00:00:00 --- log: started retro/10.01.12 00:18:45 --- quit: docl (farmer.freenode.net irc.freenode.net) 00:18:46 --- quit: SimonRC (farmer.freenode.net irc.freenode.net) 00:21:17 --- join: docl (n=luke@97-120-211-224.ptld.qwest.net) joined #retro 00:21:17 --- join: SimonRC (n=sc@fof.durge.org) joined #retro 00:43:40 --- quit: crc (Read error: 110 (Connection timed out)) 01:00:36 --- quit: retro-commit (Read error: 110 (Connection timed out)) 04:32:35 --- join: crcx (i=d8012b82@gateway/web/freenode/x-ogmcmygskdpsjqbr) joined #retro 04:51:53 Morning crc crcx 05:17:50 good morning 05:18:36 did you find the gist with the .inline class? 05:19:04 Good morning, how are you? I didn't find the inline. 05:19:18 * js4 is porting ll-test.retro to use rut. 05:19:34 * js4 is planning on starting tests for kernel/core 05:20:14 http://gist.github.com/272946 05:20:34 Thank you! 05:21:18 I am debating ripping my extensions to <# # #> out and just checking in the standard. 05:22:03 I'll look at the ans spec too I guess... But I really think they got sign wrong. They shoulda asked me, I was like... oh... 8? 10?? 06:06:00 crcx: the checking tracker is not on this channel anymore? 06:20:57 my home machine's network connection is down 06:23:16 Oh, that stinks! ANyway, updated ll-test.retro to use rut 06:24:17 Think I'll start on some unittests for the core... meanwhile... my retro10 fork queue is all red again, and as far as I know I haven't done anything to retro10 since the last time I tried to make red problems go away. 06:53:06 * crcx has no idea why git doesn't seem to work as expected 09:10:58 * crcx wonders if it's time to look at other vcs systems 10:24:01 I wish git hub would give me more information about what the issue is. I can fix files. I don't know how to fix "red". 12:27:27 I hate this. I deleted my fork. I'll wait a day and make a new fork, hopefully it will forget anything about my old one in the mean time. 12:29:47 yes I did spend the last 2 hours trying to learn how to politely tell github "make mine look like that" 13:39:01 --- quit: crcx ("Page closed") 15:00:06 --- join: erider (n=chatzill@unaffiliated/erider) joined #retro 15:00:24 hi 15:08:48 --- quit: yiyus (farmer.freenode.net irc.freenode.net) 15:08:48 --- quit: erider (farmer.freenode.net irc.freenode.net) 15:08:49 --- quit: virl (farmer.freenode.net irc.freenode.net) 15:10:31 --- join: erider (n=chatzill@unaffiliated/erider) joined #retro 15:10:31 --- join: virl (n=virl__@chello062178085149.1.12.vie.surfer.at) joined #retro 15:11:11 --- join: yiyus (i=12427124@je.je.je) joined #retro 15:24:47 Hi erider. I was reading the PFE docs... 15:33:04 js4: and what to you find 15:40:36 better question is what did you learn 15:49:57 Well.... what do y'all think about replacing the UglyCamelCalled IfDefined with the hideous commitee "horse" named [defined] [undefined] [if] [else] and [then]? 15:51:35 I learned about the challenges of the various threading models, and a couple interesting new ones. 15:52:34 PFE is trying to do something pretty difficult -- C source ANS forth that is compatible (as much as possible) with all the CPU's, host OS's, and old forth source in the world. 15:53:20 that is interesting 16:05:57 http://www.forthfreak.net/index.cgi?ForthBestPractices 16:07:24 yep, already book marked, frequently consulted. I like the excerpts from Brodie's "Thinking Forth" 16:08:57 PFE and gForth are def not following the 1x rules. 16:09:38 And it's the 1x rules that make me think I should rip my extensions to <# # #> out. 16:12:16 yeah I like rule one 16:12:28 it is easier to read 16:13:12 * erider was playing with xor 16:15:23 I like xor. It's always bummed me that xor isn't universal. only nand and nor are. 16:20:34 bbiab, going to the market... 17:38:54 --- join: crc_ (n=charlesc@m715e36d0.tmodns.net) joined #retro 17:38:54 --- mode: ChanServ set +o crc_ 17:39:44 Hi Crc_ 17:40:05 --- nick: crc_ -> crc 17:40:14 hi 17:40:23 * crc setup a tether connection from his phone 17:40:42 O.O don't stay on too long, could be a big phone bill! 17:42:23 unlimited data :) 17:42:31 and unlimited minutes, so it should be ok 17:42:38 ^_^ 17:48:04 --- join: probonono (n=User@unaffiliated/probonono) joined #retro 17:49:02 --- quit: virl (Remote closed the connection) 17:52:14 factoring makes word definitions smaller right 17:56:40 abstractly that is usually the case. 17:57:23 I can probably dream up pathalogical counter examples, and depends on what you mean by "smaller". 17:57:39 I am thinking about some of the rules and how to remove swap 17:59:11 what are you reading? The ForthBestPractices page? 17:59:34 yeah 18:08:47 js4: guess you would not use swap or over 18:14:37 --- topic: set to 'Retro Forth | Logs @ http://tr.im/Kept | http://retroforth.org | Stable Release 10.3.1 - January 2010' by crc 18:19:50 swap, rot, or other operations which touch more than one stack item when touching just one item would suffice 18:26:19 I dunno about that... actually these days, phylisophically, I think over is better than swap. 18:36:36 js4: if the goal is to refactor then I don't see any issue with what they are saying 18:38:06 I always thought that after all of the word definitions you can use the program as if you were speaking English 18:39:09 like this in the code of docl "3 elements shown hidden before" 18:39:11 I think the world, even just the world of programming, is very complex. I am very reluctant to take any rule as 100% correct 100% of the time. There are always additional factors to consider. 18:39:36 true 18:40:51 You can factor all programs down to a network of nands. I don't think you should though. Forexample... 18:41:30 The ngaro VM doesn't do over. Turns out over is one of the most commonly executed operations in retro when you meta-compile the system. 18:41:51 And the place where all that over action was going on is in compare. 18:42:43 I think that just dealing with two items on the stack is not a bad idea 18:43:06 Obviously you *can* make forth without a "real" over in your cpu. But is over more or less important than So the question is complicated. Yes trying to juggle 6 items on the stack probably is a mistake. But I think you can over react in the opposite direction. There's no simple rule or answer. 18:44:48 over (lea esi, [esi-4] mov [esi], eax mov eax, [esi+4]) 18:45:03 : over push dup pop swap ; 18:45:17 : swap over push nip pop ; 18:45:42 : dup over over nip ; 18:45:49 : drop over nip nip ; 18:46:08 I think I am going to try to pass the word what it needs 18:46:10 But *should* you throw away swap dup and drop? 18:46:19 no 18:46:50 Yes. My rule is "if it's hard to get the stack juggling right, re think what you are doing, who calls you, etc." 18:46:59 dup and drop no but you should use swap only when necessary 18:47:57 I think it is hard to follow what is going on when there is a lot of juggling 18:47:59 So yeah, the fundamental problem with swap I see is two parts: 1) swap itself isn't actually doing anything. 2) swap writes two locations at the same time. 18:48:21 I have to draw a picture somethings 18:48:25 I agree. And juggling doesn't make anything happen. It's the other words that do. 18:48:53 So if you draw a picture, have to think much about the stack, have to call 2pick, then probably there is a problem. 18:49:43 otoh, if you are comparing two strings, ofcourse you have to mess with 4 things - two pointers and two characters. 18:50:34 So.... it's just not simple. And I love guidelines from great hackers, like Chucks 1x stuff, but I won't buy any of it as a hard and fast law. 18:53:01 js4: yeah but it is a good starting point to build on 18:53:56 I like the way a C author described it. He said code can "smell bad". Doesn't mean it's wrong, just that you should be suspicious. For me "Stack juggling smells bad". 18:54:55 lol 18:55:00 Or maybe it was an XP book... I forget now... anyway... 18:57:23 Personally I worry about seeing 2xxx or pick or roll. If I see alot of them, and maybe locals and variables... I get really worried. 19:02:37 I have to go talk to you tomorrow 19:02:44 --- quit: erider ("ChatZilla 0.9.86 [Firefox 3.5.7/20091221164558]") 20:50:42 --- quit: crc (Read error: 110 (Connection timed out)) 23:18:26 * docl just got home from the cryonics party 23:20:23 Hope you had a good time, kinda wish I coulda been there. 23:20:30 yeah, it was great 23:20:34 small but friendly 23:21:09 you should definitely come to the next thing 23:21:27 I think they are planning a cryo training session in june 23:21:35 I would like that. Are there any other forthers/retro-forthers anywhere near us? 23:21:42 Who is doing the training? 23:23:14 well, chana and aschwin de wolf will be there. I think aaron drake is coming, he's the main standby guy for alcor. 23:23:50 chana and aschwin are really nice. they picked me up after work and dropped me off, even though it's like an hour drive to portland. 23:24:21 I've been out of touch, I don't even recognize the names... I used to be really into it. I got transport/standby tech training... all that... 23:24:24 I got to meet carlos mondragon, one of the previous alcor presidents. cool guy. 23:24:35 Yeah, Carlos is prety cool 23:24:36 you absolutely need to reconnect 23:24:40 yeah... 23:25:05 they could probably help you find a job 23:26:11 you seem to have a good head on your shoulders. not everyone in cryonics does, I gather. 23:28:09 There are many different kinds of people involved. From Timothy Leary (now dead and unfrozen) to average joe high school teachers, with us somewhere in the middle I guess. 23:28:37 there was another meeting that I missed for people in their teens and twenties, in florida. 23:29:01 I am *SO* pissed about Leary.... and about Hienlin.... 23:29:20 yeah... it's so weird 23:29:34 like they were lying about being interested in the future the whole time, or something 23:29:38 People are *weird* about death and dieing. 23:30:17 there seems to be more of a surge in families and couples getting involved lately though. that's a good sign. 23:30:33 Leary, if I remember right, was probably due to some relative or lover or something. They killed him in my opinion. Hienlin decided not to be frozen. 23:30:41 makes it more of something for normal or quasi-normal people rather than just a few computer nerd guys 23:30:52 yeah 23:31:10 heinlein, and practically every other sci-fi author out there 23:31:28 --- join: virl (n=virl__@chello062178085149.1.12.vie.surfer.at) joined #retro 23:31:30 well, I suppose some cryonicists do write sci-fi 23:31:38 The diference is Hienlein talked like he was the sort to do it. 23:31:52 Asimov was kinda in the "old should get out of the way" camp I think. 23:31:56 exactly... 23:32:08 yeah, asimov was a little weird like that 23:32:12 old-fashioned humanist 23:32:13 meh... whatever... people are weird about it. 23:32:34 darn it, I like being human. I just don't want to be a dead one! 23:32:45 anyway.... I finaly found a colorforth that will run under windows on my machine. What an inpenetrable system... 23:32:53 heh 23:33:31 spent hours trying to figure out how to write a simple hello world. I did manage to change the color of (part of) the screen. And add 2 + 3. 23:33:41 yeah 23:33:50 I remember trying colorforth 23:35:19 at the other end of the spectrum is win32forth. a monster, lots of docs, but seems like a dead comunity. 23:38:05 Hey.... no jvm port? or is midp one of the jvm things? 23:38:28 good question 23:38:54 midp does seem to be written in java... 23:39:10 I guess midp is aka J2ME 23:41:11 Hmmm... bug or wasted time in the midp port looks like. I dunno for sure, I am no java expert... but VM_CALL again... 23:41:14 case VM_CALL: 23:41:14 ip++; rsp++; 23:41:14 address[rsp] = ip++; 23:59:59 --- log: ended retro/10.01.12