00:00:00 --- log: started forth/06.12.02 01:09:35 --- quit: Razor-X (brown.freenode.net irc.freenode.net) 01:09:36 --- quit: segher (brown.freenode.net irc.freenode.net) 01:09:36 --- quit: Raystm2 (brown.freenode.net irc.freenode.net) 01:09:36 --- quit: TreyB (brown.freenode.net irc.freenode.net) 01:09:39 --- quit: Quartus_ (brown.freenode.net irc.freenode.net) 01:09:39 --- quit: warpzero (brown.freenode.net irc.freenode.net) 01:09:39 --- quit: Snoopy42 (brown.freenode.net irc.freenode.net) 01:09:39 --- quit: Shine (brown.freenode.net irc.freenode.net) 01:09:39 --- quit: arke_ (brown.freenode.net irc.freenode.net) 01:09:40 --- quit: nighty_ (brown.freenode.net irc.freenode.net) 01:09:40 --- quit: madwork (brown.freenode.net irc.freenode.net) 01:09:40 --- quit: ayrnieu (brown.freenode.net irc.freenode.net) 01:09:40 --- quit: nighty (brown.freenode.net irc.freenode.net) 01:09:42 --- quit: madgarden (brown.freenode.net irc.freenode.net) 01:09:42 --- quit: Zarutian (brown.freenode.net irc.freenode.net) 01:09:42 --- quit: timlarson (brown.freenode.net irc.freenode.net) 01:09:43 --- quit: neceve (brown.freenode.net irc.freenode.net) 01:09:43 --- quit: JasonWoof (brown.freenode.net irc.freenode.net) 01:09:44 --- quit: ccfg (brown.freenode.net irc.freenode.net) 01:11:43 --- join: JasonWoof (n=jason@unaffiliated/herkamire) joined #forth 01:11:43 --- join: Shine (n=Frank_Bu@xdsl-81-173-248-101.netcologne.de) joined #forth 01:11:43 --- join: arke_ (n=Chris@pD9E07331.dip.t-dialin.net) joined #forth 01:11:43 --- join: Razor-X (n=user@user-11faaoj.dsl.mindspring.com) joined #forth 01:11:43 --- join: nighty_ (n=nighty@sushi.rural-networks.com) joined #forth 01:11:43 --- join: neceve (n=claudiu@unaffiliated/neceve) joined #forth 01:11:43 --- join: Snoopy42 (i=snoopy_1@dslb-084-058-112-107.pools.arcor-ip.net) joined #forth 01:11:43 --- join: segher (n=segher@dslb-084-056-193-056.pools.arcor-ip.net) joined #forth 01:11:43 --- join: Raystm2 (n=NanRay@adsl-68-93-115-120.dsl.rcsntx.swbell.net) joined #forth 01:11:43 --- join: madgarden (n=madgarde@bas2-kitchener06-1096751791.dsl.bell.ca) joined #forth 01:11:43 --- join: Quartus_ (n=Quartus_@209.167.5.2) joined #forth 01:11:43 --- join: TreyB (n=trey@cpe-66-87-192-27.tx.sprintbbd.net) joined #forth 01:11:43 --- join: madwork (n=foo@204.138.110.15) joined #forth 01:11:43 --- join: ayrnieu (n=julian@pdpc/supporter/sustaining/ayrnieu) joined #forth 01:11:43 --- join: warpzero (n=warpzero@wza.us) joined #forth 01:11:43 --- join: nighty (n=nighty@66-163-28-100.ip.tor.radiant.net) joined #forth 01:11:43 --- join: Zarutian (n=Zarutian@194-144-84-110.du.xdsl.is) joined #forth 01:11:43 --- join: timlarson (n=timlarso@user-12l325b.cable.mindspring.com) joined #forth 01:11:43 --- join: ccfg (n=ccfg@dsl-roigw1-fe8ade00-21.dhcp.inet.fi) joined #forth 01:11:43 --- mode: irc.freenode.net set +oo JasonWoof Quartus_ 01:22:14 --- join: virl (n=virl@chello062178085149.1.12.vie.surfer.at) joined #forth 01:26:01 --- quit: JasonWoof ("off to bed") 01:27:10 --- quit: ayrnieu (Connection timed out) 01:40:58 --- join: TreyB_ (n=trey@cpe-66-87-192-27.tx.sprintbbd.net) joined #forth 01:40:58 --- quit: TreyB (Read error: 104 (Connection reset by peer)) 02:01:36 --- nick: arke_ -> arke 02:30:14 --- join: ayrnieu (n=julian@pdpc/supporter/sustaining/ayrnieu) joined #forth 03:36:24 --- join: Topaz (n=top@sown-86.ecs.soton.ac.uk) joined #forth 03:43:46 --- join: Crest (n=crest@p54895984.dip.t-dialin.net) joined #forth 03:46:33 --- quit: ayrnieu (Read error: 110 (Connection timed out)) 04:18:37 --- quit: Topaz ("Leaving") 04:26:21 --- join: fik (n=fik__@200.181.84.119) joined #forth 04:39:38 --- join: Cheery (n=Cheery@a81-197-54-146.elisa-laajakaista.fi) joined #forth 04:46:05 --- join: tathi (n=josh@pdpc/supporter/bronze/tathi) joined #forth 04:46:05 --- mode: ChanServ set +o tathi 05:29:12 --- quit: segher (Nick collision from services.) 05:29:23 --- join: segher (n=segher@dslb-084-056-164-170.pools.arcor-ip.net) joined #forth 06:10:51 --- quit: Crest ("Leaving") 07:25:13 --- quit: tathi ("leaving") 07:42:06 --- join: ayrnieu (n=julian@162M34.oasis.mediatti.net) joined #forth 07:51:20 --- quit: fik (Remote closed the connection) 07:58:23 --- join: cmeme (n=cmeme@boa.b9.com) joined #forth 07:58:24 --- quit: cmeme (Remote closed the connection) 07:59:24 --- join: cmeme (n=cmeme@boa.b9.com) joined #forth 08:02:28 --- quit: cmeme (Read error: 131 (Connection reset by peer)) 08:03:02 --- join: cmeme (n=cmeme@boa.b9.com) joined #forth 08:03:02 --- quit: cmeme (Remote closed the connection) 08:04:03 --- join: cmeme (n=cmeme@boa.b9.com) joined #forth 08:20:43 --- quit: Cheery ("Download Gaim: http://gaim.sourceforge.net/") 08:52:09 --- quit: virl (Remote closed the connection) 08:55:05 Hey. 09:13:44 --- quit: nighty_ (Read error: 110 (Connection timed out)) 09:14:23 --- join: nighty_ (n=nighty@sushi.rural-networks.com) joined #forth 09:47:41 --- join: zpg (n=user@user-514d7663.l2.c2.dsl.pol.co.uk) joined #forth 09:47:56 hi 09:50:05 --- join: Crest (n=crest@p54895984.dip.t-dialin.net) joined #forth 09:50:07 hi 09:52:47 hey Crest 10:34:04 What's up? 10:35:21 Not too much -- you? 10:35:35 Busy, nothing too gripping. 10:36:10 :) 11:03:11 --- join: jackokring (n=jackokri@static-195-248-105-144.adsl.hotchilli.net) joined #forth 11:03:12 --- join: Snoopy42_ (i=snoopy_1@dslb-084-058-138-249.pools.arcor-ip.net) joined #forth 11:10:41 --- quit: Snoopy42 (Read error: 145 (Connection timed out)) 11:11:02 --- nick: Snoopy42_ -> Snoopy42 11:22:45 --- quit: Quartus_ (Read error: 104 (Connection reset by peer)) 11:34:54 --- join: Cheery (n=Cheery@a81-197-54-146.elisa-laajakaista.fi) joined #forth 11:43:07 --- join: tathi (n=josh@pdpc/supporter/bronze/tathi) joined #forth 11:43:07 --- mode: ChanServ set +o tathi 11:45:18 --- quit: segher (Nick collision from services.) 11:45:29 --- join: segher (n=segher@dslb-084-056-190-108.pools.arcor-ip.net) joined #forth 11:58:12 --- quit: Shine (Nick collision from services.) 11:58:17 --- join: Shine_ (n=Frank_Bu@xdsl-84-44-129-8.netcologne.de) joined #forth 11:58:30 --- nick: Shine_ -> Shine 12:51:38 --- quit: madgarden (Read error: 104 (Connection reset by peer)) 12:52:20 --- join: madgarden (n=madgarde@bas2-kitchener06-1096751791.dsl.bell.ca) joined #forth 13:12:03 --- quit: jackokring (Read error: 104 (Connection reset by peer)) 13:18:42 --- join: jackokring (n=jackokri@static-195-248-105-144.adsl.hotchilli.net) joined #forth 13:21:39 --- join: virl (n=virl@chello062178085149.1.12.vie.surfer.at) joined #forth 13:23:59 --- quit: madgarden (Read error: 104 (Connection reset by peer)) 13:24:35 --- join: madgarden (n=madgarde@bas2-kitchener06-1096751791.dsl.bell.ca) joined #forth 13:30:09 --- join: JasonWoof (n=jason@unaffiliated/herkamire) joined #forth 13:30:10 --- mode: ChanServ set +o JasonWoof 13:32:19 --- join: Quartus_ (n=Quartus_@209.167.5.2) joined #forth 13:32:19 --- mode: ChanServ set +o Quartus_ 13:32:43 dang, I think I deleted my kernel 13:33:42 Hmm 13:35:22 last night 13:35:35 either it was really stupid. or I've been using the wrong kernel 13:35:52 we shall see. I'm starting it up on a CD so I can see what I have 13:40:07 nah, I've still got a kernel. 13:40:22 looks like I forgot to update my yaboot config whin I installed my new system 13:40:33 guess I've been using the old kernel with the new gentoo for a while 13:40:34 oops 13:49:19 nah, I must have coppied my new kernel into the old location 13:49:36 my root partition is now reiserfs which yaboot can't read 13:51:31 boot from live disto and update 13:51:40 yeah, that's what I'm doing 13:52:02 to grub? 13:52:33 I got my kernel in place 13:52:39 but I messed up the yaboot config 13:53:44 error on boot should give options of possible roots but they don't have any falback as yet 13:55:32 I could type in some kind of boot path into yaboot 13:55:36 but I don't know what to type 13:55:40 and it doesn't have history 13:55:47 so I have to type the full thing out every time 13:55:54 and it's stuck in QWERTY which I suck at 13:56:11 root=/hdxx on lilo 13:56:17 I can still bang out 60 WPM on QWERTY. 13:56:56 yaboot 13:56:58 yaboot 13:56:59 --- join: marble (n=glass@cpc1-bolt6-0-0-cust18.manc.cable.ntl.com) joined #forth 13:57:06 not lilo, not grub 13:57:29 besides, I forget the partition numbers anyway 13:57:33 maybe bzimage boot kernal has no rasierfs except after module loader loaded 13:57:53 well, yeah that's where this all got started 13:57:57 yaboot doesn't support reiserfs 13:58:07 should be able to get partition number from fdisk 13:58:08 so I put my kernels on an old partition 13:58:22 yesterday I was cleaning crap off that old partition and didn't think I was using the kernels on it 13:58:36 yay :) now it's booting 13:58:38 heh. 13:58:48 yeah, I have a partition that I use only for kernels and such 13:58:48 I had a copy of my current kernel on my root partition 13:59:01 so I just had to boot up, find the partition numbers, mount the two partitions 13:59:13 create the /boot directory on the old partition 13:59:16 copy the kernel over 13:59:42 then I found that my yaboot config was looking for a kernel named hdd.vmlinux, so I had to reboot a couple more times to fix that. 13:59:49 yeah, I should 14:00:02 and at one point I think I did 14:00:11 that's why I didn't think I was using the kernels on my old root partition 14:00:43 I'm not that great with partition allocation 14:01:07 I've still got a 5GB hfs partition on this HD, so I think I've still got Mac OS 9 14:01:31 --- quit: JasonWoof ("switching to primary computer") 14:01:53 --- join: JasonWoof (n=jason@unaffiliated/herkamire) joined #forth 14:01:53 --- mode: ChanServ set +o JasonWoof 14:02:11 ok, that was exciting. now time to eat 14:17:03 --- join: fik (n=fik__@200.181.84.119) joined #forth 14:18:39 That's okay JasonWoof. Two days ago, I put one too many in a windows colorforth block. The editor crashes when it tries to open a block with one two many words on it. So I thought i'd remove the last word on the block with a hex editor. I did something wrong in the hex editor and now i've lost all of my last two months of changes. 14:18:53 Think I've figured out a way around it tho. 14:19:16 --- quit: jackokring (Read error: 113 (No route to host)) 14:19:57 I'll get the first 10 blocks from a previous version and just stick it on the front of the broken image. This way, I can use colorForth itself to see where I broke the thing. 14:21:58 And when is someone going to develop a hex editor that opens with... " Okay, I'll open that file, but what do you want to call the working copy?" 14:22:31 And defaults to the current name + something. 14:23:17 After all, I must be protected from myself. :) 14:26:18 Also, reminding myself to make a batch file that "BYE" can call that will execute a 7zip task to back up the current blocks.bin and icons.bin with every exit of the environ. 14:26:55 that way it can auto version control of sorts. 14:28:06 Glypher does such a thing on exit. If you change first word in block 0, it will create a file with that name. I use the date and time. 14:31:57 yah, date & time makes more sense 14:35:10 To open a previous version, you change it's name to TEMPLATE 14:35:50 The file has the first word as TEMPLATE and the second is my yymmddhhmmss file name. 14:36:05 When I want to copy the file I swap the first two words. 14:36:41 If I want to make a new on, just add a new file name. 14:36:47 on+e 14:38:36 I suppose I could automate that if I had access to Glyphers internals. 14:39:18 Just always close with a copy file, autozip the current to archive. 14:40:03 sounds like working with Color Forth is fun. I can't imagine how it would be to work with multiple developers and without version management like SVN on a bigger project 14:40:41 There are very few colorForth developers. :) 14:41:47 and AFAIK it was designed as a one-man system :-) 14:42:19 I have a thing I use, Next-block, it copies the current block into the next clear block for editing. But I've been backing up the files by hand and I should automate that. 14:42:28 Ya. 14:42:59 working with Forth blocks instead of files sounds strange for me 14:46:02 The windows version uses a binary file that is initially all zeros, replacing each word with the words you edit in. This is why going over the limit is a no-no. The editor won't open a block with too many words, the system just exits on such errors. I think that's a bug. You shouldn't be allowed to add the extra word. 14:47:42 you should use punch cards to write your programs, one card per block, then you'll see the limits ;-) 14:48:07 Good thinking. :) 14:48:39 I miss the rythm of the Halworth sorter. 14:48:40 would be in the spirit of colorForth 14:50:19 I've never used punch cards, my computer live started with the C64, which had a tape drive and a nice interactive built-in BASIC interpreter 14:51:15 version control is different 14:51:23 herkforth never overwrites anything 14:51:27 Sure, I know. 14:51:39 whenever you save it creates a new binary with everything in it 14:52:02 that's a waste of space 14:52:11 not that much 14:52:26 images are about 500KiB 14:53:12 I save as I work (just as I do when editing text files) then when I get something significant done, or decide to stop for the day/hour I backup the latest version to my server 14:53:28 do you review changes? 14:53:46 I pretty much only look at older versions when I break something 14:53:55 (I wanted to asked about version control and forth a while ago, but I forgot) 14:54:00 colorForth edits the file on disk as it goes, but it's constantly overwriteing... 14:54:21 marble: we're discussing a special case where the whole system is lumped into a single binary file 14:54:31 marble: thus making normal version controll impossible 14:54:46 Wellm the windows version does, I meant to say. Native, you SAVE to write back. 14:55:08 marble: most forth is text-file based, and thus I'd expect people to use normal version control software (svn, git, cvs, arch, etc) 14:55:18 fair enough -- it that's in keeping with the forth spirit, then it's probably worth thinking about 14:55:24 Darcs :) 14:56:02 marble: it seems to be a matter of personal taste whether people use version control if they are the only developer 14:56:17 I do 14:56:23 but some people seem to think it's not worth the bother 14:56:42 * marble uses my repository to figure out where I'm up to 14:56:52 'course some of them come in here and bitch that they accidentally deleted days/weeks/months of work... 14:57:01 hehe. 14:58:11 with the bodies of functions being so much smaller, a commit on every one would feel too heavyweight 14:59:58 --- quit: zpg (Read error: 113 (No route to host)) 15:00:04 heh 15:00:18 with svn I usually commit once per sitting 15:00:42 once I get my new thing or change working and tested 15:01:16 your commit log messages chart your progress? 15:02:04 yeah, but I rarely look at them 15:02:10 Be cool if every change was recorded and you could UNDO them. 15:02:26 I rarely look at the history, and when I do it's to find out what change broke something 15:02:44 Raystm2: of course you can do that 15:03:01 that's the point of keeping a history 15:03:21 you can view the history, and restore any file to any point in the history 15:03:48 and you don't have to treat your working directories as precious 15:04:05 and you have a central PLACE for your things 15:04:06 yep :) once it's commited, it's safe 15:04:18 yeah, it certainly helps me stay organized 15:04:26 Sure, I understand that. I'm talking every keystroke, tho. 15:04:27 and it's nice to have the stuff publically available 15:04:34 (a place for everything and everything in its place :) 15:04:38 so I can just give a link where people can see what I'm blabbering about 15:04:52 Raystm2: You'll drown under the avalanche of information :) 15:05:10 Raystm2: in archy you can undo forever even accross editing sessions 15:06:24 man, now I really want to try git 15:06:26 marble: just rewind to the point and copy and branch, I suppose. 15:06:45 JasonWoof: sounds like I should have a look, thanks. 15:06:59 archy is interesting 15:07:08 Raystm2: have you read The Humane Interface? 15:07:16 -- Jef Raskin 15:07:30 No, think i'll treat myself this Christmas. 15:07:39 Been meaning too. 15:07:39 archy is an attempt to implement what he describes in THI 15:07:43 good book 15:07:47 Ya. 15:11:24 * marble thinks about literate programming 15:15:13 Thats what I was doing, marble. I was adding comments between the words of code to make them into self documenting. Just didn't pay attention to the word count. 15:15:50 White words are ignored and can come anywhere. Very convenient. 15:19:58 I had intended to htmlizer the code, then publish to the site. 15:20:28 it would have read well? 15:20:35 Ya, like a book. 15:20:48 Fully self documenting. 15:23:36 throw in a call-graph of all the word, and you might have me sold :) 15:25:12 I'd like to get the htmler to list the front 18 blocks of assembler, and then continue the html links into the kernel. 15:26:24 That way, while reading a block, you could always go up and down the dependantcies. 15:28:50 what's the htmliser written in? 15:29:32 Howerd Oakford has one in colorForth. Tathi and Mark Slicker each did one in c. 15:29:57 I see you've done some research ;) 15:30:48 I've been around a while. 15:35:16 why does someone write a colorForth htmlizer in C instead of colorForth? Is colorForth too difficult to use for such a simple task? 15:35:17 git is cool... 15:35:26 I just followed a bit of the tutorial 15:40:40 There are htmlers in both languages. colorForth, tho very easy to use, seems to give programmers trouble when they first encounter it. 15:41:05 JasonWoof, http://git.or.cz/ ? 15:41:19 hey. 15:41:37 Hi Quartus. ;) 15:47:50 fik: yes 15:48:21 Shine: I think tathi wrote the htmlizer in C because he couldn't run colorforth 15:48:30 Shine: he runs a PPC box 15:54:36 --- join: Aurinto (i=TheBlueW@ts001d0585.wdc-dc.xod.concentric.net) joined #forth 15:54:54 --- part: Aurinto left #forth 15:55:16 --- join: TheBlueWizard (i=TheBlueW@ts001d0585.wdc-dc.xod.concentric.net) joined #forth 15:58:38 --- quit: Quartus_ (Read error: 104 (Connection reset by peer)) 16:04:57 --- quit: Cheery ("Download Gaim: http://gaim.sourceforge.net/") 16:06:34 yeah, actually I started out to modify Mark Slicker's 16:06:59 what was wrong with his? 16:07:00 but it turned out to be easier and quicker to just start over 16:07:12 heh 16:07:26 hmm 16:07:28 tathi: have you tried git? 16:07:37 JasonWoof: not yet. planning to though 16:07:40 looks like it does everything we've been looking for 16:07:46 tathi knew he could get all the like words to link to the first usage. 16:08:02 oh, wait, his didn't make the word names themselves into links 16:08:09 Right. 16:08:16 it put an underscore in dark blue after each word, and that was the link 16:08:19 very very annoying 16:08:19 Yours is prettier, as well. 16:08:26 yeah 16:08:27 ya. 16:08:56 I just did everything in the tutorial, except for the remote stuff 16:09:02 the branch thing is pretty cool 16:09:38 heh. yeah, branching is kind of the central concept in a distributed version control system 16:09:44 so they tend to be good at it 16:09:44 when you make a branch you don't get another working copy 16:09:52 (probably possibly to, but you don't have to) 16:10:03 Hi ho :) 16:10:03 ah 16:10:12 you just say you want one and whala, your working copy is the branch 16:10:17 and you hit a command to switch to other branches you have 16:10:23 Hi nighty_ :) 16:10:26 and it swaps the files in your working copy for you 16:10:31 has anyone made a forth for the HP49G/49G+/50G ? 16:10:50 you can also make a branch by importing remote changes 16:11:09 (if you choose... you can import remote changes into your current tree) 16:11:33 and it comes with this very cool graphical history browser thing 16:11:36 quite impressive 16:12:18 in the tutorial I created a branch, modified stuff in the branch, commited. then modified in the master branch, and commited 16:12:29 then pulled the changes from the branch into the master 16:12:30 quite easy 16:12:47 now I want to try making a conflict 16:31:14 --- quit: tathi ("leaving") 17:04:57 --- join: Quartus_ (n=Quartus_@209.167.5.1) joined #forth 17:04:57 --- mode: ChanServ set +o Quartus_ 17:15:38 --- part: marble left #forth 17:55:56 --- join: snowrichard (n=richard@12.18.108.162) joined #forth 17:56:11 hi 18:03:46 --- quit: snowrichard ("Leaving") 18:11:15 nighty - just use SysRPL :-) 18:12:03 --- join: mbishop (n=martin@unaffiliated/mbishop) joined #forth 18:12:23 git handled conflict beautifully 18:12:52 now I've gotta figure out having some kind of public repository 18:26:04 holy crap it has amazing features 18:27:08 it has code to load change sets into an e-mail 18:27:13 and code to extract from e-mail and apply 18:27:19 it has code to track down a bug 18:27:35 you say this revesion is not working and revision X did work 18:28:04 And it mocks you mercilessly until you reboot? 18:28:05 and it says OK, and put's you at a revision half way inbetween, and you tell it if that one works or not 18:28:17 Quartus: only the windoze version ;) 18:29:00 then it puts you halfway between the revisions left to try 18:30:17 so if there was a bug introduced somewhere in the last 100 revisions, you can easily find which rivision it came from (assuming you can easily test if your working copy has that bug) by just testing 7 revisions 18:30:27 and git picks them for you, and updates your working copy to those revisions 18:32:07 * JasonWoof looks for that cool windoze haiku 18:32:49 Yesterday it worked 18:32:49 Today it is not working 18:32:49 Windows is like that 18:32:51 JasonWoof, sounds good, except odds are pretty good that the bug is introduced immediatly following the last revision that is reported to actually be working. 18:33:38 Quartus: well, if you know where it is then just look at that one 18:34:07 this is obviously for the case where you have no idea when it broke 18:34:16 you could take a wild stab at when it broke 18:34:32 if you started 200 revisions back instead of 100 it would only require you to test once more 18:34:50 I guess. Suggests a significant lack of grokking of the code base. 18:34:57 eh? 18:35:01 I don't follow 18:35:08 I also don't know what you mean "reported working" 18:35:22 I mean, good lord, a bug that hasn't been detected for 200 revisions? 18:35:26 the scenario I was imagining was that you just discovered a bug 18:35:37 And a developer who then has no clue as to where to look? 18:35:46 Quartus: this was designed for revision control on the linux kernel 18:36:11 200 revisions ago could be a day or two ago for all I know 18:36:24 Revision control is dandy, I just question the value of a binary search bug locator. 18:36:33 at the rate the 2.6 kernel goes, 200 revisions is about 5 minutes :) 18:36:39 heh 18:37:26 Quartus: how would you find a bug that could have been introduced anywhere in 200 revisions, and you have no idea what file it's in, or who changed it? 18:37:46 or 2000 revisions for that matter 18:38:08 You mean it's my first day with the source, and the other developers are all in line for the next Star Wars movie? 18:39:10 oh, I think also you can tell it certain files and/or directories that you think the bug is in, and it'll only consider revisions where those files have changed 18:40:16 And then you're only a few minutes away from isolating it yourself, binary search revision analysis notwithstanding. 18:41:48 It just strikes me as one of those features that sounds useful, but wouldn't actually be. 18:42:44 ok 18:42:54 to me it sounds like the sort of thing that I do by hand sometimes 18:42:57 I've done it with herkforth 18:43:01 something was broken 18:43:13 so I hop back a few versions to see if it works there 18:43:18 if not I hop back 10 18:43:22 if not 50 18:43:29 if so forward 20 18:43:31 etc etc 18:43:48 If it's your code base, you know the last time you touched whatever it is that's now breaking. It's very seldom a random hunt. 18:44:18 right 18:44:25 I don't think I'll use it much if at all for my owr code 18:44:26 own code 18:44:41 but I think it would be useful if 100s of people were changing stuff 18:47:26 heh, guess I'm still in that early phase where I get excited about everything 18:47:51 I think the graphical viewer thing will be a major time saver though 18:48:29 it's got a text box that will highlight all commits that add or remove the string you type 18:48:38 or that modify the file you type 19:09:22 ayrnieu nighty - just use SysRPL :-) , it is still RPL no ? 19:10:14 nighty - no, it has much more interesting semantics, to a Forth person. 19:10:37 ttp://tunes.org/~nef/logs/forth/ 19:11:21 I don't particularly appreciate the channel being logged and displayed publicly, but maybe it is just me 19:11:34 It's been logged for years. 19:11:40 I know 19:11:58 I guess it is just me then :) 19:12:01 it is just you. 19:12:11 clog is on a dozen or more channels. 19:12:23 That is what I found out 19:12:55 * TheBlueWizard hands nighty_ a tin foil hat :) 19:13:10 and #Emacs, before someone grew distraught about the fundamental intransience of all information according to Quantum Mechanics. 19:14:05 heh 19:14:11 ayrnieu: isn't sysRPL running in a virtual machine ? 19:14:21 ayrnieu: or emulation 19:14:30 ayrnieu: to be more precise 19:16:45 nighty - on the HP49G, it ran unemulated along with all the rest of the machine code in the ROM. It certainly ran in a sort of 'virtual machine' -- everything does. On the HP50, they replaced the Saturn processor with ARM+emulation -- but they've some manner of poking through the emulation. 19:17:36 If you mean that you want to have a 'native Forth', with none of HP's own operation system, then no: I don't think anyone has made progress on that. Otherwise, use SysRPL. 19:17:49 ayrnieu: ok 19:19:12 I like the log 19:19:21 sometimes I find answers to my questions there via google 19:20:40 --- part: TheBlueWizard left #forth 19:26:13 --- join: crc (n=crc@pool-70-110-218-43.phil.east.verizon.net) joined #forth 19:26:33 --- mode: ChanServ set +o crc 19:37:25 --- quit: fik (Remote closed the connection) 21:32:03 --- join: slava (n=slava@modemcable244.105-201-24.mc.videotron.ca) joined #forth 21:40:15 --- join: arke_ (n=Chris@pD9E04D02.dip.t-dialin.net) joined #forth 21:46:06 hey 21:51:29 hi 21:51:34 i'm debugging factor on windows 21:51:39 fun? 21:51:44 trying to reproduce a reported crash, with no success 21:53:54 The worst kind of bug. 21:55:44 yes 21:56:43 --- quit: arke (Read error: 110 (Connection timed out)) 21:58:45 i reproduced it 21:58:57 Good! 21:58:58 now, time to get gdb working on windows 21:59:00 THAT is the worst kind of bug. 21:59:18 I haven't used a debugger for ... ten years? Long time, anyway. 21:59:43 i use gdb to poke around in memory, and disassemble (latter being irrelevant for this bug) 21:59:58 i hardly ever use it to debug c code, with breakpoints and all 22:00:02 mostly just as a monitor. 22:01:21 I imagine it's of some use, but I've never run aground with the simpler tools I do use. Good to know the tools are out there if I do. :) 22:01:53 i presume you use @, !, some kind of DUMP, and a SEE. that's pretty much the subset of gdb i use :) 22:02:00 does quartus have a disassembler? 22:02:05 It does. 22:02:15 i haven't got around to writing my own yet 22:02:19 it seems like a huge undertaking 22:02:25 given that i have to do it three times 22:02:30 Why three? 22:02:42 x86, x86-64 (which will reuse most of the code from x86), powerpc 22:02:48 and i'm planning an arm port for next year. 22:03:07 You're in a position of, I assume, writing them in Factor, so you can't leverage existing code. 22:03:19 i wrote the assemblers myself 22:03:27 but i studied forth assemblers first 22:03:43 I ported existing assemblers/disassemblers already in Forth, so that helped. 22:03:59 gforth's x86 assembler is useless. its source is in blocks, about 200 of them, completely undocumented. 22:04:47 In arch/386 there are .fs files, sequential, not in blocks. 22:05:09 Originally blocks, in the case of the assembler. 22:05:10 hmm, i swear it was in blocks when i looked. 22:05:15 ah 22:05:24 do you see any value in blocks? 22:05:46 i was reading the logs today where Raystm2 was talking about how he clobbered his colorforth block file in a hex editor and lost a few weeks of work. 22:05:46 No, save perhaps in some extremely limited embedded systems where the development is on-board only. 22:05:52 thank goodness for text files and revision control 22:06:27 The disassembler is pretty good; I use it unchanged in RetroForth over the ANS layer. 22:14:32 --- quit: cmeme (Excess Flood) 22:16:08 ugh, gdb won't attach to the process 22:17:07 oh, its operator error 22:17:11 works now. 22:17:13 --- join: cmeme (n=cmeme@boa.b9.com) joined #forth 22:17:17 :) 22:21:09 the bug is in code GC 22:21:22 --- quit: cmeme (Excess Flood) 22:21:22 your garbage collector? 22:21:32 i have two, one for data and one for code 22:21:37 the latter is very new 22:22:00 basically the bug is that if you reload a certain source file over and over in a loop, eventually factor crashes in the code GC 22:22:09 Ah. 22:22:29 code GC uses a typical allocate/free-style memory block manager, i suspect the bug is at that layer 22:22:40 something with the way the free list is being handled 22:23:04 Why a code GC? 22:23:17 so that you can reload code without running out of memory eventually 22:23:39 during development 22:24:18 --- join: cmeme (n=cmeme@boa.b9.com) joined #forth 22:24:58 for example if i reload the source file which defines =, then about 700kb of code needs to be recompiled 22:25:31 because all callers of =, and everything else in that file, are recompiled 22:25:59 in forth, you'd either restart the system, or dump a new image, or something like that. i used to do that too 22:26:06 Right. 23:09:58 --- join: crest_ (n=crest@p54894F6A.dip.t-dialin.net) joined #forth 23:10:07 --- quit: slava (Read error: 145 (Connection timed out)) 23:12:21 --- quit: Crest (Read error: 60 (Operation timed out)) 23:52:51 --- quit: nighty_ (Remote closed the connection) 23:59:59 --- log: ended forth/06.12.02