the parser
didn't need them, and they're only used by the parser.
I'm not saying I don't understand _why_ you'd want two forms of the 'print'
function, I'm just saying this way makes my eyes bleed, and does a great job
of summing up why I do not now and never will use python except under
duress.
> I rather take all Pythons significant whitespace rules (which aren't
> that bad, and wouldn't require me to change my coding style) repeatedly
> than Perl6 whitespace rules (which conflict with almost every line I
> write) once.
Yeah, it's not python's whitespace that makes me avoid it like the plague.
"Ooh, we're OO, except that we're functional, except, when we just make
shit up, except... at least we're not perl!"
I sweat, I've found someone able to describe what they _liked_ about python
without comparing it to perl. It makes python-ruby script wars difficult:
me: ruby is awesome
them: perl sucks!
me: Bartender! Check, please!
> Evil rules. Pure evil. Hates doesn't come close to describe it.
>
>
> Abigail (not touching Perl6 with a 10-feet pole).
I am looking forward to perl 6 like some people look forward to new episodes
of 'Cops'. If Ruby looks good now, just imagine how good it will look when
Perl 6 comes out. "See, we've added new punctuation, to help solve the fact
that there was too much punctuation, and in order to make it easier to
understand, we've tripled the number of valid statement types." Like a 100
car pileup on an 8-lane highway.
I just hope Ruby is ready to accept all of the refugees. :)
From: peter (Peter da Silva) Date: 21:29 on 13 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) > Yes, a trailing comma determines whether 'print' adds a carriage return for > you. It's BASIC! > "Ooh, we're OO, except that we're functional, except, when we just make > shit up, except... at least we're not perl!" I hate all object oriented languages that don't even TRY to at least do as good a job as Smalltalk.
From: Darrell Fuhriman Date: 21:31 on 13 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) > Yeah, it's not python's whitespace that makes me avoid it like the plague. > "Ooh, we're OO, except that we're functional, except, when we just make > shit up, except... at least we're not perl!" Bah. You think python is bad by itself, you should see what the bastards at ESRI did to the poor thing. Rant on that to follow within the next couple days. Darrell
From: Abigail Date: 21:43 on 13 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) --hoZxPH4CaxYzWscb Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Thu, Oct 13, 2005 at 03:22:26PM -0500, Luke Kanies wrote: >=20 > I sweat, I've found someone able to describe what they _liked_ about pyth= on > without comparing it to perl. It makes python-ruby script wars difficult: >=20 > me: ruby is awesome > them: perl sucks! > me: Bartender! Check, please! One thing that bothers me about Python is that one of things they claim as a good thing about their language is that "it isn't Perl". And I have the same problem with both PHP and Ruby. Whenever I read a piece that's supposed to promote Python, PHP or Ruby, somewhere in the beginning they must remark how much better they are than Perl. They're just like Go players. They can't explain what's so great about their game either. But it's better than chess! Abigail --hoZxPH4CaxYzWscb Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.0 (GNU/Linux) iD8DBQFDTscABOh7Ggo6rasRAggUAJ9vRUAJCUjsTfUidZOYtQeHFd5PvQCghQPu +Q6vR5SsfeCLE5S4SaNrXGk= =3Gxz -----END PGP SIGNATURE----- --hoZxPH4CaxYzWscb--
From: Luke Kanies Date: 22:30 on 13 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) On Thu, 13 Oct 2005, Abigail wrote: > One thing that bothers me about Python is that one of things they claim > as a good thing about their language is that "it isn't Perl". And I have > the same problem with both PHP and Ruby. Whenever I read a piece that's > supposed to promote Python, PHP or Ruby, somewhere in the beginning they > must remark how much better they are than Perl. > > They're just like Go players. They can't explain what's so great about > their game either. But it's better than chess! An ironic comparison, I think, since the things that make ruby and go great are similar: The thing doesn't get in the way of the goal of the thing. Ruby has a relatively minimal (relative to, of course, LISP), and Go has a very simple ruleset. Ruby does a very good job of getting the hell out of your way, such that you can see the program, or the intent, behind the syntax, and Go does a good job of getting the hell out of your way so that you see that strategy behind the game. Or something; I haven't played Go (yet). It's always easier to compare something you understand to something the audience understands, so it's naturally easy to compare Ruby to perl and Go to chess, and most people are unfortunately poor at explaining, um, anything. I like Ruby because it seems to have the best combination of the features I care the most about, which largely amount to having a minimalist syntax without being so minimalist that it makes me feel like I'm actually writing in binary (LISP), along with allowing insane amounts of introspection (this is the feature that really just makes me go ga ga), and finishing up by being very dynamic (i.e., no static type-checking constantly -- I can make an array of different types of objects, and then call the same method on each object in turn, without having the parser think it's smarter than me and forcing me to do all kinds of stupid casting to tell the parser that I know what I'm doing) and not making me manage my own damn memory. I _hate_ doing things the computer can do for me.
From: David Champion Date: 23:19 on 13 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) * On 2005.10.13, in <Pine.OSX.4.58.0510131548190.17665@xxxxxx.xxxxx>, * "Luke Kanies" <luke@xxxxxxx.xxx> wrote: > > care the most about, which largely amount to having a minimalist syntax > without being so minimalist that it makes me feel like I'm actually writing > in binary (LISP), along with allowing insane amounts of introspection (this Ook. Ook?
From: Luke Kanies Date: 23:48 on 13 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) On Thu, 13 Oct 2005, David Champion wrote: > * On 2005.10.13, in <Pine.OSX.4.58.0510131548190.17665@xxxxxx.xxxxx>, > * "Luke Kanies" <luke@xxxxxxx.xxx> wrote: > > > > care the most about, which largely amount to having a minimalist syntax > > without being so minimalist that it makes me feel like I'm actually writing > > in binary (LISP), along with allowing insane amounts of introspection (this > > Ook. Ook? I think I lost something in the translation here...
From: David Champion Date: 08:05 on 14 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) * On 2005.10.13, in <Pine.OSX.4.58.0510131748070.17927@xxxxxx.xxxxxxx.xxx>, * "Luke Kanies" <luke@xxxxxxx.xxx> wrote: > > > > > > care the most about, which largely amount to having a minimalist > > > syntax without being so minimalist that it makes me feel like I'm > > > actually writing in binary (LISP), along with allowing insane > > > amounts of introspection (this > > > > Ook. Ook? > > I think I lost something in the translation here... http://www.dangermouse.net/esoteric/ook.html http://en.wikipedia.org/wiki/Ook! Ook! is isomorphic to Brainfuck (which is, itself, just P'' plus I/O operators), but is held to be interesting for (a) using fewer (3 vs. 8) syntactic elements and (b) looking like a transcript of a directors' meeting of my precious organization, and is anyone hiring in the Chicago area? However, because Ook!'s "syntactic elements" must occur in pairs, they're not really syntactic elements, are they? They're spellings, so (a) is a red herring -- three pairs is really 2^3 = 8 semantic tokens, just like Brainfuck, only it takes me longer to write. To fill the obHate prerequisite. I mean, Ook! is certainly more pronounceable than Brainfuck, but how often do you read your code aloud over the telephone, and wouldn't that be like picking pecks of Peter Piper's peppers anyway? So "Ook. Ook?" is the operator for "can we please move on?" Which isn't meant for you specifically, Luke, it's just that you're the one who said "minimalist" and gave me the excuse. I picked it over Brainfuck because if "Ook. Ook?" is going to throw someone, ">" certainly doesn't carry enough context to mean anything. ... I got my LISA conference plea/brochure today. Now that I've covered the *software* hate quotient, are you someone I can ask to PLEASE MAKE THEM STOP HOLDING IT AT THE TOWN & COUNTRY? Fashion Valley sounds like a television program for thirteen-year-olds. I'll give you that the overwhelming (no, really) sense of apocalypse and brimstone last time was rather exciting, up until I had to occupy that godawful conference center or sit in my corner room with virtually no cellular reception staring at the horrid child in the picture frame, just like the horrid child in every third room's picture frame, eking bandwidth by the bps. It's lucky I had dispensation to rent a car, or (with their room service) I'd never have found it possible to drown my troubles in a dusty bottle of Graham's 20. Can I bear to attend again? It probably depends on the weather, frankly, so bravo! to whomever picked San Diego in December, but a double dose of derision for not telling the T&C to take a hike.
From: Luke Kanies Date: 14:37 on 14 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) On Fri, 14 Oct 2005, David Champion wrote: > http://www.dangermouse.net/esoteric/ook.html > http://en.wikipedia.org/wiki/Ook! Ah. > So "Ook. Ook?" is the operator for "can we please move on?" Which isn't > meant for you specifically, Luke, it's just that you're the one who said > "minimalist" and gave me the excuse. I picked it over Brainfuck because > if "Ook. Ook?" is going to throw someone, ">" certainly doesn't carry > enough context to mean anything. Of course, I specifically allowed that a language (LISP) might be so minimalist that it makes it harder. > ... > > I got my LISA conference plea/brochure today. Now that I've covered > the *software* hate quotient, are you someone I can ask to PLEASE MAKE > THEM STOP HOLDING IT AT THE TOWN & COUNTRY? Fashion Valley sounds like > a television program for thirteen-year-olds. I'll give you that the > overwhelming (no, really) sense of apocalypse and brimstone last time > was rather exciting, up until I had to occupy that godawful conference > center or sit in my corner room with virtually no cellular reception > staring at the horrid child in the picture frame, just like the horrid > child in every third room's picture frame, eking bandwidth by the > bps. It's lucky I had dispensation to rent a car, or (with their room > service) I'd never have found it possible to drown my troubles in a > dusty bottle of Graham's 20. No, I can't get them to change it; we can only pray that this year they have net access in the rooms (probably not). Apparently they got some great deal if they signed up through eternity; I do know that they get painfully self-righteous when you bring it up. > Can I bear to attend again? It probably depends on the weather, > frankly, so bravo! to whomever picked San Diego in December, but a > double dose of derision for not telling the T&C to take a hike. LISA: Not On Fire This Year BTW, you can catch a train from the mall downtown. Cars not needed.
From: peter (Peter da Silva) Date: 23:22 on 13 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) > I like Ruby because it seems to have the best combination of the features I > care the most about, which largely amount to having a minimalist syntax > without being so minimalist that it makes me feel like I'm actually writing > in binary (LISP), That's what has kept me from really digging in to Ruby. It's got too much syntax. What kind of introspection can it REALLY support if it doesn't define control structures in terms of introspection like Lisp-family languages or Smalltalk?
From: Luke Kanies Date: 23:51 on 13 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) On Thu, 13 Oct 2005, Peter da Silva wrote: > That's what has kept me from really digging in to Ruby. It's got too much > syntax. What kind of introspection can it REALLY support if it doesn't > define control structures in terms of introspection like Lisp-family > languages or Smalltalk? I think I need more information here... Can you point me somewhere that would help me explain what you are talking about? I believe Ruby's control structures are intrinsic to the language, but the only control structures I ever seem to use are 'if', 'case', and blocks. I guess I don't really see how introspection is related to how control structures are defined, but then, I've already shown many times that I'm a newbie when it comes to language design. Kinda sad, since I'm now writing and maintaining a (thankfully not general purpose) language.
From: peter (Peter da Silva)
Date: 00:41 on 14 Oct 2005
Subject: Re: Significant whitespace (was Re: Blogging sucks)
I was being a bit tongue-in-cheek... but...
> I think I need more information here... Can you point me somewhere that
> would help me explain what you are talking about?
OK: in Smalltalk, Postscript, Lisp, Forth, and most other languages that
are aggressively introspective and reflective, control structures are
defined in terms of operations that can be performed within the language.
They may be implemented in the compiler or interpreter in a more efficient
way, but that's an implementation detail. In practice you could build
them out of other control structures and blocks if you wanted to.
Smalltalk:
(expression) ifTrue: [ block ] ifFalse: [ block ].
Postscript:
expression { block } { block } ifelse
Lisp:
(cond (expression code)
(expression code)
(expression code)
(t code))
Forth is a bit harder to see, but you can take my word for it that it's
got the same kind of capability to implement new control structures.
Basically, if you can implement control structures this way, then it's
inelegant to define them such that they don't work as if they're implemented
this way, this allowing them to be replaced or extended cleanly. Since
nobody would ever leave inelegant components in a language ...
From: Luke Kanies
Date: 21:14 on 19 Oct 2005
Subject: Re: Significant whitespace (was Re: Blogging sucks)
On Thu, 13 Oct 2005, Peter da Silva wrote:
> OK: in Smalltalk, Postscript, Lisp, Forth, and most other languages that
> are aggressively introspective and reflective, control structures are
> defined in terms of operations that can be performed within the language.
>
> They may be implemented in the compiler or interpreter in a more efficient
> way, but that's an implementation detail. In practice you could build
> them out of other control structures and blocks if you wanted to.
>
> Smalltalk:
> (expression) ifTrue: [ block ] ifFalse: [ block ].
>
> Postscript:
> expression { block } { block } ifelse
>
> Lisp:
> (cond (expression code)
> (expression code)
> (expression code)
> (t code))
>
> Forth is a bit harder to see, but you can take my word for it that it's
> got the same kind of capability to implement new control structures.
>
> Basically, if you can implement control structures this way, then it's
> inelegant to define them such that they don't work as if they're implemented
> this way, this allowing them to be replaced or extended cleanly. Since
> nobody would ever leave inelegant components in a language ...
Every time I'm convinced I understand this, I immediately think that I
don't.
I have a simplistic understanding of, for instance, how one can define
control structures within the language, rather than having to hard-code them
into the parser. I just don't understand how one can easily recognize
whether a language supports this feature.
From: Peter da Silva Date: 04:41 on 20 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) > I have a simplistic understanding of, for instance, how one can define > control structures within the language, rather than having to > hard-code them > into the parser. I just don't understand how one can easily recognize > whether a language supports this feature. It's pretty easy. If the control structures look like they could be implemented this way, then the language probably supports the feature (though not always... the Plan 9 shell "rc" seems to be a counterexample).
From: Chris Nandor Date: 23:12 on 13 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) At 15:22 -0500 2005.10.13, Luke Kanies wrote: >I'm not sure who wins, this crap from perl, or the fact that a trailing >comma affects how 'print' functions in python: Please. That is not perl. It is a new language based on perl, called perl6, where 6 is not a version number but part of the name.
From: peter (Peter da Silva) Date: 23:32 on 13 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) > Please. That is not perl. It is a new language based on perl, called > perl6, where 6 is not a version number but part of the name. That's nothing. The only thing in common betwen smail and smail3 was the five shared letters in the name. I have a hate for people who implement a new program and pretend it's a new version of an old one.
From: Chris Nandor Date: 23:35 on 13 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) At 17:32 -0500 2005.10.13, Peter da Silva wrote: >> Please. That is not perl. It is a new language based on perl, called >> perl6, where 6 is not a version number but part of the name. > >That's nothing. The only thing in common betwen smail and smail3 was the >five shared letters in the name. > >I have a hate for people who implement a new program and pretend it's a new >version of an old one. On this, we agree! Maybe we can lobby Larry to change the name; has anyone tried this yet?
From: Luke Kanies Date: 00:00 on 14 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) On Thu, 13 Oct 2005, Chris Nandor wrote: > Maybe we can lobby Larry to change the name; has anyone tried this yet? Diamond: Like perl, but much harder
From: Abigail Date: 00:03 on 14 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) --wTWi5aaYRw9ix9vO Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Thu, Oct 13, 2005 at 03:35:22PM -0700, Chris Nandor wrote: > At 17:32 -0500 2005.10.13, Peter da Silva wrote: > >> Please. That is not perl. It is a new language based on perl, called > >> perl6, where 6 is not a version number but part of the name. > > > >That's nothing. The only thing in common betwen smail and smail3 was the > >five shared letters in the name. > > > >I have a hate for people who implement a new program and pretend it's a = new > >version of an old one. >=20 > On this, we agree! >=20 > Maybe we can lobby Larry to change the name; has anyone tried this yet? It has been suggested, yes.=20 I think Larry prefers 'perl6' above anything else. Abigail --wTWi5aaYRw9ix9vO Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.0 (GNU/Linux) iD8DBQFDTufXBOh7Ggo6rasRAje/AJ9zEqouKlOXPecJdOairK9PMN6PFQCdFvfe sEYERrCHJy3itwQ87DqILgg= =pcHt -----END PGP SIGNATURE----- --wTWi5aaYRw9ix9vO--
From: Luke Kanies Date: 23:53 on 13 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) On Thu, 13 Oct 2005, Peter da Silva wrote: > I have a hate for people who implement a new program and pretend it's a new > version of an old one. I know a guy who rewrote a tool from scratch and insists on calling the tool <tool>2, instead of just renaming it. He can't seem to stand leaving off the numeral, even though literally no one anywhere uses the first version (it was only used on his own network). Pick a new name, or drop the 2, unless you're transitioning (or running Debian, apparently).
From: Luke Kanies Date: 23:47 on 13 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) On Thu, 13 Oct 2005, Chris Nandor wrote: > Please. That is not perl. It is a new language based on perl, called > perl6, where 6 is not a version number but part of the name. Hah! I guess I'd have to agree.
From: Chris Nandor Date: 23:09 on 13 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) At 21:41 +0200 2005.10.13, Abigail wrote: >Abigail (not touching Perl6 with a 10-feet pole). Should we have a separate hate-perl6@ mailing list? :-)
From: David Champion Date: 23:22 on 13 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) * On 2005.10.13, in <p0623091fbf748ba50352@xxx.x.x.xxxx>, * "Chris Nandor" <pudge@xxxxx.xxx> wrote: > > Should we have a separate hate-perl6@ mailing list? :-) Not necessary; that's implied.
From: Abigail
Date: 09:07 on 14 Oct 2005
Subject: Re: Significant whitespace (was Re: Blogging sucks)
--0z5c7mBtSy1wdr4F
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable
On Thu, Oct 13, 2005 at 09:41:12PM +0200, Abigail wrote:
>=20
> my %hash; # Empty hash.
> if %hash{1} {print "Found it"} # No output.
> if %hash {1} {print "Found it"} # Prints 'Found it'.
Someone pointed out that I was incorrect. It turns out to be _worse_:
my %hash; # Empty hash.
if %hash{1} {print "Found it"} # No output.
if %hash {1} {print "Found it"} # Syntax error.
if %hash {1}
{print "Found it"} # Prints 'Found it'.
Abigail
--0z5c7mBtSy1wdr4F
Content-Type: application/pgp-signature
Content-Disposition: inline
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.0 (GNU/Linux)
iD8DBQFDT2dcBOh7Ggo6rasRAsn+AKCYqJuUeQS85ytm++FRADTELvpbSwCdFSLK
58rHily0coznlBqA4qXBsdc=
=JG7S
-----END PGP SIGNATURE-----
--0z5c7mBtSy1wdr4F--
From: A. Pagaltzis
Date: 22:11 on 16 Oct 2005
Subject: Re: Significant whitespace (was Re: Blogging sucks)
* Abigail <abigail@xxxxxxx.xx> [2005-10-14 10:10]:
> Someone pointed out that I was incorrect. It turns out to be
> _worse_:
>
> my %hash; # Empty hash.
> if %hash{1} {print "Found it"} # No output.
> if %hash {1} {print "Found it"} # Syntax error.
> if %hash {1}
> {print "Found it"} # Prints 'Found it'.
That’s really just obfuscation based on expectations from Perl5.
Once you know that there must be no whitespace between the hash
name and the key lookup operator in Perl6, the parsing is
completely straightforward.
Of course, you are free to dislike that syntax rule.
Can you throw in whitespace in those examples if you use the new
%hash«key» syntax instead?
(And if not, would it at least be possible to parse the examples
umambiguously if it *were* allowed? If the answer to that
question is yes, then I would lobby for dropping the {} hash
lookup operator altogether and permitting whitespace with the «»
operator.)
Regards,
From: Abigail
Date: 22:51 on 16 Oct 2005
Subject: Re: Significant whitespace (was Re: Blogging sucks)
--Xb8pJpF45Qg/t7GZ
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable
On Sun, Oct 16, 2005 at 11:11:45PM +0200, A. Pagaltzis wrote:
> * Abigail <abigail@xxxxxxx.xx> [2005-10-14 10:10]:
> > Someone pointed out that I was incorrect. It turns out to be
> > _worse_:
> >=20
> > my %hash; # Empty hash.
> > if %hash{1} {print "Found it"} # No output.
> > if %hash {1} {print "Found it"} # Syntax error.
> > if %hash {1}
> > {print "Found it"} # Prints 'Found it'.
>=20
> That???s really just obfuscation based on expectations from Perl5.
And C. And Python. And Java. And Pascal. And about almost every language
on the sun that allows whitespace between an aggregate and its indexing
operator.
> Once you know that there must be no whitespace between the hash
> name and the key lookup operator in Perl6, the parsing is
> completely straightforward.
"Once you know". That's the point. "Once you know" all the behaviour of
software, no software is hateful. It's just a matter of knowing it. Once.
I hate Perl6 not because I don't know the rules, I hate Perl6 because
it robs me of a coding style I've used for the past 20 years, in any langua=
ge
I've used. Switching to Perl6 seems less logical than switching to Windows.
> Of course, you are free to dislike that syntax rule.
>=20
> Can you throw in whitespace in those examples if you use the new
> %hash??key?? syntax instead?
Question marks? No idea to which syntax you are referring.
> (And if not, would it at least be possible to parse the examples
> umambiguously if it *were* allowed? If the answer to that
> question is yes, then I would lobby for dropping the {} hash
> lookup operator altogether and permitting whitespace with the ????
> operator.)
I've heard about the ??:: operator, replacing the ?: one. But ????, no idea.
Abigail
--Xb8pJpF45Qg/t7GZ
Content-Type: application/pgp-signature
Content-Disposition: inline
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.0 (GNU/Linux)
iD8DBQFDUstrBOh7Ggo6rasRAkD0AKClY6cJO2gWrttDFz70OzbD9FbUhgCggy9l
oy5grhxlv1zVmmccFtixx1U=
=AacV
-----END PGP SIGNATURE-----
--Xb8pJpF45Qg/t7GZ--
From: A. Pagaltzis
Date: 23:12 on 16 Oct 2005
Subject: Re: Significant whitespace (was Re: Blogging sucks)
* Abigail <abigail@xxxxxxx.xx> [2005-10-16 23:55]:
> > Can you throw in whitespace in those examples if you use the
> > new %hash??key?? syntax instead?
>
> Question marks? No idea to which syntax you are referring.
>
> > (And if not, would it at least be possible to parse the
> > examples umambiguously if it *were* allowed? If the answer to
> > that question is yes, then I would lobby for dropping the {}
> > hash lookup operator altogether and permitting whitespace
> > with the ???? operator.)
>
> I've heard about the ??:: operator, replacing the ?: one. But
> ????, no idea.
In both cases I am talking about the %hash<<key>> syntax (per
ASCII-transliterated version of the operator).
Bonus hate:
Software which does not cope with UTF-8, like your MUA, is
massively hateful.
But you're using a sufficiently recent mutt, so it is
configurable to cope. Adding `//TRANSLIT` to your $charset
should suffice to make to make a best-effort attempt to
transliterate undisplayable glyphs to something available in
your charset. No, I'm not sending UTF-8 out of flowery
notions, I actually need it. Yes, my MUA is configured
correctly.
Regards,
From: David Cantrell Date: 23:55 on 16 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) On Mon, Oct 17, 2005 at 12:12:39AM +0200, A. Pagaltzis wrote: > Bonus hate: > > Software which does not cope with UTF-8, like your MUA, is > massively hateful. > > But you're using a sufficiently recent mutt, so it is > configurable to cope. Adding `//TRANSLIT` to your $charset > should suffice to make to make a best-effort attempt to > transliterate undisplayable glyphs to something available in > your charset. No, I'm not sending UTF-8 out of flowery > notions, I actually need it. Yes, my MUA is configured > correctly. You mean that if I configure mutt "correctly" my vt-320 will magically support all those funny heathen characters? How does that work?
From: A. Pagaltzis Date: 00:03 on 17 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) * David Cantrell <david@xxxxxxxx.xxx.xx> [2005-10-17 01:00]: > You mean that if I configure mutt "correctly" my vt-320 will > magically support all those funny heathen characters? How does > that work? No, it will transliterate things. Ie it will display « as << and » as >>, will turn curly quotes into straight ones, etc. Much better than just getting question marks. Regards,
From: peter (Peter da Silva) Date: 01:15 on 17 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) > Software which does not cope with UTF-8, like your MUA, is > massively hateful. [...] > I'm not sending UTF-8 out of flowery > notions, I actually need it. Programming languages that include characters not in USASCII are hateful. Whether it's Perl6 or Applescript. Whether the characters are optional or not. If you want APL, you know where to find it.
From: David Cantrell Date: 12:24 on 17 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) On Sun, Oct 16, 2005 at 07:15:42PM -0500, Peter da Silva wrote: > Programming languages that include characters not in USASCII are hateful. Hear hear! Programming languages that LET YOU include characters not in ASCII are hateful. That's all of them. If you want to use a funny heathen character, put it in some species of external resource file.
From: Abigail
Date: 08:34 on 17 Oct 2005
Subject: Re: Significant whitespace (was Re: Blogging sucks)
--NzB8fVQJ5HfG6fxh
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable
On Mon, Oct 17, 2005 at 12:12:39AM +0200, A. Pagaltzis wrote:
> * Abigail <abigail@xxxxxxx.xx> [2005-10-16 23:55]:
> > > Can you throw in whitespace in those examples if you use the
> > > new %hash??key?? syntax instead?
> >=20
> > Question marks? No idea to which syntax you are referring.
> >=20
> > > (And if not, would it at least be possible to parse the
> > > examples umambiguously if it *were* allowed? If the answer to
> > > that question is yes, then I would lobby for dropping the {}
> > > hash lookup operator altogether and permitting whitespace
> > > with the ???? operator.)
> >=20
> > I've heard about the ??:: operator, replacing the ?: one. But
> > ????, no idea.
>=20
> In both cases I am talking about the %hash<<key>> syntax (per
> ASCII-transliterated version of the operator).
Ah, that one. I think that this week[*] they are going for:
%hash<key>
However, that's another subject for hate. Currenly, in Perl5, one can do:
$hash {EXPRESSION}
or:
$hash {bare_word}
as a short-cut for:
$hash {"bare_word"}
And perl is smart enough to figure out what you want. Not so in
Perl6. Perl6 will be a massive attempt to get all the DWIM from Perl
(the thing that makes Perl Perl), and make the programmer a slave of
the compiler. So, no more auto-quoting of hash keys. Instead, you got to
tell the compiler you have a bare word. If you use <> to index in a hash,
it's going to be a bare word. If you don't use a bare word, you use {}.
Fuckers. If I wanted Java, I know where to find it.
[*] Actually, that was last week. I've no idea what this weeks syntax
is going to be.
Abigail
--NzB8fVQJ5HfG6fxh
Content-Type: application/pgp-signature
Content-Disposition: inline
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.0 (GNU/Linux)
iD8DBQFDU1PqBOh7Ggo6rasRAneSAJ9NROCKflsNIpnEHvhAiMpsdRGKoACfWdeE
OXh9QPxnhg6hP7PUIfJ9Zyw=
=UOmv
-----END PGP SIGNATURE-----
--NzB8fVQJ5HfG6fxh--
From: Juerd
Date: 08:58 on 17 Oct 2005
Subject: Re: Significant whitespace (was Re: Blogging sucks)
Abigail skribis 2005-10-17 9:34 (+0200):
> > In both cases I am talking about the %hash<<key>> syntax (per
> > ASCII-transliterated version of the operator).
> Ah, that one. I think that this week[*] they are going for:
> %hash<key>
Single angle brackets are for word lists, similar to Perl 5's qw//
operator.
<foo bar baz>
is just an easy way to write
('foo', 'bar', 'baz')
Double angle brackets are also for word lists. Symmetric with quotes,
these interpolate (expand) variables. << and >> can be written as « and
» if you wish. I prefer the ASCII version.
<<foo bar $quux>>
is just an easy way to write
"foo bar $quux".split
# not necessarily ("foo", "bar", "$quux")
Hashes are subscripted with the .{} method:
%hash.{$key}
The . can be left out in almost all cases, resulting in just
%hash{key}
However, if one chooses to leave out the dot, whitespace is no longer
allowed, while with the dot, you may have whitespace on its LHS:
%hash{key}
%hash.{key}
%hash .{key}
%hash<key1 key2 ... keyn> and %hash<<key1 key2 ... keyn>> are just
shortcuts: %hash<> is a shortcut for %hash{<>}, and %hash<<>> is a
shortcut for %hash{<<>>}. (Here too, the actual operators are .<> and
.<<>>, and the dot is optional.)
> [*] Actually, that was last week. I've no idea what this weeks syntax
> is going to be.
This joke is no longer funny, as the design is currently so stable that
basic syntax no longer changes every week.
Juerd
From: peter (Peter da Silva)
Date: 12:44 on 17 Oct 2005
Subject: Re: Significant whitespace (was Re: Blogging sucks)
> <<foo bar $quux>>
> is just an easy way to write
> "foo bar $quux".split
> # not necessarily ("foo", "bar", "$quux")
If you're going to provide "a handy way to write" something, at least make it
a "handy way to write" the version of that something that doesn't make it
easier for people to write exploits.
If the number of elements in <<foo bar $quux>> (or any similar construct)
depends on what's in $quux, that's not just hateful... it's evil. You might
as well go back to the shell.
> This joke is no longer funny, as the design is currently so stable that
> basic syntax no longer changes every week.
The fact that you can write this, seriously, makes the joke funnier.
From: Juerd Date: 13:12 on 17 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) Peter da Silva skribis 2005-10-17 6:44 (-0500): > If the number of elements in <<foo bar $quux>> (or any similar construct) > depends on what's in $quux, that's not just hateful... it's evil. You might > as well go back to the shell. Agreed. Juerd
From: Jonathan Stowe
Date: 23:01 on 16 Oct 2005
Subject: Re: Significant whitespace (was Re: Blogging sucks)
On Sun, 2005-10-16 at 22:11, A. Pagaltzis wrote:
> I would lobby for dropping the {} hash
> lookup operator altogether and permitting whitespace with the «»
> operator.)
Well except there isn't a « or » on any keyboard I have seen recently,
are we reinventing APL here ?
/J\
From: A. Pagaltzis Date: 23:14 on 16 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) * Jonathan Stowe <gellyfish@xxxxxxxxx.xxx> [2005-10-17 00:05]: > Well except there isn't a « or » on any keyboard I have seen > recently, are we reinventing APL here ? You can use ASCII transliterations of all operators, like <<>> for «». But in Vim you enter them as digraphs. Don’t like that part much, either. Would have preferred they not stray from ASCII. Regards,
From: Martin Ebourne Date: 23:43 on 16 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) On Mon, 2005-10-17 at 00:14 +0200, A. Pagaltzis wrote: > * Jonathan Stowe <gellyfish@xxxxxxxxx.xxx> [2005-10-17 00:05]: > > Well except there isn't a =C2=AB or =C2=BB on any keyboard I have see= n > > recently, are we reinventing APL here ?=20 >=20 > You can use ASCII transliterations of all operators, like <<>> > for =C2=AB=C2=BB. But in Vim you enter them as digraphs. That's pure evil. I can handle whitespace nonsense, but that's just put me right off perl6. All languages should handle input and output in all character sets, but none of them should be written in anything but ASCII. At least for the next 20 years. Maybe then things will be different, but for now the breakage is just huge. Well, we've even seen it tonight, something as simple as email. Cheers, Martin. PS. Is there a hardware hate list around? I couldn't find one. And I have some serious hardware hate. I hope you don't mind the intrusion: Spent most of the day fixing in place and wiring up a new cctv camera. Finally got it all finished and perfect. Plugged it in, ENOWORK. Turns out the fucker takes a 9V supply, unlike all other cctv cameras which take 12V. So now it's landfill and I've got to start all over again. Of course, when I tested it initially I used the power supply that came with it, so it worked fine. And there was no big fat red warning telling of the difference of this power supply from the others. It's even exactly the same type of power connector. Bollocks.
From: Juerd Date: 23:52 on 16 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) Martin Ebourne skribis 2005-10-16 23:43 (+0100): > All languages should handle input and output in all character sets, but > none of them should be written in anything but ASCII. At least for the > next 20 years. Maybe then things will be different, but for now the > breakage is just huge. Well, we've even seen it tonight, something as > simple as email. Finally, someone who thinks Perl 6 will be released within 20 years. Juerd
From: A. Pagaltzis Date: 00:01 on 17 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) * Martin Ebourne <lists@xxxxxxx.xx.xx> [2005-10-17 00:45]: > All languages should handle input and output in all character > sets, but none of them should be written in anything but ASCII. You *can* use ASCII transliterations. %hash<<key>> means the same as %hash«key», and from what I’ve seen there are only a few operators which have a non-ASCII variant at all. > At least for the next 20 years. I would like to agree, but I fear that if we all take this stance then things will be just as broken in 20 years, and 20 years after that, and another 20 later. I18n is one of those things that is going to be so painful until it becomes completely pervasive that I don’t know what else other than forcing breakage might motivate people to get cracking. I really don’t know if Perl6 is making the right decision here or not. It may turn out a disaster. Only time will tell. If nothing else, Perl6 is a grand experiment that people after it will be able to learn a lot from. Its role in life may be to serve as a warning to others… we’ll see what transpires. Regards,
From: Juerd
Date: 00:16 on 17 Oct 2005
Subject: Re: Significant whitespace (was Re: Blogging sucks)
A. Pagaltzis skribis 2005-10-17 1:01 (+0200):
> You *can* use ASCII transliterations. %hash<<key>> means the same
> as %hash«key», and from what I???ve seen there are only a few
> operators which have a non-ASCII variant at all.
There are no operators in standard Perl 6 that do not have a non-ASCII
variant. Of course, you can define your own operators without having to
specify ASCII variants. But that's up to you.
Furthermore, all non-ASCII operators in standard Perl 6 are also
in iso-8859-1 (latin-1).
Currently, only these glyphs are used:
¥ (Y)
« (<<)
» (>>)
I have no idea how my current konsole, ssh, screen, mutt and vim are
configured, and frankly do not care much anymore. It's not worth the
pain.
Juerd
From: A. Pagaltzis Date: 00:35 on 17 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) * Juerd <juerd@xxxxxxxxxxx.xx> [2005-10-17 01:20]: > A. Pagaltzis skribis 2005-10-17 1:01 (+0200): > > You *can* use ASCII transliterations. %hash<<key>> means the > > same as %hash«key», and from what I???ve seen there are only > > a few operators which have a non-ASCII variant at all. > > There are no operators in standard Perl 6 that do not have a > non-ASCII variant. I didn’t say that. I said what you are saying: that all operators are available as ASCII, and only a few are *also* are available as other glyphs. > I have no idea how my current konsole, ssh, screen, mutt and > vim are configured, and frankly do not care much anymore. It's > not worth the pain. The whole notion of a terminal (and terminal emulator) is obsolete and just makes things more hateful. Regards,
From: Luke Kanies
Date: 04:46 on 17 Oct 2005
Subject: Re: Significant whitespace (was Re: Blogging sucks)
On Mon, 17 Oct 2005, Juerd wrote:
> =A5 (Y)
> =AB (<<)
> =BB (>>)
*stuttering shock*
Those actually show up correctly in my rxvt on OS X...
Not that that will get me using perl 6, either.
--=20
The optimist proclaims that we live in the best of all possible worlds,
and the pessimist fears that this is true.
-- James Branch Cabell 1879-1958
---------------------------------------------------------------------
Luke Kanies | http://reductivelabs.com | http://config-mgmt.blogspot.com
From: Matthew Garrett Date: 05:21 on 17 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) On Sun, Oct 16, 2005 at 10:46:17PM -0500, Luke Kanies wrote: > On Mon, 17 Oct 2005, Juerd wrote: > > > Â¥ (Y) > > « (<<) > > » (>>) > > *stuttering shock* > > Those actually show up correctly in my rxvt on OS X... Given that large swathes of people are able to deal with character sets more complicated than iso 8859-1 (which seems to have existed for, ooh, about 20 years), it really shouldn't be that much of a shock that some number of applications actually support it properly. Stuff like â˜, on the other hand...
From: A. Pagaltzis Date: 06:37 on 17 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) * Matthew Garrett <mjg59@xxxx.xxxx.xxx> [2005-10-17 06:25]: > Stuff like â˜, on the other hand... Mostly a font issue. The infrastructure is in place. I can’t *see* the character, but it’s gonna show up in this mail unmangled. Several projects exist which aim to create fonts with glyphs for all or at least a lot of Unicode, but it’s just an incredibly vast undertaking to make even one font that covers just the BMP’s 65536 characters (even though there are holes in the code point set), let alone all of the astral planes. And until distributors provide it OOTB, it’s a *MASSIVE* pain to get UTF-8 support set up. It took me a lot of sweat to get all the pieces just right so mailing in English, Greek and German all work equally well. As a bonus, Chinese spam gets displayed correctly as well… Regards,
From: Luke Kanies Date: 07:01 on 17 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) On Mon, 17 Oct 2005, A. Pagaltzis wrote: > Mostly a font issue. The infrastructure is in place. I can=E2=80=99t > *see* the character, but it=E2=80=99s gonna show up in this mail > unmangled. What I don't understand is what stinking moron decided that other character sets would use a different coding for the apostrophe character. I get this in Aristotle's emails (gibberish instead of '), and I get it all the time i= n Firefox on web sites, and it just doesn't make sense to me. Really. Did you _have_ to use that coding for something else? Where do you live? --=20 Measure with a micrometer. Mark with chalk. Cut with an axe. --------------------------------------------------------------------- Luke Kanies | http://reductivelabs.com | http://config-mgmt.blogspot.com
From: peter (Peter da Silva) Date: 07:52 on 17 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) > What I don't understand is what stinking moron decided that other character > sets would use a different coding for the apostrophe character. Steve Jobs, Bill Gates, and some unsung idiot at Adobe all independantly came up with the idea of cramming "smart quites" into the character set, all in different ways.
From: A. Pagaltzis Date: 15:26 on 17 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) * Peter da Silva <peter@xxxxxxx.xxx> [2005-10-17 08:55]: > Steve Jobs, Bill Gates, and some unsung idiot at Adobe all > independantly came up with the idea of cramming "smart quites" > into the character set, all in different ways. I’m sending neither MacRoman nor Windows-1252 (either of which would be a massively hateful thing to do), so this doesn’t apply. Prior to Unicode it was quite customary to do what they did, too. Just another reason to put a bullet in all the 8-bit charsets. Regards,
From: A. Pagaltzis Date: 15:23 on 17 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) * Luke Kanies <luke@xxxxxxx.xxx> [2005-10-17 08:05]: > What I don't understand is what stinking moron decided that > other character sets would use a different coding for the > apostrophe character. This is a straight apostrophe: ' This is a curly apostrophe: ’ > I get this in Aristotle's emails (gibberish instead of '), It’s because your mailer sucks. (Are you surprised?) (It might have been the sending mailer too, but I know my mail configuration is correct.) > and I get it all the time in Firefox on web sites, and it just > doesn't make sense to me. Because either the webserver/CMS config is borked so it sends the wrong charset (surprise!) or whoever wrote the HTML document used mixed charsets (well, can’t blame them too hard, charsets aren’t stuff your mum or the ACME Corp CEO should have to to understand). Software configured to use anything other than UTF-8 (or, at the very least another UTF-*) is hateful. I can’t wait until single-byte charsets are a thing of the past. Regards,
From: David Cantrell Date: 15:36 on 17 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) On Mon, Oct 17, 2005 at 04:23:12PM +0200, A. Pagaltzis wrote: > Software configured to use anything other than UTF-8 (or, at the > very least another UTF-*) is hateful. I can???t wait until > single-byte charsets are a thing of the past. Variable width character sets are themselves hateful. I'll go further and say that they are a spectacularly stupid idea, and that whoever decreed them needs shootin'. Yes, being able to represent more than 220-odd characters is a Good Idea. So damnit, just use 32-bit - or 64-bit - characters. Then you'll be able to seek!
From: Matthew Garrett Date: 15:56 on 17 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) On Mon, Oct 17, 2005 at 03:36:16PM +0100, David Cantrell wrote: > Variable width character sets are themselves hateful. I'll go further > and say that they are a spectacularly stupid idea, and that whoever > decreed them needs shootin'. Yes, being able to represent more than > 220-odd characters is a Good Idea. So damnit, just use 32-bit - or > 64-bit - characters. Then you'll be able to seek! And have to rewrite huge quantities of software. UTF-8 has its disadvantages, but there's very little hateful about it - it achieves what it sets out to do perfectly, and does so without many tools having to be rewritten.
From: David Cantrell Date: 18:04 on 17 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) On Mon, Oct 17, 2005 at 03:56:40PM +0100, Matthew Garrett wrote: > On Mon, Oct 17, 2005 at 03:36:16PM +0100, David Cantrell wrote: > > Variable width character sets are themselves hateful. I'll go further > > and say that they are a spectacularly stupid idea, and that whoever > > decreed them needs shootin'. Yes, being able to represent more than > > 220-odd characters is a Good Idea. So damnit, just use 32-bit - or > > 64-bit - characters. Then you'll be able to seek! > And have to rewrite huge quantities of software. UTF-8 has its > disadvantages, but there's very little hateful about it - it achieves > what it sets out to do perfectly, and does so without many tools having > to be rewritten. Well, apart from all those tools that have to process utf-8 text, of course. Like M*As, string manipulation libraries, terminals, ...
From: Matthew Garrett Date: 18:12 on 17 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) On Mon, Oct 17, 2005 at 06:04:41PM +0100, David Cantrell wrote: > On Mon, Oct 17, 2005 at 03:56:40PM +0100, Matthew Garrett wrote: > > And have to rewrite huge quantities of software. UTF-8 has its > > disadvantages, but there's very little hateful about it - it achieves > > what it sets out to do perfectly, and does so without many tools having > > to be rewritten. > > Well, apart from all those tools that have to process utf-8 text, of > course. Like M*As, string manipulation libraries, terminals, ... MTAs shouldn't be interpreting any utf-8. MUAs should only care about getting the content-type header correct, and optionally converting to the "simplest" character set. Terminals and string manipulation libraries need rewriting, yes. The number of them that exist is quite small compared to the number of applications that store strings. UCS-4 breaks little assumptions like "A byte with a null in is the end of a string". That's fairly hateful.
From: peter (Peter da Silva) Date: 19:22 on 17 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) > MTAs shouldn't be interpreting any utf-8. MTA is actually an overloaded term, seeing as it incorporates mail transport agents and mail routing agents. In addition software that implements mail transport and routing includes spam filters, virus checkers, vacation and other automatic response software, gateways, and so on. > UCS-4 breaks little assumptions like "A byte with a null in is the end > of a string". Any software that deals with "bytes" rather than "characters" is going to have to be rewritten. Right now it's being rewritten to use UTF-8 using libraries that incorporate all kinds of mind-bogglingly heavy l11n and i18n code that has immediate, obvious, and significant performance costs. Even setting LOCALE to "C" doesn't get you back to the performance you had with software that dealt with characters as atomic objects at the machine level. Metaphorically changing (sizeof(char)) to 4 is about the only way to get that performance back. The cost of reading and writing 4 times as many bytes per character is swamped by the 10x or greater cost of i18n and l11n code.
From: David Cantrell Date: 14:19 on 18 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) On Mon, Oct 17, 2005 at 06:12:25PM +0100, Matthew Garrett wrote: > MTAs shouldn't be interpreting any utf-8. MUAs should only care about > getting the content-type header correct, and optionally converting to > the "simplest" character set. MUAs also need to transmogrify any funny heathen characters in the subject line. MDAs need fiddling so that they can examine those subjects usefully. There are approxiamtely thirty trillion other programs which need updating anyway - anything which deals with information at any level higher than the byte. cat would probably be OK. tr, sed, awk, cut, head, tail all need fixin because of their assumptions that character == byte, but then they need fixing regardless of whether we use hateful utf-8 or less hateful large equal-length characters. > Terminals and string manipulation > libraries need rewriting, yes. The number of them that exist is quite > small compared to the number of applications that store strings. > > UCS-4 breaks little assumptions like "A byte with a null in is the end > of a string". That's fairly hateful. And replaces it with "0x00000000 is the end of a string". This is still == 0. Of course, having a magic number of any kind mark the end of a string is hateful, and is only needed because certain languages aren't clever enough to store length and text.
From: A. Pagaltzis Date: 16:09 on 17 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) * David Cantrell <david@xxxxxxxx.xxx.xx> [2005-10-17 16:40]: > Variable width character sets are themselves hateful. I'll go > further and say that they are a spectacularly stupid idea, and > that whoever decreed them needs shootin'. Yes, being able to > represent more than 220-odd characters is a Good Idea. So > damnit, just use 32-bit - or 64-bit - characters. Then you'll > be able to seek! Yuck, then you’re saddled with endianness issues. Plus null bytes can then be part of the data, so most charset-oblivious software breaks. Not worth it, considering that 99.99% of text processing is either gluing strings together without looking inside, or processing them character-by-character. Blindly indexing into a string without having scanned it previously is so rare it doesn’t merrit consideration. Regards,
From: peter (Peter da Silva) Date: 19:12 on 17 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) > Yuck, then you’re saddled with endianness issues. Good, that'll shake out the last of the big-endian systems. > Plus null bytes > can then be part of the data, so most charset-oblivious software > breaks. I thought breaking 8-bit-only software was a good thing. > Not worth it, considering that 99.99% of text processing > is either gluing strings together without looking inside, > or processing them character-by-character. Processing them character by character in UCS-4 is so much easier than doing it in UTF-8. So is gluing them together. > Blindly indexing into a > string without having scanned it previously is so rare it doesn’t > merrit consideration. Blindly indexing into a file without having scanned it previously is so common that you don't even remark on it happening. A file, remember, is a string.
From: A. Pagaltzis Date: 19:47 on 17 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) * Peter da Silva <peter@xxxxxxx.xxx> [2005-10-17 20:15]: > > Plus null bytes can then be part of the data, so most > > charset-oblivious software breaks. > > I thought breaking 8-bit-only software was a good thing. I said charset-*oblivious*. A lot of software passes around strings without ever processing them. It would be pretty pointless to force that sort of code to deal with encoding issues; just make sure null termination continues to work and the software will happily work with Unicode as well as it does otherwise. A mailer OTOH, in its interaction with the terminal and with the editor used to compose mail, needs to understand charsets, at least insofar as that it needs to know how to convert them. > > Not worth it, considering that 99.99% of text processing is > > either gluing strings together without looking inside, or > > processing them character-by-character. > > Processing them character by character in UCS-4 is so much > easier than doing it in UTF-8. So is gluing them together. Walking through the string is marginally more difficult with UTF-8 than single-byte charsets, granted. That said, I’ve written the code to do it straight from spec, twice, and got it right on the first try both times. But where you get the idea that gluing them together is difficult, I have no idea. What’s difficult about strcat?! > > Blindly indexing into a string without having scanned it > > previously is so rare it doesn’t merrit consideration. > > Blindly indexing into a file without having scanned it > previously is so common that you don't even remark on it > happening. > > A file, remember, is a string. I don’t see how that is relevant. I can’t think of any common use case where you’d be seeking blindly into a text file; though plenty where you’d be seeking blindly within binary files. But I’m not talking about heaps of bytes. Text is not the same thing as a sequence of bytes, even though it’s stored in one. A heap of bytes can be interpreted as text only if you know what encoding it’s in. If you see 0x41 in a heap of bytes, that’s only “capital A†because ASCII says so. That doesn’t mean every heap of bytes should be encoded as if it were a bunch of UTF-8 codepoint. It’s not, it’s a heap of bytes. But text is a heap of characters, and text should be encoded into a heap of bytes by way of the UTF-8 encoding. Now if you’re processing text files, you don’t index blindly anyway, you scan for newlines (cf. character-by-character processing). And that works in UTF-8 as well as it always has with ASCII: the extra bytes of a multibyte character never conflict with the <127 codepoints, so you can just scan for 0x10. Regards,
From: peter (Peter da Silva) Date: 20:01 on 17 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) > I said charset-*oblivious*. A lot of software passes around > strings without ever processing them. It would be pretty > pointless to force that sort of code to deal with encoding > issues; just make sure null termination continues to work and the > software will happily work with Unicode as well as it does > otherwise. Except that it bloody well doesn't. If you'd ever used 8-bit-oblivious software on a Unicode-oriented system for any length of time you'd know that.
From: peter (Peter da Silva) Date: 18:58 on 17 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) > > Software configured to use anything other than UTF-8 (or, at the > > very least another UTF-*) is hateful. I can???t wait until > > single-byte charsets are a thing of the past. > Variable width character sets are themselves hateful. I'll go further > and say that they are a spectacularly stupid idea, and that whoever > decreed them needs shootin'. Yes, being able to represent more than > 220-odd characters is a Good Idea. So damnit, just use 32-bit - or > 64-bit - characters. Then you'll be able to seek! Agreed. Don't piddle around with UCS-2, go straight to UCS-4. In fact ISO-10646 is way too conservative for my tastes. I think each distinct set of case transformation rules should have four 16-bit planes allocated to it, so that truly internationalised characters will be able to reliably toggle case when c&0x10000000 by flipping c&0x20000000. The current set of wishy-washy unified characters with c&0x10000000 == 0 should be left to rot like the hateful legacy things that they are.
From: Philip Newton Date: 06:33 on 18 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) On 10/17/05, Peter da Silva <peter@xxxxxxx.xxx> wrote: > In fact ISO-10646 is way too conservative for my tastes. I think each > distinct set of case transformation rules should have four 16-bit planes > allocated to it, so that truly internationalised characters will be able > to reliably toggle case when c&0x10000000 by flipping c&0x20000000. The > current set of wishy-washy unified characters with c&0x10000000 =3D=3D 0 = should > be left to rot like the hateful legacy things that they are. This "toggle" you speak of implies that you believe there are only two case= s. Thanks to Unicode (and legacy character sets, and existing alphabets which use digraphs which made their way into those legacy character sets, and round-tripping between alphabets which use digraphs and those which don't), we have three, though: upper-case, lower-case, and title-case. The difference between upper-case and title-case becomes apparent with such characters as "nj" (ASCII surrogate for U+01CC LATIN SMALL LETTER NJ), which becomes "Nj" in title-case but "NJ" in upper-case. I'm not sure whether the hate in this case is for the coded character sets that allowed digraphs in as single characters, or the fact that given they exist, software ignores them by blindly uppercasing rather titlecasing when appropriate. (That's not even considering the characters where casing depends on context; the most famous examples are probably Greek sigma (which has two lower-case forms, depending [roughly speaking] on whether it's the last letter of the word) and the letter "i" (which has two upper-case forms, one with dot and one without, depending on the language).) -- Philip Newton <philip.newton@xxxxx.xxx>
From: Peter da Silva Date: 07:24 on 18 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) On Oct 18, 2005, at 12:33 AM, Philip Newton wrote: > The difference between upper-case and title-case becomes apparent with > such characters as "nj" (ASCII surrogate for U+01CC LATIN SMALL LETTER > NJ), which becomes "Nj" in title-case but "NJ" in upper-case. So? There's lots of extra bits available. Give them 8 planes. You can have upper- lower- title- and author-case. > the letter "i" (which has two upper-case > forms, one with dot and one without, depending on the language).) No, there's latin-letter-i which uppercases to latin-letter-I, and turkic-letter-dotted-i which uppercases to turkic-letter-dotted-I, and turkic-letter-dotless-i which uppercases to turcik-letter-dotless-I. Well, it's not encoded that way in Unicode, but that's because of the limited imagination I already mentioned.
From: demerphq Date: 09:37 on 19 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) On 10/17/05, Peter da Silva <peter@xxxxxxx.xxx> wrote: > > > Software configured to use anything other than UTF-8 (or, at the > > > very least another UTF-*) is hateful. I can???t wait until > > > single-byte charsets are a thing of the past. > > > Variable width character sets are themselves hateful. I'll go further > > and say that they are a spectacularly stupid idea, and that whoever > > decreed them needs shootin'. Yes, being able to represent more than > > 220-odd characters is a Good Idea. So damnit, just use 32-bit - or > > 64-bit - characters. Then you'll be able to seek! > > Agreed. Don't piddle around with UCS-2, go straight to UCS-4. > > In fact ISO-10646 is way too conservative for my tastes. I think each > distinct set of case transformation rules should have four 16-bit planes > allocated to it, so that truly internationalised characters will be able > to reliably toggle case when c&0x10000000 by flipping c&0x20000000. The > current set of wishy-washy unified characters with c&0x10000000 =3D=3D 0 = should > be left to rot like the hateful legacy things that they are. What do you mean by "reliably toggle case"? My understanding is that while western european alphabets tend to have two cases, many languages have more than two. Thus I believe it is actually meaningless to "toggle case" in an internationalized context. Anyway, if you look at alphabets and languages as software they are all hateful. Unfortunately sacrificing backwards compatibility to resolve the problems in this sphere is completely nonviable. :-) -- perl -Mre=3Ddebug -e "/just|another|perl|hacker/"
From: A. Pagaltzis Date: 09:45 on 19 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) * demerphq <demerphq@xxxxx.xxx> [2005-10-19 10:40]: > What do you mean by "reliably toggle case"? My understanding is > that while western european alphabets tend to have two cases, > many languages have more than two. Thus I believe it is > actually meaningless to "toggle case" in an internationalized > context. And many others have no concept of "case" at all. Regards,
From: Peter da Silva Date: 04:45 on 20 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) > Anyway, if you look at alphabets and languages as software they are > all hateful. Unfortunately sacrificing backwards compatibility to > resolve the problems in this sphere is completely nonviable. :-) Just so long as you don't try and argue that the pile of crap that's come out of the political process is something other than a pile of crap, and that just increasing sizeof(char) wouldn't be infinitely less hateful. That doesn't mean it wouldn't be hateful, it's just that being better than what's been developed so far is terribly easy.
From: Aaron Crane Date: 08:49 on 20 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) Peter da Silva writes: > Just so long as you don't try and argue that [...] just increasing > sizeof(char) wouldn't be infinitely less hateful. It would be pretty hard, though: in C, sizeof(char) is 1 by definition. Or, to put it another way: sizeof measures in multiples of char.
From: Aaron J. Grier Date: 15:40 on 24 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) On Mon, Oct 17, 2005 at 03:36:16PM +0100, David Cantrell wrote: > Variable width character sets are themselves hateful. but they're easier than fixed-width charsets to read.
From: A. Pagaltzis Date: 15:47 on 24 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) * Aaron J. Grier <agrier@xxxxxxxxx.xxx> [2005-10-24 16:45]: > On Mon, Oct 17, 2005 at 03:36:16PM +0100, David Cantrell wrote: > > Variable width character sets are themselves hateful. > > but they're easier than fixed-width charsets to read. Aren’t you confusing charsets and fonts? Regards,
From: David Cantrell Date: 16:38 on 25 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) On Mon, Oct 24, 2005 at 04:47:39PM +0200, A. Pagaltzis wrote: > * Aaron J. Grier <agrier@xxxxxxxxx.xxx> [2005-10-24 16:45]: > > On Mon, Oct 17, 2005 at 03:36:16PM +0100, David Cantrell wrote: > > > Variable width character sets are themselves hateful. > > but they're easier than fixed-width charsets to read. > Aren???t you confusing charsets and fonts? <FOOM> That sound was a joke WHIZZING past your head.
From: Luke Kanies Date: 16:02 on 17 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) On Mon, 17 Oct 2005, A. Pagaltzis wrote: > This is a straight apostrophe: ' > This is a curly apostrophe: =E2=80=99 Can you explain to me why you or anyone else would bother using a curly apostrophe? Is the key more conveniently placed? Do you just like annoyin= g people who are stuck with stupider interfaces and/or software? > > I get this in Aristotle's emails (gibberish instead of '), > > It=E2=80=99s because your mailer sucks. (Are you surprised?) I'm guessing Pine has no relationship to whether this works; I'm thinking instead that it's the terminal itself. But I don't know, and I'm not likel= y to spend the hours and hours it would take to find out and then fix every terminal I have. > Because either the webserver/CMS config is borked so it sends the > wrong charset (surprise!) or whoever wrote the HTML document used > mixed charsets (well, can=E2=80=99t blame them too hard, charsets aren=E2= =80=99t > stuff your mum or the ACME Corp CEO should have to to understand). Grr. It just seems stupid to screw with a character that was already working fine in the lower bits. I feel similarly but not quite as bitter about people who do stupid "smart" quotes, because wow they're bloody annoying to handle with parsing programs. --=20 That was just a drill of the emergency y2k system. Had this been a real emergency, we would've also dumped a bucket of spiders on you and yelled out "civilization is collapsing!" --------------------------------------------------------------------- Luke Kanies | http://reductivelabs.com | http://config-mgmt.blogspot.com
From: A. Pagaltzis Date: 16:21 on 17 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) * Luke Kanies <luke@xxxxxxx.xxx> [2005-10-17 17:05]: > I'm guessing Pine has no relationship to whether this works; > I'm thinking instead that it's the terminal itself. But I > don't know, and I'm not likely to spend the hours and hours it > would take to find out and then fix every terminal I have. If it was the terminal then the character would show up screwy, but it wouldn't get mangled in your reply. > It just seems stupid to screw with a character that was already > working fine in the lower bits. I feel similarly but not quite > as bitter about people who do stupid "smart" quotes, because > wow they're bloody annoying to handle with parsing programs. What I'm using *is* a smart quote, I just think that's a stupid name. The straight quotes were a typewriter invention, along with monospace fonts. Someday computers will catch up to all the features that paper had about 400 years ago. PS.: this mail with straight quotes for your pleasure. Regards,
From: Juerd Date: 16:37 on 17 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) A. Pagaltzis skribis 2005-10-17 17:21 (+0200): > The straight quotes were a typewriter invention, along with > monospace fonts. And both have proven to work extremely well in many circumstances. Also, the apostrophe is not a single quote. You send us RIGHT SINGLE QUOTATION MARK (U+2019), in places where you should use APOSTROPHE (U+0027). It's hateful, but acceptable to send us special quotes, but not to replace apostrophes. Apostrophes are straight. (Careful readers will note that APOSTROPHE is the ASCII character abused for single quote. I don't mind, and even prefer, this abuse.) Next time, please get things right (or straight) before you send the wrong thing and belittle your audience's settings. Juerd
From: Juerd Date: 16:43 on 17 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) Juerd skribis 2005-10-17 17:37 (+0200): > Also, the apostrophe is not a single quote. You send us RIGHT SINGLE > QUOTATION MARK (U+2019), in places where you should use APOSTROPHE > (U+0027). I stand somewhat-corrected. Several sources say that U+2019 is okay to be used for apostrophes. Who would have thought that with all this redundancy, they'd get two semantically so very different things in a single codepoint? My point stands, though, and is now extended to unicode-hate. Juerd
From: A. Pagaltzis Date: 17:05 on 17 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) * Juerd <juerd@xxxxxxxxxxx.xx> [2005-10-17 17:45]: > Who would have thought that with all this redundancy, they'd > get two semantically so very different things in a single > codepoint? Yes – the one and sole thing I really hate about Unicode. Regards,
From: Matthew Garrett Date: 16:44 on 17 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) On Mon, Oct 17, 2005 at 05:37:43PM +0200, Juerd wrote: > Also, the apostrophe is not a single quote. You send us RIGHT SINGLE > QUOTATION MARK (U+2019), in places where you should use APOSTROPHE > (U+0027). U+2019 RIGHT SINGLE QUOTATION MARK Notes: • this is the preferred character to use for apostrophe (From the Unicode spec) > Next time, please get things right (or straight) before you send the > wrong thing and belittle your audience's settings. Zounds. U+0027 is sufficiently overloaded with a large number of historical meanings that it's preferable to use a more precise Unicode code point wherever possible.
From: Juerd Date: 16:45 on 17 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) Matthew Garrett skribis 2005-10-17 16:44 (+0100): > U+0027 is sufficiently overloaded with a large number of historical > meanings that it's preferable to use a more precise Unicode code point > wherever possible. U+2019, having two very different uses, isn't quite more precise. Juerd
From: Matthew Garrett Date: 16:51 on 17 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) On Mon, Oct 17, 2005 at 05:45:34PM +0200, Juerd wrote: > Matthew Garrett skribis 2005-10-17 16:44 (+0100): > > U+0027 is sufficiently overloaded with a large number of historical > > meanings that it's preferable to use a more precise Unicode code point > > wherever possible. > > U+2019, having two very different uses, isn't quite more precise. It does at least have the advantage that you know it's going to look curly.
From: Abigail Date: 18:41 on 17 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) --C7zPtVaVf+AK4Oqc Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Mon, Oct 17, 2005 at 04:51:21PM +0100, Matthew Garrett wrote: > On Mon, Oct 17, 2005 at 05:45:34PM +0200, Juerd wrote: > > Matthew Garrett skribis 2005-10-17 16:44 (+0100): > > > U+0027 is sufficiently overloaded with a large number of historical= =20 > > > meanings that it's preferable to use a more precise Unicode code poin= t=20 > > > wherever possible. > >=20 > > U+2019, having two very different uses, isn't quite more precise. >=20 > It does at least have the advantage that you know it's going to look=20 > curly. That's about as dumb as using <BLOCKQUOTE> in HTML for the indentation effect. =46rom http://www.unicode.org/standard/principles.html: Interpreting Characters and Rendering Glyphs The difference between identifying a code point and rendering it on screen or paper is crucial to understanding the Unicode Standard's role in text processing. The character identified by a Unicode code point is an abstract entity, such as "LATIN CHARACTER CAPITAL A" or "BENGALI DIGIT 5." The mark made on screen or paper -- called a glyph -- is a visual representation of the character. The Unicode Standard does not define glyph images. The standard defines how characters are interpreted, not how glyphs are rendered. The software or hardware-rendering engine of a computer is responsible for the appearance of the characters on the screen. The Unicode Standard does not specify the size, shape, nor style of on-screen characters. If you want to send to something that looks curly, send me a PDF or=20 an image. If you want to indicate you're contracting two words, use a apostrophe. And let *me* decide whether I want them straight or curly. Abigail --C7zPtVaVf+AK4Oqc Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.0 (GNU/Linux) iD8DBQFDU+JGBOh7Ggo6rasRAvhRAJ9qV6zjZsKAyBNKA3+fnd6rbMlXLwCgl+Y0 DtiLQ+T9F6F3UZzUoymj9HI= =vQVy -----END PGP SIGNATURE----- --C7zPtVaVf+AK4Oqc--
From: A. Pagaltzis Date: 18:55 on 17 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) * Abigail <abigail@xxxxxxx.xx> [2005-10-17 19:45]: > If you want to send to something that looks curly, send me a > PDF or an image. If you want to indicate you're contracting two > words, use a apostrophe. And let *me* decide whether I want > them straight or curly. Meanwhile everyone is using apostrophes as straight quote marks. This would have been sane if Unicode had conceded the overloaded meaning of U+0027 and defined a separate apostrophe character distinct from U+2019. As I said, the one hateful aspect about it. Of course, that new apostrophe codepoint would show up just as mangled in non-Unicode-aware apps as the curly quotes do. Regards,
From: Abigail Date: 19:02 on 17 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) --u3/rZRmxL6MmkK24 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Mon, Oct 17, 2005 at 07:55:14PM +0200, A. Pagaltzis wrote: > * Abigail <abigail@xxxxxxx.xx> [2005-10-17 19:45]: > > If you want to send to something that looks curly, send me a > > PDF or an image. If you want to indicate you're contracting two > > words, use a apostrophe. And let *me* decide whether I want > > them straight or curly. >=20 > Meanwhile everyone is using apostrophes as straight quote marks. Yeah. And guess what? I've been using computers since 1981, and it hasn't bothered me once. I'm human, and I understand context. It's the same character used for different purposes, but what purpose it is,=20 is always clear. > This would have been sane if Unicode had conceded the overloaded > meaning of U+0027 and defined a separate apostrophe character > distinct from U+2019. >=20 > As I said, the one hateful aspect about it. >=20 > Of course, that new apostrophe codepoint would show up just as > mangled in non-Unicode-aware apps as the curly quotes do. My terminal doesn't do Unicode (and the ones that do suck in different ways - I'm sticking with rxvt). Nor does the font (6x13) I use has glyphs= =20 above 255.=20 Abigail --u3/rZRmxL6MmkK24 Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.0 (GNU/Linux) iD8DBQFDU+c9BOh7Ggo6rasRAoODAJ9tMkPo0uC+H+K1kN0mRfZ4AjkmpwCeP86q EigfRyqE0lKVhh7daODUz30= =+wgf -----END PGP SIGNATURE----- --u3/rZRmxL6MmkK24--
From: A. Pagaltzis Date: 19:50 on 17 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) * Abigail <abigail@xxxxxxx.xx> [2005-10-17 20:05]: > My terminal doesn't do Unicode (and the ones that do suck in > different ways - I'm sticking with rxvt). You may or may not want to try urxvt alias rxvt-unicode. Which is what I’m using. Regards,
From: David Cantrell Date: 14:27 on 18 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) On Mon, Oct 17, 2005 at 07:55:14PM +0200, A. Pagaltzis wrote: > Meanwhile everyone is using apostrophes as straight quote marks. > This would have been sane if Unicode had conceded the overloaded > meaning of U+0027 and defined a separate apostrophe character > distinct from U+2019. > > As I said, the one hateful aspect about it. If that's the only thing about the Unicode charset that you find hateful then you've not looked at it very hard. There's a picture of a fucking snowman in there.
From: David Cantrell Date: 00:22 on 19 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) On Tue, Oct 18, 2005 at 09:00:40PM +0300, Jarkko Hietaniemi wrote: > > If that's the only thing about the Unicode charset that you find hateful > > then you've not looked at it very hard. > I find especially hateful people who whine about Unicode. I think most people here think that Unicode is a Good Idea. It's just a good idea badly implemented. Putting a fucking snowman in the character set is Stupid. Then there's the 'Floral Heart', U+2766, about a thousand different arrow symbols, ... Just look at all the shit in the Dingbats block! Sorry, but if you want decorative symbols, pay a graphic designer for some graphics, they do not belong in a character set. Variable width characters are Stupid too. > much more attentive to the whining once those who whine come up with > anything even approximating the width and breadth of Unicode (dozens > of scripts, hundreds of languages, thousands of glyphs), while still > more or less successfully navigating the cultural/political murky > waters. Those who whine should stick to their bloody A-Z and just > shut up. > > Sorry, David, you hit a sore point there. Nothing personal. Assuming for no good reason that I am a whiney little shit *is* kinda personal. But I forgive you. ObHate: replies sent both to me and to the list, because usually the first one I get is the one sent directly to me, so its identical twin gets shuffled off to Mr. Dave Null, and I miss the benefits of the reply-to munging which I asked Siesta to do for me. Which means that Jarkko is going to get two copies of this, with different msgids, and one of them with this extra helping of vitriol.
From: Yoz Grahame Date: 01:46 on 19 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) On 10/19/05, David Cantrell <david@xxxxxxxx.xxx.xx> wrote: > On Tue, Oct 18, 2005 at 09:00:40PM +0300, Jarkko Hietaniemi wrote: > > > If that's the only thing about the Unicode charset that you find hate= ful > > > then you've not looked at it very hard. > > I find especially hateful people who whine about Unicode. > > I think most people here think that Unicode is a Good Idea. It's just a > good idea badly implemented. > > Putting a fucking snowman in the character set is Stupid. Then there's > the 'Floral Heart', U+2766, about a thousand different arrow symbols, > ... Just look at all the shit in the Dingbats block! Sorry, but if you > want decorative symbols, pay a graphic designer for some graphics, they > do not belong in a character set. > > Variable width characters are Stupid too. You've missed a Bloody Huge Wodge of Unicode stupidity, which is that it does CJKV character sets so badly that most East Asians ignore it (which is why Ruby's Unicode support doesn't exist): http://london.pm.org/pipermail/london.pm/Week-of-Mon-20050822/033034.html But at least the snowmen are happy. -- Yoz
From: Earle Martin Date: 19:08 on 19 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) On Mon, Oct 17, 2005 at 07:55:14PM +0200, A. Pagaltzis wrote: > Meanwhile everyone is using apostrophes as straight quote marks. Which reminds me. Some people do "quoted" text ``like this''. Which is one of the most moronic things in the world. It crops up a lot in UNIX-type system documentation, which says a lot.
From: Chris Devers Date: 19:18 on 19 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) This message is in MIME format. The first part should be readable text, while the remaining parts are likely unreadable without MIME-aware tools. --0-1993319727-1129745909=:707 Content-Type: TEXT/PLAIN; charset=ISO-8859-1 Content-Transfer-Encoding: QUOTED-PRINTABLE On Wed, 19 Oct 2005, Earle Martin wrote: > On Mon, Oct 17, 2005 at 07:55:14PM +0200, A. Pagaltzis wrote: > > Meanwhile everyone is using apostrophes as straight quote marks. >=20 > Which reminds me. Some people do "quoted" text ``like this''. Which is=20 > one of the most moronic things in the world. It crops up a lot in=20 > UNIX-type system documentation, which says a lot. Hey, at least the quotes are balanced this way. Thank heaven for easily parseable miracles.=20 Quoting `like this', now that is a monstrosity. --=20 Chris Devers =17h`c8=F5=12=166=A1=BC=F5 --0-1993319727-1129745909=:707--
From: Rafael Garcia-Suarez Date: 20:05 on 19 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) On 10/19/05, Earle Martin <hates-software@xxxxxxxx.xxx> wrote: > Which reminds me. Some people do "quoted" text ``like this''. Which is on= e > of the most moronic things in the world. It crops up a lot in UNIX-type > system documentation, which says a lot. And in pod2html output. I have a vim macro specially designed to fix those.
From: Peter da Silva Date: 04:38 on 20 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) > Which reminds me. Some people do "quoted" text ``like this''. Which is > one > of the most moronic things in the world. It crops up a lot in UNIX-type > system documentation, which says a lot. That's because on a C/A/T phototypesetter back in the '70s when most of the documents that people have been cargo-culting from for 30 years were written, ``...'' came out with the right kind of curly quotes.
From: Luke Kanies Date: 23:07 on 18 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) On Mon, 17 Oct 2005, A. Pagaltzis wrote: > If it was the terminal then the character would show up screwy, > but it wouldn't get mangled in your reply. Okay, I'll have to trust you on that. > What I'm using *is* a smart quote, I just think that's a stupid > name. > > The straight quotes were a typewriter invention, along with > monospace fonts. Someday computers will catch up to all the > features that paper had about 400 years ago. All typography is a relatively recent invention; hell, punctuation itself isn't all that old, relative to typewriters, and certainly any sort of established standard is painfully recent, so specifically using a more annoying form of punctuation because of bitterness towards typewriters seems unjustifiably cruel to those of us who have to suffer through them. > PS.: this mail with straight quotes for your pleasure. And it is much appreciated.
From: A. Pagaltzis Date: 00:33 on 19 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) * Luke Kanies <luke@xxxxxxx.xxx> [2005-10-19 00:10]: > All typography is a relatively recent invention; hell, > punctuation itself isn't all that old, relative to typewriters, Excuse me? Typography as we know it today exists since Gutenberg; even the history of commercial font foundries goes back a century before the typewriter. Further, a lot of the rules of this relatively new form of typograhpy borrow from prior conventions established when copying manuscripts was laborious manual work. And the choice and placement of quotes is a grammatic rule. Half a millenium of prior art and craft was pragmatically sacrified to the limitations of constructing a mechanical device. But a computer does not have those. We can do better now. > and certainly any sort of established standard is painfully > recent, so specifically using a more annoying form of > punctuation because of bitterness towards typewriters seems > unjustifiably cruel to those of us who have to suffer through > them. It's only a problem because of lackluster Unicode support. Which I'm relying on to begin with because as noted elsewhere, I write mail in Greek and German as well as English, and sometimes a combination of several of those within a single mail. I use curly quotes because writing in Unicode means they're available to me anyway. What I've found, if it's of interest, is that in mail, Unicode is somewhat hit-and-miss yet; a substantial amount of people who get my mail have problems, maybe as many as a third. (Also, mailing list archives are frothing hateful about this.) But it works much more reliably on the whole than you'd expect. Most GUI mailers, in particular, work smoothly. (I suppose that explains why I've had more complaints on this list than anywhere else, by a *long* shot.) On the web, OTOH, support is solid. If the publishing setup (at least the webserver, if it's just static files) at a site is configured correctly, all browsers do the right thing. Regards,
From: Luke Kanies Date: 21:55 on 19 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) On Wed, 19 Oct 2005, A. Pagaltzis wrote: > Excuse me? Typography as we know it today exists since Gutenberg; > even the history of commercial font foundries goes back a century > before the typewriter. Further, a lot of the rules of this > relatively new form of typograhpy borrow from prior conventions > established when copying manuscripts was laborious manual work. > And the choice and placement of quotes is a grammatic rule. > > Half a millenium of prior art and craft was pragmatically > sacrified to the limitations of constructing a mechanical device. There have been huge changes in typography over that half of a millenium. The existence of a history in no way implies that that history was consistent, and if typogrophy itself is anything like punctuation (which seems likely) there have probably been drastic changes every generation or so. While it's downright difficult to find "authoritative" sources, here's what I could come up with: According to Lynne Truss's delightful 'Eats, Shoots & Leaves', the apostrophe was introduced into English in the 16th century, which puts it at a minimum of 50 years past Gutenberg. She has nothing that I can find to say about the typography of said character, but considering how drastically the usage of the apostrophe has changed (from elided letters to possessives and back again) I would be surprised if there were a universal, consistent history of its typography. I can't seem to find much of anything online about it, other than pages telling me how important it is that I spend lots of extra effort typing a very specific puncuation mark that, as far as I can tell, provides no benefit. > But a computer does not have those. We can do better now. Well, some of us can. I personally don't think I've ever seen those characters outside of MSWord or something -- I get a heckuva lot of funky blocks in all of my web browsers, and it's already obvious that my mail reader doesn't show them correctly. I'm downright pedantic about some aspects of language, but I can clearly say that typography is not one of them. Notice that I offensively used two hyphens in the middle of a sentence? That's at least as bad as using a straight apostrophe, I assume, but, well, I don't care. And I'm going to continue using three periods to indicate an ellipsis. All because it's more important that I communicate easily and successfully than that I agree with a relatively arbitrary standard of typography, which is such a stupid phrase that I feel dumber writing it. > > and certainly any sort of established standard is painfully > > recent, so specifically using a more annoying form of > > punctuation because of bitterness towards typewriters seems > > unjustifiably cruel to those of us who have to suffer through > > them. > > It's only a problem because of lackluster Unicode support. Which I'm > relying on to begin with because as noted elsewhere, I write mail > in Greek and German as well as English, and sometimes a > combination of several of those within a single mail. I use > curly quotes because writing in Unicode means they're available > to me anyway. > > What I've found, if it's of interest, is that in mail, Unicode is > somewhat hit-and-miss yet; a substantial amount of people who get > my mail have problems, maybe as many as a third. (Also, mailing > list archives are frothing hateful about this.) But it works much > more reliably on the whole than you'd expect. Most GUI mailers, > in particular, work smoothly. I'm quite impressed that you don't mind that up to a third of the recipients of your email, and all archives of your email, are worse off because your adherence to a specific typographic system. > (I suppose that explains why I've had more complaints on this > list than anywhere else, by a *long* shot.) POssibly. But it's also on-topic on this list. :) > On the web, OTOH, support is solid. If the publishing setup (at > least the webserver, if it's just static files) at a site is > configured correctly, all browsers do the right thing. I haven't had that same luck; or at least, I know that many of my browsers (I bounce around machines and browsers pretty regularly) often display crap instead of an apostrophe. And let me tell you, I would _always_ prefer to see a typewriter apostrophe than crap.
From: peter (Peter da Silva) Date: 19:04 on 17 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) > Can you explain to me why you or anyone else would bother using a curly > apostrophe? Because stupid GUI software automatically turns the one into the other, instead of treating it as a presentation issue. I mean, if the editor can tell when to do the conversion on *input*, then the display device can just as easily do it on *output*. Same with smart-quotes, smart-ellipses, and so on. I'm surprised Windows doesn't convert "file" into "file" just to be difficult.
From: Philip Newton Date: 06:38 on 18 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) On 10/17/05, Peter da Silva <peter@xxxxxxx.xxx> wrote: > I'm surprised Windows doesn't convert "file" into "file" just to be = difficult. Mac Keynote apparently does this. It looked exceedingly odd to me when this happened in monospaced text that represented code -- having a variable called $filename seemed most inappropriate. I thought this had happened by mistake, and I mentioned it to the presenter; he showed me that this was done automatically and on-the-fly: moving after the fi and deleting backwards one character turned it into $flename automatically rather than deleting the entire fi "ligature" (which was apparently merely on-screen and not in the underlying storage). Apparently, similar software is also used to typeset some books and magazines; not only does it mess up with people who align things with spaces (since "file" is one character-width shorter than "pile" in the cases I've seen involving monospace characters), but if you interpret it literally, you'll get code that doesn't even compile since it uses non-ASCII characters in identifiers. (Unless you have software that does allow this, which was already hated on further up.) I mean, first we had typesetting software automatically turning two consecutive hyphens into en or em dashes, which looks weird in command-line invocations such as "foo --help"; that seems to be on the decline now. I'm not sure whether this automatic-ligaturing in code is better or worse, though. -- Philip Newton <philip.newton@xxxxx.xxx>
From: A. Pagaltzis Date: 07:14 on 18 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) * Philip Newton <philip.newton@xxxxx.xxx> [2005-10-18 07:40]: > I mean, first we had typesetting software automatically turning > two consecutive hyphens into en or em dashes, which looks weird > in command-line invocations such as "foo --help"; that seems to > be on the decline now. I'm not sure whether this > automatic-ligaturing in code is better or worse, though. It still does – that’s why I use `alias man='LC_CTYPE=C man'` on my UTF-8 terminals. OTOH, SmartyPants, mentioned elsewhere in the thread, does this correctly: while it will convert double and triple dashes and straight quotes and apostrophes to the typographically correct characters, text between <code></code> or <pre></pre> tags is left untouched. The idea is willing, but the execution is weak, in most cases. Regards,
From: Phil!Gregory Date: 21:15 on 17 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) * Luke Kanies <luke@xxxxxxx.xxx> [2005-10-17 10:02 -0500]: > On Mon, 17 Oct 2005, A. Pagaltzis wrote: > > > I get this in Aristotle's emails (gibberish instead of '), > > > > It’s because your mailer sucks. (Are you surprised?) > > I'm guessing Pine has no relationship to whether this works; I'm thinking > instead that it's the terminal itself. It's probably Pine. A. Pagaltzis sent an email in UTF-8. Your quoted reply contained UTF-8 characters, but had been labeled (by your MUA, presumably) as ISO 8859-1. Thus, previously properly-rendered UTF-8 gets mangled on its trip through your software. Charset specification via MIME has been around for a while now. Programs should be able to get this sort of thing right. (I took the liberty of correcting the curly quotation mark. Otherwise, my mailer would have encoded each byte of the original as a separate character, doubling or tripling its length.)
From: peter (Peter da Silva) Date: 07:50 on 17 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) > And until distributors provide it OOTB, it’s a *MASSIVE* pain to > get UTF-8 support set up. Get a Mac. It just works. Out of the box. I still think even having it as on option in Perl6 is so fundamentally misguided it makes Perl4 seem sane.
From: Juerd Date: 08:28 on 17 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) Peter da Silva skribis 2005-10-17 1:50 (-0500): > Get a Mac. And install Linux on it :) I heard tho "Get a Mac" mantra so often that I started believing it would actually not be as hateful as the rest of software. A little less than a year later, I know much better and have installed Linux on it (Kubuntu breezy), because dpkg is better at handling system packages than the .app crap that lacks centralized upgrades, and KDE is so much better at providing useful GUIs. OS X is beautiful and lacks user choice. User friendly, however, I think it is not. It has been a very frustrating experience. One of the last things I did with OS X, before installing Kubuntu yesterday, was burning the Kubuntu installation CD that I had downloaded as an iso file. Right click the iso file, see if there's anything that looks like "Burn" in the context menu. No, there is not. Okay, then see if I can drag it onto something. But there's nothing to drag it to. No icon for the blank cd I inserted. I found no program in my entire Applications folder that looked like it had to do with burning a CD, so I resorted to Google. It turns out that you have to use a Disk Utility to open the image (which also mounts it, without that being either asked for or necessary). It has a Burn button. It burned and verified correctly. But were I given the choice, I'd have disabled the verification to save a few minutes. This last OS X experience convinced me that installing Linux was the Right Thing to do. Juerd
From: peter (Peter da Silva) Date: 12:38 on 17 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) > I heard tho "Get a Mac" mantra so often that I started believing it > would actually not be as hateful as the rest of software. It's software. What did you fucking expect? But if you want Unicode to Just Work, it's your best option by far. > A little less > than a year later, I know much better and have installed Linux on it > (Kubuntu breezy), because dpkg is better at handling system packages > than the .app crap that lacks centralized upgrades, Darwinports or Fink. As for KDE... personally, I prefer Windowmaker, and I detest both KDE and Gnome, but I'll readily admit that there are things I miss about traditional UNIX windowing environments. But having shit Just Work instead of having to go through an Amish Barn Raising every time I install a new application (and if you try and tell me ANY package manager on Linux is going to change that I'll laugh) is worth having to *run an application to burn CDs*. When I go crook at GUI stuff on my Mac and post about it to this list, it's almost always stuff that I'd never loose my cool over on any X11-based environment... not because X11 does it better but because X11 puts so much Amish Barn Raising shit in my way I'd never have tried... or it's stuff even I'll admit is a bit geeky. Where Mac OS X pisses me off is low level shit like the lack of tape support and the fact that UFS isn't the default file system. And Linux sure as hell isn't going to scratch my UFS itch with its filesystem-of-the-week and none-of-them-support-dump dance. OK, raw disk is a crappy and hateful API for backups, but it's a hell of a lot better than not HAVING one.
From: David Cantrell Date: 12:46 on 17 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) On Mon, Oct 17, 2005 at 06:38:33AM -0500, Peter da Silva wrote: > Where Mac OS X pisses me off is low level shit like the lack of > tape support and the fact that UFS isn't the default file system. And even if it did make UFS the default it would still suck, because it doesn't support large filesystems. Want a 2TB fs? You've got no choice but to use HFS+.
From: Peter da Silva Date: 17:18 on 17 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) On Oct 17, 2005, at 6:46 AM, David Cantrell wrote: > And even if it did make UFS the default it would still suck, because it > doesn't support large filesystems. Want a 2TB fs? You've got no > choice > but to use HFS+. OK, I'd assume that if they were to make UFS the default they'd start by picking up the latest version from FreeBSD (like they did for Panther). That one uses 64-bit block pointers, and while all the userland tools haven't been upgraded yet they'd be rolling their own anyway. Not doing that would be hateful, of course, and on topic.
From: Juerd Date: 13:10 on 17 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) Peter da Silva skribis 2005-10-17 6:38 (-0500): > It's software. > What did you fucking expect? It to suck less than Linux+KDE. > But if you want Unicode to Just Work, it's your best option by far. Except that Terminal.app came configured to use iso-8859-1 by default, and the default fonts seem to be very limited in the number of glyphs they can display correctly. I have this feeling -again- that you haven't seriously given recent Linux distributions a serious try and chance. > Darwinports or Fink. I'm lazy. I've tried these, though, out of desperation. They sucked in interesting ways, and a lot of software that I wanted wasn't available through them. I've never liked ports systems because I am as impatient as I am lazy. > As for KDE... personally, I prefer Windowmaker, and I detest both > KDE and Gnome, but I'll readily admit that there are things I miss > about traditional UNIX windowing environments. The only problem I have with KDE is that its default installation is extremely bloated. (And, but this is true for almost all open source software, that its Dutch translations are stupid.) I think Kubuntu did a good job cutting much of the crap away from KDE and cleaning up some interfaces, for example Konqueror's. > But having shit Just Work instead of having to go through an Amish > Barn Raising every time I install a new application (and if you try > and tell me ANY package manager on Linux is going to change that I'll > laugh) is worth having to *run an application to burn CDs*. I like making people laugh. Just to make your day: try Debian's dpkg/apt system! You'll LOVE it! The only weird technological dance that I had to do to make X work well on my Mac mini's brand new Kubuntu installation was to set the refresh rate to the weird value that my monitor wants. But this is hardware hate mostly, though it would have been very nice if Kubuntu just asked me during installation. At least Kubuntu got the resolution right, while OS X decided that 1024x768 would be good enough for me, as a default. > And Linux sure as hell isn't going to scratch my UFS itch with its > filesystem-of-the-week and none-of-them-support-dump dance. Most Linux distributions have defaulted to ext2 (now ext3, which is ext2 plus a journal) for ages. What is this filesystem-of-the-week thing you mention? Ext2-and-3-do-support-dump. Juerd
From: Luke Kanies Date: 15:59 on 17 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) On Mon, 17 Oct 2005, Juerd wrote: > I have this feeling -again- that you haven't seriously given recent > Linux distributions a serious try and chance. Being a heavy user of both OS X and Linux (Debian), I can definitely say that neither are without their hates. I wouldn't want to run an OS X server doing anything complicated, because of how difficult it is to maintain all of the packages, but I wouldn't use Linux for anything other than serving if 1) I didn't basically have to for the dev work I do, and 2) it didn't make having 30 terminal windows open so much easier than on OS X. Either way, though, I can definitely say that my single Linux box takes something like 10x the amount of effort to maintain than both my and my wife's laptops, and it certainly gets less usage overall. > I like making people laugh. Just to make your day: try Debian's dpkg/apt > system! You'll LOVE it! While I think apt is the best or second best in most ways (emerge has some key advantages, especially in user feedback), the day that I 'love' a package manager is the about 3 years after I officially stopped maintaining my computers. Still plenty of hate there.
From: A. Pagaltzis Date: 16:23 on 17 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) * Luke Kanies <luke@xxxxxxx.xxx> [2005-10-17 17:00]: > the day that I 'love' a package manager is the about 3 years > after I officially stopped maintaining my computers. On this we can definitely agree. Regards,
From: Phil!Gregory Date: 21:22 on 17 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) * Peter da Silva <peter@xxxxxxx.xxx> [2005-10-17 06:38 -0500]: > > I heard tho "Get a Mac" mantra so often that I started believing it > > would actually not be as hateful as the rest of software. > > But if you want Unicode to Just Work, it's your best option by far. Not if you dip into Terminal.app. When using my friends' Macs to ssh to my computers, Terminal.app has caused me quite a lot of problems. In its default configuration, it seems to regard U+2500 (─) and friends as double-width characters, which tends to mess up any box drawing they might be used for. I recall fiddling with the settings to try to get them to work properly; the best I could do was to force everything to be single-width, which breaks all real double-width characters, but fortunately for me, I'm a dumb Westerner and don't read Chinese/Japanese/etc.
From: peter (Peter da Silva) Date: 21:42 on 17 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) > Not if you dip into Terminal.app. When using my friends' Macs to ssh to > my computers, Terminal.app has caused me quite a lot of problems. Terminal emulation is a really tough problem. I wrote a terminal emulator once. My goal was just to have something that worked right on RSX-11, VMS, UNIX, and ANSI BBSes, and passed the vt100 torture test with readable output. That ended up taking something like eighteen months, maybe a few weeks solid hacking if I'd not been doing it in my free time, and it was only trying to support two character sets and 8-bit characters. LOTS of good righteous hate there, from editors that depended on bugs in the implementation, to bugs designed as features, to workarounds for the crappy BIOS and the evil of depending on a commercial serial driver because Microsoft's sucked so badly. I managed to hate every kind of software there was, including my own, before I was done.
From: Luke Kanies Date: 23:50 on 17 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) I just want to point out this software, since it seems so topical: http://www.michelf.com/projects/php-smartypants/
From: Martin Ebourne Date: 00:52 on 18 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) On Mon, 2005-10-17 at 17:50 -0500, Luke Kanies wrote: > I just want to point out this software, since it seems so topical: > > http://www.michelf.com/projects/php-smartypants/ Yeuch. I don't like that already. More pants than smart it seems. Martin.
From: Rhesa Rozendaal Date: 01:20 on 18 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) Luke Kanies wrote: > I just want to point out this software, since it seems so topical: > > http://www.michelf.com/projects/php-smartypants/ > > If that can handle things like: "Foto's en agenda's" properly, then it's smarter than Word. That source of hate turns those apostrophes into single-quotes, as if "s en agenda" is a paraphrase. For the record, I'm too lazy¹ to type properly curled quotes, even though I am a big fan of them in printed material. All you ascii freaks should try reading War and Peace some time - typeset entirely in Courier! Bah. Rhesa ¹even though it's as simple as ralt+v and ralt+b. -- aà ãáäąääãà - just because my keyboard lets me.
From: A. Pagaltzis Date: 02:48 on 18 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) * Rhesa Rozendaal <rhesa@xxxxxx.xx> [2005-10-18 02:25]: > Luke Kanies wrote: > >I just want to point out this software, since it seems so topical: > > > >http://www.michelf.com/projects/php-smartypants/ > > If that can handle things like: "Foto's en agenda's" properly, > then it's smarter than Word. That source of hate turns those > apostrophes into single-quotes, as if "s en agenda" is a > paraphrase. With Foto's en agenda's it produces Foto’s en agenda’s so it passes that test. In fact, with 'Foto's en agenda's' I get ‘Foto’s en agenda’s’ so yeah, it’s smarter than Word. At least that’s what I get using the original Perl version of SmartyPants: http://daringfireball.net/projects/smartypants/ It’s from the same author who designed Markdown and wrote the original implementation. The online converter for Markdown has a “Filter†option that runs only SmartyPants over the given text, so you can check for yourself what it does with any given input: http://daringfireball.net/projects/markdown/dingus (Which is what I used to do the above tests.) Regards,
From: Patrick Carr Date: 22:36 on 18 Oct 2005 Subject: typography of the apostrophe (was Re: Significant whitespace (was Re: Blogging sucks)) On Oct 17, 2005, at 9:48 PM, A. Pagaltzis wrote: > * Rhesa Rozendaal <rhesa@xxxxxx.xx> [2005-10-18 02:25]: > >> If that can handle things like: "Foto's en agenda's" properly, >> then it's smarter than Word. That source of hate turns those >> apostrophes into single-quotes, as if "s en agenda" is a >> paraphrase. >> > > With > > Foto's en agenda's > > it produces > > Foto’s en agenda’s > > so it passes that test. In fact, with > > 'Foto's en agenda's' > > I get > > ‘Foto’s en agenda’s’ > > so yeah, it=92s smarter than Word. This is a huge HUGE hate, stemming from the fact that I'm (a) =20 involved with two alumni boards and (b) extremely pedantic. With the =20 awesomeness that is Word's smartquote "feature," a list of meeting =20 attendees turns into: Steve B. Alum `85 Alison M. Bee '91 Jimmy Cerebus `82 Maude Dechere '63 etc. I'm not that angry that it doesn't know that those are class years =20 and not numbers that might happen to be in quotes, because I can see =20 it being non-trivial to take all cases into account. However, in standard U.S. typography, the single open quote is so =20 rarely used (only quotes within quotes) that that particular =20 "smartness" should be off by default. Even at the beginning of word, =20 it's still more likely to be an apostrophe signifying elision. `Tis a =20= hateful, hateful thing, I tells ya. On web sites where I have the capability I swap ' with ’ on the =20= fly because it's pretty and looks better with a proportional font. If =20= I were to ever need a single open quote, I would input ‘ by hand. Now that I think about it, one could consider extending the hate to =20 proportional typefaces even having a straight quote mark glyph. If =20 only browsers had been built around latex. Pat Carr
From: A. Pagaltzis Date: 01:11 on 19 Oct 2005 Subject: Re: typography of the apostrophe (was Re: Significant whitespace (was Re: Blogging sucks)) [ Note: if you're replying to this message, please mind the To: and Cc: -- this is crossposted to the Hates-Software and Markdown lists. ] * Patrick Carr <pmc1@xxxxxxx.xxx> [2005-10-18 23:40]: >On Oct 17, 2005, at 9:48 PM, A. Pagaltzis wrote: >>* Rhesa Rozendaal <rhesa@xxxxxx.xx> [2005-10-18 02:25]: >> >>>If that can handle things like: "Foto's en agenda's" >>>properly, then it's smarter than Word. That source of hate >>>turns those apostrophes into single-quotes, as if "s en >>>agenda" is a paraphrase. >>> >> >>With >> >> Foto's en agenda's >> >>it produces >> >> Foto’s en agenda’s >> >>so it passes that test. In fact, with >> >> 'Foto's en agenda's' >> >>I get >> >> ‘Foto’s en agenda’s’ >> >>so yeah, it’s smarter than Word. > > This is a huge HUGE hate, stemming from the fact that I'm (a) > involved with two alumni boards and (b) extremely pedantic. > With the awesomeness that is Word's smartquote "feature," a > list of meeting attendees turns into: > > Steve B. Alum `85 > Alison M. Bee '91 > Jimmy Cerebus `82 > Maude Dechere '63 > etc. > > I'm not that angry that it doesn't know that those are class > years and not numbers that might happen to be in quotes, > because I can see it being non-trivial to take all cases into > account. Yeah. SmartyPants documents it as a known (and unfixable) bug that *all* of these will turn into opening curly quotes. (Which is because it does not try to pair up quotes, unlike Word, but rather uses the context to guess what kind of quote it is, which is why it almost always guesses right, unlike Word. With this case being the inevitable exception.) > However, in standard U.S. typography, the single open quote is > so rarely used (only quotes within quotes) that that particular > "smartness" should be off by default. Even at the beginning of > word, it's still more likely to be an apostrophe signifying > elision. `Tis a hateful, hateful thing, I tells ya. Sounds like a fair point. > On web sites where I have the capability I swap ' with ’ > on the fly because it's pretty and looks better with a > proportional font. If I were to ever need a single open quote, > I would input ‘ by hand. > > Now that I think about it, one could consider extending the > hate to proportional typefaces even having a straight quote > mark glyph. If only browsers had been built around latex. Regards,
From: Martin Ebourne Date: 11:27 on 18 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) Rhesa Rozendaal <rhesa@xxxxxx.xx> wrote: > For the record, I'm too lazy=C2=B9 to type properly curled quotes, even > though I am a big fan of them in printed material. All you ascii > freaks should try reading War and Peace some time - typeset entirely > in Courier! Bah. > > =C2=B9even though it's as simple as ralt+v and ralt+b. All this talk by mac users how theirs is the only OS where i18n just works and sometimes you risk believing them. I'm running Gnome (FC4) which has displayed and handled all these characters in this thread faultlessly. And now I find how to type them too. I've been round the keyboard, lots of funky things out there. All the programs work, from email through to the editor. Even good old xterm works, though it can't display some of the more esoteric characters due to the fixed font not having them. I've most certainly never done anything to this setup to make any of this lot function, besides setting en_GB as my locale. As far as I can make out the whole i18n thing on Gnome just works out of the box. Makes sense really, because Gnome's top priorities have always included accessibiliy, i18n, and just working. I don't get to read the chinese spam, but I'm pretty sure that's just cause I don't have the right fonts installed. Come to think of it, I don't think I want them either. Cheers, Martin.
From: Peter da Silva Date: 14:11 on 18 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) > All this talk by mac users how theirs is the only OS where i18n just > works and sometimes you risk believing them. It's not the OS, it's the window system and applications. The Mac has had good text support from day one, and so all=20 applications, even 68000 apps that you have to run under Basilisk=20 because Classic doesn't have 68000 support, Just Work. With X11, you have about four different and incompatible sets of=20 extensions to support i18n developed over the years for different=20 toolkits. The problem is that X11 was designed originally with academic=20= research as a primary goal. They didn't know what kinds of user=20 interfaces would be good, so they didn't implement a user interface...=20= they just implemented a display server that could handle overlapping=20 windows, text, bitmaps, and some drawing primitives, and an input=20 handler that could pass mouse and keyboard events to the application,=20 and some tools to let you pass configuration information and objects=20 between applications (without defining either) and left everything else=20= to libraries running in the application. The result is a system that's insanely latency sensitive for a=20 networked window system, and imposes ridiculous levels of latency for a=20= local window system, and that forces you to implement everything,=20 including all policy, all i18n and l11n support, in the application. So Gnome apps work well with Gnome apps, and KDE apps work well with=20 KDE apps, and GNUstep apps work well with GNUstep apps. The Athena=20 widgets were ugly as sin, but they were a consistent and simple toolkit=20= that was available in a huge variety of skins... pretty much all apps=20 that didn't roll their own used Xaw and you could just relink it to=20 Xaw3d or Xaw95 and get the look and feel you wanted... and then Motif=20 came along and lots of people (including some idiot free software=20 developers who missed the clue train) started using Motif, In X11R5 or=20= X11R6 (I forget which) they threw out (or at least obsoleted) server=20 support for composed characters and put responsibility for that on=20 applications, so everything broke until apps were rewritten to do THAT=20= work all over again. Then, of course, we got Motif-clones and Openlook=20= and the Gnome and KDE toolkits. There's toolkits to emulate Plan 9, and=20= GNUstep. There's still lots of apps like XV that cut their own UI. You can "just use Gnome", but, damn... The Mac is the only credible platform that has the advantage of a=20 fascist developer that was willing to break apps that did their own=20 thing, and that has attempted to support i18n from an early stage. So I=20= can call a file "=E4=B9=85=E7=9F=B3=E8=AD=B2 =D0=BA=D0=B0=D1=82=D0=B0=D0=BB= =D0=BE=D0=B3" or "=EB=96=B3=EB=8B=A4!! =EA=B7=B8=EB=85=80!!.mp3" and = while many UNIX=20 apps are unhappy with it (which is why I'm less than enamored of UTF8,=20= sometimes breaking everything is what you need to do and making the=20 whole system UCS-4 (or at least UCS-2) would make it a LOT easier to=20 transition these apps to the New World Order) it does come up in=20 Terminal: [absinthe:~] peter% touch "=E4=B9=85=E7=9F=B3=E8=AD=B2 =D0=BA=D0=B0=D1=82=D0= =B0=D0=BB=D0=BE=D0=B3" [absinthe:~] peter% file "=E4=B9=85=E7=9F=B3=E8=AD=B2 =D0=BA=D0=B0=D1=82=D0= =B0=D0=BB=D0=BE=D0=B3" =E4=B9=85=E7=9F=B3=E8=AD=B2 =D0=BA=D0=B0=D1=82=D0=B0=D0=BB=D0=BE=D0=B3: = empty It's a pity Plan 9 didn't get open-sourced early and end up just=20 becoming "what UNIX is". It could have. While 8=C2=BD is almost as=20 aggressively minimalist a window system as X11, at least it supported=20 multi-byte characters from the get-go.=
From: Martin Ebourne Date: 14:57 on 18 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) Peter da Silva <peter@xxxxxxx.xxx> wrote: > It's not the OS, it's the window system and applications. Ok, there's varying definitions of OS. I was including the window system, not meaning just the kernel. > So Gnome apps work well with Gnome apps, and KDE apps work well with KDE = apps Gnome and KDE seem to interoperate fairly well these days, getting better all the time. And on Linux there's not much reason to use any other GUI apps, so that pretty much covers it. All that other stuff is legacy crap. > The Mac is the only credible platform that has the advantage of a > fascist developer that was willing to break apps that did their own > thing, and that has attempted to support i18n from an early stage. So > I can call a file "=E4=B9=85=E7=9F=B3=E8=AD=B2 =D0=BA=D0=B0=D1=82=D0=B0= =D0=BB=D0=BE=D0=B3" or "=EB=96=B3=EB=8B=A4!! =EA=B7=B8=EB=85=80!!.mp3" and = while many > UNIX apps are unhappy with it (which is why I'm less than enamored of > UTF8, sometimes breaking everything is what you need to do and making > the whole system UCS-4 (or at least UCS-2) would make it a LOT easier > to transition these apps to the New World Order) it does come up in > Terminal: > > [absinthe:~] peter% touch "=E4=B9=85=E7=9F=B3=E8=AD=B2 =D0=BA=D0=B0=D1=82= =D0=B0=D0=BB=D0=BE=D0=B3" > [absinthe:~] peter% file "=E4=B9=85=E7=9F=B3=E8=AD=B2 =D0=BA=D0=B0=D1=82= =D0=B0=D0=BB=D0=BE=D0=B3" > =E4=B9=85=E7=9F=B3=E8=AD=B2 =D0=BA=D0=B0=D1=82=D0=B0=D0=BB=D0=BE=D0=B3: e= mpty Works fine for me on linux too. Missing those characters in the fixed font though (and the first couple in my normal font too). As to weird characters in filenames, all my CD collection is ripped on Linux, and there's plenty of classical and world music with all sorts of weird characters, which are all entered correctly and present in the filenames. I've yet to have a problem with a program on Linux/Gnome which doesn't work with them. Even the shell and xterm. The only grief I've had actually comes when I try to copy them onto the music player, which of course is FAT32, that doesn't grok them. I have a small shell script to do the copying for me which fixes that up. Cheers, Martin.
From: Jonathan Stowe Date: 14:59 on 18 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) On Tue, 2005-10-18 at 14:11, Peter da Silva wrote: > [absinthe:~] peter% touch "ä¹…çŸ³è² ÐºÐ°Ñ‚Ð°Ð»Ð¾Ð³" > [absinthe:~] peter% file "ä¹…çŸ³è² ÐºÐ°Ñ‚Ð°Ð»Ð¾Ð³" > ä¹…çŸ³è² ÐºÐ°Ñ‚Ð°Ð»Ð¾Ð³: empty > I was just shocked by bash running in gnome-terminal - creating a file with the the above name it actually tab completed the name correctly ...
From: peter (Peter da Silva) Date: 01:23 on 17 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) > I would like to agree, but I fear that if we all take this stance > then things will be just as broken in 20 years, and 20 years > after that, and another 20 later. Restricting the character set used syntactically by a programming language is not going to make I18N any harder. There's no reason the programming language can't allow these characters in strings, comments, or any other place where it might cause problems for I18N. Hell, Perl would benefit from a little smaller collection of syntacticaly meaningful symbols as it is.
From: peter (Peter da Silva) Date: 01:18 on 17 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) > All languages should handle input and output in all character sets, but > none of them should be written in anything but ASCII. At least for the > next 20 years. Maybe then things will be different, but for now the > breakage is just huge. Well, we've even seen it tonight, something as > simple as email. The first versions of Smalltalk uses special characters that were available on all the input and output devices that every person at Xerox used on a regular basis. They still decided that was a bad idea, and backed out of it very quickly. If the insular folks at Xerox Parc gave it up as a bad idea, it must be a very bad idea indeed.
From: peter (Peter da Silva) Date: 01:13 on 17 Oct 2005 Subject: Re: Significant whitespace (was Re: Blogging sucks) > Once you know that there must be no whitespace between the hash > name and the key lookup operator in Perl6, the parsing is > completely straightforward. While it's nice to see Perl authors having to deal with the same kind of problems everyone else who tries to deal with Perl as if it was a sane language has to put up with, making it less sane is hardly a step in the right direction. > Of course, you are free to dislike that syntax rule. The behaviour is hateful whatever the reason.
From: peter (Peter da Silva) Date: 21:13 on 13 Oct 2005 Subject: Re: Blogging sucks It isn’t anything like Python at all. In the way that I personally find Python hateful, this is hateful. There is only one level of indentation. Once a block is indented far enough to be considered a code block, all further indentation is literal. What if I'm indenting something that isn't a code block? "Like a quotation in an indented block of text?" -- Peter da Silva Not to mention that *all* alternative solutions are *really* hateful. indent << \! Any non-hateful solution is isomorphic to a "here" document. !
From: A. Pagaltzis Date: 22:16 on 13 Oct 2005 Subject: Re: Blogging sucks * Peter da Silva <peter@xxxxxxx.xxx> [2005-10-13 22:15]: > What if I'm indenting something that isn't a code block? Who cares? HTML has no generic “indent a block†mechanism, so that the > Not to mention that *all* alternative solutions are *really* > hateful. > > indent << \! > Any non-hateful solution is isomorphic to a "here" document. > ! Since when are heredocs non-hateful? You have the choice between a terminator that does not appear within the verbatim text (the epitome of elegance) or escaping any appearances of the terminator within the verbatim text (the paragon of beauty). At the beginning of a line within a heredoc the parser is in an ambiguous state: what follows may either be another verbatim line or may be a hedoc end marker. If it’s the former but looks like the latter, the parse will disambiguate wrongly, so escaping is required. The solution is to require each line to be prefixed invidually, so the parser can never guess wrongly by design, so that you never have to escape anything. Regards,
From: A. Pagaltzis Date: 22:28 on 13 Oct 2005 Subject: Re: Blogging sucks * A. Pagaltzis <pagaltzis@xxx.xx> [2005-10-13 23:20]: > * Peter da Silva <peter@xxxxxxx.xxx> [2005-10-13 22:15]: > > What if I'm indenting something that isn't a code block? > > Who cares? HTML has no generic “indent a block†mechanism, so > that the Wetware is hateful too. This was supposed to say > Who cares? HTML has no generic “indent a block†mechanism, so > a simplified HTML authoring syntax lacking one is hardly a > surprise. Regards,
From: peter (Peter da Silva) Date: 23:19 on 13 Oct 2005 Subject: Re: Blogging sucks > * Peter da Silva <peter@xxxxxxx.xxx> [2005-10-13 22:15]: > > What if I'm indenting something that isn't a code block? > Who cares? HTML has no generic “indent a block†mechanism, so > that the <div style="margin: 0.5i"> I don't quite understand what point you're trying to make here. Not only are there a number of mechanisms that could be used. <blockquote> Traditionally, of course, blockquote has been an option. </blockquote> but applying a style to a div is more versatile. </div> > Since when are heredocs non-hateful? Assuming for the sake of argument that a non-hateful solution exists (which I am not at all sure of), if it's going to allow you to include verbatim text at all it must be isomorphic to "here documents", since no other solution actually lets you include verbatim text. > You have the choice between > a terminator that does not appear within the verbatim text (the > epitome of elegance) or escaping any appearances of the > terminator within the verbatim text (the paragon of beauty). You have the choice between a terminator that doesn't appear within the verbatim text and a terminator that doesn't appear in the verbatim text. Or you could use a terminator that doesn't appear in the verbatim text. If you're putting escapes in the text, it's not verbatim, is it? > At the beginning of a line within a heredoc the parser is in an > ambiguous state: what follows may either be another verbatim line > or may be a hedoc end marker. In the case of traditional "here documents", the newline is part of the end marker. In the case of the slashcode "here documents" (which are hateful for other reasons) the leading and trailing angle brackets are part of the end marker. In the case of the paired terminators used by IOS or the "ed/ex" family of text editors the end marker is identical to the start marker. > If it’s the former but looks like > the latter, the parse will disambiguate wrongly, so escaping is > required. The solution is to require each line to be prefixed > invidually, so the parser can never guess wrongly by design, so > that you never have to escape anything. But then you're not including the text verbatim, you're escaping the newline. You have to separately modify every line of the text whether you put the escape at the beginning (as in Fortran) or at the end (as in the C preprocessor). Which is hateful.
From: A. Pagaltzis Date: 00:40 on 14 Oct 2005 Subject: Re: Blogging sucks * Peter da Silva <peter@xxxxxxx.xxx> [2005-10-14 00:20]: > > * Peter da Silva <peter@xxxxxxx.xxx> [2005-10-13 22:15]: > > > What if I'm indenting something that isn't a code block? > > > Who cares? HTML has no generic “indent a block†mechanism, so > > that the > > <div style="margin: 0.5i"> > I don't quite understand what point you're trying to make here. Not only are > there a number of mechanisms that could be used. > <blockquote> > Traditionally, of course, blockquote has been an option. > </blockquote> > but applying a style to a div is more versatile. > </div> The div solution is inappropriate. The markup is meaningless, and for the rendered of authoring syntax to litter a document with style attributes is hateful because you’ll have to work extra hard to make CSS work. Likewise it is hateful for it to invent CSS class names that may or may not clash with the existing stylesheet, and passing in class names for it to use requires configuration, which is another kind of hate. The Markdown solution is that if you want `<div style="margin: 0.5i">`, then you write `<div style="margin: 0.5i">`. It’s only trying to make the common tasks less tedious, not to invent punctuation or other syntax to replace all HTML constructs. For blockquotes, guess what? > Email-quoted text is <blockquote>d in the HTML output. > Works in arbitrary conjunction with all other Markdown block > constructs: > > * Lists. > * echo 'Literal code blocks' > foo > * etc. > > > You can nest it. You just write something that looks very much like email, and out comes a sensibly formatted HTML document. But you’d know that if you just took a look at the (brief) syntax documentation instead of dismissing it off hand. Regards,
From: Earle Martin Date: 07:01 on 14 Oct 2005 Subject: Re: Blogging sucks On Fri, Oct 14, 2005 at 01:40:08AM +0200, A. Pagaltzis wrote: > The markup is meaningless What meaningful markup would you suggest be created, then, to describe an indented area?
From: A. Pagaltzis Date: 21:10 on 16 Oct 2005 Subject: Re: Blogging sucks * Earle Martin <hates-software@xxxxxxxx.xxx> [2005-10-14 08:05]: > On Fri, Oct 14, 2005 at 01:40:08AM +0200, A. Pagaltzis wrote: > > The markup is meaningless > > What meaningful markup would you suggest be created, then, to > describe an indented area? There is no meaningful markup to do this in HTML, so I have no suggestion. Whatever markup was generated would have to be chosen by convention, and Markdown makes no such choices because it aims to be zero-configuration. You can always write tags into your source document for anything Markdown has no own syntax for. It works quite well in practice. Regards,
From: peter (Peter da Silva) Date: 01:10 on 17 Oct 2005 Subject: Re: Blogging sucks > There is no meaningful markup to do this in HTML, so I have no > suggestion. Will you please explain what the presence or absence of this markup in HTML has anything to do with the matter at hand? I don't get it. Honestly, I don't get it. You're arguing as if markdown and raw HTML are the only options. If that's the case... why?
From: A. Pagaltzis Date: 02:51 on 17 Oct 2005 Subject: Re: Blogging sucks * Peter da Silva <peter@xxxxxxx.xxx> [2005-10-17 02:15]: > > There is no meaningful markup to do this in HTML, so I have no > > suggestion. > > Will you please explain what the presence or absence of this > markup in HTML has anything to do with the matter at hand? I > don't get it. Honestly, I don't get it. You're arguing as if > markdown and raw HTML are the only options. If that's the > case... why? I don’t know that anyone was arguing any of the things you are hinting at, or what exactly the suggested arguments were. The question it all started with is: if you want to indent something in Markdown, how do you do that? And the answer is: then you write raw HTML, because Markdown deliberately has no syntax for indenting, because HTML has no element to express indented text. Regards,
From: peter (Peter da Silva) Date: 07:47 on 17 Oct 2005 Subject: Re: Blogging sucks > The question it all started with is: if you want to indent > something in Markdown, how do you do that? No, the question was "if you want to indent something without it turning into code, how do you do that". That was in response to your comment about indented code. > And the answer is: > then you write raw HTML, because Markdown deliberately has no > syntax for indenting, because HTML has no element to express > indented text. OK, that's not what I expected, but it's still nonsense. 1. HTML has no element to express "indented code", either. 2. HTML *does* have an element to express indented text. It's got several. Pick one. 3. In any case, that's like refusing to implement square roots in a runtime library because some processors don't have it in hardware. Who the hell cares what HTML does, it's machine code as far as a higher level input format it concerned.
From: Stephen Deken Date: 03:25 on 14 Oct 2005 Subject: Re: Blogging sucks On Oct 13, 2005, at 4:16 PM, A. Pagaltzis wrote: > The solution is to require each line to be prefixed > invidually, so the parser can never guess wrongly by design, so > that you never have to escape anything. Wait, what? Require each line to be prefixed so that you never have to escape anything? Is there some semantic difference between 'escaping' and 'prefixing' that I'm just not getting? Because it seems like you're saying each line should be individually escaped so that we don't have to escape anything. --sjd;
From: Chris Devers Date: 11:54 on 13 Oct 2005 Subject: Re: Blogging sucks This message is in MIME format. The first part should be readable text, while the remaining parts are likely unreadable without MIME-aware tools. --0-1668461769-1129200855=:651 Content-Type: TEXT/PLAIN; charset=ISO-8859-1 Content-Transfer-Encoding: QUOTED-PRINTABLE On Wed, 12 Oct 2005, Peter da Silva wrote: > > This way may lie madness, but what about an upload filter that=20 > > auto-escapes angle brackets for anything wrapped in <pre> or <code>=20 > > tags? Surely some of the blogging toolkits have such abilities, as=20 > > plugins if not as core functionality, no? >=20 > What if you've got </pre> or </code> as part of the code you're=20 > entering? Yo may be writing about HTML. I *did* say "this way may life madness"... :-) Yes, there are always going to be degenerate cases like that, and no, I=20 don't think there's any great way around that with a simple markup. Your=20 suggestion has exactly the same drawback that mine does.=20 If you're going to use anything other than HTML, it might as well be=20 some variant of Wiki syntax, if only because they're all intended to be=20 simple and there's starting to be a lot of people that are already=20 comfortable with editing some dialect of WikiText.=20 =20 It'll still be prone to the problem you cite here, but at least it gets=20 a bit less likely, and adds benefits of simplicity & familiarity... --=20 Chris Devers =C3=91=EBU=9B=E06Q=B1Ruj --0-1668461769-1129200855=:651--
From: Stephen Deken
Date: 03:36 on 13 Oct 2005
Subject: Re: Blogging sucks
> <pre>
> for(i = 0; i < 10; i++) {
> puts("Hey! Where'd my code go?");
> if(!you_can_see_this())
> break;
> }
> if(i>=10)
> puts("Oh, there it is!");
> </pre>
You'd see it all. Go on, try it. HTML is less hateful than you think.
In this case.
At least, you see it all with WebKit, Gecko, and IE for Mac. I can't
vouch for IE in Windows at the moment but I believe it also works
there.
<pre> is shpeshul.
--sjd;
From: Peter da Silva Date: 05:30 on 13 Oct 2005 Subject: Re: Blogging sucks > You'd see it all. Go on, try it. HTML is less hateful than you > think. In this case. 1. If HTML treats <pre> THAT specially, it's more hateful than I think. 2. I've already been burned by that exact example in existing html-enhanced web markup. I don't actually care if the browser or the server fucked it up.
From: Abigail Date: 08:09 on 13 Oct 2005 Subject: Re: Blogging sucks --KFztAG8eRSV9hGtP Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Wed, Oct 12, 2005 at 11:30:17PM -0500, Peter da Silva wrote: > >You'd see it all. Go on, try it. HTML is less hateful than you=20 > >think. In this case. >=20 > 1. If HTML treats <pre> THAT specially, it's more hateful than I think. No, <pre> is not special. < in HTML only defines markup if it's followed by !, by something that can be a NAME, or followed by / followed by a NAME. And that's actually an SGML feature - not something special for HTML. Any browser that doesn't display 'a < b && c > d' is broken. And while many browsers used to be broken, in 10 years of struggling, they managed to fix that. Unlike the naive HTML parsing of gazillions of people. > 2. I've already been burned by that exact example in existing=20 > html-enhanced web markup. I don't actually care if the browser or the=20 > server fucked it up. Browsers suck. I've yet to find a main-stream browser that can actually parse HTML2. Abigail --KFztAG8eRSV9hGtP Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.0 (GNU/Linux) iD8DBQFDTgg4BOh7Ggo6rasRArP2AJ9cAu7LvuJiFycxDZGlvy+eAgM47wCcDzl1 CLnWcKOE6M3mi1LprPAeTok= =DSI0 -----END PGP SIGNATURE----- --KFztAG8eRSV9hGtP--
From: A. Pagaltzis Date: 04:35 on 13 Oct 2005 Subject: Re: Blogging sucks * Chris Devers <cdevers@xxxxx.xxx> [2005-10-13 03:05]: > <p> It could be better, sure, but for minimal HTML like this, a > lot of the "solutions" I've heard of seem more cumbersome than > the original problem they were supposed to solve. </p> Have a look at Markdown. Markdown source is plaintext that looks like email and Does What You Mean. If you need more elaborate markup than the constructs it supports you just write HTML straight into the document. Publishing code snippets is trivial too, if you use one of the code constructs (quoting with backticks if it’s inline or indenting by 4 spaces or a tab if it’s a block) it escapes all your angle brackets for you. Really, it is about as close to non-hateful as HTML authoring tools can get. Regards,
From: Luke Kanies
Date: 17:56 on 13 Oct 2005
Subject: Re: Blogging sucks
On Thu, 13 Oct 2005, A. Pagaltzis wrote:
> Have a look at Markdown.
>
> Markdown source is plaintext that looks like email and Does What
> You Mean. If you need more elaborate markup than the constructs
> it supports you just write HTML straight into the document.
> Publishing code snippets is trivial too, if you use one of the
> code constructs (quoting with backticks if it=E2=80=99s inline or
> indenting by 4 spaces or a tab if it=E2=80=99s a block) it escapes all
> your angle brackets for you.
>
> Really, it is about as close to non-hateful as HTML authoring
> tools can get.
Yeah, like Restructured Text. But none of the hosted blogs seem to
support it, or anything like it. Which means, like I'm currently doing,
I have to resort to just writing the posts offline, converting it
manually, and then pasting the results.
--=20
It's very hard to predict things . . . Especially the future.
-- Prof. Charles Kelemen, Swarthmore CS Dept.
---------------------------------------------------------------------
Luke Kanies | http://reductivelabs.com | http://config-mgmt.blogspot.com
From: peter (Peter da Silva) Date: 21:09 on 13 Oct 2005 Subject: Re: Blogging sucks > Yes, it is not really. It's a simple nod to the people who might want to > include one certain string in their code. A silly workaround is better > than no workaround. It's not just a silly workaround, it's one that makes slashdot markup incompatible with *ML, so why base it on *ML? > >That's completely unparsable by any credible general-purpose *ML parser. > So? How is that a problem in any way? If you're not going to make it *ML, why make people deal with *ML instead of bbcode or something?
From: Chris Nandor Date: 23:11 on 13 Oct 2005 Subject: Re: Blogging sucks At 15:09 -0500 2005.10.13, Peter da Silva wrote: >> Yes, it is not really. It's a simple nod to the people who might want to >> include one certain string in their code. A silly workaround is better >> than no workaround. > >It's not just a silly workaround, it's one that makes slashdot markup >incompatible with *ML, so why base it on *ML? First, it is a workaround no one has ever used except for me, as best I can tell, so who really cares? Second, why not? Again, who cares? >If you're not going to make it *ML, why make people deal with *ML instead >of bbcode or something? Why? Who cares?
From: peter (Peter da Silva) Date: 23:30 on 13 Oct 2005 Subject: Re: Blogging sucks > First, it is a workaround no one has ever used except for me, as best I can > tell, so who really cares? Second, why not? Again, who cares? I care. It's the thin end of the wedge. Next thing you know people are going to come up with <a %end="]" href="blah"]<html>...[/a %end=">">. And Forth does that kind of thing SO much better. It'll all end in tears! PS: I implemented a markup language \[i]in Forth\[n] around 1982, that defined Forth words for markup... and anything inside \x...x was passed to the Forth outer intepreter.
From: Chris Nandor Date: 23:33 on 13 Oct 2005 Subject: Re: Blogging sucks At 17:30 -0500 2005.10.13, Peter da Silva wrote: >> First, it is a workaround no one has ever used except for me, as best I can >> tell, so who really cares? Second, why not? Again, who cares? > >I care. And why should that make ME care? You've not shown how it is actually bad, is what I am saying.
From: velut luna. immanis. Date: 23:38 on 13 Oct 2005 Subject: Re: Blogging sucks On 10/13/05, Peter da Silva <peter@xxxxxxx.xxx> wrote: > PS: I implemented a markup language \[i]in Forth\[n] around 1982, that > defined Forth words for markup... and anything inside \x...x was passed > to the Forth outer intepreter. That's nothing. One time, I ate a whole jar of olives. I mean a WHOLE JAR= . -v -- +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ Don't ever tell anybody anything. If you do, you start missing everybody.
From: Luke Kanies Date: 00:01 on 14 Oct 2005 Subject: Re: Blogging sucks On Thu, 13 Oct 2005, velut luna. immanis. wrote: > On 10/13/05, Peter da Silva <peter@xxxxxxx.xxx> wrote: > > > PS: I implemented a markup language \[i]in Forth\[n] around 1982, that > > defined Forth words for markup... and anything inside \x...x was passed > > to the Forth outer intepreter. > > That's nothing. One time, I ate a whole jar of olives. I mean a WHOLE JAR. My wife did that when she was young; she still can't eat olives. Peter, can you still not write in Forth? I know I haven't been able to write in the shell since I wrote that shell script using arrays and references. *shudder*
Generated at 10:26 on 16 Apr 2008 by mariachi