00:00:00 --- log: started forth/04.09.11
00:44:34 --- join: mur_ (~mur@smtp.uiah.fi) joined #forth
00:55:50 --- quit: mur (Read error: 110 (Connection timed out))
01:04:26 --- join: Topaz (jonny@spc1-horn1-6-0-cust217.cosh.broadband.ntl.com) joined #forth
01:24:14 --- quit: kazuya ("leaving")
01:49:28 --- join: crc (crc@65-pool1.ras11.nynyc-t.alerondial.net) joined #forth
02:53:22 --- join: crc_ (crc@36-pool1.ras11.nynyc-t.alerondial.net) joined #forth
03:03:19 --- quit: crc (Nick collision from services.)
03:03:25 --- nick: crc_ -> crc
03:12:10 --- join: mur (~mur@mgw2.uiah.fi) joined #forth
03:22:20 --- quit: mur_ (Read error: 110 (Connection timed out))
03:34:38 --- quit: crc ("Time for bed... Goodnight!")
03:44:42 --- join: crc (crc@55-pool1.ras11.nynyc-t.alerondial.net) joined #forth
03:54:26 --- join: saon (Ecoder@c-24-129-95-254.se.client2.attbi.com) joined #forth
03:55:32 <saon> mornin' forthers
03:57:15 <crc> Good morning
04:02:05 --- quit: Topaz (Remote closed the connection)
04:21:12 --- join: qFox (C00K13S@82-169-140-229-mx.xdsl.tiscali.nl) joined #forth
04:21:34 --- join: mur_ (~mur@kyberias.uiah.fi) joined #forth
04:28:58 --- join: Tomasu (~moose@S010600045a4c73cc.ed.shawcable.net) joined #forth
04:32:19 --- quit: mur (Read error: 110 (Connection timed out))
04:58:17 --- nick: Tomasu -> Tomasu-
04:58:29 --- nick: Tomasu- -> Toma
04:59:16 --- nick: Toma -> Tomasu
05:16:11 --- quit: mur_ (Remote closed the connection)
05:16:44 --- join: mur (~mur@smtp.uiah.fi) joined #forth
05:46:30 --- quit: saon ("Leaving")
06:47:33 --- join: tathi (~josh@pcp02123722pcs.milfrd01.pa.comcast.net) joined #forth
07:01:16 --- join: saon (Ecoder@c-24-129-95-254.se.client2.attbi.com) joined #forth
07:04:53 --- join: mur_ (~mur@kyberias.uiah.fi) joined #forth
07:13:18 * Tomasu is away: dlrrp
07:16:03 --- quit: mur (Read error: 110 (Connection timed out))
07:35:05 --- quit: saon ("Leaving")
07:58:20 --- quit: crc ("Time for bed... Goodnight!")
08:11:42 --- quit: mur_ (Remote closed the connection)
08:12:44 --- join: mur (~mur@mgw2.uiah.fi) joined #forth
08:29:27 <I440r> sc
08:35:20 --- join: Herkamire (~jason@h000094d30ba2.ne.client2.attbi.com) joined #forth
08:35:20 --- mode: ChanServ set +o Herkamire
08:43:54 --- join: zruty (~chuck@64-121-15-14.c3-0.sfrn-ubr8.sfrn.ca.cable.rcn.com) joined #forth
08:46:52 <Herkamire> hi
08:47:05 --- quit: tgunr (Excess Flood)
08:48:58 --- join: tgunr (~davec@vsat-148-63-4-107.c001.g4.mrt.starband.net) joined #forth
09:56:27 --- quit: tathi ("Lost terminal")
10:11:03 --- join: mur_ (~mur@smtp.uiah.fi) joined #forth
10:13:52 --- quit: mur (Read error: 60 (Operation timed out))
10:23:11 --- quit: zruty ("Lost terminal")
10:26:17 --- join: zruty (~chuck@64-121-15-14.c3-0.sfrn-ubr8.sfrn.ca.cable.rcn.com) joined #forth
10:36:13 <qFox> anyone here familiar with collection logic?
10:37:46 <I440r> whussat ?
10:37:54 <qFox> {a,b,c}
10:38:46 <qFox> i need to know whether the result of a "power" collection (i dont know whether thats proper) of P{a,{a}} is equal to {{a},{a,{a}},0} or {a,{a},{a,{a}},0}
10:45:35 <arke> o.O no idea
10:48:49 <I440r> you lost me a {
10:48:51 <I440r> hehe
10:52:30 <arke> whats easier, software or hardware protection?
10:54:08 <qFox> the answer was neither
10:54:22 <qFox> it was {{a},{{a}},{a,{a}},0}
10:54:40 * qFox makes note not to trust classmate again
10:54:46 <qFox> he told me {{a}} wasnt allowed
10:55:54 <arke> hehe
10:57:41 <Herkamire> arg.
10:58:03 <Herkamire> anybody know html/css well enough to tell me how to make a table cell that contains an image and absolutely no space around it
10:58:12 <qFox> cellspacing=0
10:58:18 <qFox> cellpadding=0
10:58:27 <qFox> i believe you want the first one, perhaps you want both.
10:58:40 <qFox> (as part of the <table> tag)
10:58:58 <Herkamire> I have cellspacing="0" cellpadding="0"
10:59:26 <Herkamire> css for the table cell says: margin: 0; padding: 0; border: 0; line-height: 0;
10:59:39 <Herkamire> it works fine in firefox
10:59:56 <Herkamire> but Safari shows a space below the images
11:00:09 <Herkamire> (line-height required for it to work in firefox)
11:00:12 <qFox> if it works in ff, and you have the w3 engine any browser _should_ display teh same
11:00:16 <Herkamire> get's rid of the little gaps below the images
11:00:52 * arke offers gmail invites to anyone interested
11:01:52 <qFox> Herkamire> you run that page in the w3 engine?
11:01:54 * arke should join #gentoo and say FREE GMAIL INVITE TO THE FIRST 6 PEOPLE TO MSG ME CHILD PR0N lol
11:01:57 <qFox> doctype thing
11:02:17 <arke> except that im not into child pron
11:02:19 <Herkamire> Safari uses khtml to render
11:02:34 <arke> but it would be interesting to see how many sick perverts are hanging out in #gentoo :)
11:02:38 <Herkamire> IE 5.1 on the mac renders it without spaces
11:03:20 <qFox> "ie5.1 on the mac", i find it funny every time i see it (at my university they have that too)
11:03:50 <Herkamire> first time it's been opened in months I think
11:03:59 <Herkamire> I've always hated IE
11:04:31 <qFox> well as far as i know, all browsers _should_ display html pages in a certain doctype the same, also in a faster engine (not that that matters anything to html, but still)
11:04:51 <qFox> but i am aware that this theory is slightly flawed with certain browsers so..
11:06:02 <Herkamire> qFox: you don't really know what you're talking about
11:06:12 <qFox> ...
11:06:16 <Herkamire> you can be quite specific with css
11:06:31 <Herkamire> but browsers are allowed to (and have) different defaults for everything
11:06:41 <qFox> yes but there will always be browserincompatibility, as you see yourself...
11:07:07 <qFox> but its my understanding that a website rendered in a certain doctype, _should_ always be displayed the same in any browser supporting that mode
11:07:23 <Herkamire> that's not true at all
11:07:44 <Herkamire> maybe it's true if you specify the exact size for every attribute on everything
11:07:48 <Herkamire> which basically nobody does
11:07:57 <Herkamire> and if they have all the fonts you specify
11:08:13 <Herkamire> and you have to specify things in certain ways
11:08:31 <Herkamire> eg specifying fonts in point sizes is relative, and displays differently on different systems
11:09:52 <Herkamire> html/css was not designed to specify exactly what things look like.
11:09:53 --- join: goofygurl (~agrzw@ip68-224-208-118.lu.dl.cox.net) joined #forth
11:09:54 <Herkamire> that's PDF
11:10:03 <goofygurl> i found u I440r
11:10:06 <qFox> yes, but should it not display table marges equally?
11:10:13 <Herkamire> PDF was designed to get something to look exactly the same on different systems, and in print
11:10:25 <Herkamire> qFox: it's not the table margins
11:10:32 <qFox> the cellspacing.
11:10:34 <qFox> whatever
11:10:44 <Herkamire> I think it's trying to align the bottom of the images with the bottom of normal letters like M
11:10:57 <Herkamire> and leaving space below it for the tails of pgjy etc
11:11:18 <Herkamire> line-height: 0px; fixes that for firefox, but not in Safari
11:11:43 <Herkamire> maybe I just have to tell it to align the image with text differently (even though there's no text in there
11:16:36 <slava> hi
11:25:08 --- quit: goofygurl (Read error: 60 (Operation timed out))
11:45:39 <Herkamire> grrrrr
11:45:57 <Herkamire> I removed the DTD (html 4.01 Strict) and it displays properly in Safari
11:46:13 <Herkamire> hi slava
11:47:17 <arke> try xhtml strict :)
11:50:13 <slava> arke, i see you've been in #squeak lately :)
11:54:56 <slava> arke, be careful, you might upset AlonzoTG and he'll come in here as 'water' and yell at you :)
11:57:35 --- join: Robert__ (~snofs@c-bf5a71d5.17-1-64736c10.cust.bredbandsbolaget.se) joined #forth
11:57:41 --- quit: mur_ (Remote closed the connection)
11:58:14 --- join: mur (~mur@mgw2.uiah.fi) joined #forth
12:02:36 <qFox> well he wont as such ;)
12:05:37 --- quit: Robert (Read error: 60 (Operation timed out))
12:09:42 --- join: saon (Ecoder@c-24-129-95-254.se.client2.attbi.com) joined #forth
12:16:27 --- join: Topaz (~top@spc1-horn1-6-0-cust117.cosh.broadband.ntl.com) joined #forth
12:16:32 --- join: mur_ (~mur@mgw2.uiah.fi) joined #forth
12:18:45 --- quit: mur (Read error: 60 (Operation timed out))
12:20:33 --- quit: zruty ("Lost terminal")
12:22:26 --- quit: saon ("Leaving")
12:36:15 --- join: imaginator (~George@georgeps.dsl.xmission.com) joined #forth
12:46:53 <Herkamire> arke: good idea, but that didn't fix it
13:23:09 --- quit: tgunr (Excess Flood)
13:24:57 --- join: tgunr (~davec@vsat-148-63-4-107.c001.g4.mrt.starband.net) joined #forth
13:50:49 --- quit: mur_ (Remote closed the connection)
13:51:37 --- join: mur (~mur@uiah.fi) joined #forth
14:16:56 --- join: mur_ (~mur@mgw2.uiah.fi) joined #forth
14:17:56 --- join: TheBlueWizard (TheBlueWiz@modem-041.nyc-tc03b.fcc.net) joined #forth
14:17:56 --- mode: ChanServ set +o TheBlueWizard
14:21:16 --- join: madgarden (~madgarden@Kitchener-HSE-ppp3575808.sympatico.ca) joined #forth
14:21:51 <slava> hi madgarden 
14:21:56 <madgarden> hi ya
14:24:32 --- join: saon (~Ecoder@c-24-129-95-254.se.client2.attbi.com) joined #forth
14:25:22 <slava> i'm trying to come up with the best way to parse wiki pages
14:25:23 --- quit: saon (Client Quit)
14:25:28 --- quit: madgarden_ (Read error: 60 (Operation timed out))
14:25:31 --- join: saon (Ecoder@c-24-129-95-254.se.client2.attbi.com) joined #forth
14:25:58 <slava> hi saon
14:26:06 <saon> hey slava
14:26:52 --- quit: mur (Read error: 238 (Connection timed out))
14:31:45 --- quit: qFox ("this quit is sponsored by somebody!")
14:35:04 <slava> whoa i had a great idea
14:35:10 <slava> WikiWords should actually be words
14:35:16 <slava> that display that wiki topic
14:38:15 <slava> : (wiki-parser) ( text -- )
14:38:15 <slava>     [
14:38:15 <slava>         scan dup wiki-word? [
14:38:15 <slava>             <a href= dup a> write </a>
14:38:15 <slava>         ] [
14:38:16 <slava>             write
14:38:18 <slava>         ] ifte " " write
14:38:20 <slava>     ] with-parser ;
14:39:21 <Herkamire> ahhhhhhh...
14:39:34 <Herkamire> this page describes the html rendering problem I'm having, and how to fix it: http://devedge.netscape.com/viewsource/2002/img-table/
14:40:39 <Herkamire> except that the virtical-align thing didn't seem to work for me.
14:40:42 <Herkamire> I tried that
14:40:49 <Herkamire> display: block; works great
14:41:07 <slava> hmm i'm not happy with the above wiki parser either since it munges whitespace
14:41:11 <Herkamire> except of course that it's sometimes inconvenient to have it be a block level element
14:41:20 <Herkamire> eg I had to make an extra two tables in my design
14:42:53 <Herkamire> google to the rescue again
14:47:36 --- quit: TheBlueWizard (Nick collision from services.)
14:48:26 --- join: TheBlueWizard (TheBlueWiz@modem-008.nyc-tc03b.fcc.net) joined #forth
14:48:26 --- mode: ChanServ set +o TheBlueWizard
14:53:05 --- quit: saon ("Leaving")
14:59:07 --- quit: TheBlueWizard (Nick collision from services.)
14:59:31 --- join: TheBlueWizard (TheBlueWiz@modem-147.nyc-tc03a.FCC.NET) joined #forth
14:59:31 --- mode: ChanServ set +o TheBlueWizard
15:45:06 --- quit: mur_ (Remote closed the connection)
15:45:56 --- join: mur (~mur@uiah.fi) joined #forth
15:53:46 --- nick: Robert__ -> Robert
15:57:26 <TheBlueWizard> all bye
15:57:42 --- part: TheBlueWizard left #forth
16:19:02 * Tomasu is back (gone 09:05:44)
16:49:16 --- quit: Topaz ("Leaving")
16:52:36 --- join: mur_ (~mur@mgw2.uiah.fi) joined #forth
17:01:52 --- quit: mur (Read error: 238 (Connection timed out))
17:16:44 --- quit: mur_ (Remote closed the connection)
17:17:44 --- join: mur (~mur@mgw2.uiah.fi) joined #forth
17:18:00 --- join: saon (Ecoder@c-24-129-95-254.se.client2.attbi.com) joined #forth
17:20:29 * Tomasu is away: foed
17:52:21 --- join: Falala (~lala@207.140.211.88) joined #forth
17:54:26 * arke is away: work
18:13:13 --- quit: saon ("Leaving")
18:18:44 <Herkamire> rrrrrr perl
18:21:48 <Herkamire> mmm. never mind
18:21:52 <Herkamire> perl is cool again
18:24:35 <Falala> perl == linenoise
18:32:45 <slava> he
18:32:46 <slava> heh*
18:36:36 --- join: saon (Ecoder@c-24-129-95-254.se.client2.attbi.com) joined #forth
18:38:16 <Falala> seriously
18:38:25 <Falala> have you ever seen clean perl?
18:38:34 <Falala> maybe in a jewelry store, but that's it
18:39:04 <Falala> slava: you develop JEdit, right? how old are you?
18:40:54 <slava> yes i do
18:40:55 <slava> i'm 20
18:41:08 <Falala> whoa
18:41:17 <Falala> you must have started writing jedit when you were like, 15?
18:41:22 <slava> 14
18:41:40 <Falala> was it your first major project?
18:41:57 <slava> yes
18:44:40 <Falala> how did you manage to create something on such a large scale at such a young age
18:44:46 <Falala> was it tough in the early stages?
18:45:16 <slava> it started off small and simple
18:45:39 <Falala> did you have help?
18:45:50 <slava> i wrote most of it mysef
18:46:59 --- quit: saon ("Leaving")
18:47:23 <Falala> what did your CS teachers in high school think? :)
18:47:47 <slava> i never did well in high school
18:48:12 <Falala> shocking
18:51:42 <Falala> man, i wish i could throw together something as cool as jedit
18:51:55 <Falala> my comp sci teacher would really worship the ground i walk on
18:58:58 --- nick: Falala -> iolairesliver
18:59:08 --- nick: iolairesliver -> falala
18:59:24 <slava> falala, are you doing a CS degree?
19:01:26 <falala> slava: i'm 15
19:01:29 <falala> in high school
19:02:06 <slava> cool
19:02:09 <slava> you're interested in forth?
19:02:34 <falala> i wanna learn it
19:02:46 <slava> do you know other languages?
19:02:55 <falala> i've already mastered C++, C, perl, java, and ppc & x86 asm
19:03:01 <falala> and some python and ruby
19:03:17 <slava> that's a lot. when i was 15 all i knew was java
19:03:29 <slava> you should learn forth
19:03:33 <slava> and some dynamic language
19:03:47 <falala> give me an example of a dynamic langauge you'd recommend
19:03:57 <slava> lisp, scheme, smalltalk, dylan
19:04:11 <slava> take a look at www.squeak.org. a nice free smalltalk implementation
19:04:37 <slava> i'm making my own language, but its not fully ready yet. factor.sourceforge.net
19:04:37 <falala> i talked to my cs teacher about lisp before and he said it isn't worth my time
19:04:42 <slava> why did he say that?
19:04:47 <falala> no one gets paid to write lisp anymore
19:04:51 <slava> a lot of people do
19:04:53 <falala> he said to focus on the newer stuff
19:05:00 <slava> and the ones that do get paid much more than a java programmer
19:05:01 <slava> what is newer?
19:05:16 <falala> slava: java, c++, python, xml, html, etc
19:05:16 * Tomasu is back (gone 01:44:47)
19:05:33 <slava> the latter two are not programming languages, and arguably c++ is less advanced than lisp
19:06:02 <falala> well, my teacher calls them skillsets
19:06:26 <slava> anybody can learn html or xml its not a special skill
19:06:40 <falala> he's like, "in today's IT world if your skillset is older than 5 years it is worthless"
19:06:50 <slava> that's definately true
19:06:59 <falala> so you have to learn the newest languages as soon as they become popular
19:07:07 <imaginator> Why?
19:07:13 <falala> imaginator: to make money
19:07:17 <slava> the thing is, lisp nowadays is totally different from lisp in the 60's.
19:07:21 <falala> employers aren't going to want to hire some crusty old lisp hacker ;)
19:07:32 <slava> saying that lisp is 'old' is like saying taht java is old because its similar to algol 60.
19:07:34 <falala> i mean, lisp is only good for like AI work, and that's all dried up
19:07:43 <imaginator> not
19:07:46 <slava> falala, like i said, a lot of people get paid a lot to program lisp.
19:07:57 <slava> falala, you just don't hear about it because java's being hyped
19:07:58 <imaginator> There are companies that sell commercial lisp systems that are still in business.
19:08:16 <slava> falala, in a few years java will totally disapper and be replaced by some other 'new' technology
19:08:21 <slava> but lisp will still exist
19:08:58 <falala> slava: i don't think so. java's been growing exponentially since its introduction in like 94(?)
19:08:59 <I440r> slava - for ANY given problem... what is THE best solution ?
19:09:09 <falala> oh yeah, i forgot to mention, C#
19:09:20 <falala> my teacher told me that's the most important one to learn right now
19:09:21 <slava> falala, if you haven't noticed, java and c# are basically 70's style languages
19:09:30 <falala> because he said in a few years, EVERYTHING is going to run on .Net
19:09:32 <slava> falala, i doubt C# will overtake java any time soon
19:09:46 <falala> like, there won't be a computer allowed to connect to the internet that doesn't run .NET
19:09:50 <slava> falala, do you realize your teacher is only an HS teacher because he couldn't get a real job? so his opinion is not really worth listening to.
19:09:53 <falala> microsoft is going to make sure of it
19:09:57 <falala> i don't doubt it either
19:10:04 <slava> falala, they can't do that, its impossible
19:10:04 <falala> slava: ........
19:10:10 <slava> falala, think about it
19:10:16 <slava> if this teacher was a genious programmer or CS researcher
19:10:19 <slava> he wouldn't be teaching HS
19:10:21 <slava> he'd be at university at least
19:10:31 <slava> any windbag can read a VB book and go teach HS
19:10:53 <slava> but if you're clueless and teaching at university you'll be in a situation where the students know more than you.
19:10:53 <falala> did you show this same attitude to YOUR hs teachers? :P
19:10:57 <slava> falala, yes
19:11:10 <falala> slava: and you must have been proven wrong tiem and again in class
19:11:41 <falala> this guy used to write VB databases for like, local banks and law firms
19:11:43 <slava> falala, C# and java are great for solving easy problems. there's many IDEs, a lot of developers willing to work for low wages, a lot of pre-built APIs, etc.
19:11:44 <falala> i think he knows his stuff
19:11:59 <slava> falala, writing business code in VB is not the hallmark of programming
19:12:14 <slava> falala, has he ever solved any hard problems? like writing compilers, speech recognition, 3d graphics...?
19:12:53 <falala> slava: he wrote a little game in VC++ once, it's on his site, it's pretty neat, it does simple DOOM-style graphcis in a 500kb binary
19:12:55 --- join: kc5tja (~kc5tja@66-74-218-202.san.rr.com) joined #forth
19:12:55 --- mode: ChanServ set +o kc5tja
19:13:22 <falala> slava: why would anyone write a compiler if we already have GCC & Visual Studio? :)
19:13:39 <I440r> falala, BECAUSE they are not good enough
19:13:43 <I440r> nuff sed
19:13:45 <slava> falala, if you want to follow in his footsteps and be mediocre, stick to the current 'next big thing', like c#, java, etc. that is a good set of skills to have but it shouldn't be your only skill.
19:14:02 <falala> slava: if C# is going to be unimportant why is MS pushing it so hard
19:14:19 <kc5tja> falala: Because they want it to be important.
19:14:21 <falala> seriously, everyone i've talked to about programming (alot of smart ppl) are like, "C# is going to be the biggest language ever"
19:14:24 <kc5tja> Why does any business push anything?
19:14:30 <slava> falala, C# will be popular no doubt.
19:14:34 <kc5tja> When Windows first was introduced, it was wholesale *rejected* by the industry.
19:14:36 <falala> it makes sense too. have you ever tried programming for .net?
19:14:44 <slava> falala, lets try a thought experiment :)
19:14:49 <falala> it's a trusted compiler type system i think and it allows for really rapid prototyping
19:14:51 <slava> falala, you're writing a program in C#, or C++, or whatever
19:14:56 <falala> you can toss together an app for windows in record time
19:14:59 <slava> falala, you run your program
19:15:04 <slava> falala, there's a bug -- the program crashes
19:15:08 <slava> falala, what do you do?
19:15:21 <falala> uh, fire up the debugger?
19:15:22 <slava> falala, a C# developer starts making changes to the source, recompiles, restarts the app, tests again, repeat cycle
19:15:23 <I440r> falala,  no dood. dragging and dropping gra[phical blobs where you want them is not coding
19:15:35 <slava> falala, a lisp or smalltalk developer fixes the bug, and resumes execution at that point to test the fix
19:15:42 <slava> falala, there is no recompile/restart cycle
19:15:58 <falala> I440r: well it's easier to do that than like, enter all the values for the windows and widgets etc by hand
19:16:00 <slava> I440r, that is not the point here
19:16:05 <slava> I440r, gui builders are a valid technology
19:16:08 <falala> yeah
19:16:21 <I440r> slava i didnt say they werent. i said it wasnt coding
19:16:30 <slava> falala, the fundamental problem with C# (and not specifically C#) is it doesn't support iterative development.
19:16:48 <slava> falala, you're stuck with a 'black box' compiler. you feed it your source, it spits out either errors or an exe
19:16:49 <falala> slava: huh? i think it does. it's actually partly based on smalltalk
19:16:51 <slava> falala, you run the exe
19:16:55 <slava> falala, it crashes
19:16:57 <slava> falala, repeat...
19:16:59 <I440r> most c coders mix their pre-coded ingredients and stuff em in the easy bake oven and out pops 400 megs of hello world
19:17:02 <falala> slava: i read some interview with the guy behind C# and he said they use a VM like Smalltalk
19:17:11 <slava> falala, yes but do you know how development is done with smalltalk?
19:17:17 <slava> falala, you code, test and debug at the same time
19:17:20 <kc5tja> falala: A VM does not a Smalltalk make.  
19:17:23 <falala> although i don't know much about smalltalk or lisp.... but i think that's how it goes
19:17:34 <slava> falala, you can update the code, inside a running program.
19:17:41 <falala> kc5tja: but in this case, like, the benefits slava's describing are part of C# because of the vm
19:17:54 <kc5tja> C# is based firmly in the C/Java camp, where there is a discrete edit-time, compile-time, run-time, debug-time, and the cycle continues.
19:17:55 <falala> slava: why are you focusing on this little detail, is it really that important?
19:17:57 <slava> falala, c# and java do support some dynamic features
19:18:12 <slava> falala, however, its not developed to such an extent as a truly dynamic language
19:18:13 <kc5tja> With Smalltalk and Forth, all those times are integrated (or, at least, overlapped so heavily that they're inconsequential).
19:18:32 <slava> falala, also, about the gui builder thing
19:18:37 <slava> falala, in VS.net, you make your forms in a 'form editor'
19:18:39 <falala> slava: microsoft says C# is going to be the next-generation development environment. they have BILLIONS to spend on R&D. do you think that like, they would let these features of LISP/smalltalk fall by the wayside?
19:18:41 <slava> falala, THEN run your app and test it
19:18:45 <falala> if they are truly important?
19:18:50 <slava> falala, in smalltalk, the gui builder *is* the app
19:18:55 <slava> falala, you can tweak your GUI while the app runs
19:19:09 <slava> falala, you've fallen victim to marketing.
19:19:16 <slava> falala, what MS ads say, and what reality is, is quite different
19:19:35 <falala> uh
19:19:46 <falala> but seriously, i think my claim is correct
19:19:52 <slava> (but i don't want to have a debate on MS being evil or not, etc, i don't really care about that)
19:19:53 <falala> can you say anything that would dispute that
19:19:59 <slava> which claim?
19:20:02 <falala> the R&D
19:20:28 <slava> they could spend a lot on r&d, but the fact is, lisp and smalltalk exist *today*, not 10 years later when MS implements feature X Y and Z etc.
19:20:59 <I440r> sc
19:21:05 <slava> also, a good programming environment is based on a small set of well-designed principles
19:21:19 <slava> 10 billion dollars worth of features is not the central concept
19:21:39 <falala> hey I440r, what are your opinions on commenting? what style of commenting is best? :)
19:22:07 <slava> falala, first write self documenting code... then write comments to describe non-obvious interactions and side effects.
19:22:08 <I440r> falala read my sources... they say all
19:22:17 <slava> falala, don't write comments like: i++; /* add 1 to i */
19:22:20 <falala> lolol
19:22:30 <falala> my cs teacher PENALIZES us if we miss even one comment
19:22:36 <falala> we have to write a comment next to every line of code
19:22:43 <slava> because your CS teacher is thick
19:22:45 <I440r> good man
19:22:47 <kc5tja> Which is bullshit.
19:22:48 <I440r> its overkill
19:22:51 <falala> like
19:22:54 <I440r> but it makes you comment
19:22:58 <falala> int c;   // counter variable for second loop
19:23:02 <slava>     /* blank line; does nothing */
19:23:02 <I440r> gets you into the mode of "comment this code"
19:23:04 <slava> falala, hmm
19:23:09 <slava> falala, its better to have: int counter;
19:23:17 <slava> falala, that's self-documenting. 'c' is opaque
19:23:18 <I440r> LATER you can find a happy medium
19:23:25 <slava> falala, 'counter' means what it says
19:23:28 <falala> slava: no. you're wrong. counter variables must be short and succint
19:23:34 <slava> falala, why?
19:23:36 <kc5tja> "Must" be?
19:23:37 <kc5tja> Why?
19:23:40 <kc5tja> Says who?
19:23:40 <slava> falala, doesn't your editor have abbrev expansion? :)
19:23:41 <kc5tja> God?
19:23:46 <slava> kc5tja, the CS teacher, clearly
19:23:50 <falala> using 'counter' is like redundant
19:23:56 <slava> falala, how is it redudant?
19:24:01 <slava> falala, it can avoid having to write a comment
19:24:02 <falala> i mean, if they're one-shot variables, what's the point of an expansive name
19:24:04 <slava> falala, compare
19:24:08 <slava> int counter;
19:24:08 <falala> look at this:
19:24:09 <slava> with
19:24:13 <slava> int c; /* a loop counter */
19:24:15 <kc5tja> They're only one-shot variables -- what's the point of a comment?
19:24:16 <slava> cleraly #1 is simpler
19:24:25 <falala> http://www.joelonsoftware.com/articles/fog0000000073.html
19:24:31 <falala> that should clear up your thought process a bit
19:24:44 <falala> we did a discussion of that aritcle in class a while back, i think ti's brilliant
19:24:50 <slava> can you point a specific part of the article?
19:24:52 <slava> i'm not reading the whole thing
19:25:12 <falala> Some signs of a good programmer: good programmers have a habit of writing their { and then skipping down to the bottom of the page and writing their }s right away, then filling in the blank later. They also tend to have some kind of a variable naming convention, primitive though it may be... Good programmers tend to use really short variable names for loop indices. If they name their loop index CurrentPagePositionLoopCounter it is sure sign that they have not w
19:25:30 <slava> well clearly CurrentPagePositionLoopCounter is excessive!
19:25:38 <slava> however, 'pagePos' is better than, say, 'p'.
19:25:54 <slava> as for writing { and writing the } straight away, that is a matter of taste
19:25:56 <falala> only if the variablle is to be used more than just a few times
19:26:00 <slava> i don't do that, because i know how many outstanding } i have
19:26:16 <slava> falala, when iterating through an array, i'd use 'int i' not 'int arrayIndex'
19:26:27 <slava> falala, beacuse the 'i' has no further meaning, it *is* the loop index
19:26:37 <slava> falala, however, 'int pagePos' might be a loop index, but it has another meaning -- the page position.
19:26:52 <falala> also
19:27:00 <falala> variables should be preceded by the type
19:27:07 <falala> like a string should be szMyString
19:27:11 <slava> no
19:27:12 <kc5tja> falala: You should be kickbanned for something like that.
19:27:15 <I440r> ufg fuck no
19:27:16 <falala> kc5tja: ???
19:27:18 <falala> kc5tja: the article?
19:27:22 <slava> kc5tja, don't make threats like that...
19:27:25 <falala> it's called Hungarian Notation
19:27:29 <slava> falala, not everybody likes it
19:27:33 <I440r> falala,  we know what its called
19:27:33 <falala> our teacher says it is the proper way to name variables
19:27:40 <falala> it is the way MS coders do it
19:27:42 <kc5tja> Hungarian notation is UNIVERSALLY accepted across the ENTIRE computer science and non-Microsoft industrial fields as being wholesale redundant and a continual source of code maintenance issues.
19:27:45 <slava> falala, your teacher claims that his opinions are fact.
19:27:45 <I440r> and it does NOTHING except fill your sources with bullshit visual clutter
19:27:49 <falala> !!!!
19:27:58 <slava> falala, sure, listen to and respect his opinions but don't take it to heart
19:28:03 <slava> falala, there's more than one right way of coding
19:28:09 <slava> falala, your teacher is not the ultimate expert in the field
19:28:11 <falala> MICROSOFT uses it in all their code! how will you get a job coding if your variables don't conform to MS's guidelines????
19:28:19 <I440r> its a microsoft invention
19:28:21 <slava> falala, who cares about MS?
19:28:23 <I440r> of corse they use it
19:28:23 <falala> slava: he probably knows more than any of you guys...
19:28:34 <falala> slava: he used to write databases for like, doctor's offices and stuff
19:28:38 <slava> falala, that's easy work
19:28:38 <falala> he's actually been in the fieled
19:28:40 <falala> field
19:28:41 <kc5tja> falala: Microsoft requires it.  Nobody else does.  I highly encourage you to find a job that actively encourages Hungarian.  You won't find a one.
19:28:43 <slava> falala, i've wokred in the field
19:28:49 <kc5tja> Hint: I've been coding for 20 years.  :)
19:28:51 <slava> falala, i've written a lot of open source as well
19:28:51 <falala> slava: like waht?
19:29:04 <slava> falala, like, stuff much more complex than your HS teacher's 'databases' :)
19:29:09 <falala> kc5tja: why would it be so bad, if Microsoft encourages it
19:29:15 <slava> falala, its not bad per se
19:29:17 <falala> they are the biggest most important software company in the world
19:29:17 <slava> falala, its just redundant
19:29:20 <falala> that makes NO sense
19:29:37 <slava> falala, if MS is so good, why is windows so unsecure? :)
19:30:02 <falala> slava: he wrote a file index system and jpg thumbnail system on his website in PHP
19:30:05 <imaginator> I suppose since China is the largest country I should become Chinese.
19:30:14 <kc5tja> falala: Because (a) MIcrosoft isn't God, and (b) lpsz is meaningless in a 32-bit environment, where "long pointers" mean nothing.  Long pointers were important in the 16-bit x86 architecture days when segments had to be explicitly handled.  Not anymore.  Consequently, achFoo and pszFoo both refer to the same base kind -- which is correct?
19:30:15 <falala> slava: it's not
19:30:24 <slava> falala, right now i'm writing a *web server*...
19:30:35 <falala> slava: it's just there are so many copies of Windows in the world, many are bound to be installed improperly, or unsecure due to admin failure
19:31:09 <falala> kc5tja: hey, my cs teacher *forces* us to write "lp" before all our pointers....
19:31:18 <slava> <imaginator> I suppose since China is the largest country I should become Chinese.
19:31:19 <falala> he marks off 5 points for each misnamed var
19:31:21 <slava> ^-- lol
19:31:29 <slava> falala, i mark your CS professor down 100 points :)
19:31:38 <falala> i'm going to be coding this way for at least 4 more years in hs
19:31:44 <kc5tja> falala: Your teacher is a moron.  Obviously sold by false facts.  Instead of reading Microsoft Press books on the subject, he should be reading Computer Science academia papers and journals.
19:31:46 <I440r> falala, tell your cs teacher he is destroying any possability you will ever be a good coder
19:31:49 <falala> how am i going to break the habit then
19:31:55 <imaginator> slava: glad you're amused :)
19:31:59 <I440r> hjeh
19:31:59 <kc5tja> falala: However, I note you still didn't answer my question.  :)
19:32:19 <falala> slava: wait, you didn't tell me what kind of PAID coding you did
19:32:41 <falala> what am i going to say... "some dudes in a forth channel on irc told me everything you (and microsoft and all the textbooks) is completely wrong"
19:32:44 <falala> heh
19:32:54 <slava> falala, most recent is serial port work (credit card verification), various business apps (gui forms talking to databases)
19:33:05 <slava> falala, what text books?
19:33:12 <falala> slava: you got paid real money for that? not, like, monopoly money?
19:33:43 <slava> falala, if you want to end up on /ignore, keep being dumb
19:34:01 <falala> jeez
19:34:05 <falala> it was a joke
19:34:10 <slava> so was mine :)
19:34:20 <falala> slava: so like, were these contracts, or what
19:34:21 <imaginator> falala: why did you venture into #Forth? You said you had mastered C++, C, PPC and x86 asm correct?
19:34:29 <slava> just stop being so damn "i'm right and you're wrong" when you're clearly quite inexperiened compared to everyone else here
19:34:33 <falala> imaginator: i'm thinking of trying to learn forth
19:34:38 <falala> imaginator: to broaden my horizons
19:34:46 <imaginator> falala: with your experience you should be able to create your own language.
19:34:52 <falala> imaginator: uhhhhhh no
19:34:55 <slava> falala, when i was 15 i was quite arrogant too
19:35:02 <slava> falala, and i made my own languages when i was 15 :-P
19:35:22 <falala> slava: how'd you get those jobs writing code for money, were those internships or what
19:35:29 --- join: snowrichard (~richard@adsl-068-209-159-248.sip.shv.bellsouth.net) joined #forth
19:35:39 <falala> slava: i doubt you wrote your own compilers at 15.....
19:35:44 <slava> falala, people saw my work online -- emailed me -- we negotiated terms -- i sent them code they sent me money
19:35:44 <falala> my TEACHER can't even do that
19:35:44 <imaginator> falala: so what does "mastered" mean to you?
19:35:48 <slava> falala, interpreters not compilres
19:35:52 <slava> falala, and i'm writing a compiler right now
19:36:00 <slava> falala, its not hard at all if you know what you're doing
19:36:07 <slava> falala, and *optimizing* compiler can be quite hard
19:36:12 <falala> imaginator: well, you know, i learn all the different commands in the language, like "switch" and "if" and stuff in C, "mov" "add" etc in asm
19:36:18 <slava> falala, that's not mastery
19:36:25 <slava> falala, to master a language, you have ot know how to code algorithms
19:36:33 <slava> falala, could you produce a quicksort without looking at a book?
19:36:36 <falala> slava: so Jedit got your footin the door?
19:36:41 <falala> slava: did you make ALOT of money doing this or what
19:36:46 <slava> falala, yes
19:37:03 <slava> now i'm getting into x86 assembly and C, and hope to get work doing embedded and such
19:37:17 <falala> slava: yes to the "alot of money"?
19:37:20 <slava> yes
19:37:27 <falala> like, >$10k?
19:37:32 <slava> yes
19:37:36 <falala> whoa
19:37:46 <falala> for like, a week's worth of work?
19:37:48 <falala> or what
19:38:04 <slava> falala, the contracts i've done were between 30 and 50 dollars an hour US
19:38:08 <slava> falala, which is not much
19:38:22 <slava> falala, compared to what some consultants get (100$/hour etc).
19:38:26 <falala> my goal is to make as much money as possible
19:38:34 <slava> falala, mine is not.
19:38:38 <slava> falala, i want to be a good programmer, foremost.
19:38:41 <falala> so forgive me if i like, disregard stuff that seems worthless
19:38:52 <slava> falala, and hint: you won't make a lot of money by listening to your CS teacher and emulating him
19:39:08 <falala> slava: well i won't make money writing LISP compilers or wahtever
19:39:14 <kc5tja> falala: Forth compilers and interpreters are trivially easy to write.  I did write my first Forth environment when I was 17 years old.  It didn't work all that great, but it did work.  :)
19:39:15 <slava> falala, if you want to make money, instead of following the current big thing (C#, windows, spyware, etc) make the NEXT big thing
19:39:23 <falala> i only want to learn forth and lisp briefly just so i get more ideas about programming, just to say i know something about em
19:39:47 <slava> falala, you won't make money writing a lisp compiler, but then you won't make money writing a C# compiler either. but you could make money making a 'killer app' that everybody wants to have
19:39:53 <falala> but my teacher says you gotta follow the latest and greatest
19:39:55 <falala> like C#
19:39:55 <kc5tja> falala: It's better if you learned them in-depth.  Learning them "briefly" will only show you how to do what you already know in those languages -- which defeats the purpose of learning them.
19:40:13 <slava> falala, C# is not the greatest like i already explained
19:40:16 <slava> falala, can i ask you something?
19:40:21 <falala> yes?
19:40:30 <I440r> falala, whats the best. THE latest and "greatest" or "tried and true"
19:40:30 <slava> have you ever done any substantial programming? not counting HS assignments
19:40:43 <falala> slava: well, i wrote a game in VB
19:40:45 <falala> and some web stuff
19:40:55 <falala> but i'm 15 and i figure i have time still
19:40:57 <slava> falala, how many lines of code in the game?
19:40:58 <imaginator> How many thousands of lines of code have you written?
19:41:07 <falala> slava: it was like a tile-based RPG kind of thing
19:41:19 <falala> slava: maybe around... hmmm.. a few thousand?
19:41:23 <slava> falala, i know lines of code is a very imprecise metric, but a lot of people in this channel have clocked millions of lines
19:41:47 <slava> and that's WITHOUT the copy and paste you usually see in VB programs :)
19:41:53 <falala> slava: so by the time you were 15, how many "millions" of lines had you written? :P
19:42:26 <slava> falala, i didn't claim to know everything there is about programming when i was 15.
19:42:43 --- quit: snowrichard ("Leaving")
19:42:56 <slava> when i was 15, i think my finished projects were a few interpreters, a GUI scientific calculator, and maybe the first jedit release
19:43:06 <slava> not much else
19:43:18 * kc5tja did make his first home-made CPU implementation when he was 13 years old though.  THAT was fun!
19:43:25 <slava> kc5tja, cool!
19:43:26 * kc5tja wishes he could invest the time to make something again.
19:43:42 <kc5tja> It was a piece-of-crap 4-bit CPU that could barely add 2+2.  :)  But it did do it.
19:43:55 <kc5tja> Had all of 128 bytes of memory (organized 256 words of 4-bits each).
19:43:56 <slava> kc5tja, 4004 clone? :)
19:44:04 <kc5tja> slava: No.  4004 would have run circles around this thing.
19:44:06 <slava> that wouldn't even run forth :)
19:44:12 <falala> my cs teacher actually didn't start writing stuff until he was a junior in college
19:44:20 <falala> and he's like 29 now
19:44:35 <slava> falala, i'm not doign CS in college
19:44:41 <falala> slava: why not?
19:44:53 <slava> falala, because i'd rather do math for my undergraduate
19:44:59 <falala> lame
19:45:04 <slava> falala, and do my masters (or if i do it) PhD in CS
19:45:04 <kc5tja> falala: I'll tell you what.  If you really want to learn to program, here is what you need to do.  Write some kind of code.  Any code.  Print it out on real, honest to goodness, dead-tree paper.  File it away for about 1.5 years.
19:45:14 <falala> slava: you sound like a professional academic
19:45:17 <kc5tja> Then, go back, and look at the code again.  Upgrade/improve the code.  
19:45:26 <slava> falala, no, i don't want to be in academia once i graduate
19:45:39 <slava> falala, however a masters degree or PhD is essential if you want non-trivial programming work
19:45:51 <falala> uh, really?
19:45:51 <slava> falala, in my opinion, anyway
19:46:07 <slava> falala, how do you think advanced algorithms work? magic? no, its all math and hard theretical CS.
19:46:12 <falala> is that because employers demand it or the skills you learn are important in themselves?
19:46:17 <slava> falala, the latter.
19:46:31 <slava> falala, suppose you're asked to code a program that simulates the flow of air aroudn the wing of an aircraft.
19:46:41 <kc5tja> Employers actually kind of prefer bachelors because they get paid less.  They do more medial work too.  Ask me.  I know.  :)
19:46:43 <slava> falala, *that* is much harder than doing GUI forms in vs.net
19:46:45 <kc5tja> And I don't even have a degree.
19:46:57 <falala> math is not my strong suit
19:46:58 <Herkamire> ok, I think I'm finished with my new website design.
19:47:02 <Herkamire> http://herkamire.com/w/services.html
19:47:06 <slava> falala, you'll need to brush up.
19:47:09 <falala> i kind of suck at math to be honest
19:47:15 <Herkamire> can somebody with windoze send me a screenshot of that page please?
19:47:31 <falala> but most of the math geniuses in school aren't even good programmers
19:47:38 <slava> Herkamire, in konqueror there's a rendering flaw
19:47:40 <falala> so i think the two must be like diametrically opposed
19:47:42 <Herkamire> BTW, I know the "Contact" link is broken
19:47:45 <slava> falala, because they're math geniuses, not good programmers
19:48:00 <Herkamire> slava: oh dear. can I see a screenshot?
19:48:02 <slava> falala, what about games -- everybody loves games. games are 50% graphics/sound, 50% math
19:48:05 <slava> Herkamire, sure hang on
19:48:05 <falala> slava: i thougth as a programmer/hacker all you needed to realy know was boolean algebra and logical induction
19:48:24 <slava> falala, ugh, graph theory, discrete math at least
19:48:29 <slava> falala, calculus, linear algebra too
19:48:31 <falala> no way
19:48:34 <slava> falala, abstract algebra
19:48:35 <slava> etc
19:48:37 <slava> they're all useful
19:48:47 <falala> i read some interview with ESR online and he was like "as hackers we rarely use hard math such as calculus"
19:48:54 <falala> ESR == Eric S raymond
19:48:55 <slava> falala, because ESR is mediocre
19:49:00 <slava> falala, what did he write?
19:49:01 <slava> fetchmail?
19:49:01 <slava> lol
19:49:04 <falala> slava: he's one of the top linux hackers
19:49:12 <falala> do you need calculus to hack the linux kernel?
19:49:14 <falala> i'm curious
19:49:17 <slava> falala, no
19:49:20 <slava> but most people don't hack the kernel
19:49:34 <falala> do you need to calculus to do embedded work"?
19:49:42 <falala> to write web databases
19:49:43 <slava> depends on what kidn of work
19:50:07 <Herkamire> you aren't likely to use calculus unless you do some serious simulations of phisics or electronics or something
19:50:08 <kc5tja> Herkamire: It looks good!  Although I'm on Linux, not Windows.  :)
19:50:13 <slava> Herkamire, factor.sf.net/foo1.png
19:50:18 <Herkamire> slava: thanks
19:50:21 <slava> Herkamire, its near the bottom of the window
19:50:27 <kc5tja> slava: If you ask madwork, he'll say it's a lot more than 50% math.  :)
19:50:28 <slava> Herkamire, or graphics of course
19:50:30 <Herkamire> kc5tja: cool :)
19:50:41 <falala> generally speaking hard math is only a prerequisite when the work involves some kind of scientific application in which the code is a means and not an end in itself
19:50:51 <slava> falala, is the code ever an end in itself?
19:50:57 <slava> falala, isn't that 'masturbatory programming'?
19:51:05 <falala> slava: i think if you look at things like the linux kernel... in a way, yes
19:51:06 <slava> falala, your code must always do something useful otherwise its useless, cleraly.
19:51:26 <slava> falala, yes but the linux kernel still involves math.
19:51:30 <falala> my cs teacher says for most things math is useless, and we shouldn't worry about it
19:51:31 <slava> falala, eg, you know the O(1) scheduler?
19:51:34 <slava> falala, how do they prove its O(1)?
19:51:40 <falala> like if you go for the MCSE, you won't need calculus
19:51:45 <slava> falala, here's news
19:51:49 <slava> MCSE will get you a shitty job
19:51:50 <slava> no more
19:51:58 <falala> dude you're a moron!! MCSE's make the big bucks
19:52:00 <slava> nope
19:52:04 <kc5tja> falala: ESR is a decent programmer and technician, and I certainly don't claim to be an expert at anything myself.  But one thing is for sure, until you start doing fluid dynamics math in computers, and making them go *fast*, then you're unwashed.  :)  (BTW, so am I, although I did do a pseudo-simulation of the Wankel rotary engine once)
19:52:06 <falala> where have you been living
19:52:15 <kc5tja> falala: Where I work, we don't hire MCSEs.
19:52:18 <falala> if you keep going for PhDs and MS's in super fancy math and algorithms you're not going to make jack!
19:52:25 <slava> falala, actually, you're wrong
19:52:41 <falala> employers don't want fancy algorithms that can iron their cat....
19:52:49 <slava> falala, depends on the employer.
19:52:59 <slava> falala, if all you want to be is a web monkey, you have the right attitude
19:53:07 <kc5tja> falala: We've found, repeatedly, that employees who have MCSEs have ego issues and therefore are incompatible with the remainder fo the workforce here.  Moreover, we also find that their solutions to problems work only temporarily; they are never permanent solutions.
19:53:21 <slava> kc5tja, i guess MCSE gives them a false sense of security
19:53:22 <falala> wtf....
19:53:27 <slava> kc5tja, like "yeah i know this, i'm MCSE..."
19:53:32 <kc5tja> slava: Yeah, exactly.
19:53:36 <falala> my teacher says MCSE is the top of the food chain among programming
19:53:39 <kc5tja> They have a ton of book knowledge, but no *practical* knowledge.
19:53:53 <slava> falala, if you say one more like that begins with 'my CS teacher' i'll start ignoring you
19:53:58 <kc5tja> Your teacher has been proven wrong on this very channel no less than four times that I recalled.  This will be the fifth.
19:54:01 <slava> falala, you're well aware i'm not interested in your teacher's opinions
19:54:02 <falala> he's like, "as soon as you can get MCSE and go into the workplace and make tons of money"
19:54:06 <slava> falala, he is wrong
19:54:16 <kc5tja> falala: Sorry, that's patently wrong.
19:54:18 <falala> slava: i think you should talk to him face to face and you'd see how wrong you are
19:54:31 <falala> maybe i could even get him into this irc chan... hmmmm
19:54:37 <kc5tja> You can get a lot of money if (a) you're hired at a MS-certified work place, and (b) your manager is really benign.
19:54:43 <slava> falala, that would be quite entertaining, please try to get him here
19:54:54 <falala> actually
19:54:54 <kc5tja> I get paid $10/hour to do Linux technical support -- a job *no* MCSE could ever do with their level of training.
19:54:58 <falala> maybe you can look at his website
19:55:01 <kc5tja> BTW, I also support Windows boxes too.
19:55:17 <falala> let me msg you the webiste slava
19:55:19 <falala> k?
19:55:21 <kc5tja> We are a CIS-certified data center facility too.
19:55:22 <slava> why not paste it here?
19:55:32 <falala> i think this channel is logged and i don't want any trouble
19:55:54 <falala> kc5tja: $10/hour?
19:55:58 <kc5tja> Yep.  That's it.
19:56:08 <kc5tja> THe IT industry is *severely impacted*.
19:56:10 <falala> kc5tja: wait, you're one of these fancy programmers in here and that's all you're making?
19:56:15 <kc5tja> Yep.
19:56:17 <kc5tja> Get used to it.
19:56:21 <slava> falala, don't insult people based on income
19:56:29 <falala> uh, slava over there just said he's raking in the big bucks
19:56:30 <kc5tja> I don't consider it an insult.
19:56:32 <slava> falala, i have friends that live in public housing projects
19:56:37 <kc5tja> The truth hurts sometimes.
19:57:11 <kc5tja> I should easily be making a good $16/hr at the very least for what I'm doing.  But alas, nobody else was hiring at the time I got the job.  Some people are moving to greener pastures now, but they aren't reporting everything is all rosy either.
19:57:23 <falala> slava said he is making $50/hour
19:57:23 <kc5tja> Unless you have that degree, you get *nothing* for pay.
19:57:28 <falala> and he even said that's *low* for him
19:57:32 <slava> falala, contracts are different
19:57:34 <slava> falala, you work for a few hours
19:57:37 <slava> falala, and you're done
19:57:38 <slava> falala, its not 9-5
19:58:02 <falala> kc5tja: so you recommend going to college and getting that degree?
19:58:13 <kc5tja> Strongly.
19:58:25 * kc5tja is himself saving up to go back to school for a compsci degree.
19:58:26 <falala> even when your'e a l33t hax0r and you know it
19:58:31 <slava> falala, except you're not
19:58:36 <kc5tja> Even a bachelors will get me into jobs that I just have no chance in competing in right now.
19:58:43 * kc5tja has no certs, no degrees at the moment.
19:58:46 <kc5tja> And I'm 29 years old.
19:58:55 <falala> kc5tja: despite the fact that you've taught yourself everything you would learn at a compsci class...?
19:59:03 <kc5tja> Half my working lifetime is spent without certs/degrees, and I am below poverty line.  It's not a coincidence.  :)
19:59:09 <kc5tja> Yep.
19:59:23 <falala> so , why?
19:59:28 <falala> i mean, it's not like you don't know the stuff
19:59:33 <falala> does the piece of paper really mean all that much
19:59:42 <slava> falala, its a filter
19:59:46 <slava> falala, they look at 50 resumes
19:59:54 <kc5tja> It's not what you know, it's what you're certified to know.  MCSE is a nice certification to have, but it isn't everything.  There's also Cisco certs, Novell certs, and of course, college degrees.  The more you have, the more you learn, the better off you are.
19:59:55 <slava> falala, the ones without degrees or <tech X> experience get thrown away
19:59:59 <kc5tja> Oh, and the Linux certs too.
20:00:26 <kc5tja> Linux is growing like a fungus in the IT industry.
20:00:41 * kc5tja also wants to get experience with IBM mainframes too.  I hear those are great to have on the resume.
20:00:43 <fridge> MS says it grows more like cancer
20:00:44 <falala> so why is it that i go to like guys' websites and look at their resume and i see like "1996-1999 Sr. Server Programmer, D&E SHaw" etc
20:00:48 <kc5tja> Anyway, I have to go check up on a server.  I'll be right back.
20:00:52 <falala> and he was 18 in 1996
20:00:58 <fridge> than fungus
20:01:05 <kc5tja> fridge: Not my problem.  :)
20:01:13 <slava> falala, beacuse he got hired out of sheer luck
20:01:41 <falala> and what about like John Carmack
20:01:46 <falala> he dropped out of college his freshman year
20:02:03 <slava> just because some smart guy dropped out doesn't mean its smart to drop out
20:02:08 <slava> he didn't get smart by dropping out
20:02:27 <falala> i think the rules don't apply for l33t hax0rs like us
20:03:17 <falala> look at JWZ (Jamie Zawinski) he doesn't have any degrees, and he made millions at Netscape
20:03:26 <falala> he's one of the best hackers around
20:03:31 <slava> falala, there's only a handful of people in that situation
20:03:39 <slava> falala, is there hordes of college dropouts becoming millionares?
20:03:47 <slava> falala, most do to mcdonalds etc
20:04:13 <falala> like i said, rules don't apply to special people
20:04:26 <slava> what makes you think anybody here (you included) is "special"?
20:04:34 <falala> why not?
20:04:43 <falala> you can code as well as John Carmack, can't you
20:04:48 <falala> from the way you speak, i'd assume that
20:04:50 <slava> i never claimed that
20:05:06 <slava> so you never gave me that CS teacher's home page
20:05:24 --- quit: warpzero (Client Quit)
20:05:43 <falala> anyway, my plan is, i'm going to drop out of high school when i'm like 17, maybe, and get a job working for some local MCSE shop
20:05:47 <falala> and make loads of money
20:05:51 <slava> hahaha
20:06:00 <slava> your plan just won't work
20:06:06 <slava> what kind of job are you hoping to get?
20:06:12 <falala> i'm goin to be Director of IT
20:06:17 <slava> doing what?
20:06:24 <falala> whatever they do :P
20:06:31 <slava> what will make them hire you?
20:06:35 <slava> and not somebody with a degree and experience?
20:06:55 <falala> slava: because i'm l33t
20:07:04 <slava> falala, how will they know?
20:07:14 <falala> they can tell by the way i talk
20:07:21 <falala> and the penny arcade tshirts i wear
20:07:28 <falala> :P
20:07:59 <slava> falala, first, note that your teacher has a masters degree in math *AND* education
20:08:03 <slava> falala, so he's not a HS dropout
20:08:16 <falala> lol
20:08:21 <falala> i didn't know about that
20:08:30 <falala> it's probably just on their to pad the resume
20:08:31 <slava> i don't see any programming on his resume at all
20:08:33 <falala> he probably made it up
20:08:41 <slava> that's illegal and people *do* check
20:08:58 <slava> all i see is this:
20:09:04 <slava> "Maintained Pascal program and dBase III database, which were used to test and track college freshmen for library research aptitude. "
20:09:07 <slava> did he make that up too? :)
20:09:13 <Herkamire> what's MCSE?
20:09:21 <slava> Herkamire, it teaches you how to reboot windows boxes
20:09:23 <imaginator> Microsoft Certified Snot Engineer
20:09:30 <kc5tja> Herkamire: Microsoft Certified Systems Engineer
20:09:43 <Herkamire> hell
20:09:44 <kc5tja> There is also MCSD -- Microsoft Certified Software Developer too.
20:09:46 <slava> falala, http://www.minich.com/portfolio/mathmsthesis.pdf
20:10:12 <Herkamire> I was afraid the M stood for Micro$oft
20:10:23 <slava> falala, that's his master's thesis.
20:10:42 <slava> falala, so...
20:10:46 <slava> your CS teacher is a snot-nosed academic
20:10:51 <slava> who has never done programming in the industry
20:11:02 <slava> except for some dbase III one liners on a green screen PC in the 1840's
20:12:21 <slava> in fact, he seems to specialize in web design
20:12:26 <slava> wait i have a theory
20:12:30 <slava> falala IS the cs teacher
20:12:31 <slava> lol
20:13:18 <kc5tja> I won't go so far as to partake in ad hominem attacks against either the CS teacher or falala -- but the evidence does speak heavily against CS teacher's credentials for coding expertise.
20:15:04 --- join: warpzero (~warpzero@dsl.103.mt.onewest.net) joined #forth
20:15:08 <slava> hi warpzero 
20:15:47 <Herkamire> I bet dropping out of highschool would make it harder to get any technical computer job
20:15:58 <warpzero> hey hey
20:17:05 <slava> warpzero, what's new?
20:17:07 <warpzero> it doesn't
20:17:23 <warpzero> not much, you
20:17:29 <Herkamire> warpzero: really?
20:17:34 <warpzero> really
20:17:39 <kc5tja> warpzero: Since I never dropped out of highschool, I can't tell.
20:17:44 <warpzero> well i did
20:18:03 <kc5tja> But I did drop out of college four times because I couldn't afford it each time.  And I can tell you, it's all but impossible for someone to hire me.  Without a degree, I'm virtually unhirable.
20:18:08 <Herkamire> I asumed they would question 1) your intelligence 2) your tendancy to stick to things
20:18:12 * kc5tja was damn lucky to find my current source of employment.
20:18:22 <kc5tja> Anyway, I'll be right back again -- I ahve to perform facilities monitoring.
20:18:23 <warpzero> Herkamire: well im at college so
20:18:35 <warpzero> Herkamire: they don't really care
20:19:00 <Herkamire> warpzero: oh. skipping ahead isn't so bad
20:19:35 <warpzero> yeah
20:19:42 <warpzero> i just couldn't take highschool anymore
20:19:42 <Herkamire> warpzero: falala was considering dropping out of highschool at age 17 to become a MCSE
20:19:45 <warpzero> it was horrible
20:21:03 <Herkamire> and while I suspect that there is a certain amout of "hire anybody who is willing to do it", I suspect that employers would choose people who didn't bail on school before the end of highschool
20:21:23 <warpzero> okay so do you guys think name[type]value is better or maybe something like name:type=value
20:21:34 <slava> type name = value
20:21:41 <warpzero> that doesn't work
20:21:53 <warpzero> whitespace is list delimeter
20:23:32 <slava> is anybody familiar with SSA form?
20:30:30 <kc5tja> back
20:31:20 <kc5tja> slava: Sort of.
20:31:35 <slava> kc5tja, i wonder if it can be applied to a concatenative language
20:32:16 <kc5tja> It can, and if you're thrifty enough, you can even optimize out all the SWAPs DUPs and whatnot.
20:32:31 <kc5tja> The problem is, you need to know how many arguments a word takes, and what it produces.
20:32:41 <slava> that's what i'm interested in. not as much optimizng out SWAP and DUPs, but redundant type checks.
20:32:48 <kc5tja> Knowing that, you can almost directly convert the concatenative format to a tree, and from there, to SSA.
20:35:32 <slava> interesting.
20:36:17 --- join: saon (~Ecoder@c-24-129-95-254.se.client2.attbi.com) joined #forth
20:36:26 <slava> hi saon
20:36:42 <saon> hello
20:37:11 <falala> anyway
20:37:30 <falala> if you want to make money, you gotta learn C#
20:37:32 <falala> bottom line
20:37:48 <falala> look at kc5tja , he's a l33t forth coder, lisp guru or whatever, but he said it himself, he's living on the poverty line
20:37:54 <kc5tja> falala: Umm...no...sorry.  Sure, C# is a nice language to know, but it is far from the bottom line.
20:37:56 <falala> get an MCSE, and you're golden!
20:37:56 <slava> falala, if all i was interested in was money, i'd be a drug dealer
20:38:03 <slava> falala, an MCSE isnt even about programming
20:38:05 <slava> falala, its sysadmin stuff
20:38:08 <falala> slava: yeah right, that's a dangerous job
20:38:16 <falala> slava: you couldnt' be a drug dealer, you're a nerd
20:38:25 <falala> you need the right contacts, you need to avoid being arrested/shot at
20:38:26 <falala> etc
20:38:29 <kc5tja> falala: Behave yourself, or you WILL be kick-banned.
20:38:36 <slava> falala, i'll just take tips from you
20:38:39 <falala> kc5tja: chill dude
20:38:48 <falala> slava: ??
20:38:53 <falala> slava: seriously, do you think you could be a drug dealer?
20:38:54 <slava> falala, on the drug dealing
20:38:56 <slava> falala, no
20:39:02 <slava> falala, i was just making a point
20:39:08 <falala> slava: if money *was* your main prerogative
20:39:11 <kc5tja> The point is, C# is a neat language.  As far as I can see, it is exactly like Java.  It was introduced to compete with Java, and yet, Java still rules the Internet coupe.  And it always will.
20:39:28 <falala> kc5tja: you're a MORON!!
20:39:33 <falala> kc5tja: C# is completelyu different
20:39:34 <slava> falala, be polite.
20:39:35 <falala> it's revolutionary
20:39:39 <slava> falala, how so?
20:39:40 <kc5tja> falala: Prove it to me.
20:39:41 <falala> look at any Microsoft Press clippings
20:39:45 <falala> they even SAY SO
20:39:46 <falala> RIGHT THERE
20:39:47 <falala> IN PRINT
20:39:50 <slava> falala, name some revolutionary C# features
20:39:58 <falala> Bill Gates talkks about the revolutionary C# features in his KEYNOTE SPEECHES
20:40:09 <slava> falala, can you answer my quesiton please?
20:40:09 <kc5tja> falala: Name me ONE non-Microsoft press clipping that supports the "revolutionary" claims of C#.
20:40:15 <falala> szLinePtr = REVOLUTIONARY;
20:40:26 --- part: falala left #forth
20:40:27 <kc5tja> falala: Don't you think it odd that you're only seeing Microsoft's side?
20:40:29 <slava> lol
20:40:43 --- mode: kc5tja set +b falala*!*@*
20:40:46 <kc5tja> End of discussion.
20:40:47 <slava> lol
20:41:11 <slava> i bet he'll be in here tomorrow with another nick
20:41:18 <kc5tja> Probably.
20:41:23 <kc5tja> If so, I'll ban his entire domain.
20:41:36 <slava> in efnet #java they ban aol
20:41:37 <kc5tja> I don't care.
20:42:07 <kc5tja> Well, I suppose I could ban all of att.net -- that's where he's coming from.
20:42:27 <slava> i notice our friend alonzo is still in the ban list ;)
20:42:39 <kc5tja> And he always will be, as long as I'm chanop here.
20:43:02 <kc5tja> If I find out it's no longer here, and that someone here removed him, and it's not I440r (since this is technically his channel), I'll ban them too.
20:43:16 <slava> I440r, this is your channel?
20:43:21 <slava> I440r, when did you start it?
20:44:28 <Herkamire> oh my god
20:44:35 <slava> yes?
20:44:39 <I440r> i dont own it now
20:44:46 <I440r> everyoen here owns it and futhin manages it
20:44:50 <Herkamire> just reading some of what falala said
20:44:51 <slava> the channel 0wnz j00!
20:44:58 <slava> Herkamire, not the sharpest knife in the draw?
20:45:08 <Herkamire> I thought everybody know that gates was full of #*$#"
20:45:23 <I440r> i started it like a year before the first release of isforth
20:45:54 * kc5tja never, ever thought he'd have to ban people from this channel.
20:46:04 <kc5tja> I never had to before.
20:46:17 <kc5tja> And as I440r knows, when things start getting irrational, I usually just leave.
20:46:21 <kc5tja> But this is bullshit.
20:46:43 <I440r> kc5 is not one to kick/ban ppl without cause
20:48:30 <Herkamire> that was rediculous
20:49:14 <Herkamire> yelling, calling kc5tja a moron because he doesn't believe MS's media hype
20:49:27 <I440r> lol
20:49:28 <kc5tja> I can't help but somehow feel dirty.
20:49:33 <I440r> i missed the rukus ?
20:49:38 <I440r> whow won ?
20:49:51 <kc5tja> falala thinks he won because he left #forth first.
20:50:03 <I440r> i said i thunked he was a troll
20:50:07 <I440r> even if he is only 15
20:50:18 <kc5tja> After he left, I promptly banned him from ever coming back.
20:50:23 <Herkamire> slava: I like your line "falala, if all i was interested in was money, i'd be a drug dealer"
20:50:26 <kc5tja> At least with the nick of falala.
20:50:37 <I440r> :)
20:50:53 <kc5tja> If you want to make money, offer Internet hosting solutions to porn providers.
20:50:55 <kc5tja> Seriously.
20:51:02 <kc5tja> You make more money that way than selling drugs.
20:51:34 <Herkamire> yeah, just gotta find a colo facility that allows it
20:51:49 <I440r> yup
20:51:51 <Herkamire> they'll use your bandwidth though
20:52:06 <kc5tja> Herkamire: Precisely why colo facilities love pornsters.  They pay through the nose for the bandwidth.
20:52:27 <kc5tja> We actually host about 15 to 20 ISPs that in turn host porn providers.
20:52:54 <kc5tja> Not a one of them has an ugly car, that's for sure.  :)
20:56:24 <Herkamire> weird...
20:56:36 <Herkamire> just looked at my bandwidth usage graphs for the first time in quite a while
20:56:42 <slava> hey guys i got to the bottom of the falala mystery
20:57:12 <slava> i portscanned him and found a huge number of unpatched services on his windows box -- i threated to hack him in /query, and he confessed to being a troll:
20:57:32 <kc5tja> slava: Hahah!
20:57:33 <slava> <fala> ok, look, i was just trolling you
20:57:33 <slava> <fala> honestly i don't mean to insult you
20:57:33 <slava> <slava> i'm not insulted
20:57:33 <slava> <fala> please don't hack me
20:57:45 <slava> <fala> no, i'm just a troll, i don't even go to HS
20:58:24 <kc5tja> It's good to have a "troll" on *our* side.  >:D
20:58:34 <kc5tja> (and that's meant in a good way)
21:00:20 <slava> no, he's just a dumbass trying to piss people off. we get a lot of those in #java
21:02:02 <kc5tja> Ahh
21:02:14 <kc5tja> Well, he really didn't piss a whole lot of people off here.
21:02:14 <kc5tja> :)
21:02:28 <kc5tja> I rather politely dealt with him -- I'm proud of myself.
21:02:34 <kc5tja> Now if only I can do things that way in real life... :)
21:02:46 <Herkamire> kc5tja: :)
21:04:16 --- quit: saon ("Leaving")
21:11:17 <imaginator> That was so strange.
21:11:40 <kc5tja> imaginator: Yep.
21:11:43 <imaginator> I suspected he was a troll, and someone from here that was playing a game.
21:14:45 <Herkamire> weird... looks like the bandwidth to my server dropped about a month ago from about 80Kbps to about 7Kbps.
21:15:11 <Herkamire> it's stayed down for three weeks
21:15:33 <Herkamire> and it looks like it was up around 60-80 for at least 8 months
21:16:24 <Herkamire> I really should get around to installing some sort of apache log analizer thing
21:16:45 <kc5tja> Heheh :D
21:17:11 <Herkamire> it dropped in the middle of my vacation too
21:17:24 <slava> DOS maybe?
21:17:28 <Herkamire> I still have all my customers afaik
21:17:34 <Herkamire> slava: not on me.
21:17:47 <Herkamire> perhaps I have a virus that's helping pull one on others
21:18:06 <slava> north korea tests nuke
21:18:11 <slava> http://www.reuters.com/newsArticle.jhtml?type=topNews&storyID=6211081
21:21:46 <Herkamire> I'm starting to think that the kind of programmer I aspire to be, is not terribly well suited to working in the business world
21:21:56 <slava> why not?
21:21:58 <Herkamire> and good programming in general
21:22:19 <slava> most programming problems are simple because the tools to solve them already exist, and this covers most business programming
21:22:36 <slava> but this doesn't mean its bad to do
21:22:39 <Herkamire> you're such a java head
21:22:47 <Herkamire> sorry
21:23:01 <Herkamire> thing is that the existing tools generally aren't perfectly suited to the task
21:23:03 <slava> there's nothing *wrong* with the fact that its easy to make an online shopping cart, etc.
21:23:23 <Herkamire> what's "wrong" is all the crappy crappy carts out there
21:23:52 <Herkamire> I think the path to making really good software often involves starting over and doing it right
21:24:12 <slava> the thing is, for a shopping cart or whatever, it doesn't matter of it uses 10mb of memory or 10kb, or how elegant the code is
21:24:16 <kc5tja> I HATE STORESENSE!!!!
21:24:20 <kc5tja> Oh, I'm sorry, did I say that out load?
21:24:30 <Herkamire> roger that
21:24:40 <kc5tja> I hate StoreSense and the Java horse that it rode in on.  It's been nothing, let me repeat this -- NOTHING, but trouble for us.
21:24:42 <imaginator> If it uses less memory, it probably is simpler, and that probably means it has fewer bugs.
21:25:21 <kc5tja> imaginator: Not necessarily true.  I have nothing against garbage collected systems, which usually consume about 2x the amount of memory during a typical run than a non-GC program of the same kind.
21:25:30 <kc5tja> But if it uses less *code*...I'll agree to that.
21:25:32 <slava> imaginator, how many bugs can there really be in a shopping cart?
21:25:37 <imaginator> too many
21:25:41 <slava> we're not talking about a space probe here
21:25:41 <Herkamire> slava: many
21:25:57 <kc5tja> slava: Enough to offer our customers non-stop issues.  *sigh*
21:25:59 <Herkamire> slava: memory and CPU are not the only costs of crappy code
21:26:07 <imaginator> User clicks back, the browser doesn't handle some HTML properly, the browser sends crap or non-standard crap.
21:26:46 <kc5tja> User clicks on items to purchase, gets proper shipping charges on the screen, clicks "Order Now!", order comes through, and only 1/3rd the required shipping is actually billed to the customer.
21:26:48 <slava> imaginator, the issues with HTML generation and HTTP request handling are taken care of by the application server, not the shopping cart.
21:27:17 <kc5tja> But only for ONE customer, and therefore, is next to impossible to track down why it's doing it only for him, and not anyone else.
21:27:22 <Herkamire> slava: great, now you have two huge buggy programs to deal with
21:27:42 --- join: crc (crc@175-pool1.ras11.nynyc-t.alerondial.net) joined #forth
21:28:02 <slava> Herkamire, would you rather everybody duplicated the same functionality over and over again?
21:28:24 <imaginator> Then add cookies into the mix.  And people that turn them off.
21:28:26 <slava> code reuse is a *good* thing, especially for the business apps we're discussing here.
21:28:50 <Herkamire> slava: it's good to reuse code for what it's good for
21:29:05 <Herkamire> managers tend to tell you to keep using code you have instead of write something similar
21:29:45 <slava> here comes the mythical manager, the bugbear of all programmers world-wide!
21:30:05 <Herkamire> sometimes it's clear that in the long run it will be easier to re-implement something, or change the way you do something, but there isn't time in the schedule to change things around, so it never happens
21:30:12 <kc5tja> Hey, if it is anything like the managers I've had to deal with, they are.  :)
21:30:16 <Herkamire> the stuff you're reusing just gets slowly more and more complex
21:30:22 <Herkamire> impossible to maintain
21:30:30 <slava> the fact is, if you're doing web apps or whatever in a business, they're not paying you to code for the sake of coding -- they're paying you to implement a solutoin which will directly lead to the business's profitability (or lack thereof)
21:30:46 <Herkamire> slava: the manager isn't mythical. I've had one for 3 years
21:30:53 <slava> if your manager is incompetent, writing better tools won't help
21:31:01 <kc5tja> slava: The problem is, StoreSense is coding a finished, commercial, OTS product.
21:31:18 <kc5tja> That is, if anything, one of the key areas where a fully customized solution is required for maximum quality control.
21:31:23 <slava> kc5tja, i was never defending storesense
21:31:25 <Herkamire> my manager isn't incompetant
21:31:36 <Herkamire> he runs a successful web development company
21:32:01 <Herkamire> I'm just saying that business invirenmonts don't lend themselves to producing excelent software
21:32:02 <slava> well, its sucessful in part because of the programming at least, right? so its not all wasted
21:32:15 <slava> sometimes, "good enough" is all that's needed
21:33:18 <I440r> actually, thinking about it, that guy knew ALL the wrong answers
21:33:33 <Herkamire> I've programmed for them for about three years, and tathi for about two. I learned a lot from it, but I don't think there's very little code we created there that I care to keep.
21:33:41 <Herkamire> maybe a couple KB.
21:33:51 <I440r> herk for what?
21:33:56 <slava> the thing is, it doesn't matter if your shopping cart is 700 lines of beautiful Haskell or 10000 lines of ugly java, as long as it *works*
21:34:44 <imaginator> slava: until you have to fix a bug in it.  The more lines of code, the more difficult it becomes.
21:34:49 <Herkamire> I440r: we had some neat code to handle passing requests to another script and for displaying with a sort of template system so pages look allike
21:35:09 <Herkamire> slava: that's not true.
21:35:16 <slava> imaginator, its possible to have a code base that's both large, and well debugged
21:35:19 <I440r> :)
21:35:23 <Herkamire> slava: next week the client complains about some little thing and you have to fix it
21:35:34 <Herkamire> slava: no it's not
21:35:43 <Herkamire> :)
21:35:52 <slava> do you find the linux kernel unbearably buggy?
21:36:04 <Herkamire> there is no bug-free 10,000 line java program. no way in hell
21:36:13 <Herkamire> my computer crashed last week
21:36:21 <Herkamire> I was Pissed
21:36:26 <imaginator> Code can only be certified as perfect if it provides the proper result/effect with ALL possible inputs, without corrupting outside states.
21:36:49 <slava> and again, there's the distinction between 'bug free' and 'good enough'
21:36:58 <slava> NO program is bug free unless its trivial
21:37:02 <kc5tja> imaginator: OR, if it can be proven to be correct for all possible inputs, hence the appeal of formal design methods.
21:37:33 <Herkamire> slava: I don't understand what you're getting at
21:37:54 <slava> that 'perfect code' is an unpractical ideal in a business environment
21:37:54 <Herkamire> in my experience working for a web development company, most of the work was maintaining code.
21:38:01 <Herkamire> adding little things, and fixing little things
21:38:09 <Herkamire> slava: exactly
21:38:14 <slava> and how do you propose to avoid this?
21:38:23 <slava> code is never a done work
21:38:29 <Herkamire> I'm not.
21:38:31 <kc5tja> slava: If all programs can be broken into smaller units that are trivial in and of themselves, then by definition, a large program can be bug-free too.
21:38:40 <kc5tja> That's the principle of unit testing, and the principle of formal methods.
21:38:41 <slava> kc5tja, you neglect interactions between the units
21:38:50 <Herkamire> I'm saying the value of good code (especially in business) is that it's easy to maintain
21:38:54 <Herkamire> and change
21:38:55 <kc5tja> slava: Again, that's where unit testing and/or FMs come into play.
21:39:30 <warpzero> holy fuck
21:39:30 <slava> Herkamire, yes, but its a case of diminishing returns after a while
21:39:39 <warpzero> guys check out the news about north korea
21:39:41 <Herkamire> slava: what is?
21:39:44 <slava> Herkamire, in a business environment, you're not expected to spend 3 weeks perfecting 25 lines of code to the fullest
21:40:00 <Herkamire> slava: yep. that's sorta my point
21:40:11 <Herkamire> I want to create a great OS
21:40:24 <Herkamire> but I figure it'll probably take 10 years
21:40:31 <Herkamire> to do it right and make it really great
21:40:38 <Herkamire> that's too long for the business world
21:40:50 <slava> then work jobs in the meantime. you'll find it easy.
21:41:04 <Herkamire> they need to see a return in a year or two or have it be huge returns and a guarantee
21:41:13 <Herkamire> that's what I'm doing
21:41:13 <imaginator> warpzero: what about it?  Has it reached google's news?
21:41:36 <slava> Herkamire, so what's the problem? :)
21:42:00 <Herkamire> no problem.
21:42:14 <warpzero> imaginator: well North Korea might have just done a yield test
21:42:15 <slava> you sound like you believe business programming simply shouldn't exist
21:42:18 <warpzero> imaginator: yahoo has it
21:42:18 <Herkamire> I'm just saying I'm slowly coming to the opinion that the business world is not terribly conducive to writing good code
21:42:35 <slava> Herkamire, yes, its a known fact that this is true
21:43:08 <slava> and i don't think clean code should be the #1 priority of a business
21:43:21 <Herkamire> I do
21:43:26 <slava> otherwise they'd spend 3 years perfecting every little component, and go bankrupt in the meantime because they have nothing to offer
21:43:29 <Herkamire> mmmm... actually maybe not
21:43:44 <Herkamire> quick ugly hacks tend to make more money
21:43:50 <Herkamire> you need to be cheap up front
21:44:02 <slava> a quick ugly hack can be refactored over time.
21:44:21 <Herkamire> and if it get's logerithmically harder to work on then that's fine, because you work hourly.
21:44:55 <Herkamire> and the more money you get out of them, the less they will even think of having somebody do it over, because they are painfully aware of how much money they've poured into it already.
21:45:25 <Herkamire> the client doesn't want to pay a bunch of money for you to do it over. they figured they already payed for it.
21:45:50 <Herkamire> your bos doesn't want you to spend all that time doing it over when he get's lots of money from any little change because it takes you too long
21:46:06 <Herkamire> if you mean for money then I agree.
21:46:14 <Herkamire> if you mean ethically, then I dissagree.
21:46:25 <Herkamire> ethically, clean code should be your #1 priority
21:46:40 <imaginator> Well engineered software keeps customers for the long term doesn't it?
21:46:45 <slava> not to the point of being anal-retentive about it.
21:47:17 <slava> if I spend 10 years designing a web frontend to a local gift card store, then I'm not performing my ethical responsibility
21:47:20 <Herkamire> slava: I didn't say you have to be anal-retentive about it.
21:47:33 <imaginator> slava: why would it take 10 years?
21:47:41 <slava> for it to be 'perfect'
21:47:49 <kc5tja> imaginator: IBM still has mainframe customers from 30 years ago.  :)
21:47:49 <Herkamire> imaginator: not really. that's the thing.
21:47:51 <slava> when i could spend a week or two cooking up some simple .jsps
21:48:11 <slava> even though they might not be perfectly factored etc
21:48:48 <Herkamire> if you make really good software that works perfectly, does everything people want in that domain... then you probably won't do so well
21:49:04 <Herkamire> you need to sell upgrades...
21:49:15 <slava> that's not the issue
21:49:38 <slava> the thing is, if somebody using 'uncool' tools like prebuilt libraries, Java, etc can do a 'good enough' equivalent in 30% of the time, he'll get the job
21:49:39 <imaginator> There are dozens of tools for the web that do the same thing.  People reinventing database accessors, and databases, and protocols, and ...
21:49:41 <Herkamire> you need to sell services
21:50:22 <imaginator> If there was one perfect tool for each solution it would be easier to move forward, but now we have different language bindings, different operating systems ...
21:51:51 <imaginator> And the motivation is to outdo the next guy, and not to improve societies access to information.
21:53:03 <imaginator> But now as it is, we have debates over the design of an API, the protocol it uses, and on and on.
21:53:25 <imaginator> Maybe there just isn't a perfect solution.  :/
21:53:50 <Herkamire> maybe the people that do good research don't code it
22:01:18 <imaginator> slava: does your Forth-like language use garbage collection?
22:01:26 <slava> imaginator, yes
22:02:00 <imaginator> For everything or just strings? 
22:02:15 <slava> for all heap allocated objects, which includes strings.
22:03:09 <imaginator> I've decided my idea of having 'foo'  push each character onto the %esp stack with a length doesn't make life easier, so I'm thinking about writing a garbage collector.
22:03:25 <imaginator> Did you use any books to implement your GC?  Do you use the Boehm-GC?
22:03:37 <slava> i implemented a traditional copying collector.
22:03:48 <slava> using online resources, and bugging #lisp :)
22:04:15 <imaginator> heh
22:04:23 <slava> its quite simple, 200 liens of C or so
22:04:34 <imaginator> I wrote this http://www.xmission.com/~georgeps/engineering/prototype/garbage_collector-2.c a while ago, but moving beyond that has been difficult.
22:04:43 <imaginator> wow, just 200?
22:04:53 <slava> yes
22:05:07 <slava> yours looks like a reference counter?
22:05:29 <imaginator> All variables that use an object need to specifically state that they use it.
22:05:33 <slava> i think copying gc is actually simpler...
22:05:44 <imaginator> What's the theory behind it?
22:05:56 <slava> ok, the fundamental operation is 'allot'
22:06:03 <slava> it increments a pointer in the heap, by the requested size
22:06:07 <Herkamire> one problem with reusing code, is it takes a long time to figure out how to use it. and often you later find out that it doesn't do what you need.
22:06:11 <slava> eventually, the pointer reaches the end of the heap
22:06:13 <slava> then you collect:
22:06:26 <Herkamire> so you really should spend a long time researching to find what to use.
22:06:35 <slava> first, you swap the 'current' heap with an unused one the same size
22:06:39 <Herkamire> I figure if I'm going to spend a long time, I'd rather use that time to write it myself
22:06:40 <slava> (so yes, this has 2x memory overhead)
22:06:47 <slava> then, you copy the root objects from the old heap to the new heap
22:06:59 <slava> this includes anything referenced by stacks, global variables, and so on
22:07:13 <slava> so, your new heap looks like this so far:
22:07:22 <slava> | roots    | empty space                 |
22:07:26 <slava>         ^ 'here' pointer
22:07:32 <slava>            ^ 'here' pointer
22:07:34 <slava> :)
22:07:41 <slava> now, you set a 'scan' pointer to 0
22:07:46 <slava> and, while scan < here
22:08:02 <slava>   copy all objects referenced by object in newheap at 'scan'
22:08:08 <slava>   increment 'scan' by size of object at 'here'
22:08:34 <slava> so 'here' grows and 'scan' chases it, until all objects reachable from the roots have been copied from the old heap to the new heap
22:08:40 <slava> now there is one implementation detail
22:08:53 <slava> if an object is referenced more than once, you have to make sure you don't copy it more than once
22:09:06 <slava> the first tiem you copy an object, you overwrite it in the old heap with a 'forwarding pointer' that is tagged as such
22:09:24 <slava> subsequent times this object is referenced, you don't copy it, you simply fixup the reference with the forwading pointer's address
22:11:33 <imaginator> This assumes that the heap is linear right?  What if my heap is from the system's malloc, and may be spread out?  Would I need to implement a private heap and then realloc as needed?
22:12:07 <slava> imaginator, yes, it implies managing your own heap
22:12:33 <slava> imaginator, in factor, i just mmap two 'large enough' private mappings, and at any time one of them is in use
22:12:47 <slava> here is the source http://cvs.sourceforge.net/viewcvs.py/factor/Factor/Factor/native/gc.c?rev=1.18&view=auto
22:12:51 <imaginator> My forth isn't traditional.  Everything gets compiled into an IC and then run (by jmp'ing into it), and next returns from a run().
22:13:18 <slava> does your data have runtime tags?
22:13:25 <slava> copying gc relies on total knowledge of heap structure
22:13:32 <imaginator> runtime tags?
22:13:35 <slava> it must know what is a pointer, what is not, and if its a pointer, how large is the object its pointing to.
22:14:19 <slava> if this canot be guaranteed (eg, if at runtime an integer looks like a pointer) you must use a conservative collector which does not move objects
22:14:25 <imaginator> no, a pointer just gets pushed onto the stack, like numbers, and anything else (except for floating point (which I haven't implemented)).
22:14:59 <slava> ok. there's theory to conservative gc's (how to do it safely, etc) but i'm not familiar with it
22:15:04 <slava> you'll have to do mark/sweep collection
22:15:24 <slava> in factor, each 32-bit value has a 3-bit tag, in the least significant bytes
22:15:30 <slava> tag 0 means the rest of the bits are an integer value
22:15:36 <slava> other tags are pointers to various object types
22:15:39 <imaginator> That's clever.
22:15:57 <slava> the advantage is that some objects don't need a header in the heap at all
22:16:02 <slava> like a linked list cell only uses 8 bytes
22:16:07 <imaginator> Now that I think about it, I recall reading about a Lisp system or maybe ML with 30-bit integers, because it used a similar scheme.
22:16:14 <slava> because all pointers to it have the same tag so its size is implicitly known
22:16:21 <slava> imaginator, yeah, its a standard lisp technique
22:16:58 <slava> another thing tagging allows is a relocatable image
22:17:17 <imaginator> How so?
22:17:18 <slava> when its reading an image, it can determine which values are pointers, and safely update them with the base address
22:17:33 <imaginator> ah, that's neat.
22:17:55 <slava> gforth simply writes two images at different base addresses and looks for differences but i think that's a gross hack
22:18:14 <kc5tja> Hack as it may be, it works.
22:18:20 <slava> sure.
22:18:48 <imaginator> slava: did you implement your own bignum library?
22:18:52 <slava> imaginator, the downside of all this is that it is not possible to have pointers to the inside of an object, etc.
22:18:56 <slava> imaginator, no pointer arithmetic, etc
22:19:01 <slava> imaginator, no, i used scheme48's
22:19:25 <slava> http://cvs.sourceforge.net/viewcvs.py/factor/Factor/Factor/native/s48_bignum.c?rev=1.6&view=auto
22:19:32 <imaginator> hmm, well I definitely planned on pointers.
22:19:55 <slava> then you'll need a conservative gc
22:20:08 <imaginator> That code looks like Lisp, but it's C.  Strange.
22:20:12 <slava> if this is just for strings, i'm sure a simpler solution is possible
22:20:19 <slava> imaginator, it was written by MIT people, what do you expect )
22:20:21 <slava> :)*
22:20:48 <slava> hmm you could do a conservative gc that allows pointers inside objects
22:20:52 <slava> i think boehm does
22:22:46 <kc5tja> slava: I think this is another argument in favor of vector-based languages.  :)
22:22:54 <Herkamire> why is there space between my <div>s
22:22:55 <slava> how so?
22:23:01 <kc5tja> Anyway, I need to get home.  I'll be back online in about an hour or so.  Thanks.
22:23:23 <imaginator> Do the newer schemes use libgmp?
22:23:36 <kc5tja> slava: Because it's not possible to muck with pointers at all -- in order to reference a component you MUST have the base pointer -- ergo, for any one object, at least one base pointer must exist to it.
22:23:43 <kc5tja> Anyway, I'm outta here.  Peace.
22:23:59 <slava> that is not unique to vector languages. any 'safe' language has that property, eg java or smalltalk
22:24:11 <imaginator> later kc5tja 
22:24:34 <imaginator> This s48_bignum.c looks pretty good.
22:24:48 <imaginator> And the licensing isn't invasive like libgmp.
22:24:53 <slava> imaginator, yeah i was able to adopt it with minimal changes.
22:25:11 <slava> imaginator, you might want to get the original instead of using factor's
22:28:21 <imaginator> As I plan to eventually port my Forth-like language to raw hardware (via Bochs or qemu) I'm somewhat concerned that if I depend on Boehm-GC it will be ackward to manage and port.
22:28:54 <slava> what language are you implementing it in?
22:29:06 <imaginator> C and GAS (asm)
22:29:13 <slava> ok
22:31:33 <slava> i'm learning asm at the moment
22:32:15 <imaginator> Do you use a hash table for storage of word addresses and variables?
22:32:22 <slava> yes
22:32:38 <slava> words are objects, there's a few more fields than just a code address, but yeah
22:32:50 <imaginator> I'm starting to have doubts about my hashing trie.  It uses too much memory.
22:33:01 <slava> heh
22:33:20 <slava> i use a simple hashtable. its quite inefficient, because the number of buckets doesn't increase with load
22:33:23 <slava> i have to fix that some day
22:33:25 <imaginator> I used my 4-bit hashing trie in my Smalltick object system and it gobbled up memory, especially with many objects and long method names.
22:33:49 <slava> http://cvs.sourceforge.net/viewcvs.py/factor/Factor/Factor/library/hashtables.factor?rev=1.6&view=auto
22:33:56 <imaginator> Knowing when to increase to prevent to many ->nexts via more root buckets is ackward.
22:34:28 <imaginator> And then the hashing attacks take advantage of it, so some people use randomization with keys, but that introduces more checks for a fetch from the association.
22:34:33 <imaginator> to/too
22:34:43 <slava> hashing attacks?
22:34:53 <slava> if attackers can insert arbitrary words in your dictionary
22:34:56 <slava> or define arbitrary variables
22:35:02 <slava> you've got bigger problems than a potential DoS :)
22:35:41 <imaginator> yes, there are some people that attack servers by creating strings that hash to the same bucket, and thus you get an endless ->next to ->next and possibly with lots of rebuilding of the table if the code does that.
22:36:42 <imaginator> I guess a length check fixes most of that. 
22:38:00 <Herkamire> AAAAAAAAAAAARG
22:38:18 <Herkamire> when I stick borders on my divs then the background images work right
22:38:24 <Herkamire> but when I don't there's gaps
22:38:41 <Herkamire> it's not the images. the backgrounds don't meet
22:38:51 <Herkamire> but if I stick in:  border: 1px solid red;
22:38:59 <Herkamire> then there's no space
22:39:00 <slava> i hate html
22:39:01 <slava> and css
22:39:51 <imaginator> slava: you make a good point about "bigger problems than a potential DoS."  
22:40:25 <imaginator> What has Chuck Moore created lately?
22:40:44 <Herkamire> change one character so it's:   border: 0px solid red;   and I get 10 pixel gaps between my divs
22:41:24 <imaginator> Didn't Moore say in some chat (possibly here) that he wanted to build an HTML-like thing in Forth that would be better?
22:44:22 <imaginator> Touchie subject or no one knows?
22:44:42 <slava> we're still waiting for the meta-compiled colorforth :)
22:44:43 <I440r> guide
22:44:48 <I440r> bah wrong window lol
22:45:07 <imaginator> I wonder if he bit off more than he can chew.
22:45:49 <imaginator> He thinks files are a bad idea, but I disagree.  Blocks seem ackward for general apps and users to manipulate.
22:46:51 <imaginator> Then the code for his BMP reader/writer in the videos seems crazy, however short, but he noted that it has a bug.
22:47:12 <slava> did he fix it? :)
22:47:23 <imaginator> I don't know.  :)  And he doesn't like dealing w/ strings, so Forth lacks good string handling.
22:47:40 <imaginator> I think he said "They aren't very useful to me."  When someone asked about strings.
22:48:49 <imaginator> I liked seeing Jeff Fox's GUI.   The idea that it was so few words appealed to me, but now I wonder if it was inflexible and buggy.
22:48:59 <slava> jeff fox's gui?
22:49:19 --- join: slava_ (~slava@CPE00096ba44261-CM000e5cdfda14.cpe.net.cable.rogers.com) joined #forth
22:49:25 <slava> oops
22:49:31 --- part: slava_ left #forth
22:50:22 <imaginator> I'm looking for the link.
22:51:04 <imaginator> http://www.ultratechnology.com/scope.htm
22:51:52 <slava> looks cool
22:52:10 <imaginator> This I found introduced more doubt: "It is also because the thermal bug on F21d requires that a lot of the instructions be no-ops, no operation instructions."
22:52:19 <imaginator> (on that same page)
22:52:22 <slava> heh
23:00:44 <imaginator> What happened to itv?
23:06:22 <imaginator> The internet archive pages for it make it sound like a panacea.  I'm left wondering why Moore, Fox, and others weren't able to do what they set out to do.
23:21:56 <arke> imaginator: on colorforth.com he sets his list of priorities for CF. first is TCP/IP support, but theres already several others working on it, so he decided to go with the second one, which is metacompiled CF.
23:22:11 <arke> imaginator: i assume he has a job which takes up alot of his time so he doesnt have much time to tinker with CF
23:26:14 <imaginator> The pages at colorforth.com I see haven't been updated in 2 years.
23:27:02 <arke> thats true too :/
23:30:32 <crc> http://meme.b9.com/cview.html?channel=forth&date=040812 has some mention of the colorforth that can metacompile itself
23:33:30 --- join: qFox (C00K13S@82-169-140-229-mx.xdsl.tiscali.nl) joined #forth
23:41:27 <Herkamire> ok, how's it work now? http://herkamire.com/w/services.html
23:43:45 <qFox> in ie the line is broken at the top
23:45:00 <I440r> slava u there ?
23:45:14 <qFox> http://www.phil.uu.nl/~zee/herk.png
23:45:15 <I440r> i figured out anotehr eason why 100 million cell sort takes so long
23:45:27 <crc> looks ok here (Mozilla 1.6, Windows XP)
23:45:34 <I440r> thats 400 million bytes allocated for the buffers - damned if it aint mostly in SWAP lol
23:46:02 <Herkamire> qFox: thanks
23:46:22 <qFox> ie6 btw
23:55:02 <arke> hi kc5tja 
23:59:59 --- log: ended forth/04.09.11