pouet - new platforms

category: general [glöplog]
gasman, no offense meant... but given many years of practical experience with restoring lost prods to viewable form i seriously doubt that knowledgeable people will suddenly step up to take care of such jobs, more likely the brunt of the workload will have to be picked up by the usual few suspects or else it just will not happen at all
added on the 2024-02-28 13:09:18 by havoc havoc
how does renaming the category lead to more maintenance work? I don't get it, sorry for my ignorance.
added on the 2024-02-28 13:18:05 by v3nom v3nom
best advice i can give is to get involved with maintenance and/or preservation, then you will naturally develop an understanding for such things
added on the 2024-02-28 14:27:47 by havoc havoc
Well lets not convolute things too much here, the extension of scope to glsl/shadertoy prods does bear a few kinks to be worked out.

However as a first step I think it'd be good if we can all acknowledge that literally all "javascript" prods are in fact "browser" prods, or does anyone know of a javascript prod that consists of just a javascript file to be fed to a runtime other than a browser? With that in mind I don't see how anyone can argue against a simple rebranding, keeping everything as is, except the name/icon. In the future this would also be inclusive to webassembly and the odd css prods, since they too cannot run stand-alone and their native execution environment is a browser.

The discussion about the scope being extended to glsl prods can and should be had, but doesn't really affect the aforementioned points imho. GLSL-Sandbox has been very stable and a stand-alone player-html can easily be derived (already did that a few times), for shadertoy prods it's a bit more complex as it has been evolving a bit in the past but utilizing their export format it shouldn't be too hard to build (or use an existing) player.
Considering the alternatives:

  1. introduce platforms for shadertoy, glsl-sandbox, vertexshaderart and whatever will eventually be used (webgpu is coming)
  2. not support these prods

I think it's still a worthwhile discussion.

The argument of user-friendliness while valid has its limits considering that to run any prod natively I need some kind of idea of how to operate the environment/platform, in fact I'd argue that one big aspect for browser-based prods is that I don't have to download a zip and run the program somehow, the preferred, primary way to run browser prods is via the link to the hosted online version (which btw. could also be hosted and served on scene.org ala github.io).
added on the 2024-02-28 15:05:08 by LJ LJ
gasman, no offense meant... but given many years of practical experience with restoring lost prods to viewable form i seriously doubt that knowledgeable people will suddenly step up to take care of such jobs, more likely the brunt of the workload will have to be picked up by the usual few suspects or else it just will not happen at all

Well, right now, Shadertoy does exist, and happens to be run by a scener who I wouldn't consider to be one of the "usual few suspects", so extrapolating your current real-world experience to a hypothetical "what if iq got hit by a bus" world doesn't seem entirely valid. FWIW, I believe Shadertoy is important and active enough both inside and outside the scene that there would be no trouble finding someone to take over, enthusiastically or not. But let's say you're right, and the demoscene just doesn't have enough web-monkeys to sustain the sort of friendly web experience we all want - well, we'll just have to live with whatever sub-standard web experience we can put together with the resources we do have. And meanwhile, as an archivist you can be content that you've done everything necessary to keep the files alive for the day when there's a resurgence of interest and someone does step up to improve the situation.

You might as well ask "What if the authors of VICE gave up maintaining it for new Windows versions and nobody cared enough to take on that job" - well, then we'd have to live with C64 demos being less convenient to run. That doesn't change anything about the validity of C64 as a platform or the value of preserving them. Or look at Flash demos - running them in 2024 requires a whole lot more hoop-jumping and expertise than it did in 2004, due to factors outside the demoscene's control. Someone could do the archive.org thing and make them a one-click launch away, and the fact that they haven't is indicative of the quantity of fucks people give about Flash demos at this moment. But still, Pouet and scene.org have done everything to ensure that they're still around in the event that there's a future resurgence of interest in them.

Don't get me wrong - "what if iq got hit by a bus" is absolutely a valid thing to worry about as an archivist, but "how do we ensure that the experience of shader code watching remains completely unchanged in the event that iq gets hit by a bus" isn't.
added on the 2024-02-28 15:20:42 by gasman gasman
lj: the problem with renaming to browser is that there will be plenty of fools who'll use it as an excuse to try to move the goalposts for that category by posting their (or other's) crappy website, animations, pictures, and other random types of junk that to me (and hopefully also you) clearly wouldn't belong in any category on pouet but that they'd like to put on the site to gain attention (or such, please don't ask me to give a psychological analysis of such individuals).

gasman: i'd be concerned about iq getting hit by a bus, but not at all because of the reasons you are projecting, rather just because he's a nice person who always treated me well. not sure what to do with the rest of your argumentation, i don't feel it applies to me or my role here at pouet or as a preserver of demoscene productions in general.
added on the 2024-02-28 15:51:23 by havoc havoc
I have a hard time believing that hundreds of users are just waiting for the gates to open to submit crappy random websites to pouet, there already is "wild" and "animation/video" as a potential target for that, music disk websites(not saying they're crappy) are already being submitted and accepted via "javascript", so again no goalpost seems to have been moved?

I think the point gasman is trying to make is simply that its worth archiving and giving a platform to the prods even if running them eventually becomes tricky, which tbf applies to a ton of the more obscure platforms already. Even for the less obscure fixed hardware platforms it's not all sunshine and rainbows, when's the last time you tried to run Playstation2 demos? Last time I tried I wasn't able to run the vast majority of them on real hardware.
added on the 2024-02-28 16:28:35 by LJ LJ
fine, let me try a tl;dr version then: In a hypothetical future where Shadertoy prods are as obsolete and unloved as Flash demos are today, they can be preserved in exactly the same way that we're preserving Flash demos (namely, as zipfiles on scene.org so that people who enjoy code archaeology can dust them off and jump through the necessary hoops to run them).
added on the 2024-02-28 16:37:47 by gasman gasman

How many DOS demos run without issues on a stock DOSBox configuration?

Yes, almost all of them work.

(the stock configuration of DosBox is only bad because of the cycles=auto setting. Well, turning EMS/XMS on/off is a necessity for real hardware as well).
added on the 2024-02-28 17:03:04 by bitl bitl
Just as a point of order: There are already 155 Shadertoy links from prods, of which eight are the main download link (the rest are probably extra links on intros?), plus Shadertoy itself as a prod. E.g. this or this.

On a more subjective note: I do regret that like most online prods, my mirror does not properly store these toys, only the HTML (none of the supporting JavaScript, images, etc.). I see Shadertoy has an API, so perhaps the best thing to do would be to query for and download the shaders, even though that form isn't viewable in itself.
added on the 2024-02-28 17:41:34 by Sesse Sesse
lj: you might have a hard time believing it but i've had hard times sifting through thousands upon thousands of prods with broken downloads and having to delete hundreds of them for not fitting the criteria for prods that should be uploaded to this website. the reason you don't see such prods so much on this site anymore today is because we've eradicated them over the years, and keep on deleting them as soon as they are noticed.
added on the 2024-02-29 00:19:23 by havoc havoc
The Nintendo 3DS now have 10 prods referenced on Pouët.
Here is an (eventual) icon for a new category: https://hitchhikr.net/3ds.png
added on the 2024-03-01 13:00:57 by hitchhikr hitchhikr

I think there are enough Mega65 releases now for a new platform icon and categoree..

What do you think?
added on the 2024-03-04 19:25:41 by magic magic
people should not regard scenecity, as scenecity is not fine enough
added on the 2024-03-05 00:16:42 by gentleman gentleman
So uuuh, what category should I upload my html+css only demo to?
added on the 2024-04-01 04:49:11 by LJ LJ
People continue to be confused and put their modern Mac OS ARM prods under the "MacOS" platform which is meant for classic 68K macs. Please do something about it, it's annoying to have to send edit requests for all these prods asking to put them under "MacOS Intel", which is also wrong but a bit less wrong :(
So uuuh, what category should I upload my html+css only demo to?

hmm, I think browser would be a fitting category
added on the 2024-04-01 10:31:18 by v3nom v3nom
Which category is browser? :) there's only JavaScript, which this production isn't
added on the 2024-04-01 13:00:22 by wayfinder wayfinder
Put it there anyway for now, we'll deal with the difference later.
added on the 2024-04-01 13:25:12 by Gargaj Gargaj
Hello everyone, in my opinion Pouet.net is already a rich site, I use Rapsberry Pi and in its versions the model changes, not the concept, it can start Linux distributions ARM64, RISC OS, BSD, the old ARM32 models, the Linux project it is Unix-like and is part of the Free Software Foundation, what changes are the mother distributions, such as: Slackware, Debian, Red Hat, Arch Linux, Gentoo, SuSE, others.
AmigaOS, Microsoft (Windows, MS-DOS), Atari and all their derivatives are closed source code, current companies (like those of the past) are the owners and have all copyright rights, even companies that are not in vogue, many of them they are silent (not active) but exist for the market.
Linux is truly the free novelty within everyone's reach, for a community based on the "force awakening" like yours it is a thorny and complicated topic, because it means liberating (not liberalizing) the "market", probably infinite "n" could be born categories, in my opinion the Pouet.net community must become more efficient in the functionality of the site, there is already "a lot of irons in the fire", do not consider copyright and free licenses as rivals, but keep in mind that the latter outclasses the former, this leads to greater difficulty for small proprietary companies, while Big Tech uses the EEE "Embrace, Extend and Extinguish" method.
I write this as a Debian user!
added on the 2024-04-01 13:37:32 by crackount crackount
+1'ing the MacOS proposal. Could add a 'MacOSX ARM', but it's not really called Mac OS X anymore. Maybe change 'MacOS' to 'MacOS Classic' or something similar, and rename the current 'MacOSX Intel' to 'macOS Intel', and add 'macOS ARM'?
added on the 2024-04-05 13:13:58 by shiz shiz
Bump. We have more and more macOS ARM productions being added. We have the icon: BB Image

What's the hold-up?
added on the 2024-06-11 16:19:01 by Blueberry Blueberry