00:00:00 --- log: started forth/08.10.13 00:24:14 --- quit: uiu (Read error: 110 (Connection timed out)) 01:00:46 --- join: qFox (i=C00K13S@132pc222.sshunet.nl) joined #forth 01:11:32 --- quit: JasonWoof ("off to bed again") 01:14:11 --- quit: proteusguy (Read error: 110 (Connection timed out)) 01:14:55 --- join: proteusguy (n=proteusg@61.7.144.97) joined #forth 01:18:41 --- quit: ASau (Read error: 110 (Connection timed out)) 01:53:46 --- join: snowrichard (n=richard@12.169.182.169) joined #forth 01:55:50 hello 01:55:57 crc? 01:56:24 --- quit: snowrichard (Client Quit) 02:14:19 --- join: pierre- (n=pierre@89.179.89.191) joined #forth 02:23:54 Privet! 02:24:22 It seems, that I'm going to SPB in near future. 02:52:12 --- join: toutou (n=posteint@41.201.104.124) joined #forth 02:52:28 --- part: toutou left #forth 03:21:38 --- join: snowrichard (n=richard@12.169.182.169) joined #forth 03:23:24 hello 03:25:33 --- quit: snowrichard (Client Quit) 03:34:13 --- quit: ASau` ("ERC Version 5.2 (IRC client for Emacs)") 03:36:02 --- join: ASau (n=user@host214-231-msk.microtest.ru) joined #forth 03:55:08 --- quit: mathrick (Read error: 104 (Connection reset by peer)) 03:55:48 --- join: mathrick (n=mathrick@users177.kollegienet.dk) joined #forth 04:15:07 --- quit: proteusguy (Read error: 110 (Connection timed out)) 04:16:28 --- join: LOOP-HOG (n=jasondam@97-115-89-97.ptld.qwest.net) joined #forth 04:16:33 hi 04:17:01 Good afternoon. 04:17:12 I am ok, how are you? 04:17:26 Fine. 04:17:48 sorry that I don't remember, but do you use Forth much? 04:18:34 I use it enough. 04:18:56 I have a question 04:19:17 --- join: proteusguy (n=proteusg@61.7.144.97) joined #forth 04:19:21 supposed that I wanted to code a custom internet app, but wanted to use Forth instead of PHP, is that possiable? 04:19:37 I would set it up on a unix server 04:20:28 Sure. 04:20:47 CGI is the same for every language. 04:21:20 CGI is how a website communicates to the server 04:21:38 and it doesn't matter then if you are using PHP or Forth or Perl? 04:21:43 Yes. 04:21:44 or even C? 04:21:47 Yes. 04:22:07 There's C-based framework for web applications. 04:22:26 What Forth would you recommend? 04:22:37 Is there a forth wordset to already let me start on this? 04:24:20 Hm. 04:24:28 also I would probably need to access SQL for the database 04:25:08 I wrote UDP service, which saved data in Postgres via libpq. 04:25:32 udp service with a sql server in between? 04:25:43 isnt that a huge bottleneck? 04:25:48 ? 04:25:52 how does Postgres differ from SQL 04:25:56 or am i missunderstanding 04:26:16 net -> udp -> libpq -> Postgres 04:26:25 I mean, how does MySQL differ from postgre SQL ? 04:26:37 oh, its accepting udp packages and saving it in sql. nvm :p 04:26:37 MySQL is worse. 04:26:50 please tell me how so 04:27:39 ASau> in all fairness, that statement seems a bit biased. 04:27:45 but whatever :) 04:27:47 PostgreSQL is like Oracle. 04:28:06 It has means to run huge databases with complex processing. 04:28:37 You can't do many things in MySQL, unless you go for FFI. 04:28:59 For Postgres you have PL/pgSQL. 04:29:03 the choice between mysql and postgres is a matter of context... 04:29:17 Sure. 04:29:41 heh. i keep telling myself not to get involved in this discussion and i keep writing it anyways :p i'll stop 04:29:43 When you could use BDB, but have no competent programmers, 04:29:43 or don't care much, you use MySQL. 04:30:58 It would probably be a few hundred or a few thousand users max 04:31:49 what forth was this written in? 04:31:53 Gforth. 04:32:07 in that case, go with whatever sql server you're most comfortable with. 04:32:16 I can't force myself in studying PFE FFI capabilities. 04:32:40 what facilities does Gforth use to accept CGI? 04:33:01 LOOP-HOG: just read what CGI is. 04:33:02 what does Gforth provide in regards to my problem? 04:33:08 FFI. 04:33:10 Common Gateway Interface 04:33:20 FFI? 04:33:32 How are you going to connect to database? 04:34:13 I don't know yet 04:34:29 http://en.wikipedia.org/wiki/FFI 04:34:46 Choose what fits the context. 04:35:26 that is good advice 04:35:36 http://en.wikipedia.org/wiki/Common_Gateway_Interface 04:35:41 Read what CGI is. 04:36:08 foreign function interface? 04:36:14 Yes. 04:37:22 that gives me some reading to do 04:38:16 I wrote simple reporting tool using /bin/sh, awk and psql to query database. 04:38:36 ok 04:38:45 CGI is easy. 04:38:57 CGI EASY = -1 ( good ) 04:39:24 what about FFI, is that easy too? 04:40:21 That's a bit harder. 04:40:38 If you can wait untill the evening, I may come with actual code. 04:40:51 "Evening" = 8:00pm. 04:40:56 I can be back here this evening. I am getting a little sleepy 04:41:03 so, 8:00pm PST? 04:41:07 MSD. 04:41:14 Mountain 04:41:22 Moscow. 04:41:34 duh 04:41:52 It is in 4,5 hours. 04:42:21 I can take a nap and be back up in 5 hours 04:42:23 is that ok 04:43:03 Well... If that's easier to you, you can be back in any time before midnight MSD. 04:43:29 that would be before 2:00pm my time I think 04:44:07 Or... 04:44:15 You can ask me in mail. 04:44:58 could I get your email address 04:45:10 asau at inbox ru 04:45:44 ok 04:45:46 thanks 04:45:51 I'll log out now 04:45:56 bye' 04:45:59 Bye. 04:46:03 --- part: LOOP-HOG left #forth 04:58:37 ASau: well,you at Moscow? so cool 05:36:18 --- quit: pierre- (Read error: 110 (Connection timed out)) 06:40:47 um. i've got the seaforth thing :D 06:41:42 * qFox pokes segher 06:43:16 What do you mean? 06:43:26 Intellasys evaluation board? 06:44:43 qfox: pong 06:44:45 aye 06:45:02 cool, played with it yet? 06:45:16 well, i decided i'd read the eula first 06:45:26 heh 06:45:32 i'm not about to "open this package and you agree to sell us your soul" 06:45:46 luckily its only one page :p 06:45:50 and a bit 06:46:28 * qFox shuts down metroid 06:48:20 oh good, the max penalthy is $5000. i'm game :p 06:53:51 qFox: EULA said sell your soul? 06:54:53 if you think my soul is worth $5000, then i guess so 06:55:17 ... 06:56:30 oh good, the manual starts from scratch. i needed a refresher course :p 06:56:52 I never look EULA... 06:57:01 trust me, this is my first 06:57:12 :D 06:57:22 apart from "certain" snippets for chrome :p 06:57:26 Bloody 06:57:54 chrome ugly 06:58:08 ugly but fast 06:58:32 V8 is fast 06:59:09 but flash not good, so I stay in firefox 3 06:59:47 oh i hadnt tested that 07:02:48 --- nick: ASau -> ASau` 07:03:23 --- nick: gogonkt_ -> gogonkt 07:09:54 qfox: refresher course... well, seaforth isn't quite ansi forth, heh 07:10:39 nope but its been quite some time since i've done anything in forth 07:10:51 obviously this is another level as well 07:11:55 i guess i should thank you, since i doubt i would have found out about this without you mentioning it :) 07:11:57 so thanks 07:13:47 :-) 07:22:04 --- join: snowrichard (n=richard@12.169.182.169) joined #forth 07:22:32 --- join: pierre- (n=pierre@89.179.89.191) joined #forth 07:22:43 hi pierre- 07:23:21 hello crc 07:25:48 --- quit: snowrichard (Client Quit) 07:37:58 --- quit: proteusguy (Read error: 110 (Connection timed out)) 07:38:38 --- join: proteusguy (n=proteusg@61.7.144.97) joined #forth 07:49:16 --- join: ENKI-][ (n=weechat@c-71-234-190-248.hsd1.ct.comcast.net) joined #forth 07:54:28 --- join: ASau (n=user@193.138.70.52) joined #forth 08:58:56 so far, it seems, i'm working with 8 item stacks (one for each core) and the uh 64 (?) words for each core 08:59:36 there's a data and a return stack (both 8 items..?) and ventureforth offers some tail recursion and other optimizing tricks 08:59:47 i guess that's nothing new though 09:00:04 but i'm only starting at the interesting part now, halfway through the manual :p 09:00:16 (the inter-node communication) 09:00:38 so far, ventureforth seems pretty straightforward, except that it's run-time branching words dont eat their arguments 09:00:53 which is fine, i guess. 09:11:09 i'm just starting learning forth, but i've done a lot of os development in various languages. i'm considering writing a toy OS in forth wherein the dictionary is used for all permanent storage (i.e., words in the dictionary act as commands, files, etc.) 09:11:17 is this something that's doable? 09:12:25 also, should i use something like retroforth to have asm in my forth code, or should i extend the dictionary with asm/c code, or should i just write a forth compiler from scratch? 09:13:33 qfox: you have 10 items on the data stack, and 9 on the return stack. there are separate registers for TOS NOS RTOS (named T S R resp.), and a circular list of eight items for each of data and return stack 09:14:58 you can treat that as if it was a short stack, but you can do evil little tricks as well 09:15:26 oh T and S and R are not part of the stack? 09:15:37 i missunderstood that then, i thought they were copies of the them 09:15:38 exactly 09:15:45 ok, 10 and 9 it is 09:15:49 they are separate registers 09:16:16 well, not really 10 and 9; there's not even _really_ a stack. but you get it, i guess 09:16:25 ENKI-][> it is doable, its up to the programmer how to use dictionary space :) 09:16:43 and i'm not sure how to answer the second question, so i wont. 09:16:43 qFox: cool. i figured as much, but i thought i'd ask 09:17:03 qFox: i'm considering doing versioning that way too, sort of. 09:17:13 segher> yeah :) 09:17:33 it says stacks are implemented as an array though. but i guess that doesnt matter at all from the forth point of view 09:18:01 ENKI-][> how do you mean, versioning? 09:18:14 writing a forth implementation doesn't seem THAT difficult, in comparison to other languages one could try to implement, though i'm sure there are caveats here and there (prolog looked really easy too, until i tried it) 09:18:59 qFox: from what i understand, you can overwrite a word's definition in the dictionary with a colon definition, but it only shadows, and if you remove one definition, you have the older definition still there 09:19:46 qFox: i'm thinking that one can use that for backtracking between versions of functions and files, and maybe choose a specific version 09:20:09 Did you try that? 09:20:21 that said, i can only think of scrolling through the versions by pushing the xt then deleting the word until the word is undefined 09:20:30 ASau: it's just an idea. 09:20:35 I did. 09:20:36 ah, yes, some implementations use words like FENCE to prevent the core from being deleted 09:20:39 ASau: i don't know if they are still done that way 09:20:40 And this isn't easy. 09:20:56 hm 09:21:14 why exactly? using multiple defs of a word just means a specific way if implementing ' 09:21:19 unless i'm overseeing something right now 09:21:23 ASau: you mean versioning, or writing an implementation of forth? 09:21:33 --- join: Quartus (n=neal@CPE001b115d994a-CM001947482b20.cpe.net.cable.rogers.com) joined #forth 09:21:33 --- mode: ChanServ set +o Quartus 09:21:34 Versioning. 09:21:35 qFox: if you redefine a word, it shadows the old definition 09:21:39 as long as ' searches for a word new to old (like it should anyways), thins should be fine? 09:21:42 aye 09:21:52 I suggest not writing yet another implementation. 09:21:54 qFox: i'm wondering if you can choose a specific shadowed definition. 09:22:08 : one 2 ; : one 1 ; one => 1 09:22:19 No, you can't. 09:22:33 well, not directly that i know of 09:22:35 well, "you can" :p 09:22:37 <_<\ 09:22:42 but its not recommended, no 09:22:51 it would require a big slow wrapper, and it wouldn't be transparent anyway 09:23:05 aye, you can have a dictionary layout with a huge header 09:23:12 and do version control there 09:23:19 If you want both versions, use two names. 09:23:27 and have a special tick that searches for version number as well 09:23:32 Unless you use rather inefficient dictionary and know the 09:23:32 way to get specific definition, you can't do that. 09:23:36 or you can have a normal dictionary with shadowing, and copy-push-delete, i'm guessing 09:23:45 And if you do need different versions, use different names. 09:23:49 mm 09:24:18 Or two wordlists, if for some reason I can't see, you badly need to use the same name for two different meanings. 09:24:21 it's just an idea i'm playing with. i imagine it's possible, but probably neither useful nor easy 09:25:08 Quartus: i'm talking about versions of the same meaning. like cvs or svn would let you keep old versions of the same source, or wikipedia allows reversions or views of a specific version 09:25:29 if you really want versioning, i would really go for a word header kind of way 09:25:32 I still can't see the point. 09:25:33 and a special tick 09:25:42 Quartus: i don't think i "badly need" it, but i am considering it 09:25:46 ENKI-][: overloading is herecy here. 09:25:51 ASau: i see 09:25:55 :p 09:26:29 ASau: this is in the context of a standalone forth with the dictionary being the forth image on disk, and the dictionary word definitions taking the place of a file system. 09:26:50 ASau: so some of the words will be code, and some will just return strings. 09:27:24 I don't see the point. 09:27:29 and some of the other oses i've worked on have involved versioned files on odd filesystems 09:27:48 ASau: there isn't a point per-se. i'm just playing around. i don't know that it would be especially useful 09:27:51 so you want to have words act as file handlers or something? 09:27:58 qFox: yeah 09:28:07 right ok 09:28:11 and when you modify a file, you redefine the word 09:28:13 :P 09:28:28 Sound like ancient "loads" word. 09:28:33 mm 09:29:17 well, the idea i guess is that you could use the return stack for piping if you had a pretty good wrapper around "file" words 09:29:33 sort of like unix piping 09:30:01 then the interactive interp would just append .s or whatever and run it at the end of every line 09:30:04 maybe 09:30:18 You're new to Forth, I take it. 09:30:22 yep 09:30:37 Strongly recommend you learn Forth, rather than writing one. 09:30:39 --- join: JasonWoof (n=jason@c-65-96-160-164.hsd1.ma.comcast.net) joined #forth 09:30:39 --- mode: ChanServ set +o JasonWoof 09:31:01 what better way is there to learn a language than write an implementation of it? 09:31:02 :P 09:31:13 Almost any way is better, as regards Forth. 09:31:20 The better way is not writing it. 09:31:21 but yeah, i'm planning to learn it more first 09:32:12 i suppose using a dictionary for a filesystem with piping isn't very practical then? 09:32:27 Yes. 09:32:32 okay 09:32:39 hm 09:33:10 ENKI-][: write a web browser 09:33:19 so i suppose one would need to write a separate filesystem layer, not using the dictionary, and have the dictionary be a file...\ 09:33:30 meh. i'm not big on the web. 09:33:35 ENKI-][: what are you trying to accomplish? 09:33:44 an OS. 09:33:52 just a toy one to learn forth 09:33:53 It is stupid. 09:33:55 Good grief. 09:34:03 easy guys :p 09:34:24 people try to do the weirdest things to learn forth 09:34:29 lol. 09:34:34 i wrote an os to learn D too 09:34:42 you learn a language by reading about it, reading example code and writing applications 09:35:08 i was going to try one for prolog, but i realized i'd have to implement prolog first, so i wrote a distributed filesystem in prolog instead. 09:35:13 ENKI-][: and what can your D OS do? 09:36:04 ASau: it was a toy os. it had keyboard and disk support, a really simple slow filesystem based on hashing, a memory manager, and vga console support with ansi color sequences. 09:36:20 Learn Forth, ENKI-][. It's more challenging and more useful than these other time-wasters you're proposing. 09:36:26 So, not even like CP/M? 09:36:27 ASau: it had support for a filesystem on a ramdisk and on a partition too 09:36:57 ASau: it had it's own executable format based on treewalking, but i never finished that bit. 09:37:11 apparently it was the first os written in D that worked. 09:37:48 i've actually never seen code in D and am not sure whether i actually heard it as being an actual language, next to the joke of A B C so there ought to be a D 09:37:55 but i tend to shy away from doing useful things :P 09:37:58 (not surprising, though) 09:38:29 qFox: d is sort of like java crossed with c, with some elements of lua. it's pretty nice. 09:38:48 hm, i hope certain elements of lua were left out then :p 09:39:02 anyways, do what you want. we're just telling you that there are easier ways of learning forth 09:39:24 everyone eventually writes their own forth, its kind of a forth programmers thing, i guess 09:39:44 qFox: it's mostly a redesign of object orientation in c, since the owner of digitalmars implemented a c++ compiler and decided that c++ was awful and that he could do it better. 09:40:16 I started writing a forth interpreter too early. and as a result, I still don't know forth that well 09:40:23 To paraphrase Heinlein -- writing a toy system is not necessarily something to be ashamed of. But do it in private and wash your hands afterwards. 09:40:34 heh 09:40:35 i doubt that i'd write a forth from scratch first. i'm considering writing an OS in it though, if it's practical to do 09:40:54 lol 09:41:02 thing is, a forth system can be an OS 09:41:06 given the right building blocks 09:41:06 true 09:41:24 hence the order is forth system first, OS second 09:41:24 If you embark on writing anything, let alone an OS, in Forth, before you know Forth well, you will create a huge bolus of stinky code that you actually *should* be ashamed of, with reason. 09:41:36 and will be! 09:41:39 lol 09:41:48 understandably. 09:42:21 it's very easy to write crappy forth code 09:42:23 yep 09:42:26 in that way it's kinda like C++ 09:42:26 so i've heard 09:42:47 that may be the only way in which it's like C++... 09:42:50 Even knowing Forth well, a toy OS as a project is both not difficult, and useless. 09:42:52 :P 09:43:55 well, the other idea i had was a markov chain algorithm that actually interprets tokens as words, and redefines the words to update the token frequency 09:44:10 but i haven't really thought that one through all the way 09:44:14 The above applies to writing a toy Forth, too. 09:44:19 when new to forth I strongly recommend writing code that you can throw away after a few hours at most 09:44:40 Forthlets. heh 09:44:41 once you get the hang of organizing your code in a way that works for forth, you will be very happy not to have to use any code that you wrote before this 09:44:47 they even trademarked it :p 09:44:50 hehe 09:45:16 forth, internally, is pretty simple no? 09:45:33 it's using it that's a more delicate matter 09:45:34 ENKI-][: usually, yes 09:45:48 it's organizing your code that's tricky 09:45:56 so writing a forth implementation in something other than forth isn't too bad, right? 09:45:57 implementing things in a way that keeps everything simple 09:46:01 not making problems for yourself 09:46:30 A crude, incomplete toy Forth can be built in a couple of days by a moderately-competent programmer. Shouldn't be, but can be. 09:46:35 you have to know forth before you can write an implementation 09:46:44 hm 09:47:03 it's simple compared to other languages, but that's not saying much 09:47:09 if i knew how forth was implemented, i'd be closer to knowing forth, i think 09:47:15 You're wrong about that. 09:47:25 ah 09:47:31 knowing the syntax is just better than useless 09:47:36 what you need to know is the coding style that works 09:47:49 the philosophy of forth goes deep 09:48:06 er. i mean, knowing how it works on the back end should give a certain amount of perspective useful in trying to write stuff in it 09:48:07 There's no philosophy behind Forth itself. 09:48:13 it's pointless to just learn another syntax, unless you have to fix someone else's code that uses that syntax 09:48:19 Common laws apply. 09:48:24 Nothing special. 09:48:45 ASau: forth encourages and makes possible a very different way of coding from most other languages 09:48:54 Knowing how your particular toy Forth works internally will only lead you down the path of thinking that Forth must be implemented that way, and making semantic decisions based on whatever provincial choices were made in your toy. 09:49:08 JasonWoof: I don't know "most others", which ones do you refer to? 09:49:47 JasonWoof: does, e.g., SML count? 09:49:48 ASau: http://en.wikipedia.org/wiki/List_of_programming_languages 09:49:52 Quartus: good point 09:50:03 ASau: sml? or smil? 09:50:08 SML. 09:50:10 (i've never heard of sml) 09:50:26 if you don't know how to code in forth, the thing forth has to offer you is to learn to code in forth 09:50:31 it will stretch your brain 09:50:43 that's always a good thing :-) 09:50:45 you'll be a better programmer for it even if you never use forth again 09:51:01 Any kid can write a toy Forth, or toy OS. The value in either is minimal at best. 09:51:07 * ENKI-][ is a fan of prolog and brainfuck as well, so forth is a good addition 09:51:13 writing an os or intperpreter with a forth-like syntax without really knowing how to code in forth may not have this benefit 09:51:36 prolog is not a language :( 09:51:40 you'll probably miss the benefit in fact 09:52:00 its a depth first search algorithm, disguised in logic. 09:52:02 qFox: declarative languages are still languages. 09:52:10 qFox: but yeah, it is super slow 09:52:19 i wouldnt say that. i just dislike it :p 09:52:30 prolog is nice mainly because it is bidirectional by nature 09:52:33 qFox: it isn't "depth-first". 09:52:36 but the speed sucks 09:52:48 i thought it was depth? is it breadth? 09:52:50 qFox: you should learn it better. 09:53:03 Both wrong. 09:53:09 ehr... 09:53:21 Refer to Horn clauses. 09:53:41 i'm afraid its been too long to remember what htye were, although i do remember the term 09:53:48 It is parallelizable under some conditions. 09:53:53 mm 09:54:23 Contrary to Forth, which is tighted to single stack, 09:54:34 a single one? 09:54:38 unless you provide SIMD operations. 09:54:39 <_< 09:54:52 i thought the data stack and return stack were separate? 09:54:56 ENKI-][> you cna have as many stacks as you want :) 09:55:04 but 10 stacks is not recommended or anything 09:55:08 lol 09:55:34 thing is, there is no groundwork of how a forth should look like 09:55:40 What words operate on return stack? 09:55:46 i mean, there are things we mostly agree on 09:55:53 but forth is more free than that 09:55:57 i don't know. i just remember that the return stack is supposed to be separate 09:56:00 Try naming them all and you'll understand, what I mean. 09:58:08 ciforht used a data stack, return stack, dictionary stack, float stack and i think a fifth one 09:58:41 sheesh 09:59:03 and something called denotations, which is really confusing until you get the hang of it 09:59:10 Standard Forth has a data stack, and a return stack. There is also, conceptually, a control-flow stack, though the data stack is normally used for it. There is also an optional floating-point stack. 09:59:38 I would like to have two cell stacks (data and return), two float stacks, and loop stack 10:02:10 two float? 10:02:20 Yes. 10:03:08 why...? 10:06:00 --- join: LOOP-HOG (n=jasondam@97-115-89-97.ptld.qwest.net) joined #forth 10:06:05 hi 10:06:12 Ah! 10:06:16 :) 10:06:59 I have never installed GForth before 10:07:12 There are a few binaries, but would I use that? 10:07:22 or am I supposed to compile using C? 10:07:32 I build it from source. 10:08:03 what C do you use? 10:09:20 Stock one, which comes in base system. 10:09:45 ? 10:10:11 There's only one C compiler in base system. 10:10:16 ok 10:10:25 that makes sense, unix is C based 10:11:10 AFAIR, gforth relies on GNU extensions. 10:11:38 Which come since 2.6 or 2.8, or maybe 2.9. 10:11:57 I don't care, this is very distant past 10:13:13 I'm sure that I would need command line access 10:14:02 how fast is Gforth compared to PHP? 10:14:06 LOOP-HOG: command line access? 10:14:21 LOOP-HOG: depends what you're doing 10:14:32 LOOP-HOG: gforth is quite fast, but also low-level 10:14:58 wouldn't gforth also be more secure than PHP? 10:15:12 in what way? 10:15:22 they both allow you to execute system commands 10:15:29 gforth can easily crash and execute data 10:15:54 --- join: snowrichard (n=richard@12.169.182.169) joined #forth 10:15:56 my programs rarely crash 10:16:02 in forth? 10:16:06 hi loop 10:16:24 hi crc 10:16:27 snowrichard: I looked at your language definition yesterday 10:16:35 I'm just thinking about installing gforth onto my website server so that I can play around with CGI with it 10:16:47 the code has been updated as of this morning the forward defines work now 10:16:48 LOOP-HOG: fun 10:17:03 LOOP-HOG: if you don't have a shell account, you'll obviously have to grab a binary 10:17:20 snowrichard: I just read your syntax page you posted 10:17:20 what I said command line interface, I ment to say shell account 10:17:26