00:00:00 --- log: started forth/05.09.27 00:24:19 --- quit: Quartus (Remote closed the connection) 00:24:22 --- join: Quartus_ (n=trailer@ansuz.pair.com) joined #forth 00:34:26 --- quit: Quartus_ (Remote closed the connection) 01:48:53 --- quit: YoyoFreeBSD_ (Read error: 104 (Connection reset by peer)) 03:02:45 --- join: rostling (n=rostling@gjalp.nada.kth.se) joined #forth 03:03:49 Forth: one language, but a surprising number of incompatible and bitterly fighting fractions. 03:09:00 * rostling reads about prolog. 03:38:15 Back pain, going home. See you. 03:38:18 --- quit: rostling ("leaving") 04:50:37 --- join: amca (n=plump@as-bri-3-92.ozonline.com.au) joined #forth 05:44:08 --- join: PoppaVic (n=pete@0-1pool73-210.nas24.chicago4.il.us.da.qwest.net) joined #forth 05:44:25 G'day 05:45:16 Gday 05:45:34 How goes it? 05:45:40 Not too bad 05:45:45 Yourself? 05:45:45 Good 05:46:01 Started working up some notes/doc yesterday. 05:46:16 For fovm? 05:46:40 yah, I feel it might be time to drop back, refocus, and punt. 05:46:55 * amca nods 05:47:07 So you arent all tangled up in implementation issues 05:47:21 How are they going? 05:47:25 Right. I've got so many issues cookin', I can't get any further ahead 05:48:33 I've still got to create opcodes/tables for essential strings and i/o support, and I don't want to run too deep into the woods on this. 05:48:59 * amca nods 05:49:27 Just starting with the essential/core stuff and adding the optional/extened stuff later? 05:49:48 Well, it's a trifle more cross-grain than that. 05:50:18 Oh? 05:50:36 The core is what you ALWAYS need.But, almost everything is sorta' modular/optional 05:51:20 But.. The fun part is deciding whom belongs where, when. 05:52:30 And, I just *know* I'm gonna' scream, because I want wchar_t for message-type strings, but ascii for headers/filenames, etc. 05:52:31 Well IMHO you only have the bare minimum as core, and only add other stuff to it where necessary for efficiency reasons 05:53:03 Why not wchar_t for both and convert headers to ascii? 05:53:39 Hmm.. It is my understanding that some wchar_t can not be xlated into ascii. I believe you may be right, though. 05:54:31 Well the wchar's that are over ord(256) can create an error/exception 05:54:40 Or be ignored 05:54:40 I've never worked with wchar_t or their functions, so I am seriously nervous. 05:56:05 Maybe you ought to make a toy project that uses wchar for the datatype first for the experience? 05:56:29 Yeah, perhaps. 05:57:23 I also want to select the apropos string-support, and... I don't like this "counted strings" assumption, which we have all suffered forever. 05:57:48 "counted strings" assumption? 05:58:03 Not exactly sure how to resolve it, but I would PREFER to use string "objects" 05:58:32 yeah, like... : count ( a --- a' len ) .... 05:59:00 they always seem to _assume_ it's bytes/chars and the count is as well 06:00:40 well you just need to create a new string data type dont you? 06:00:53 yes, I do. 06:00:58 Encapsulate it so it doesnt get used improperly 06:01:27 I've also considered other 'types', but I believe that is almost a complete task in itself. 06:02:08 hello all. 06:02:31 I get to thinking it over and tend to conclude we should allow folks to shoot-off a foot, or melt their faces and be done with it. 06:02:34 hi Ray_work 06:03:21 breakfast on me. 06:03:30 someone get a towel 06:03:50 Towel? 06:04:32 amca: did you follow the source pretty well? Any issues to address or major changes to consider? 06:04:47 napkin? 06:04:56 Ray_work: or a mop ;-) 06:05:13 industrial vac 06:05:22 shop-vac 06:05:28 ya 06:05:45 PoppaVic: Sorry Ive been not looking after myself so I havent been doing anything with my brain of late 06:05:59 ahh, ok. No prob. 06:08:52 --- join: Robert (n=robert@unaffiliated/robert) joined #forth 06:10:54 Ahh! Contamination! 06:10:58 * amca points at Robert 06:11:28 * Robert cries a little. 06:11:37 * amca comforts Robert 06:11:48 OK, back pains. Be BACK later. (ha-ha) 06:12:01 um, ok 06:16:03 Robert: How is the low level vm going? 06:16:20 Which? 06:16:48 The one you mentioned the other day 06:17:10 PoppaVic: You will let me know when you have got enough docs to show me wont you 06:17:13 ? 06:17:21 amca: will email it. 06:17:33 tnx 06:18:03 Robert: You asked me if I knew any asm 06:18:17 sure. If the doc changes anything in the source, it'll be the tarball. 06:19:16 cool :) 06:25:00 Robert: Actually it was a minimal Forth, not a vm. My mistake 06:25:06 2005 Sep 27 01:10:34 Was just thinking about the potential interest in a minimal (not speed-optimised or comprehensive) Forth for Linux. 06:29:32 Robert! how are you? 06:29:51 oh missed him :) 06:39:57 * PoppaVic harumphs and fumes... 06:47:20 Ray_work: Hey Ray. 06:47:34 Trying to get some rest, so my back won't get worse. 06:47:36 wb, Robert 06:47:52 back prob's? 06:47:58 ouch 06:48:02 amca: I wouldn't call that a vm, but I haven't done anything. 06:48:09 Yeah, ouch... and bbl 06:48:16 Robert: Yeah, I got mixed up. Good luck 06:48:19 okay 07:00:02 --- join: zoly (n=l@tor/session/x-7c8cfecdddf40802) joined #forth 07:00:32 g'day 07:00:49 Hi. 07:02:07 god dag rob_ert 07:02:35 Tach, spoiler. 07:08:01 --- join: JasonWoof (n=jason@pdpc/supporter/student/Herkamire) joined #forth 07:08:01 --- mode: ChanServ set +o JasonWoof 07:08:21 hey jas 07:08:41 --- join: derv0 (n=derv0@proxy1.nscl.msu.edu) joined #forth 07:10:21 --- part: derv0 left #forth 07:14:28 hey 07:15:17 --- join: sproingie (i=foobar@64-121-2-59.c3-0.sfrn-ubr8.sfrn.ca.cable.rcn.com) joined #forth 07:15:27 --- quit: amca ("d34d") 07:16:01 calling-conventions are a real PITA 07:17:22 designing them? 07:17:30 hmm.. If we are passing around the vm-context, then... Are we not expecting no more than 1 FFI format and that any "other language" simply obeys OUR form? 07:17:55 "no "// 07:20:06 JasonWoof: they seem to be "native-code" artifacts 100%. 07:21:37 sooo... hmm. The 'syscall' (and my own 'ccall') opcodes seem to be THE artificial FFI, mostly related to the FFI ideas. 07:22:56 I get the feeling it's more of "syscall, sysreturned" and so forth. 07:23:37 or maybe even "sysinit, syscall, sysdone" 07:24:19 Not sure... It's odd, and perhaps the opcode support (native) is entirely responsible? 07:25:32 hmm.. Yeah. That could be it. Except, we have to trust in the IP and RP being correct.. I can't see a way around that. 07:26:22 JasonWoof: do you find the 'syscall' opcode runs a trifle slower than normal opcodes? 07:27:23 Seems like it would mean about 2-3 or 4 levels of nesting/calls/interrupts 07:31:06 whoa... Brainfart. I just had a pseudo epiphany.. 07:33:23 --- part: zoly left #forth 07:33:27 JasonWoof: you code in asm, right? You use what C considers "struct", right? 07:33:46 and even struct* 07:33:48 ? 07:40:54 rrr 07:41:02 heh 07:41:02 man there's a lot of confusing stuff going on here 07:41:33 fovium does not have an interface to linux syscalls generally 07:41:57 the emulator implements a few "syscalls" 07:42:03 ahhh 07:42:07 which the guest system asks for by number 07:42:08 I wondered about that. 07:42:29 currently, the syscalls are: exit emit key 07:42:59 ahh, so by "syscall", you are thinking of "fovium-implementation" 07:43:14 to call a syscall, you put it's argument(s) (if any) on the stack, then push it's syscall number, then execute the syscall opcode 07:43:30 ahhhhhhhhhhh 07:43:36 well, that's the only thing we have that's even vaguely related to syscalls 07:43:37 I misunderstood the use! 07:43:48 ah-HAH! 07:43:53 we don't have a syscall in the sense of linux syscalls 07:44:04 yeah, no FFI support 07:44:11 we don't have ffi or any possible way to make it 07:44:23 (without extenting fovium) 07:44:44 ahhh-haaaahhhh 07:44:53 I was spinning in circles 07:45:05 the only way a fovium guest can interact with the world outside it's 1MB memory block, is those three syscalls: exit emit key 07:45:26 this will be expanded at least a little 07:45:37 oh, man.. I need to rethink my func-calls. 07:45:44 eg adding some sort of access to mass storage 07:45:54 yeah, you'd need to 07:46:41 xxx constant sys-emit 07:46:44 BUT, what I was glaring it - for weeks - was func-returns and argument-prep. The former can be empty/void, the latter can live 100% inside the VM 07:46:48 yyy constant op-syscall 07:47:04 oh, man. 07:47:11 : emit sys-emit lit op-syscall op, ; 07:47:35 *sigh* THis can slightly complicate the implementation, but vastly simplifies a whole lot of other shit. 07:48:23 well, making interfaces to things with loads of parameters is ugly 07:48:28 The kicker ends up being calls/returns and "use the VM stacks!" 07:48:36 you can either suck it up and deal with it, or make your own api 07:48:49 Yep, I can see it. 07:49:12 I've seen various solutions 07:49:32 yah, but it took me awhile to note the disconnect 07:50:04 thing is, even if you do have access to something like linux syscalls, or an FFI to a library or something 07:50:07 it's still a pain 07:50:43 It always will be. BUT, we never pass other than the VM* in, and we don't *have* to accept a return. 07:51:00 we could use a member of the struct. 07:51:20 oh, I forgot you were saying stuff about passing the VM around... I don't understand this at all 07:51:43 why would you pass the vm, or reference to it to anything, and if you're passing it, what are you passing it to? 07:51:50 ahh, it's almost trivial. Asm folks may use the stack, but it's more likely they'd use a reg 07:52:17 JasonWoof: the opcodes. Every opcode can adjust whatever w/i the struct. 07:52:26 ppc linux syscalls you put the arguments in registers 07:52:45 what struct? 07:52:47 right, but there you mean "interface to OS" 07:53:04 what? 07:53:16 the struct I call Fovm_t - it's a plain old structure where we fake registers. 07:53:23 in my current project, I don't plan to have an interface to the host OS 07:53:28 right 07:53:32 I figured that 07:53:38 largely because I want it to run on wildly different host OSes 07:53:45 sure. 07:53:59 I misinterpreted your "syscall" 07:54:12 but in herkforth, I do... I wrote an interface for the ppc linux host, and openfirmware 07:54:28 yeah. it's confusing 07:54:34 I wasn't sure at first which way it would be 07:54:36 yeah, hell yes 07:54:58 You think "my space" and "their space" - and then it tends to mesh, merge and bollix 07:55:13 what? 07:55:30 you're speaking it tounges again 07:55:41 Anytime you need to think of "standalone" and then "application", shit gets confuzzled 07:56:15 ok, I feel better and thank you muchly... 07:56:36 Whew. I thought I was gonna' pull out all my hair or skull-thump the wall 08:00:39 * PoppaVic does the "tentative happy-dance" 08:30:34 --- join: Quartus (n=trailer@ansuz.pair.com) joined #forth 08:30:41 Hey all. 08:31:39 hi Quartus :) 08:31:45 wb 08:36:23 hi Quartus 08:47:58 --- quit: PoppaVic ("brb - isp sweep-beater") 08:49:23 --- join: PoppaVic (n=pete@0-1pool67-107.nas22.chicago4.il.us.da.qwest.net) joined #forth 08:49:38 damned isp 09:00:16 What's up? 09:23:58 @work 09:24:04 with you? 09:26:43 * Robert is eating an apple. 09:39:45 Working too, Raystm2. 09:39:49 oops. Ray_work 09:41:45 --- join: virl (n=hmpf@chello062178085149.1.12.vie.surfer.at) joined #forth 09:41:54 hi 09:44:16 I'm thinking "syscall" may be a huge misnomer/mistake 09:44:33 virl: Hi 09:44:35 PoppaVic: Why? 09:44:58 Robert: as described to me, it means "an engine handler" 09:45:31 ..would we not be better with a table/voc of defer-type words? 09:46:17 You mean ideally, or in practice? 09:46:40 in which context is syscall bad? 09:46:44 I'm not sure if x86 could conveniently handle access checking with large numbers of functions. 09:46:53 Not sure. Working up this new engine - it doesn't seem "consistent" 09:47:37 x86 is a number of 8086/80386 spin-offs, and should not matter. 09:47:57 That's why I asked if I should take reality into account. 09:48:16 "syscall", as JasonWoof described, means "core and final access" 09:48:55 so, we are vectoring "opcodes" to "functions" specific to - apparently, the engine itself. 09:49:22 Robert, could you translate to me what he means? 09:49:26 I'm not sure those justify the term "syscall". 09:49:46 or whats so bad about a syscall interface. 09:49:54 ..and I'm well aware that some platforms implementing ARE the "OS" 09:50:51 maybe we need "mustcall" or "mustexec", etc. Because it is not a "syscall" 09:51:20 hell, it may even be a true "defer" 09:52:50 Basically, the way to overload is "defer" - I'd love alternatives. But, "at this time", deferral means the pointed-at shit can be asm or pcode or whatever. 09:54:49 This is the sorta' stuff that "secondary" and "primitive" always used to imply. 09:56:13 Robert: welcome to my world ;-) 09:56:42 It's scary. 09:56:48 Yes, it is. 09:57:03 Robert: could you explain me what he means? 09:57:20 You almost need to think-within, work-within a "metacompiler". 09:57:21 I think you should ask him instead. 09:57:52 Or maybe even a "meta-assembler" 09:58:14 (which is nearly was where I was 8 months ago) 09:58:22 Robert: so you don 09:58:28 Robert: so you don't know it? 09:59:50 I was told it was foolish and insensible. So, I upgraded my plan. Basically, the same problems occur - but higher up. 10:00:27 and, yeah: the problems are also fewer - but still exist. 10:01:38 Hmm... I may need a nap... I can see a simple solution to several issues. 10:01:42 Not really. 10:01:55 I understand how the concept can seem a bit...overkill. 10:02:05 You can't help it. 10:02:53 for me it seems like, that he want something simple but doesn't simplifies it. 10:03:01 Every CPU is a special-issue; every Implementation-language relies on the CPU, "the OS"; and data-formats. 10:03:33 so, we try to limit actual access to the CPU-space and OS 10:05:01 looks like java 10:05:38 but ok when doing only an autotools replacement, it doesn't matter. 10:05:59 what you DO need is a way to stay internal, and also to dip outside to assorted [languages+os+cpu:=platforms]. 10:06:58 I'm pretty sure forthish has a better chance than lisp. But, I'd be hard-pressed to suggest/tell "why". 10:07:58 maybe we need a really primitive/portable "pseudo assembler"? No idea, 10:08:33 anyway, lemme call it a Knight and cogitate it. 10:08:37 --- quit: PoppaVic ("Pulls the pin...") 12:02:53 does anybody uses 'wink' sometimes? and if yes, how does it looks like? 12:05:45 what's wink? 12:10:20 an old pygmy forth dos word processor, I don't have dos or pygmy forth here so I can't test it. 12:19:34 Never tried it. 15:02:33 Hey again. 15:10:00 Hi. 15:12:57 hi.. 15:13:03 how is glypher? 15:13:27 What? 15:14:43 hi all 15:14:47 hi crc 15:14:52 Hi, crc. 15:15:14 * virl thinks that it looks fine, but it's also one of his won't-buy-because-don't-like-buying-over-inet forths 15:15:31 Oh. 15:15:33 virl, you can buy via mail-order from most places. 15:15:51 * Robert is too used to Free Software. :0 15:15:56 Glypher isn't even for sale yet... it's still in beta 15:16:10 Are you developing it? 15:16:22 when I say not buying over inet I mean it so. 15:16:30 virl, what? 15:16:32 Robert: no 15:16:42 Roger Levey is developing it 15:19:15 * crc watches Glypher's development since it's based on RetroForth 7 (with some of the improvements I made in 8) 15:19:55 Ah. 15:20:05 What license is Retro? 15:20:50 and how is it moving? could be it THE-REVIVAL-OF-THE-LONG-LONG-DEAD forth programming language? 15:21:13 Robert, it's freeware, licensed under creative commons. 15:21:31 Which, Retro or Glypher? 15:21:32 Retro is public domain 15:21:34 freeware in the sense of you can do everything with it 15:21:38 virl, I wonder why you're here if you think Forth is long dead? 15:21:59 Glyper's final license has yet to be determined 15:22:02 Quartus: There are quite a few historians out there. 15:22:02 uh my god... 15:23:07 Forth is in extremely active use; I'm not sure there's ever been a time since its inception that it hasn't been. It did have a mainstream presence in the early 80s, and doesn't as much now, but that hasn't affected its continued use. 15:26:41 "Dead" implies that it's not being used; that's demonstrably not true. 15:26:45 Quartus, it's forth is living, but that are only some movements of a zombie. forth is great and that's why I'm here, but I know that forth is somehow dead, little community, lots of very old implementations which are not used today anymore, another implementations which are bloatware(W32forth) and so unusable, then lots of magazines where only someone can find it's dead bodies in archives, because they aren't actively published anymore and 15:28:42 virl, you have freely admitted that you have no interest in commercial Forths; but that's where the action is. It is a mistake to judge the liveliness of Forth based solely on freeware implementations. 15:29:27 In terms of magazines, there's no Forth-specific one now that I know of -- but how many languages are there that are directly represented by magazines? C, and C++, probably; Perl possibly (though I think the Perl journal is dead or resting now -- does that mean Perl is dead?) 15:29:51 Much of that is market-driven. The web is the new forum for publishing technical content. Magazines are 'old hat'. 15:30:01 Python and Ruby are dancing on Perl's grave. ;) 15:30:09 no, but there are a lot of wikis, howtos about perl... 15:30:39 Lots about Forth on the web, too. 15:30:42 and I don't know why you are so defensive 15:30:44 Is OpenBIOS the currently most prominent example of Forth use? 15:31:01 Everyone I know outside the community who know about Forth, know it from there. 15:31:02 probably 15:31:10 Or possibly because they were exposed to it in the 70s. 15:31:26 virl, defensive? I'm not defending myself, I'm disagreeing with you. 15:33:23 it's a fact that forth has a small community. 15:33:51 What do you mean by 'community'? The people who participate on comp.lang.forth? 15:34:46 not only 15:35:03 Then how are you establishing this fact? 15:35:55 I wouldn't say it's small. 15:36:03 Considering how I run into random people who use it. 15:36:21 Robert, I agree. Lots of people use it who don't post to Usenet, certainly. 15:36:24 OK - not enough to (statistically) base such a conclusion on. 15:36:28 I wouldn't say it's community is like the lisp community in it's size. 15:36:33 Quartus: I don't, for one. :) 15:36:38 --- join: Amanita_Virosa (n=jenni@CPE0000e812679b-CM000a7362da55.cpe.net.cable.rogers.com) joined #forth 15:36:45 virl: But lisp is promoted by RMS. ;) 15:36:47 virl, your opinion is all well and good, but what do you base it on? 15:36:49 Hi. 15:36:55 Hi Amanita_Virosa. 15:37:28 i'm not synced yet... i can't reply until i sync. 15:37:30 oh there we go 15:37:34 synced in here 15:41:01 websites, broken links, not actively developed forth systems, forth systems < commercial forths -> bad -> worse when they are more innovative/faster/etc., specialized computer magazines, asking people about forth, counting forums/newsgroups, etc. 15:42:23 somehow it seems that forth is at a turning point where it will get of it's long sleep and will wipeout hopefully some bastards. 15:43:30 virl, if you're going to cite 'facts' about the size of a community relative to the size of another, you might want to have some numbers to show. And I don't understand what you mean by "forth systems < commercial forths -> bad -> worse when they are more innovative/faster/etc." 15:45:32 I subjectively dislike commercial forth systems, when there isn't anything around which has ca. the same power and is free. 15:46:02 The stats are badly skewed by what is taught in school. C, and now Java, generate a huge amount of web traffic from students undergoing the questionable process of CS education. 15:46:27 If you were to measure the success of a language by, say, number of programs in existence -- I think COBOL would reign supreme, followed by Fortran. 15:47:08 Hey! 15:47:10 virl, if free is your only guiding light, then what is stopping you from building a powerful system and giving it away? 15:47:14 I'm doing Prolog homework. 15:47:27 (just to prove that CS education is about obscure languages as well). 15:47:40 Robert, sure, but you take my point I'm certain. 15:48:04 Of course. For every line of Prolog there are a hundred of Java. 15:48:47 lol 15:48:50 oh my. 15:49:11 i feel like i'm in never never land 15:50:16 Quartus, ok, you have a problem with it, because you want to make money with it. 15:50:48 No, virl. I sell a Forth system, yes -- and because of that I am more aware of the nature, size, and growth of the Forth community than you are. 15:53:16 Thanks Quartus, that you need to show us, how stupid I'm, I have a bad day and wow, god must love me, because he send me this wise discussions loving angel! 15:53:40 I'm trying to do that, virl. 15:53:55 ... 15:54:15 how the heck do you sell a Forth system? 15:54:24 i have a closed Forth system that i'd sell if i knew how, lol. 15:54:27 and now there comes the point where say 'You are f***ing a******!' 15:54:56 Amanita_Virosa, you sell it the same way you sell any software -- compiler or otherwise. 15:55:25 that's really an uninspired answer. 15:55:28 i mean, who buys it? 15:55:56 It wasn't that inspirational a question. :) Who buys it? People who want to use it. (I get the feeling you'll think that's uninspired, too. ) 15:56:11 * Amanita_Virosa shakes her head sadly. 15:56:48 * virl begins to dislike Quartus 15:57:01 My customers are from a number of different walks of life; the only thing they all share in common is a desire to write Palm apps on-board the device itself. 15:57:21 do you have a lot of sales? 15:58:10 Amanita_Virosa, yes -- quite a few, over the past few years. A substantial base of developers, both previous Forth developers and new ones. 15:58:50 quite a few as in... "it's a fun hobby" or quite a few as in "it pays the bills"? 15:59:20 It's my key income stream presently, Amanita_Virosa, though I do take on contract work now and again. It's market-driven, so it fluctuates. 15:59:28 ah, i see. 15:59:52 that's actually somewhat impressive. 16:00:20 It's taken quite a lot of work, but fortunately it's something I enjoy. 16:01:25 i'm actually far more impressed by the business skills than whatever programming skill it took. 16:01:47 Oh, I wasn't even thinking of the programming skills. The heart of it is support and customer relations, for sure. 16:02:20 Having a solid product is just necessary, at that point, or the support will either overwhelm you, or alternately, the customers will go away. Both are bad. :) 16:02:30 * Amanita_Virosa nods 16:02:44 sure... but in my experience, the solid product is the easy part. 16:02:58 i have a whole lot of closed software that's not met its market potential 16:03:29 I think so too. Keeping it constrained can be challenging; knowing what to put in, and what to keep out, and maintaining that border in the face of customer requests/demands -- that's a challenge. 16:03:42 sure... 16:03:51 but if it came to that, you could just modularize 16:04:05 again, though, the software part isn't the hard part, at least, not if you're even half-decent as a programmer 16:04:30 horray Quartus is business man with a tube of money horray. 16:04:34 i wrote a useful 3D editing program in a weekend... so i'm not too scared of that part. but i'm currently completely devoid of the business skills. 16:04:47 There is a tremendous difference between software that's good enough for the developer, and software that's good enough somebody else, though. And documentation is a steep hill to climb, too. 16:05:42 perhaps. 16:05:46 And once you've got those things in place, there's marketing. 16:05:46 forth is dead; some buys it; but for a big-bang there is now a time 16:06:03 marketing is the hard part. 16:06:48 Yup. It's all kind of a juggling act. The solid product doesn't stay static, either, and you have to be sure changes don't impact the existing customers unduly, and perform full regression testing to make sure you haven't broken what was previously solid. 16:07:07 sure... but that's all stuff that any good developer already knows 16:07:38 and not the part that interests me, even. i write a lot of code that i use on a daily basis... i'm not just a tinker. 16:07:56 at the moment, my programming and business skills are drastically out of step 16:07:58 and that's the problem. 16:08:20 if you're making a living off of a program that's within my problem domain, so to speak, then you're someone i want to ask about how you did it. 16:08:26 --- quit: virl ("Client Exiting") 16:11:25 so how did you do it, lol? or more correctly... assume $software already exists... how does one make money with it? 16:11:50 sorry; phone... back momentarily 16:11:56 sure 16:14:58 Straightforward process -- set up a relationship with an order-processing outfit, package the software, put it up on the web, and market. 16:15:17 order-processing outfit? 16:15:23 For instance, RegSoft. 16:15:30 (I use them.) 16:15:36 ah okay 16:15:43 at the moment, i have a custom system built around paypal. 16:15:49 That might work too. 16:15:58 * Amanita_Virosa nods. 16:16:03 i guess i'm down to the marketing 16:16:12 The big hurdle. 16:16:26 yes, i get that sense. 16:16:34 i've seen Real Crap marketted successfully for a grand a copy 16:16:42 and good software sit idly on a shelf. 16:18:09 Agreed. 16:19:50 what sort of marketing did you use? 16:19:57 your product is easily as niche as one that i've come up with... 16:20:00 or i'd think it is, anyway 16:20:12 but i've only had one serious inquiry so far 16:22:49 brb 16:23:28 sure 16:25:47 I market in a variety of ways -- paid, unpaid, word-of-mouth (Usenet, etc.). 16:26:52 * Amanita_Virosa nods 16:26:57 ah okay, i see. 16:28:18 Banners, Google AdWords, etc. 16:28:21 * Amanita_Virosa nods 16:28:23 interesting. 16:45:05 Usenet can be a good channel. You can make ANN: style announcements, but otherwise just cooperating with ongoing discussions can raise awareness. 16:46:45 What's your product, Amanita_Virosa? 16:47:03 at the moment, i have a program that edits character poses 16:47:44 Lends itself to visual demonstration, I imagine. 16:48:02 it does, yes. 16:48:11 http://caladan.nanosoft.ca/c4/software/posemaker.php 16:48:42 Marketing is obviously two things: getting the eyeballs, and making the sale. So if you've got something that's amenable to visual demos, that's ideal for making the sale, and probably for visual ads, too. 16:48:49 * Amanita_Virosa nods. 16:48:50 indeed 16:49:31 Just yesterday I added an animation to my Forth product page at http://quartus.net/products/forth -- I'm already getting positive feedback about it. 16:50:20 very much so 16:50:23 One other recent change I made to that product page was to change it from 'here's what the product can do' to 'here's what you can do with the product'. Seems to help. 16:50:25 your animation is quite nifty. 16:50:29 * Amanita_Virosa nods 16:50:31 Thanks! :) 16:51:01 I'm reading your product description, and it reads like a blog -- I suggest turning it into a quick list of what amazing super-powers I will have at my disposal if I spend the bucks. 16:51:04 i should put up an animation for my software... and maybe redo my business page. 16:51:09 and put it on my business page. 16:51:18 well, that's the problem... it *is* on my blog. 16:51:20 lol 16:51:41 Heh. That accounts for it. :) But you see what I mean. The product landing page should do its best to close the sale. 16:51:48 indeed 16:51:52 Also, if possible, evaluation versions -- timed or otherwise -- are great tools. 16:51:53 i'll redo my business page then. 16:51:57 * Amanita_Virosa nods 16:52:05 this particular one is easy to write an evaluation version of 16:52:17 just disable file save and it's quite useless :P 16:52:18 Depending on the product, a limited 'lite' vs. a 'professional' can be a good course. 16:52:50 * Amanita_Virosa nods 16:52:59 Tutorial documentation, web-based walk-throughs, and animations make great selling tools. 16:53:08 * Amanita_Virosa nods 16:53:10 Testimonials are under-used, too. 16:53:14 indeed. 16:53:23