[00:00:22] <arg> no one should ever let Dave Winer near an XMLRPC API ever again
[00:00:44] <etrepum> no one should let Dave Winer .... ever again
[00:01:15] <arg> haha exactly
[00:01:30] <arg> same with RMS
[00:01:49] <arg> well, RMS has actually done useful stuff
[00:03:55] <etrepum> yeah
[00:04:45] <arg> he sure is irritating tho
[00:05:07] <etrepum> so are a lot of people
[00:05:25] <arg> yeah, im sure im irritating 
[00:05:27] <arg> in fact
[00:05:30] <arg> i know i am!
[00:05:34] <arg> my wife just told me
[00:06:16] <etrepum> I am too
[00:06:37] <etrepum> I think it kinda goes with the territory
[00:07:32] <arg> yeah i know
[00:08:14] <arg> (about it going with the territory, not you being annoying)
[00:08:20] <arg> :)
[00:08:34] <etrepum> haha, both are probably equally true
[00:09:42] <etrepum> pysqlite's web pages are all messed up
[00:09:55] <etrepum> it takes you right to a trac, with no download links
[00:10:35] <arg> i think i went right to sf.net
[00:10:49] <arg> i vaguely recall having the same problem
[00:10:50] <etrepum> yeah it's there in /projects/
[00:11:14] <arg> sf.net download page makes me want to stab my own eyes out
[00:11:36] <arg> when i want to just copy the link
[00:11:44] <arg> not download it to the machine im browsing on
[00:11:49] <etrepum> yeah exactly
[00:11:59] <etrepum> even if you're a subscriber their direct download stuff isn't that great
[00:12:12] <etrepum> it's IP bound and the links only show up if it thinks you're logged in
[00:12:17] <arg> i can imagine the whole thing is a mess of spaghetti php
[00:12:32] <etrepum> is there any other kind?
[00:13:03] <arg> haha no.
[00:13:05] <arg> it sucks
[00:13:16] <arg> every time someone tells me there going to do something in php i ask why not python
[00:13:24] <arg> and i've never heard a good answer
[00:13:29] <arg> other than lame hosting
[00:14:14] <etrepum> lame hosting, popularity, "support", and the "hit by a bus" problem
[00:14:33] <arg> heh, yeah
[00:14:55] <etrepum> I contend that the hit by a bus problem is going to be a problem regardless of the language that's chosen
[00:15:04] <arg> yeah, its just a fact of life
[00:15:17] <arg> for any non-trivial system
[00:15:22] <etrepum> exactly
[00:15:29] <arg> it takes a non-trivial amount of time to learn how to maintain it
[00:15:36] <arg> being non-trivial and all
[00:15:55] <etrepum> like if I wrote PHP code, I can't imagine that someone else would understand it, because I'm sure I wouldn't be using PHP idioms
[00:16:16] <arg> yeah, its like when I write Perl
[00:16:19] <arg> or C++
[00:16:31] <etrepum> in the times that I've used PHP, it feels like a broken version of Perl
[00:16:36] <arg> i dont know the relevant line noise in those languages
[00:16:40] <etrepum> and well, you know what Perl feels like
[00:16:56] <arg> ugh
[00:17:01] <arg> dont get me started
[00:17:24] <arg> i had to maintain a 70 KLOC perl system
[00:17:28] <arg> for a few years
[00:17:38] <etrepum> well I have several years of experience with Perl, so I'm somewhat desensitized to it
[00:17:53] <etrepum> I've done more or less the same, except I wrote it
[00:18:16] <arg> i didnt learn much maintaining it
[00:18:25] <etrepum> pretty soon after I wrote it, I learned Python
[00:18:32] <arg> because the fixes were usually obvious mutations of the existing code or cut-and-paste type stuff
[00:18:40] <etrepum> yeah
[00:18:42] <etrepum> debugging perl sucks
[00:18:50] <etrepum> especially if the code isn't use strict; compliant
[00:19:21] <arg> yeah all our perl is use strict;
[00:19:27] <etrepum> thank god for that
[00:19:40] <arg> its possible to write very elegant perl
[00:19:44] <arg> i've seen it
[00:20:04] <etrepum> use English; or whatever it's called gives you alternatifves to a lot of the crazy $_!@#$*( crap
[00:20:13] <arg> ah! a comment popup!
[00:20:33] <etrepum> yea popup is a popular solution
[00:20:51] <etrepum> but you still usually want to say "there are N comments" on the page
[00:20:56] <arg> i mean what the hell kind of calling convention is my $arg1, $arg2, $arg3 = shift;
[00:20:58] <arg> or whatever
[00:21:02] <arg> yeah
[00:21:06] <etrepum> yeah the calling convention is scary
[00:21:15] <arg> @_
[00:21:15] <etrepum> and the pass by value / pass by reference distinction
[00:21:23] <etrepum> and the fact that subs aren't first class
[00:21:32] <arg> its all scary
[00:21:38] <etrepum> well they are sort of first class
[00:22:07] <etrepum> but if you do if (foo) { sub bar {} } else { sub bar {} } then you will pleasantly be surprised that the latter bar is always used
[00:22:15] <etrepum> regardless of what the if statement leads you to believe
[00:22:47] <arg> oh, and the string "0 but true" has a value of zero but is true in a boolean context
[00:22:56] <arg> its special-cased in the interpreter
[00:23:16] <arg> or maybe not
[00:23:19] <etrepum> I was trying to add some platform-dependent code, so I was checking the relevant line noise special variable and then defining the right sub
[00:23:22] <etrepum> and it didn't work
[00:23:29] <arg> maybe it takes the first element and casts it to an int
[00:23:30] <etrepum> and it confused me
[00:23:44] <etrepum> but then I just eval()'ed the relevant sub and then it worked
[00:25:06] <arg> i've enjoyed reading larry walls 'apocalypses' on perl 6 tho
[00:25:41] <etrepum> I really liked the periodic table of operators
[00:26:05] <arg> haha yeah i've seen that
[00:26:16] <etrepum> http://www.ozonehouse.com/mark/blog/code/PeriodicTable.html
[00:27:08] <etrepum> just learning what all of those operators do would take a year
[00:28:53] <arg> haha i know... perl hackers seem to be very different types of people than python folks
[00:29:00] <arg> a lot more sysadmin oriented
[00:29:06] <etrepum> yeah
[00:29:08] <arg> grumpy
[00:29:15] <etrepum> I've been both
[00:30:14] <etrepum> if ruby came out a couple years earlier I'd probably have ended up with that
[00:30:37] <etrepum> but I didn't really even hear about it until after I was already stuck on python
[00:30:44] <arg> yeah same here
[00:30:54] <arg> my dad got me a c64 when i was 4
[00:31:03] <arg> used that until like 1993, when we got a 486
[00:31:14] <etrepum> how long was that?
[00:31:21] <arg> had jobs programming VB and access
[00:31:33] <arg> got the c64 in 1980
[00:31:37] <arg> well, a vic 20
[00:31:42] <arg> c64 came out a lil later i think
[00:31:57] <arg> then in college one of my bosses turned my on to unix
[00:32:05] <arg> and suggested i pick a languagew and go with it
[00:32:18] <arg> and basically said he had heard cool things about python
[00:32:27] <etrepum> haha
[00:32:32] <arg> this was right when 1.52 came out
[00:32:40] <arg> so that was that
[00:32:57] <etrepum> you're a couple years older than me, so the computers were different
[00:33:44] <arg> i used to love that thing
[00:34:29] <etrepum> I had a PCjr (8088?), then a PS/2 model 30 (8086), then a PS/2 model 70-486
[00:34:58] <etrepum> after that I put together my own boxes
[00:35:08] <etrepum> .. and switched to Mac a couple years ago
[00:35:39] <arg> nice
[00:35:44] <arg> i wish i got into the mac earlier
[00:36:02] <etrepum> I dabbled a bit in MacOS programming, and it sucked
[00:36:08] <arg> yeah?
[00:36:19] <etrepum> yeah, Carbon is painful
[00:36:21] <arg> i can imagine that i guess
[00:36:30] <etrepum> you probably wish you had gotten into NeXT :)
[00:36:43] <arg> true
[00:36:56] <etrepum> I wish I did, but there's no way a high school kid was going to get his hands on a NeXT box
[00:37:05] <arg> or been a bit older and gotten my hands on a Symbolics machine
[00:37:32] <etrepum> I had access to some SGIs at work, because I worked for a computer game company, but they were basically outdated by the time I had gotten there
[00:37:58] <etrepum> I also wish I had learned BSD instead of Linux
[00:39:39] <etrepum> but I did Perl because I was doing web stuff
[00:39:53] <etrepum> and thats what everyone did
[00:40:04] <arg> yeah
[00:40:21] <arg> you prolly got into unix sooner than I.. Im sure i'd have gotten into Perl too
[00:40:46] <etrepum> yeah I was running Linux in like 92
[00:41:02] <arg> yeah, i didnt catch on until 97 or so
[00:42:03] <etrepum> I was actually using it as a desktop OS for a while but then I had to switch to windows b/c I was doing crap in director and then flash
[00:42:15] <etrepum> but then Mac OS X came out and solved that problem
[00:42:31] <arg> yeah i used linux for my workstation at work until i bought my powerbook
[00:42:49] <arg> debian is nice
[00:43:11] <etrepum> yeah, it's either nice or horrible, depending on if you need things that are in stable or not
[00:44:00] <arg> well, yeah
[00:46:53] <etrepum> gcc needs an option that says "link statically if an .a is around"
[00:50:58] <arg> yes!
[00:51:21] <arg> i was saying that the other day
[00:51:30] <etrepum> either that or distutils could grow an option to do it and just bypass -l altogether
[00:51:36] <etrepum> maybe I'll do that
[00:51:39] <arg> ugh, distutils
[00:51:44] <etrepum> yeah, I second that
[00:51:57] <arg> its better than nothing tho
[00:52:00] <etrepum> distutils makes easy things easy and hard things harder :)
[00:52:05] <arg> it works, once you get it working
[00:53:17] <etrepum> and it's so entrenched it's hard to replace
[00:54:38] <etrepum> although I suppose it's possible to write something that exposes a distutils compatible interface
[00:54:47] <etrepum> but doesn't suck inside
[00:56:02] <etrepum> it's not really clear to me why they didn't use make like everyone else does
[00:56:02] <arg> can you do 'if key in dict' in 2.3?
[00:56:08] <etrepum> you can do it in 2.2
[00:56:14] <arg> ok cool
[00:56:14] <etrepum> maybe even 2.1
[01:00:22] <etrepum> I just added pysqlite and APSW to pythonmac.org packages
[01:01:07] <etrepum> when 2.4.1 comes out I'm going to start a 2.4 section
[01:03:51] <etrepum> it looks like /hacking/ distutils to support a static option is easy, but patching it would be hard :)
[01:04:09] <etrepum> there's a module-level function, gen_lib_options, that does it
[01:04:15] <etrepum> so you can't override it by subclassing
[01:04:20] <etrepum> you'd have to monkey-patch, wooha
[01:04:42] <arg> hehe :)
[01:04:47] <etrepum> or you could subclass the link method of the Compiler
[01:04:52] <etrepum> which is the call point for that
[01:05:09] <arg> i luckily havent had to poke too hard into distutils
[01:05:17] <arg> thanks god pyrex comes with the extensions
[01:05:22] <etrepum> yeah, unfortunately, someone had to write py2app
[01:05:33] <etrepum> and I'm probably the only one who knows how
[04:02:31] [connected at Sun Mar 13 04:02:31 2005]
[04:02:31] <> *** Looking up your hostname...
[04:02:31] <> *** Checking ident
[04:02:31] <> *** Found your hostname
[04:03:02] <> *** No identd (auth) response
[04:03:02] <> *** Your host is kornbluth.freenode.net[kornbluth.freenode.net/6667], running version dancer-ircd-1.0.35
[04:03:02] [I have joined #peak]
[04:03:02] ** kornbluth.freenode.net set the topic to http://dirtsimple.org/2004/11/generic-functions-have-landed.html
[06:39:20] ** hazmat has joined us
[06:52:49] ** hazmat has left IRC ("Leaving")
[06:53:09] ** bear has left IRC (Read error: 60 (Operation timed out))
[09:29:17] ** vlado has joined us
[14:13:57] ** bear_ has joined us
[14:15:57] bear_ is now known as bear
[14:46:39] ** bear_ has joined us
[14:49:33] ** bear has left IRC (Nick collision from services.)
[14:49:40] bear_ is now known as bear
[15:04:21] ** debugger has joined us
[18:37:51] ** vlado has left IRC ("Leaving")
[19:52:07] ** debugger has left IRC ()
[20:58:02] ** hazmat has joined us