open source pouet
category: residue [glöplog]
consider this post me volunteering publicly to do grunt work to help with the revamp effort if it ends up being useful.
hm, is there some way to sort BBS topics that doesn't result in having to plow through zillions of "general" tagged category postings to get to "code" tagged category postings that I'm missing?
reed: woah - Ted Danson really let himself go.
tomaes, the attitude that only people who already possess know-how to watch times are entitled to do so is just fucked up.
technical info for how the demo can be watched doesn't need to be a necessarity for a prod to be in db, but it would really help a lot in the future when people find something they want to see on the actual hardware.
and really, for example when i try to watch demos on my a600, there's no way to know before trying if it'll work or not, it shouldn't be like that.
technical info for how the demo can be watched doesn't need to be a necessarity for a prod to be in db, but it would really help a lot in the future when people find something they want to see on the actual hardware.
and really, for example when i try to watch demos on my a600, there's no way to know before trying if it'll work or not, it shouldn't be like that.
one thing that just crossed my mind: An "official" interchange format for "prods" would be cool.
I mean: There are other scene sites besides pouet and sharing the info to avoid duplicate efforts would make sense, right?
("bloated" XML would be fine for that purpose, I guess) (think of it as some kind of machine-readable .nfo)
I mean: There are other scene sites besides pouet and sharing the info to avoid duplicate efforts would make sense, right?
("bloated" XML would be fine for that purpose, I guess) (think of it as some kind of machine-readable .nfo)
xyz: xnfo, already implemented by Pouet. Only solves half the problem, though... sites still need to know each others' ID numbers for sceners, parties and so on in order to match data up correctly (the '5 different sceners called Abyss' problem again).
gasman: interesting but I stopped reading after "To make it more concrete, here is a valid example of what Xnfo looks like" (crap) (sry:))
also it's from 2003..
I'm not interested in over-engineering it, I just picture a machine *and* human-readable text file that states authors, involvements in the prod, general diz, how-to-run info (including hw/sw requirements), ASCII/ANSI logos (optional), ..
regarding the '5 different sceners called Abyss' problem: I do not really see it as a problem. We are not trying to police-and-track-down every one of us, do we ?
If a demo-contributor want's to have her/his real name / global unique ID known to everyone it should be her/his choice, not a quest-to-find-out for the rest of us ;)
also it's from 2003..
I'm not interested in over-engineering it, I just picture a machine *and* human-readable text file that states authors, involvements in the prod, general diz, how-to-run info (including hw/sw requirements), ASCII/ANSI logos (optional), ..
regarding the '5 different sceners called Abyss' problem: I do not really see it as a problem. We are not trying to police-and-track-down every one of us, do we ?
If a demo-contributor want's to have her/his real name / global unique ID known to everyone it should be her/his choice, not a quest-to-find-out for the rest of us ;)
i think discogs has an ok solution to bands with same names.
ASCII / ANSI logos? To me that sounds as bloated as anything that exists in xnfo (and yes, I know you say it's optional, but so is most of the stuff in xnfo).
I wouldn't say I'm much of a cheerleader for xnfo either, but the fact that the idea's been around since 2003 and we still haven't reached the goal of proper data exchange between demoscene sites should say something. Data interchange formats are easy - the tricky bit is making that data meaningful on either side, when different sites have different approaches to their data model. (As another example: what do you do when your list of platforms doesn't match the one on Pouet - let's say you have an incoming file with a platform of "mobile phone", but your site has the category split into iphone, android, symbian...?)
Duplicate names might not be a problem to you, but personally I don't want to build a site that confuses me with a Swedish punk band. Real names are a different issue entirely, and one that we're taking very seriously on Demozoo. (Knowing them is very handy for piecing together the historical record, but we're sure as hell not going to publish them against people's wishes.)
I wouldn't say I'm much of a cheerleader for xnfo either, but the fact that the idea's been around since 2003 and we still haven't reached the goal of proper data exchange between demoscene sites should say something. Data interchange formats are easy - the tricky bit is making that data meaningful on either side, when different sites have different approaches to their data model. (As another example: what do you do when your list of platforms doesn't match the one on Pouet - let's say you have an incoming file with a platform of "mobile phone", but your site has the category split into iphone, android, symbian...?)
Duplicate names might not be a problem to you, but personally I don't want to build a site that confuses me with a Swedish punk band. Real names are a different issue entirely, and one that we're taking very seriously on Demozoo. (Knowing them is very handy for piecing together the historical record, but we're sure as hell not going to publish them against people's wishes.)
yep, but bloat is affordable these days.
I acknowledge that the idea has been around for a while, yet we need to refine it <3
p.s.: That swedish punk band is from a different namespace, so, not much confusion there :)
p.p.s.: add to that a base64 encoded version of the demo itself and you got yourself one good-to-go self-contained package :D
p.p.p.s: "but the fact that the idea's been around since 2003 and we still haven't reached the goal of proper data exchange between demoscene sites should say something" -> true
I acknowledge that the idea has been around for a while, yet we need to refine it <3
p.s.: That swedish punk band is from a different namespace, so, not much confusion there :)
p.p.s.: add to that a base64 encoded version of the demo itself and you got yourself one good-to-go self-contained package :D
p.p.p.s: "but the fact that the idea's been around since 2003 and we still haven't reached the goal of proper data exchange between demoscene sites should say something" -> true
gasman: let's talk this over at revision over a beer or seven :)
gargaj: sounds like a plan :-)
oh well, last night I was already a bit silly and way past that 7 beers Gargaj mentioned :D
@steam: that (cidoc crm) is actually also quite interesting but working with that format looks like a full-time job to me
@gasman: xnfo was definitely a step into the right direction. just keep the name (and credits) and resurrect/modernize it.
(btw: even if a single .xnfo was 64k in size, for 10,000 prods that would just amount to 625 MBytes. not much these days)
A proper interchange format is only as good as the number of ppl supporting it. Therefore, it should be discussed properly
before jumping to any implementations. (it should also evolve over time, of course)
Regarding the platform problem ("a platform of "mobile phone", but your site has the category split into iphone, android, symbian..."):
I would suggest to list all required hw/sw components and their versions, e.g. the base platform and any required extensions
(like sound, gfx, or accelerators cards).
The more (interesting) info, the better.
Not every site would have to display all the info (or even make it editable).
If the hw/sw info is properly structured, an algorithm could determine e.g. a pouet platform id by looking at the data.
Another solution would be to store site-specific info (e.g. a <pouet> tag) (even it's just some auto-generated cache).
Maybe some day we'll have some kind of MAME like frontend/"app" that let's you browse through ideally every demo ever made :)
About my privacy rant: Yea yea, IDs come with the terrority. Linking to a scene-ID (group ID mandatory, user IDs optional?!) would be ok, I guess.
@steam: that (cidoc crm) is actually also quite interesting but working with that format looks like a full-time job to me
@gasman: xnfo was definitely a step into the right direction. just keep the name (and credits) and resurrect/modernize it.
(btw: even if a single .xnfo was 64k in size, for 10,000 prods that would just amount to 625 MBytes. not much these days)
A proper interchange format is only as good as the number of ppl supporting it. Therefore, it should be discussed properly
before jumping to any implementations. (it should also evolve over time, of course)
Regarding the platform problem ("a platform of "mobile phone", but your site has the category split into iphone, android, symbian..."):
I would suggest to list all required hw/sw components and their versions, e.g. the base platform and any required extensions
(like sound, gfx, or accelerators cards).
The more (interesting) info, the better.
Not every site would have to display all the info (or even make it editable).
If the hw/sw info is properly structured, an algorithm could determine e.g. a pouet platform id by looking at the data.
Another solution would be to store site-specific info (e.g. a <pouet> tag) (even it's just some auto-generated cache).
Maybe some day we'll have some kind of MAME like frontend/"app" that let's you browse through ideally every demo ever made :)
About my privacy rant: Yea yea, IDs come with the terrority. Linking to a scene-ID (group ID mandatory, user IDs optional?!) would be ok, I guess.
demozoo will blow your mind
is it in 64k?
it's in actual size
demozoo v2
be afraid, be very afraid
be afraid, be very afraid
i'm not afraid!
are we there yet?
are we there yet?
are we there yet?
Yes, we're there.
weee \o/
