The past, present and uncertain future of scene-run infrastructure
category: general [glöplog]
*** The past, present and uncertain future of scene-run infrastructure ***
I. (Self-)Introduction on the subject
As you may or may not know, for 25 years I am hosting untergrund.net, the biggest host of demoscene websites – over 500 or so. Most of this is now for historical/archiving purposes as the features of course are outdated, but on the other hand also hosting still some of the biggest scene websites, like the Amiga Demoscene Archive or chiptune.com.
Three years ago a team of sceners then tried to build “the next big thing” when it comes to providing the Demoscene with infrastructure: SceneCity. We wanted to create a privacy-focussed, non-commercial and everlasting service for community chats and video streaming. Over the years we added lots of innovations, including being the first mayor service on the net to enable h.265 streaming.
But hardly anyone cared. A bunch of people basically invested a couple of years in R&D time for a service that is not getting used.
That that tragedy is not the exception in the demoscene, it has become the norm now. “Back in the days” it was important for the demoscene to be self-sufficient and not to reply on commercial services. The scene wanted to make sure to be able to do their own rules, and to enable long term preservation.
But none of that does to seem to matter much anymore.
II. The past and present of scene infrastructure
I’ll give you a couple of examples:
III. Raising awareness
If they want future generations to be able to still understand what the demoscene is (or was), sceners need to take the health of demoscene platforms more seriously. It is not about sucking some admins genitals, it is a general matter of making yourself aware about the amount of work that goes into that infrastructure, being mindful about the subject, and to understand that abandoning or even actively sabotaging these platforms long-term will have very negative consequences for the demoscene.
In general, in the demoscene those people who are making demos – rightfully – have the highest visibility. People applaud heavily if someone has spent 100 hours on a demo to amaze and/or entertain you. Demoparty organizers already have a far lower visibility – there you might have spent 1000 hours, with people not noticing or crediting the wrong person (mainorgas) for your work (the “behind the orga scenes” some parties are offering have helped a bit). But when it comes to scene infrastructure, you can spend 10000 hours of work without any applause at all. It’s just taken for granted. And there are scene infrastructure site developers and/or administrators (not me) who are more in the region of having spent 100000 hours on scene infrastructure. And it’s not just the amount of work that is not honored, the skills that went into it also have a low visibility. Most major demoscene sites have an uptime ratio that beats AWS and Azure clouds by far. Scene.org is one of the oldest Internet distributed file storage system that did not lose its data. There is a difference between basing your demo on a commercial 3D engine vs rolling your own, and the same applies to scene infrastructure development.
IV. Missing support from key players - lead and people will follow
It’s not just the “general scene public” that is not mindful about the demoscene infrastructure. What is worse is that even all kinds of organizers of various other demoscene aspects do not appear to be thinking about their role when it comes to pushing and supporting scene-run platforms. There are podcasts that you can’t simply download from any scene website, but that ask you to got to Apple ITunes. Someone might be putting a lot of work into a regular demoscene news show – but it’s hosted on YouTube. Demoscene seminars hardly are archived at all anymore, because the host and/or audience might feel comfortable with sceners re-watching this in the future, but do not want this to be on YouTube. Most demoparty organizers do not bother to even ask SceneSat or SceneCity for streaming. They use Twitch and YouTube, as it just works, so why bother? Same thing when it comes to communication about your party, and preserving history. For most party organizers, Facebook appears to be far more important than demoparty.net. Party chat is on Discord, or Twitch, or YouTube, or all of those, with the community split.
V. Does scene infrastructure still have a future?
If people don’t learn to actively support demoscene platforms, we will sooner or later be losing the last ones that remain today. And active support can have a lot of forms. Sometimes it can just be a “thank you” or some other way of showing respect. It might be joining an administrator team, a developer team, or a moderation team. It might be financial support, legal support, content support. It might be spreading the relevance of a platform via the word of mouth. There really are quite a lot of options.
VI. Back to the projects I am involved in myself, and the question on if they have a future
To close this off, I will return to my personal infrastructure. When it comes to untergrund.net, after 25 years, I sooner or later will need to find a way to convert all of these dynamic websites into static archives. That’s especially a challenge for those sites that have search functions for their demos, or forum/bbs systems. I can not assume that the actual site owners will do that. Many have long left the scene, and some are physically dead. And then there are sites where the administrators are no longer active, but the users still are, producing new content. So transplanting those systems that might be based on a 15 year old version of PHP to a new platform could be a lot of work.
If I would ask the scene if someone would be willing to invest the time to find a solution and preserve untergrund.net’s content, would anyone raise their hands? Or would people just not bother if another 25 years of scene history would go down the drain? Sooner or later I will find out, because at a certain point the code base and especially the OS will be so old that it no longer runs on any somewhat modern server.
And when it comes to SceneCity – well. Can’t speak for the other developers, but when it comes to myself: Yes, nice that I learned a lot of stuff coding this, but it certainly would have felt better to then see the work result getting appreciated, as in: it getting used. Right now one of our team members is still actively working on building a new version of the iOS app to hopefully finally get app approval from evil Apple (who are for two years already refusing to list the app because it uses SceneID), with none of us developers being Apple users. We will soon have another great app that nobody is going to use.
And then it is now close to a year that we have filled a lot of space in a data center with a hardware pipeline to completely replace YouTube for the scene, so we can have demo captures preserved long-term, without ads, and without content censorship. What’s missing is the software. We have managed to get a very skilled (and well-known) developer on our team for this. But there are doubts: Does it really make sense to again build something that ends up as an unused “proof of concept”? Does it make sense to maintain a hardware pipeline that could host any demoscene video there is on YouTube today, with us then ending up with 10 videos? I wrote messages to some people who have uploaded demo captures to YouTube. Some have left the scene, some have deleted the original capture files so the re-compressed clips on YouTube are all that’s left, some clearly are motivated by collecting ad revenue from YouTube. The idea was to ask the scene as whole to start doing fresh captures of old and new demos, with us this time preserving the originals without having them re-compressed. But would there really be enough volunteers? Or will it again be “YouTube is good enough for now, why bother about the future…”?
So, for me personally preserving the infrastructure we have built for however long it is practical is a given thing, but of course I might get run over by a bus. And the majority of scene sites now appear in such a maintenance mode, often with some single scener being the single point of failure.
VII. Will anyone still be motivated to create new infrastructure?
Is it still possible getting new people motivated to build and extend scene infrastructure? I think in the current atmosphere of ignorance and an open preference to use commercial services instead of scene-run ones that will not happen much anymore. It is like creating a wonderful demo, and then playing it to an audience of five people – you might be OK with just knowing yourself how great the stuff was you have created, but in the long run you might question if your art should make more people happy, and you might need to switch to a different subculture for that. Besides the commercial use (not interested in that, this is a hobby project) there would be plenty of subcultures being very grateful for someone providing them such infrastructure, including those living in regions that are locked out of services provided by US corporations, or where commercial services are under close surveillance. Good luck trying to take part in demoscene communication being a Chinese, for example, where pretty much every single thing the scene is using as “why bother?” platform is banned.
To give another example: There was the plan to create a scene-internal LLM AI that is fed with all scene knowledge there is, but strictly only being available for sceners. Because similar to “Seminars on YouTube” a lot of sceners might be more open-minded having an AI around if they know for sure that the data won’t leak out.
VIII. Closing words
Again, all of this is my personal perspective. Other scene site admins and even SceneCity coders might of course have a completely different view.
Also, anyone joining this thread announcing “hey, soon I am going to launch some scene infrastructure I had been working on for long!” would already make me very happy. Because what’s worse than my own stuff not getting used is no longer having innovations in regards of scene sites at all anymore.
Also I might have forgotten some important sites. Any additions and any feedback welcome.
P.S.:
Sooner or later some people will jump into this thread saying “People don’t use SceneCity because YOU are involved”. I’m one of 6 developers. And it’s infrastructure. It doesn’t matter who has built infrastructure, what’s relevant is that it exists. I’ve last acted as a content moderator in demoscene chat and bbs forums well over 20 years ago (on IRC and FidoNet). I am not good at human-to-human communication. But I am good at building infrastructure, and with untergrund.net I have proven that I can run such infrastructure for 25 years. So please don’t make me or SceneCity the (sole) subject of this thread... Let’s please discuss the current state and future of scene infrastructure in general. Thank you!
I. (Self-)Introduction on the subject
As you may or may not know, for 25 years I am hosting untergrund.net, the biggest host of demoscene websites – over 500 or so. Most of this is now for historical/archiving purposes as the features of course are outdated, but on the other hand also hosting still some of the biggest scene websites, like the Amiga Demoscene Archive or chiptune.com.
Three years ago a team of sceners then tried to build “the next big thing” when it comes to providing the Demoscene with infrastructure: SceneCity. We wanted to create a privacy-focussed, non-commercial and everlasting service for community chats and video streaming. Over the years we added lots of innovations, including being the first mayor service on the net to enable h.265 streaming.
But hardly anyone cared. A bunch of people basically invested a couple of years in R&D time for a service that is not getting used.
That that tragedy is not the exception in the demoscene, it has become the norm now. “Back in the days” it was important for the demoscene to be self-sufficient and not to reply on commercial services. The scene wanted to make sure to be able to do their own rules, and to enable long term preservation.
But none of that does to seem to matter much anymore.
II. The past and present of scene infrastructure
I’ll give you a couple of examples:
- Visual party and scene memories: In the old days after every party I could check out pictures on slengpung. These days instead I go to “after the party” page of party (if that exist) to then follow links to dead flickr or Google Photo accounts. Slengpung has been a collective visual memory of demoparties and demosceners. When you go to a demoscener’s funeral, in my personal experience most pictures you will see posted will be coming from Slengpung. That memory is no longer fed, and it might even get deleted completely.
- In the old days we made sure that our group chats and direct messages where hosted on non-commercial services. We used FidoNet, Usenet, IRC – and in all of those cases we controlled who could read or archive our messages. On IRC, every major demogroup had their own channel. These days instead, most of the demoscene chats are running on Discord. A gigantic corporation storing all your public and non-public messages forever, potentially one day selling it to the highest bidder? Used to be unthinkable. And quite a bunch of demoscene groups moved their internal chats to Slack. And surprise, then one day Slack decided to take your group chat history hostage, making you pay if you wanted to have it back.
- In the old days the scene had their own video streaming and video archive services. There was demoscene.tv, capped, SceneSat (still somewhat active), just to name a few. Now demo captures are almost exclusively stored on YouTube. You now can watch demos with ad interruptions. AIs are getting trained on demos, no matter if you as the original author of a demo wanted that. And when it comes to streaming, the vast majority of demoparties now stream via Twitch, which is owned by Amazon. Privacy Policy is “anything goes”. If you are now caught on camera at a demoparty, you are training an AI with your face. Back in the days we used to make our own rules at demoparties, what kind of content is allowed in demos and which isn’t. These days it’s now a US corporaton making those rules. It’s now not uncommon for democompo organizers to discuss “can we show this demo without us getting banned from Twitch?”. And it has become normal that part of party streams get muted due to Content ID stuff, and DJ sets no longer streamed at all. And even when Revision got banned for having a butt shown in a demo, nobody learned any lesson.
- Back in the days, if you wanted to check what demoparties were upcoming and receive news about those, you checked demoparty.net (and I am aware and very happy that demoparty.net got modernized and is used more again). These days you have to check Facebook.
- Now we have a risk of pouet getting down, after well over 20 years. Yes, there are people unhappy about this, but the most voiced opinion appears along the lines of “RIP. It was nice while it lasted.”. Seriously? This is the oldest and biggest demo knowledge base and text communication platform that still exists, and you aren’t fighting for keeping it alive, even if it might have been a love-and-hate affair for you personally? Yes, we can download a database dump and worst-case create a static website to preserve history, but then it is still no longer a living platform. And yes, there is demozoo, but it doesn’t contain the same content as pouet, and is no replacement of the BBS system either. Also, it’s not just that people don’t appear to care much: In pouet’s case a rather small amount of trolls who probably actually love the scene and pouet are in part to blame for its potential demise by having made site moderation a frustrating task.
- And when it comes to scene.org, everybody appears to take it for granted that this site exists for 27 years now. If it shuts down, again a huge amount of memory about the scene will die. Yes, someone will put up a mirror. But who is handling server management, maintenance, cost, abuse contacts etc? This is another case of maybe people who are not experienced in running infrastructure (which obviously is the majority of sceners) not having visibility that running such a site is something that needs permanent care and resources.
- I am not an 8 Bit scener, and I understand that demozoo is trying to actively pick up oldschool content. But if CSDB or Bitfellas go down, I am sure again a lot of scene history will be lost and forgotten, in a community where the people who actually created that content might be getting closer to their “best before” date, so that re-uploading to some new and shiny site or recording their memories might not be feasible.
III. Raising awareness
If they want future generations to be able to still understand what the demoscene is (or was), sceners need to take the health of demoscene platforms more seriously. It is not about sucking some admins genitals, it is a general matter of making yourself aware about the amount of work that goes into that infrastructure, being mindful about the subject, and to understand that abandoning or even actively sabotaging these platforms long-term will have very negative consequences for the demoscene.
In general, in the demoscene those people who are making demos – rightfully – have the highest visibility. People applaud heavily if someone has spent 100 hours on a demo to amaze and/or entertain you. Demoparty organizers already have a far lower visibility – there you might have spent 1000 hours, with people not noticing or crediting the wrong person (mainorgas) for your work (the “behind the orga scenes” some parties are offering have helped a bit). But when it comes to scene infrastructure, you can spend 10000 hours of work without any applause at all. It’s just taken for granted. And there are scene infrastructure site developers and/or administrators (not me) who are more in the region of having spent 100000 hours on scene infrastructure. And it’s not just the amount of work that is not honored, the skills that went into it also have a low visibility. Most major demoscene sites have an uptime ratio that beats AWS and Azure clouds by far. Scene.org is one of the oldest Internet distributed file storage system that did not lose its data. There is a difference between basing your demo on a commercial 3D engine vs rolling your own, and the same applies to scene infrastructure development.
IV. Missing support from key players - lead and people will follow
It’s not just the “general scene public” that is not mindful about the demoscene infrastructure. What is worse is that even all kinds of organizers of various other demoscene aspects do not appear to be thinking about their role when it comes to pushing and supporting scene-run platforms. There are podcasts that you can’t simply download from any scene website, but that ask you to got to Apple ITunes. Someone might be putting a lot of work into a regular demoscene news show – but it’s hosted on YouTube. Demoscene seminars hardly are archived at all anymore, because the host and/or audience might feel comfortable with sceners re-watching this in the future, but do not want this to be on YouTube. Most demoparty organizers do not bother to even ask SceneSat or SceneCity for streaming. They use Twitch and YouTube, as it just works, so why bother? Same thing when it comes to communication about your party, and preserving history. For most party organizers, Facebook appears to be far more important than demoparty.net. Party chat is on Discord, or Twitch, or YouTube, or all of those, with the community split.
V. Does scene infrastructure still have a future?
If people don’t learn to actively support demoscene platforms, we will sooner or later be losing the last ones that remain today. And active support can have a lot of forms. Sometimes it can just be a “thank you” or some other way of showing respect. It might be joining an administrator team, a developer team, or a moderation team. It might be financial support, legal support, content support. It might be spreading the relevance of a platform via the word of mouth. There really are quite a lot of options.
VI. Back to the projects I am involved in myself, and the question on if they have a future
To close this off, I will return to my personal infrastructure. When it comes to untergrund.net, after 25 years, I sooner or later will need to find a way to convert all of these dynamic websites into static archives. That’s especially a challenge for those sites that have search functions for their demos, or forum/bbs systems. I can not assume that the actual site owners will do that. Many have long left the scene, and some are physically dead. And then there are sites where the administrators are no longer active, but the users still are, producing new content. So transplanting those systems that might be based on a 15 year old version of PHP to a new platform could be a lot of work.
If I would ask the scene if someone would be willing to invest the time to find a solution and preserve untergrund.net’s content, would anyone raise their hands? Or would people just not bother if another 25 years of scene history would go down the drain? Sooner or later I will find out, because at a certain point the code base and especially the OS will be so old that it no longer runs on any somewhat modern server.
And when it comes to SceneCity – well. Can’t speak for the other developers, but when it comes to myself: Yes, nice that I learned a lot of stuff coding this, but it certainly would have felt better to then see the work result getting appreciated, as in: it getting used. Right now one of our team members is still actively working on building a new version of the iOS app to hopefully finally get app approval from evil Apple (who are for two years already refusing to list the app because it uses SceneID), with none of us developers being Apple users. We will soon have another great app that nobody is going to use.
And then it is now close to a year that we have filled a lot of space in a data center with a hardware pipeline to completely replace YouTube for the scene, so we can have demo captures preserved long-term, without ads, and without content censorship. What’s missing is the software. We have managed to get a very skilled (and well-known) developer on our team for this. But there are doubts: Does it really make sense to again build something that ends up as an unused “proof of concept”? Does it make sense to maintain a hardware pipeline that could host any demoscene video there is on YouTube today, with us then ending up with 10 videos? I wrote messages to some people who have uploaded demo captures to YouTube. Some have left the scene, some have deleted the original capture files so the re-compressed clips on YouTube are all that’s left, some clearly are motivated by collecting ad revenue from YouTube. The idea was to ask the scene as whole to start doing fresh captures of old and new demos, with us this time preserving the originals without having them re-compressed. But would there really be enough volunteers? Or will it again be “YouTube is good enough for now, why bother about the future…”?
So, for me personally preserving the infrastructure we have built for however long it is practical is a given thing, but of course I might get run over by a bus. And the majority of scene sites now appear in such a maintenance mode, often with some single scener being the single point of failure.
VII. Will anyone still be motivated to create new infrastructure?
Is it still possible getting new people motivated to build and extend scene infrastructure? I think in the current atmosphere of ignorance and an open preference to use commercial services instead of scene-run ones that will not happen much anymore. It is like creating a wonderful demo, and then playing it to an audience of five people – you might be OK with just knowing yourself how great the stuff was you have created, but in the long run you might question if your art should make more people happy, and you might need to switch to a different subculture for that. Besides the commercial use (not interested in that, this is a hobby project) there would be plenty of subcultures being very grateful for someone providing them such infrastructure, including those living in regions that are locked out of services provided by US corporations, or where commercial services are under close surveillance. Good luck trying to take part in demoscene communication being a Chinese, for example, where pretty much every single thing the scene is using as “why bother?” platform is banned.
To give another example: There was the plan to create a scene-internal LLM AI that is fed with all scene knowledge there is, but strictly only being available for sceners. Because similar to “Seminars on YouTube” a lot of sceners might be more open-minded having an AI around if they know for sure that the data won’t leak out.
VIII. Closing words
Again, all of this is my personal perspective. Other scene site admins and even SceneCity coders might of course have a completely different view.
Also, anyone joining this thread announcing “hey, soon I am going to launch some scene infrastructure I had been working on for long!” would already make me very happy. Because what’s worse than my own stuff not getting used is no longer having innovations in regards of scene sites at all anymore.
Also I might have forgotten some important sites. Any additions and any feedback welcome.
P.S.:
Sooner or later some people will jump into this thread saying “People don’t use SceneCity because YOU are involved”. I’m one of 6 developers. And it’s infrastructure. It doesn’t matter who has built infrastructure, what’s relevant is that it exists. I’ve last acted as a content moderator in demoscene chat and bbs forums well over 20 years ago (on IRC and FidoNet). I am not good at human-to-human communication. But I am good at building infrastructure, and with untergrund.net I have proven that I can run such infrastructure for 25 years. So please don’t make me or SceneCity the (sole) subject of this thread... Let’s please discuss the current state and future of scene infrastructure in general. Thank you!
Let me briefly express my no-one-needs opinion.
Yeah, I worked on some media projects for the demoscene too. Mostly just local stuff, so you probably haven't heard of them.
But I ended up dropping them. Why? Simple - i just didn’t see the point anymore. People stopped needing it, the site activity slowed down, and in the end, it felt like I was doing all this for, like, two and a half users. Not much motivation to keep going after that.
It’s a different story with pouet.net. Sure, here might not be as many people as there were 15 years ago, but the users here are still pretty active, and the site is definitely in demand. Oddly enough, though, it seems like the admin has gotten tired of all this activity. Kind of ironic, at least to me.
P.S:
https://www.demoparty.net/ - it’s one of my favorite sites! It’s super convenient and really useful. Big thanks to everyone who created and maintains it!"
Yeah, I worked on some media projects for the demoscene too. Mostly just local stuff, so you probably haven't heard of them.
But I ended up dropping them. Why? Simple - i just didn’t see the point anymore. People stopped needing it, the site activity slowed down, and in the end, it felt like I was doing all this for, like, two and a half users. Not much motivation to keep going after that.
It’s a different story with pouet.net. Sure, here might not be as many people as there were 15 years ago, but the users here are still pretty active, and the site is definitely in demand. Oddly enough, though, it seems like the admin has gotten tired of all this activity. Kind of ironic, at least to me.
P.S:
https://www.demoparty.net/ - it’s one of my favorite sites! It’s super convenient and really useful. Big thanks to everyone who created and maintains it!"
Ranting while not providing a constructive strategy doesnt helps much...
if i go to scenecity (as a potential user)
i will see a interface where the first points are privacy and rules
while the home screen just displays a huge welcome to scenecity (i allready know that im at scenecity so it doesnt provides any useful information).
For further information i need to scroll 1 page down just to get a vague info that scenecity is something about a chat system and streaming.
Then the page ends. No more informations beside i disclaimer that i cant register here and need a scene.id (dear potential user leave and come back with a proper account).
It basically ends here. Now i need to click all the options of the interface which are mostly disclaimer stuff or unrelated to me as new user. So i guess i have to hit the "app" button just to find out that i wont find anything at scenecity.org cos everything is hidden behind a Android/IOS/whatever ÄPP and i need to create a token just to find out what it really is.
Dont you think that the entry barrier is abit to high?
if i go to scenecity (as a potential user)
i will see a interface where the first points are privacy and rules
while the home screen just displays a huge welcome to scenecity (i allready know that im at scenecity so it doesnt provides any useful information).
For further information i need to scroll 1 page down just to get a vague info that scenecity is something about a chat system and streaming.
Then the page ends. No more informations beside i disclaimer that i cant register here and need a scene.id (dear potential user leave and come back with a proper account).
It basically ends here. Now i need to click all the options of the interface which are mostly disclaimer stuff or unrelated to me as new user. So i guess i have to hit the "app" button just to find out that i wont find anything at scenecity.org cos everything is hidden behind a Android/IOS/whatever ÄPP and i need to create a token just to find out what it really is.
Dont you think that the entry barrier is abit to high?
Just wanted to add a quick note here from the side of the "somewhat still active" SceneSat.
@scamp: I totally understand your frustration and must say that I applaud your tenacity in keeping things going. This goes without putting any bias towards any side into the YOU-bit, as you mention at the end of your post, so let's just leave that bit out of this, as you say. (after all, you did try to kick me and fnord out a couple of times from the buildup of the final Breakpoint, which was the first time I met you... ;))
When it comes to SceneSat, the "somewhat still active"-bit is actually quite correct. We have no steady show schedule with consistent shows so people can know about it and actually gather around to listen to/experience. We are not present for all parties who could utilize our system to stream through, simply because they don't want to - or don't know to - ask for our help, because it's simply easier to go to YouTube/Twitch or similar, as you also mention. We DO store everything that is streamed through us, though, and have no problem republishing to these other platforms. Field-FX have been using our platform exactly like that for the last year or so.
However, the issues starts showing here.
The biggest one: time (and money, in all fairness).
We all who are in any way involved with SceneSat have sadly way too much going on in our lives outside of the scene to make it possible to keep going as much as we (well, at least me, can't speak for all the others) would like to. My home situation sadly makes it quite hard to have a space to do any shows because it would cause too much hassle for my family, since my "office/studio" is in our bedroom, which I share with my wife and youngest son, since the rest of the house is occupied by our other kids. I, like everyone else, also need to spend more time than I'd like to to work for a living, to feed me and my family and make us actually have a home at all. If only I could win that darn EuroJackpot...
The next huge one, and the main topic of this whole thread: infrastructure.
Since we started SceneSat in 2008, we've aimed at hosting it ourselves (of course with generous help of others), because we simply wanted to be in control of what we had, and also because hosting elsewhere like in the cloud, would cost a lot of money, which we obviously don't have an abundance of. I have personally put way too much money into SceneSat through the years for what's good for me, but it also has been totally worth it, because I'm proud of what we've built. Just getting a little notion anywhere about us makes me even prouder that people actually knows us and knows what we're trying to do. Also, we are lucky enough to have had (and still have) a lot of great benefactors through the years who have helped by donating money, hardware or just overall support to make us still able to run this all. If it wouldn't be for those benefactors who are supporting us via Patreon, or you all who are spending a dime or two on our Sound of SceneSat compilations, we simply wouldn't be able to be around anymore.
Also, as I now noticed the newer comment about entry barrier, this is actually also a big thing with attracting new people to the scene. Or even attracting old people. As my wife constantly says (and she's usually right about most things, tbh), we in the demoscene tend to be "grumpy, old, elitist men who do not want to change anything from the way it always has been". To some aspects, I totally agree. I see that in myself as well. But, I also see myself wanting to open things up more, make it easier to be inclusive for anyone to take part of this amazing thing we have. This goes to making it easier to navigate through the demoscene overall and find information and discover new (and old) things all the time. We could totally be better at this on a pure technical level, making it easier to share data amongst each system and that way make it easier to navigate through it all. This takes thought and time, though, as most things. And yes, I know there are APIs available at sites like Pouët, Demozoo and others, but I think we could do more there too. If more time was available, again... :D
Once again speaking from a SceneSat standpoint, we're not ONLY doing SceneSat. Much like scamp, we're also hosting a bunch of (but not as many, of course) demoscene-related sites and platforms. Modland, Slengpung, ModArchive, BitJam Radio, AmigaRemix and RKO, for instance. We do that because we want to help out and also because we can. With your help, of course. At the moment, our video recording archive from different parties are in the magnitude of 5-7TB which is constantly growing and that will be interesting to see how we will handle in the future. We are wrestling with our current, outdated platform, and the lack of proper hardware makes it a complicated issue. Especially when our lives don't allow too much time to be put into this.
However, to round of this not-that-quick-but-still-a-quick-note, I'd like to say that I'm definitely up for discussing how to keep this little scene of ours running in the future. I only wish I could add about 20 more hours per day to accommodate this, or so...
When it comes to SceneSat, I think we'll look into setting up some sort of Kickstarteresque project to help fund some new hardware, but that's still not certain in any way. We might "just" need to do a bunch of Sound of SceneSat volumes to help out with this as well. :D
But yes, let's have some real, balanced discussions around this topic, if anyone would be interested.
@scamp: I totally understand your frustration and must say that I applaud your tenacity in keeping things going. This goes without putting any bias towards any side into the YOU-bit, as you mention at the end of your post, so let's just leave that bit out of this, as you say. (after all, you did try to kick me and fnord out a couple of times from the buildup of the final Breakpoint, which was the first time I met you... ;))
When it comes to SceneSat, the "somewhat still active"-bit is actually quite correct. We have no steady show schedule with consistent shows so people can know about it and actually gather around to listen to/experience. We are not present for all parties who could utilize our system to stream through, simply because they don't want to - or don't know to - ask for our help, because it's simply easier to go to YouTube/Twitch or similar, as you also mention. We DO store everything that is streamed through us, though, and have no problem republishing to these other platforms. Field-FX have been using our platform exactly like that for the last year or so.
However, the issues starts showing here.
The biggest one: time (and money, in all fairness).
We all who are in any way involved with SceneSat have sadly way too much going on in our lives outside of the scene to make it possible to keep going as much as we (well, at least me, can't speak for all the others) would like to. My home situation sadly makes it quite hard to have a space to do any shows because it would cause too much hassle for my family, since my "office/studio" is in our bedroom, which I share with my wife and youngest son, since the rest of the house is occupied by our other kids. I, like everyone else, also need to spend more time than I'd like to to work for a living, to feed me and my family and make us actually have a home at all. If only I could win that darn EuroJackpot...
The next huge one, and the main topic of this whole thread: infrastructure.
Since we started SceneSat in 2008, we've aimed at hosting it ourselves (of course with generous help of others), because we simply wanted to be in control of what we had, and also because hosting elsewhere like in the cloud, would cost a lot of money, which we obviously don't have an abundance of. I have personally put way too much money into SceneSat through the years for what's good for me, but it also has been totally worth it, because I'm proud of what we've built. Just getting a little notion anywhere about us makes me even prouder that people actually knows us and knows what we're trying to do. Also, we are lucky enough to have had (and still have) a lot of great benefactors through the years who have helped by donating money, hardware or just overall support to make us still able to run this all. If it wouldn't be for those benefactors who are supporting us via Patreon, or you all who are spending a dime or two on our Sound of SceneSat compilations, we simply wouldn't be able to be around anymore.
Also, as I now noticed the newer comment about entry barrier, this is actually also a big thing with attracting new people to the scene. Or even attracting old people. As my wife constantly says (and she's usually right about most things, tbh), we in the demoscene tend to be "grumpy, old, elitist men who do not want to change anything from the way it always has been". To some aspects, I totally agree. I see that in myself as well. But, I also see myself wanting to open things up more, make it easier to be inclusive for anyone to take part of this amazing thing we have. This goes to making it easier to navigate through the demoscene overall and find information and discover new (and old) things all the time. We could totally be better at this on a pure technical level, making it easier to share data amongst each system and that way make it easier to navigate through it all. This takes thought and time, though, as most things. And yes, I know there are APIs available at sites like Pouët, Demozoo and others, but I think we could do more there too. If more time was available, again... :D
Once again speaking from a SceneSat standpoint, we're not ONLY doing SceneSat. Much like scamp, we're also hosting a bunch of (but not as many, of course) demoscene-related sites and platforms. Modland, Slengpung, ModArchive, BitJam Radio, AmigaRemix and RKO, for instance. We do that because we want to help out and also because we can. With your help, of course. At the moment, our video recording archive from different parties are in the magnitude of 5-7TB which is constantly growing and that will be interesting to see how we will handle in the future. We are wrestling with our current, outdated platform, and the lack of proper hardware makes it a complicated issue. Especially when our lives don't allow too much time to be put into this.
However, to round of this not-that-quick-but-still-a-quick-note, I'd like to say that I'm definitely up for discussing how to keep this little scene of ours running in the future. I only wish I could add about 20 more hours per day to accommodate this, or so...
When it comes to SceneSat, I think we'll look into setting up some sort of Kickstarteresque project to help fund some new hardware, but that's still not certain in any way. We might "just" need to do a bunch of Sound of SceneSat volumes to help out with this as well. :D
But yes, let's have some real, balanced discussions around this topic, if anyone would be interested.
@xeNusion
There are two buttons in the top menu bar: "Stream" and "Chat". There are also the direct links https://scenecity.tv and https://scenecity.chat for bookmarking.
Yes, there is an entrance barrier. But you should not have ended up on the https://scenecity.site info/help via Google. You should have discovered SceneCity by wanting to watch a party stream, or because another scener invited you to his chat room. Network effects.
But let's please not get into SceneCity usability issued or even bugs in this thread. There is a channel for reports like this on SceneCity itself.
Of the sites mentioned in my essay, SceneCity is the least important right now. If it's shut down tomorrow, not many will miss it. It's quite a different story for, say, scene.org.
There are two buttons in the top menu bar: "Stream" and "Chat". There are also the direct links https://scenecity.tv and https://scenecity.chat for bookmarking.
Yes, there is an entrance barrier. But you should not have ended up on the https://scenecity.site info/help via Google. You should have discovered SceneCity by wanting to watch a party stream, or because another scener invited you to his chat room. Network effects.
But let's please not get into SceneCity usability issued or even bugs in this thread. There is a channel for reports like this on SceneCity itself.
Of the sites mentioned in my essay, SceneCity is the least important right now. If it's shut down tomorrow, not many will miss it. It's quite a different story for, say, scene.org.
Maybe you should consider the possibility that not every type of media benefits from running on a private infrastructure. Streaming is a good example. As xenusion already started explaining, there's a massive entry barrier to the scene's own streaming services. I've never met anyone who stumbled across the SceneCity, SceneSat or CCC video offerings by accident and then decided to stick around, it just doesn't happen. In contrast to this, on Twitch or Youtube that kind of crossover does actually happen. Casual viewers are likely to already have an account on those sites, and there actually are sceners who joined the scene via such channels. So there are clear and obvious reasons to run a party's stream on those mainstream platforms. You just don't want to acknowledge it because you run one of those superfluously gated communities yourself.
Quote:
So there are clear and obvious reasons to run a party's stream on those mainstream platforms.
I completely agree and understand that sentiment, which is why we have a very transparent republishing functionality for SceneSat. The best thing is that we don't really have to handle transcoding and such things since we at the moment don't really have the hardware power to manage that properly. And also, that casual viewer stumbling onto the stream really works well on these bigger sites. Especially if the stream ends up on the frontpage or similar. Still, since they stream through us, we make sure to work as an archive and also a fallback in case we get another butt-gate situation...
...and I also agree with people who feel that it's not really welcoming to jump into any of these sites and understand how to handle it, either chatting or even just watching things. We as tech nerds tend to miss putting ourselves into the shoes of the "average user" quite often. SceneSat does NOT show a good example of a welcoming UI, which we're painfully aware of. :D
@havoc
I don't see where we disagree. I am somewhat aware that YouTube on a global scale is a bit more known and used than SceneSat, Demoscene.TV or SceneCity. But as pointed out in great length above, there are good reasons that the scene always had it's own infrastructure for over 30 years. We could have based the first scene networks on AOL. We chose BBSs.
I don't find it fair that you are badmouthing SceneCity without having ever used it or even having read the website. All apps are open source, the server code is open source, you can connect to the stream with any browser, you can join the chat using everything from Telnet over IRC up to apps. There are no gates outside of your imagination here.
My definition of gated would better match "Locks out all alternative clients, forces you to watch ads to see a demo, have to provide lots of personal information and a credit card to run a channel, and can get your account nuked for content that is streamed".
You have touched a topic that is a matter of each one's personal definition of a sub-culture: Yes, if you stream on Twitch for example, in addition to 20 sceners watching you might another 100 gamers/kids. For some that might be desired, other's might have the position "we are a subculture. our main audience is our subculture".
But it is what it is. Of course there are reasons people don't use scene run services much anymore. But it means that your art will now be under someone else's control, including how much of it is preserved long-term. I have given enough examples above of scene content that has disappeared or may disappear forever, soon.
I don't see where we disagree. I am somewhat aware that YouTube on a global scale is a bit more known and used than SceneSat, Demoscene.TV or SceneCity. But as pointed out in great length above, there are good reasons that the scene always had it's own infrastructure for over 30 years. We could have based the first scene networks on AOL. We chose BBSs.
I don't find it fair that you are badmouthing SceneCity without having ever used it or even having read the website. All apps are open source, the server code is open source, you can connect to the stream with any browser, you can join the chat using everything from Telnet over IRC up to apps. There are no gates outside of your imagination here.
My definition of gated would better match "Locks out all alternative clients, forces you to watch ads to see a demo, have to provide lots of personal information and a credit card to run a channel, and can get your account nuked for content that is streamed".
You have touched a topic that is a matter of each one's personal definition of a sub-culture: Yes, if you stream on Twitch for example, in addition to 20 sceners watching you might another 100 gamers/kids. For some that might be desired, other's might have the position "we are a subculture. our main audience is our subculture".
But it is what it is. Of course there are reasons people don't use scene run services much anymore. But it means that your art will now be under someone else's control, including how much of it is preserved long-term. I have given enough examples above of scene content that has disappeared or may disappear forever, soon.
@ziphoid
Thanks for your insightful post.
Potentially due to you being busy with replicating, you might have forgotten about what we discussed a year or so again - you could use our infrastructure. ;)
The amount of equipment we had bought for SceneCity is immense. For example we have three streaming kits that consist of a h.265 encoder two camera inputs and running OBS sitting around here. They were meant to be able to serve three demoparties at the same time (or one that wants to do multiple streams). We also have add-ons for bonded LTE/5G internet uplink for those streams, video mixers etc. All right now maybe used two times per year. Instead of here, one of those units could sit somewhere at SceneSat. Everything is better than this stuff we spent so much time building sits here unused.
And of course there is the server infrastructure with hardware transcoders, edge servers distributed all across Europe etc.
But let's not get into details of that here. If you ever find time, you know where to find me :)
Thanks for your insightful post.
Potentially due to you being busy with replicating, you might have forgotten about what we discussed a year or so again - you could use our infrastructure. ;)
The amount of equipment we had bought for SceneCity is immense. For example we have three streaming kits that consist of a h.265 encoder two camera inputs and running OBS sitting around here. They were meant to be able to serve three demoparties at the same time (or one that wants to do multiple streams). We also have add-ons for bonded LTE/5G internet uplink for those streams, video mixers etc. All right now maybe used two times per year. Instead of here, one of those units could sit somewhere at SceneSat. Everything is better than this stuff we spent so much time building sits here unused.
And of course there is the server infrastructure with hardware transcoders, edge servers distributed all across Europe etc.
But let's not get into details of that here. If you ever find time, you know where to find me :)
haven't properly read the whole thing yet but quick note to address this
i'm planning to upload an archive of my reports to scene.org and archive.org atleast. did so for mdt9k already.
Quote:
Someone might be putting a lot of work into a regular demoscene news show – but it’s hosted on YouTube
i'm planning to upload an archive of my reports to scene.org and archive.org atleast. did so for mdt9k already.
Quote:
But let's not get into details of that here. If you ever find time, you know where to find me :)
I haven't forgotten, it's just the lack of available time that has caused the non-conversation. :) I'll surely get back to you once I've burned down the pile of todos a bit further.
...and speaking of that replication of mine, I actually have cloned myself a couple of times. My two sons are extreme proof of that. Also, just to underline my dedication to SceneSat (and no need to start any discussion about being pro-Russia or anything like that now), our logo consists of a Sputnik. That also happens to be the middle name of my youngest son. :D
Quote:
AIs are getting trained on demos, no matter if you as the original author of a demo wanted that. [...] If you are now caught on camera at a demoparty, you are training an AI with your face.
Quote:
There was the plan to create a scene-internal LLM AI that is fed with all scene knowledge there is
ziphoid: recognizing your own (platform's) shortcomings is the first step towards resolving them. :)
scamp: i'm sorry i even replied now, i forgot how terribly you deal with any form of putting things in perspective. can't wait to hear from your buddies how terrible a person i am!
scamp: i'm sorry i even replied now, i forgot how terribly you deal with any form of putting things in perspective. can't wait to hear from your buddies how terrible a person i am!
offtop
psenough , is there a reason you stopped posting announcements for your news videos here on Pouet? Or did you just not have the time? I was even going to make an announcement for your next video myself, but I wasn’t sure if that would be a good idea.
psenough , is there a reason you stopped posting announcements for your news videos here on Pouet? Or did you just not have the time? I was even going to make an announcement for your next video myself, but I wasn’t sure if that would be a good idea.
Quote:
There was the plan to create a scene-internal LLM AI that is fed with all scene knowledge there is
for what purpose exactly? what would this be used for and contribute?
gargaj: Scene-internal is the keyword here. Trained on own servers, hosted on own servers, not sharing the data. Not uploading the dataset to huggingface. Being able to control the life-time of something. Making it only accessible to sceners.
lynn: I was thinking mostly a chatbot... like...
And stuff like that. Stuff where doing manual searches or hoping a human remembers would be complex, and where having all of the demoscene knowledge in a LLM would provide a benefit.
By the way, I should not have used the term "plan". "Concept" would have been better. I am aware that prior to the technical challenges, ethical questions would arise.
lynn: I was thinking mostly a chatbot... like...
- "What was that C64 demo that for the first time had FLI with all borders opened?" and it would return you a link to the prod on pouet/demozoo/csdb
- "Which members did Smash Designs have in the year 2005?"
- "Please give me an example on how to code a twister in C language"
- "How did the number of visitors from Japan to the German easterparties develop over the last 15 years?"
And stuff like that. Stuff where doing manual searches or hoping a human remembers would be complex, and where having all of the demoscene knowledge in a LLM would provide a benefit.
By the way, I should not have used the term "plan". "Concept" would have been better. I am aware that prior to the technical challenges, ethical questions would arise.
Quote:
Scene-internal is the keyword here. Trained on own servers, hosted on own servers, not sharing the data. Not uploading the dataset to huggingface. Being able to control the life-time of something. Making it only accessible to sceners.
That's okay, you do not have my permission to use anything I've ever been involved in.
havoc: I have never said a bad word about you that I also didn't tell you in person when we were sitting next to each other with me helping you on the outline video chain and streaming.
You are reacting hyper-aggressive towards me since the day I had taken the liberty to provide you feedback on pouet that I didn't like the noise level limitations preventing to have loud compos or DJ Sets at the current Outline location. A personal opinion based on facts, because I was one of the few who still danced with a 65dB limit in place with us trying not to breathe too loud while dancing :)
I find it very unreasonable how you are treating me since that. If you ever care to tell me why you are doing that, you know where to find me.
And I find it even more unreasonable to just make up bullshit about a site, just because I am involved in creating it. That's just not OK.
But let's move back to the subject...
You are reacting hyper-aggressive towards me since the day I had taken the liberty to provide you feedback on pouet that I didn't like the noise level limitations preventing to have loud compos or DJ Sets at the current Outline location. A personal opinion based on facts, because I was one of the few who still danced with a 65dB limit in place with us trying not to breathe too loud while dancing :)
I find it very unreasonable how you are treating me since that. If you ever care to tell me why you are doing that, you know where to find me.
And I find it even more unreasonable to just make up bullshit about a site, just because I am involved in creating it. That's just not OK.
But let's move back to the subject...
bitl: multiple small reasons, one being i didn't want to spam that spot with recurring news about the same thing, to leave some space for other scene things to have their visibility; the other is that news items submitted to bitfellas usually take a few days to manually approve (extra points if you guess which random hungarian scener is the last man standing doing that thankless job), so usually when they get approved i'm already on next week and if there were multiple of them queued the system approves them on the wrong date order and looks weird, so i try to avoid it and only use it once a month or so; and then there is also the issue that the main visibility for bitfellas news seems to be mostly on pouet itself (bitfellas being quite ontopic as another interesting case of demoscene infrastructure left unused by majority of sceners), so with all those things in mind i concluded i might aswell just post on the oneliner instead when it's out and hope the usual trolls don't immediately hide it with their shit talk, it's less "digital burocracy" required to announce there. ¯\_(ツ)_/¯ my expectation is that majority of people who want to see a new report will subscribe to the channel and get it directly there, or follow me on social media and get their notifications there. could maybe be skewed, who knows.
gargaj: I don't need a permission, because it was just an idea/concept. I am totally allowed to have a freedom of mind on this :)
(Also, as a side note: As you probably know recent court rulings in the EU have decided that you don't need anyone's permission to train a LLM on content that's public openly on the web. That would therefore be more of a question of finding a consensus inside the scene and/or have a vote on it, not an individual decision. But again: IT WAS AN IDEA. It's not gonna happen. Chill.)
(Also, as a side note: As you probably know recent court rulings in the EU have decided that you don't need anyone's permission to train a LLM on content that's public openly on the web. That would therefore be more of a question of finding a consensus inside the scene and/or have a vote on it, not an individual decision. But again: IT WAS AN IDEA. It's not gonna happen. Chill.)
And let me add this. If you look at our heritage, as I see it, we have the following options:
1.) Keep the scene infrastructure alive. Gargaj keeps running pouet, for example, Demozoo stays alive, Slengpung does etc. If someone wants to research something about the scene in 10 years, they can still do that manually then.
2.) Preserve all knowledge that might get deleted in an LLM. In 10 years from now, at least you have a central knowledge backup
3.) Accept that in 10 years time a scene historian will hallucinate about what the scene was like.
4.) Accept that most of our knowledge and heritage will be forgotten
I don't want to have this thread drift into a LLM discussion. But if there is one thing LLMs are good for, then it is to collect an amount of (albeit fuzzy) knowledge that's simply too much for a single person to know. That's why LLMs are perfect to diagnose rare diseases, for example. A single doctor in a village can't have read all the studies and papers published on every single topic in medical sciences ever.
But I very clearly would prefer 1.)
Again: It was an idea. I am not planning to invest the gigantic amount of time, energy and money needed to pull this one off.
1.) Keep the scene infrastructure alive. Gargaj keeps running pouet, for example, Demozoo stays alive, Slengpung does etc. If someone wants to research something about the scene in 10 years, they can still do that manually then.
2.) Preserve all knowledge that might get deleted in an LLM. In 10 years from now, at least you have a central knowledge backup
3.) Accept that in 10 years time a scene historian will hallucinate about what the scene was like.
4.) Accept that most of our knowledge and heritage will be forgotten
I don't want to have this thread drift into a LLM discussion. But if there is one thing LLMs are good for, then it is to collect an amount of (albeit fuzzy) knowledge that's simply too much for a single person to know. That's why LLMs are perfect to diagnose rare diseases, for example. A single doctor in a village can't have read all the studies and papers published on every single topic in medical sciences ever.
But I very clearly would prefer 1.)
Again: It was an idea. I am not planning to invest the gigantic amount of time, energy and money needed to pull this one off.
scamp: LOL, what a fantastic story teller you are :)
psenough: Appreciate the plan to archive the videos outside of YT.
I for example am someone who does not use YT subscriptions/notifications, because I haven't found a way or could not be bothered to manage notifications in a way that does not annoy me with all the devices I have with an YT app doing BING noises in the middle of the night etc. When I watch your show it's because I see it as a Pouet News entry.
(And yes, the route of submitting news through bitfellas sadly is too slow these days to be of much use.)
I for example am someone who does not use YT subscriptions/notifications, because I haven't found a way or could not be bothered to manage notifications in a way that does not annoy me with all the devices I have with an YT app doing BING noises in the middle of the night etc. When I watch your show it's because I see it as a Pouet News entry.
(And yes, the route of submitting news through bitfellas sadly is too slow these days to be of much use.)
Though it is long, I think it's worth reading the post. No matter what anyone thinks of the specifics, the topic is important and interesting. And it will affect the demoscene!
I am thankful for the infra. I can hardly think a live scene without sites like the ones mentioned.
I have a peek at Pouet and CSDb multiple times a day. Add to that demozoo.org and demoparty.net for upcoming events. It gives a feeling that things are still going on, constantly.
For me, trying to build some kind of social media presence is boring, often not successful and less than satisfying. I would prefer to dump the few things I do to demoparties, CSDb and Pouet.
And it's not obvious these sites will remain forever. I can understand that a lot of the new commercial sites for streaming, chat, sharing etc. are low threshold, but it can come with consequences.
Indeed the new generations of people (outside scene) seem fine with the eventual destruction of their digital trail. Some media are even built with the idea that nothing is preserved. And it could be the correct attitude for many things. Not everything that happens in the digital domain need to be preserved forever.
I am thankful for the infra. I can hardly think a live scene without sites like the ones mentioned.
I have a peek at Pouet and CSDb multiple times a day. Add to that demozoo.org and demoparty.net for upcoming events. It gives a feeling that things are still going on, constantly.
For me, trying to build some kind of social media presence is boring, often not successful and less than satisfying. I would prefer to dump the few things I do to demoparties, CSDb and Pouet.
And it's not obvious these sites will remain forever. I can understand that a lot of the new commercial sites for streaming, chat, sharing etc. are low threshold, but it can come with consequences.
Indeed the new generations of people (outside scene) seem fine with the eventual destruction of their digital trail. Some media are even built with the idea that nothing is preserved. And it could be the correct attitude for many things. Not everything that happens in the digital domain need to be preserved forever.
Quote:
That's okay, you do not have my permission to use anything I've ever been involved in.
I would like to explicitly state that you do not have my permission for that either.
