Yes, this is a provocative statement. In this statement I am strictly speaking as myself and not in my position as TASVideos Staff. THis statement is a collection of my own beliefs and observations after about 3 years being involved with TASVideos and its community at large.
As emulators entered the mainstream, save states slowly became a commonplace feature. They were convenient! We can save those long games of Sonic the Hedgehog, or forget about passwords for Metroid. Similarly, movie files showed up as a cute feature to record a sequence of inputs and show them play out in the game. DOOM Demo files were a big deal, and the ease of distribution for Demos was a huge boon for the speedrunning community that was born around the game. And what about pausing? Well, some games don't even pause! I just gotta go check the door, let me pause my game of Street Fighter.
What we didn't know at the time was what all of these features would allow when used together.
As the story goes, in November of 2003, bisqwit stumbled upon an interesting "time attack video" of "super mario bross 3". The video, created by Marimoto with the Famtasia emulator, was intriguing and fun and beautiful. Look at the skill! The long bounces and 1up chains. The daredevil jumps. It was awesome! What was this anyway? Was Morimoto that much better at SMB3 than anyone else? Well, no. This was a Tool-Assisted Speedrun.
While TAS work had already existed as early as 1994 in the DOOM demo scene, this SMB3 warp run was revolutionary, and led people (bisqwit included) to discover this craft. bisqwit himself figured that work like this was worth sharing and celebrating, and created NESVideos as a site to share them, and a tool to generate AVI dumps of the movies as played back from their emulators.
The stated goal of NESVideos, as shared in the available archive, is simple: they are art. While his views on what that entailed and the specifics on his taste on the movies itself is not something I agree with, he was correct: TAS work is a creative endeavor. A craft worth celebrating, sharing, discussing, and maybe even engaging in. The original goal is noble: NESVideos would curate high quality work for TASing, and provide a place to celebrate their authors and demystify the creative process.
Now, as we all know, bisqwit had very specific views on a TAS should be. In the archive of NESVideos, he outlines what he believes a tool assisted movie should be, and it can all be summed up in one word: entertaining. This decision led to him deciding on a very specific, and somewhat narrow, range of expression for what a TAS should even be, and these would get slowly codified into a proper ruleset as the system became formalized in the process of submission and publication. This system allowed him to curate the highest quality work, pushing authors to strive for excellence, and justified the large amount of work that publication would take at the time.
However, the world that bisqwit was in when he first found that movie is not the world we currently live in. For starters, as speedrunning becomes a larger part of gaming as a whole, naturally TASing has also become more culturally relevant within speedrunning and speedrun-adjacent communities. The advent of platforms such as YouTube and Twitch and new emulation techniques made both TAS creation and sharing available to just about anyone. Communities become more specialized as they push forward, with better tools and more advanced knowledge, creating extensive decompilations, trainers, replay devices and so on. As the artform developed and expanded, the goals, tools and views that different authors and communities desired shifted away from what the site was first made for. And at the time, TASVideos failed to recognize this shift. Furthermore, the site culture cultivated by its original vision made it uniquely hostile to this adaptation, and it took a long time to shake it off. We're still dealing with years, decades of bad decisions that led to many people, rightfully in my opinion, writing TASVideos off.
I admire the work of our current staff and their commitment to make the site the best version of itself it can be. I wouldn't be a staff member if it weren't for those before me and around me that work to make the site better. The love for the craft they show and their desire to make the site live up to its goals is noble indeed. But their work has now exposed a different, more fundamental problem: The current framework of TASVideos is fundamentally unsustainable.
As the site breaks off the binds it was born with, it finds itself more and more accepting of work. After all, that is what we want. But as we expand the rulesets and make our best effort to engage with more and more work, we find the simple truth that creativity is not to be bound. Longer and longer, as a judge, the rules feel less like what we work with and more what we work around, or, in increasingly more often cases, what we have to fight against. As the artform evolves, specializes and expands, our rulesets and guidelines feel ever less useful, ever more constraining, ever more obsolete. And we know this. Judges have never had to discuss more rulesets and changes than now. And as you would expect, different staff members have different views on what those guidelines should be, which is completely ok but also means arriving at consensus can take very long. Furthermore, thanks to the work we have already done, we have naturally been rewarded with exactly what we wanted: more and more submissions. 2023 was the single largest year in TASVideos history measured by number of submissions. 2024 so far is on pace to surpass it. The higher load on work, even if it wasn't work that particularly challenges the boundaries of our rules, has been welcome by a relatively small staff of volunteers, all of us just doing our best to help while dealing with jobs and lives and other interests. Absences are felt, burnout can be very real. We tried to enforce systems to keep us on track and contributing actively and it only made things _worse_.
While not all submissions take as long, the whole process has become significantly harder. Complicating matters, the publications that precisely push against us tend to be those that we most want to have, or at least have someone vouching for them. If we didn't want them, rejecting would be the much easier way to go. But we do, and then the process extends to the point where by the time some of these get published, the authors already shared them elsewhere. Steering the tsunami of creativity is foolish, and has led to burnout and exhaustion. Or worse: it leads to people not bothering to submit in the first place. After all, in a world where a commentated TAS that doesn't abide by our rules can gain millions of views on Youtube, why?
We cannot live like this. TASVideos will die.
... But it doesn't have to be this way. It never did.
The way the site was originally set was a response to specific cultural and technological circumstances, filtered by the creative vision of the site founders. But we don't live in those times anymore. The world of TASing does not work like it did when bisqwit stumbled upon that movie. His views, his circumstances, are not for us to follow.
As such, this is my statement:
* I believe Tool Assisted Superplays, Tool Assisted Speedruns, Tool Assisted Demonstrations and other related forms to be creative work of different kinds with unique and different goals. They turn the interactivity element of video games as a medium and build a new medium on top.
* I believe TAS work is worthy of celebration, archival, preservation, curation, and discussion.
* I believe TASVideos as a community should strive to archive TAS work and celebrate its history, development and the communities' accomplishments.
* I believe TASVideos should offer an introduction to TASing as a whole, as well as resources for those interested to both gain insight on what the craft is and how it is performed. I believe this to be fundamental for the future of the craft.
* I believe TASVideos can and should connect different TASing communities together.
* And most important to this discussion, I believe all of these goals, most of them concordant with the site's original mission, can be best accomplished without the submission and publication system currently in place. As such, I believe TASVideos should no longer have its submission or publication process.
Now, if you ask me "What should TASVideos look like if it were to abandon its submission pipeline and rules" the answer is... I don't know. I have some ideas, for sure. I think archival of movie files and knowledge bases for both the general TASing process and specific games would definitely be part of that, and these elements already exist in the currently standing TASVideos site. But I can't say for certain, and even if I did, I don't think figuring such a thing out should be my work alone. More than ever, we need people, and people are there. We ought to listen to you, to what you think the best way to achieve the site's goals is, and how we can work towards that goal.
Morimoto didn't need to go around each airship autoscroller by jumping constantly and bouncing on everything. But they did it because it was fun, and it communicated the superhuman nature of the run. In short, it was a creative decision. And it was a decision that worked so well that it made someone want to make a whole website to celebrate work like this. This small decision has rippled all the way into tool assisted work today, where autoscrollers, a part that commonly is derided in real time speedruns, can be some of the most fun a TASer could possibly have. They found a way to make autoscrollers fun.
TASVideos as it currently exists, shackled to its rules and guidelines, slowly inches forward, doing its best to escape the pit it dug itself into, with good intentions, but slowly consuming itself in the process.
And in the meantime, creators jump around us, gliding, collecting 1ups.
- ikuyo, TASVideos Judge.
As a "victim" of the rules, I understand the feeling you're trying to convey here, but I don't have any great suggestion or alternative, maybe if we could have like smaller judges to help with the judging process for some games. Like I said in another topics, speedrun communities should be a reference when judging. Maybe some grouping of the TASes for a specific game instead of just the TAS itself.
Example: I'd help to judge fighting games as it's the genre I have the most experience here. I still feel it would take a lot of work for Publishers if judgement process started being faster but I don't feel this is a reason not to speed up judging somehow. Or maybe keep the same judges and use other players as reference for help with judging faster.
Another possibility: peeps wont be able to find my many moviefile tests for Digimon Rumble Arena 2 to see faster completion with the other characters unless they go to the forum thread, Games by themselves could be TAS Archives, some of the moviefiles being published and others not but still properly archived.
Very provocative title, it really got my attention lol Dante from DMC SERIES must die.
I want all good TAS inside TASvideos, it's my motto.
TAS i'm interested:
Megaman series, specially the RPGs! Where is the mmbn1 all chips TAS we deserve? Where is the Command Mission TAS?
i'm slowly moving away from TASing fighting games for speed, maybe it's time to start finding some entertainment value in TASing.
* And most important to this discussion, I believe all of these goals, most of them concordant with the site's original mission, can be best accomplished without the submission and publication system currently in place. As such, I believe TASVideos should no longer have its submission or publication process.
Interesting take.
If we abolish the submission process then how exactly should things be verified to even be complete submissions or to be correctly assigned to a goal? Would anyone look at these submissions at all?
Without the publication process would the tasvideos youtube be rendered nonfunctional?
I do think this needs to a more cohesive process flow to really see what kind of end point you are aiming for. But would the idea of less or no judging at all, and judges becoming more of a content moderator type position be more inline with your idea? Or something like speedrun.com's process be more of what you are thinking of? Maybe something that just automatically puts the new run in the submitted category or creates a goal if it is a new one and just appends it to the game they selected and it all gets voted on by users without anything obsoleting other runs?
The title was provocative and click bait-y enough for me to click it lol.
Getting rid of the submission and publication process definitely seems like a radical suggestion from where we sit currently. But I also will say it's something I've wondered about sometimes from my five years here. I've wondered whether the judging process best accomplishes our goals of showcasing and organizing TAS work, and whether it can accommodate an expanded scope and role as for example a TAS record-keeping site.
I... find myself struggling to comprehend the full implications such a change would entail, to be honest. I would boil down my present thoughts that judging and publication serve some important roles such as ensuring at the very basic level a movie can be synced by someone other than the original author and posting a high-quality video to YouTube, which is where a lot of the broader audience gets exposure to TASes. Of course, anyone can post their TAS to YouTube and try to get traction, but TASVideos does have the benefit of an existing audience. The rules will never be perfect, but they do provide some constraints and clarity on some common types of movies that are well understood, and the current movie navigation structures are generally well-integrated with this purpose.
But that's not to say the site can't evolve... definitely something to think about.
I do think whatever direction the site goes, it's important that site staff isn't overworked for what is voluntary work in their spare time. We all should all be given the grace to do our best to take care of ourselves.
Ambassador, Moderator, Site Developer, Player
(155)
Joined: 3/17/2018
Posts: 358
Location: Holland, MI
I had the privilege to discuss this with ikuyo prior to posting and in person with DrD2k9 due to the timing of our Long Island Retro expo showing. I have a few thoughts:
I believe that curation is too core to the site to completely abolish. Some level of curation is necessary to have a place for things to be confirmed as syncing and confirmed to beat known records. dwangoAC's recent DEF CON talk is a great example of the importance of this curation layer.
From talking with DrD2K9, this should already be similar to the official process, but my opinion would be that judging should be minimized to replicating sync and either confirming with a game community that an improvement appears legitimate or if no community exists just to do a visual check that the TAS appears like a good faith attempt to set a baseline record. In this way I would love larger game communities to be handed more direct judging capability over their own games, and to have our current judges act more as the backup pool for games without baselines and final say on resolving disputes. This would allow for example the street fighter community to curate TASes of tens of different individual characters without burdening the judge system.
If that were to streamline the judging system, it leaves publication as a bottleneck. For that I would suggest having an alternate publication track channel "tasvideoschannel2" where the submitter or other community member provides the encode and we don't dig into the minutiae of the video quality. In other words the main channel could host for example the fastest character "any%" but the authors or community could supply encodes for channel 2 for as many other characters as desired or for those baseline uncontested records.
I aree with TIKevin that some degree of curation is necessary. Without that, there’s no reason to maintain the existence of the site beyond being a simple file repository for any and all things that can be labeled as a TAS, regardless of how well produced they are. That goes for both artistic and speed oriented runs.
Without somehow being able to say “these are the best TASes we’re aware of (that we have the means to reproduce),” what service (beyond file hosting) would we be offering people who want to create TASes? What meaning would publications have if there’s no curation? What impetus is there in striving for a publication if we don’t have a minimum standard of acceptance to give meaning to that publication? Without meaning to publications, why publish in the first place? And if we’re not doing some form of publishing, do we even need to exist at all?
I’m not trying to say that change isn’t an option, just that I don’t see value in changing to an uncurated approach.
Regarding current judging/publication bottlenecks (and this may sound ironic given that i just emphasized value): We currently have limitations on how frequently runs can be published (as well as strict rules on the act of publication). This timing delay is done, as I understand it, to maintain a level of highlight for publications on YouTube. Thus enhancing the value of each publication.
That said, these publication restrictions are also part of why we don’t accept more things than we currently do. For example: fighting games. I realize that there may be counter opinions, but there are multiple staff/community members who see the value in having a separate full run being allowed for each individual character in fighting games. The current issue isn’t that we think it’s wrong to publish such runs, it’s that we can’t handle the load under the current process. So, this currently isn’t accepted, but potentially could be if we had the ability to process it all.
If we were to loosen the timing restrictions on publications so that more runs could be published at a faster pace (basically as quickly as the publishing team can handle getting things published), then we may not have to be as restrictive on what types of runs are acceptable; so long as they still meet a standard minimum. (Here comes the irony…) while it may slightly lessen the value of a publication that comes from the current time exclusivity in the YouTube publication/presentation, all other aspects and value of publication could remain as they are now. I feel this approach would both further open up opportunity for authors to submit the types of runs they want, while maintaining a minimum level of curation.
Bottom line, whether we move away from our current system or not; it is the availability of volunteers (to actually do the minimal level of curation necessary for the site) that will ultimately determine whether or not the site continues or dies. More volunteers would help, period. Some may not want to volunteer because of current site policy; but to change to something different, we need the people to support the change first. So to anyone potentially interested yet hesitant, please be at least willing to join us. Just recognize that until/if a different system is a agreed upon, you need to be able to work within the confines of what currently exists.
EDIT: TiKevin83 wrote:
…judging should be minimized to replicating sync and either confirming with a game community that an improvement appears legitimate or if no community exists just to do a visual check that the TAS appears like a good faith attempt to set a baseline record.
Effectively this even further simplifies optimization requirements from the current approach of ‘doesn’t look sloppy to a casual viewer’ to ‘it’s the best known available even if improvements appear to be possible.’
In my opinion, publishing the baseline ‘best known available’ for a game is better than not publishing a game at all simply because there may be some actions in the game that are visibly not optimal.
So for judging, the bare minimum standard could be something along the lines of:
So long as continued progress toward the stated goal of a run is consistent within the run—no unnecessary extra activities taking place outside of minor suboptimal movement—and there are no other known complete runs of that goal which are better; the run is acceptable even with potentially apparent minor suboptimal actions.
Joined: 11/13/2006
Posts: 2822
Location: Northern California
Thank you so much for posting this. In all honesty, we've needed to have this discussion as a site for a while. I may not be able to be as active as I used to be given how my personal life is going, but I'll do everything in my power to be around for this going forward.
ikuyo wrote:
As such, this is my statement:
* I believe Tool Assisted Superplays, Tool Assisted Speedruns, Tool Assisted Demonstrations and other related forms to be creative work of different kinds with unique and different goals. They turn the interactivity element of video games as a medium and build a new medium on top.
Absolutely agreed. A lot of the reason that feos and I in particular worked to get Playground drafted and implemented was to be able to help showcase... Quite literally, anything that comes to us. I'm of the belief that no category is pointless, no TAS is a waste of time, and that there will always be someone interested in the product. It doesn't even have to be a final product! Even a WIP can inspire someone to work!
ikuyo wrote:
* I believe TAS work is worthy of celebration, archival, preservation, curation, and discussion.
* I believe TASVideos as a community should strive to archive TAS work and celebrate its history, development and the communities' accomplishments.
100%, and this has always been my goal for TASVideos, even before I became an admin. You mentioned earlier that the scene is different now, and that it took us way too long to adapt, and that's the absolute truth. I wish we could have done something sooner, but... Without getting into detail, there was a major factor holding us back that is no longer around to do so. That's why it took us until 2022 to revamp the site and start adapting to the modern world of speedrunning.
Just... We're not quite there yet. Not for lack of trying, but I think we could be doing better with that. We would need a more solid direction and a lot of community support, though.
ikuyo wrote:
* I believe TASVideos should offer an introduction to TASing as a whole, as well as resources for those interested to both gain insight on what the craft is and how it is performed. I believe this to be fundamental for the future of the craft.
Weirdly... I think this is one of the most difficult things on the list. Of course, I want the same thing, and I agree it's absolutely vital, but from experience I know that TASing is hard to teach, while paradoxically being easy to learn.
I think there's just this mentality among newer TASers that what we do is a lot harder than it actually is. I'm sure a lot of that mentality comes from our tech standards, but I think the entertainment factor has also contributed to it in a major way. Ultimately, the most popular runs are the ones that need the most technical knowledge to be able to do. For one thing, TASBot popularized ACE by showcasing bigger and even more complex runs every GDQ, and I think a lot of people just assume that's what all TASing is by default, but even highly regarded non-ACE runs have such insane levels of technicality to them. Super Mario World and all the Sonic games on Genesis having entire suites of Lua scripts, Pokemon games being dissected and practically automated through botting, SM64 gaining more and more new tools outside of the emulator... I'll be honest, if I hadn't gotten into TASing early, there's a very good chance I'd think it was too difficult to get into.
I'd like to just take this line and stress that I'm not blaming any of those things for implying that TASing is harder than it actually is. Especially for TASBot, they've all been exceptional promotion for the hobby and have pretty much only done good. If anything, I think TASVideos not knowing how to teach TASing is more to "blame" than anything else.
The reality is, you don't need the fancy tools like Lua scripting and disassembly. You don't even really need RAM watch. Hell, I've seen deeply optimized TASes done without even using frame advance or slowdown, just rerecords. The beauty of TASing is that every TASer finds what works for them and they can excel in it.
And yet, the problem is that you can't teach a person how to do something themselves.
I've been asked where to begin with TASing and my answer has always been to just mess around, free of any constraints, but that's ultimately the most dismissive answer that can be given even though it's correct. We have resources that explain the tools, but we don't have anything that really explains the mindset of TASing, which is by far the most important part. I've tried writing about that before and it never comes out right.
This has been a problem for a while. I really want to see it fixed, I just need ideas.
ikuyo wrote:
* I believe TASVideos can and should connect different TASing communities together.
Once again, this has been a goal of mine for TASVideos. As much as I dislike Discordification breaking up and sheltering communities, it would be wrong of me to just ignore it and continue trying to treat TASVideos as THE ONLY place for TASing. I'm pretty sure there are multiple communities spawned from TASVideos that have grown bigger than TASVideos, and we could be doing better about guiding people over to them.
I've said it before, but what I want TASVideos to be is...
Above all else, an archive and showcase of TASes and TASing resources, and a place where creators are comfortable letting us do so
A hub or jumping off point for people to find and join individual game communities if they so wish
A safe and welcoming community in and of itself for TASers, speedrunners, fans of TASing and speedrunning, and members the ROM/emulation scene in general
We're getting there, but it's very clear we still have a long way to go, even just to fix the lingering negative perception we have amongst the overall speedrunning community.
ikuyo wrote:
* And most important to this discussion, I believe all of these goals, most of them concordant with the site's original mission, can be best accomplished without the submission and publication system currently in place. As such, I believe TASVideos should no longer have its submission or publication process.
I... think it's finally time for me to talk about that. This is something I've been very quietly supporting and occasionally pushing for in private channels, so I've got a lot to say about it.
I'll start here. All emphasis is mine:
ikuyo wrote:
As the site breaks off the binds it was born with, it finds itself more and more accepting of work. After all, that is what we want. But as we expand the rulesets and make our best effort to engage with more and more work, we find the simple truth that creativity is not to be bound. Longer and longer, as a judge, the rules feel less like what we work with and more what we work around, or, in increasingly more often cases, what we have to fight against. As the artform evolves, specializes and expands, our rulesets and guidelines feel ever less useful, ever more constraining, ever more obsolete. And we know this. Judges have never had to discuss more rulesets and changes than now. And as you would expect, different staff members have different views on what those guidelines should be, which is completely ok but also means arriving at consensus can take very long. Furthermore, thanks to the work we have already done, we have naturally been rewarded with exactly what we wanted: more and more submissions. 2023 was the single largest year in TASVideos history measured by number of submissions. 2024 so far is on pace to surpass it.
The higher load on work, even if it wasn't work that particularly challenges the boundaries of our rules, has been welcome by a relatively small staff of volunteers, all of us just doing our best to help while dealing with jobs and lives and other interests. Absences are felt, burnout can be very real. We tried to enforce systems to keep us on track and contributing actively and it only made things worse.
With all that we've done lately, there's no way our current system is going to be sustainable long term. I'd even argue it hasn't been sustainable for a couple years now, we've just had an absolutely incredible staff team that have all been working overtime to keep everything afloat. It's only going to get harder to stay above water as we keep going with rule revamps and more allowances, too.
Just to remind everyone, our first submission record was set the year that submissions were opened, back in 2004. We had over 500 submissions that year, and up until recently it was the only time we had broken 500. The year we switched to the new site, 2022, we immediately shattered that record by nearly 100 submissions, breaking 600 for the first time. In 2023, we obliterated THAT record by almost 250 submissions. We got close to 900 submissions last year. Thankfully, submission volume has slowed down in 2024, but we're still far above our historical average. If we assume 2022's total of 639 submissions to be a low average going forward, then that average still outweighs every single year of the old site's operation, and the more we make things accessible, the higher that average is going to get from year to year.
We're not going to be able to keep up with that. We already can't. TASVideos is, and most likely always will be, nonprofit. This is a hobby that we all do in our free time. Everyone has lives outside of the site. Hell, I'm writing this at 3AM because it's the only time I really have to myself lately. I've been increasingly tied up in my personal life over the past year or so, and the amount of stress I've felt from being one of the top representatives of the site while knowing I don't have the free time to be around as much as I think I should be has been immensely difficult to deal with.
...And that's just me. We have a decent number of staff members that also have their own personal lives to take care of. Paying jobs, families to take care of, self-care and other hobbies, health issues at times... Oh, and on top of that, we've historically always been understaffed, particularly in publishing. We've even lessened the requirements to become staff and we still don't see people applying. Pretty much all of the recent appointees have been people we've directly reached out to.
For the site to be sustainable as it currently is, I honestly think we need double the Judges and at least three times the Publishers we have now. We currently have three Admins just to make sure that there isn't a single point of failure like there used to be, and... Well, look at my signature at the time I'm posting this, or better yet, look two paragraphs up.
That being said... Hiring enough staff to be sustainable isn't always going to work, because even with enough staff to handle everything right now, we're still growing. Eventually, we're going to reach a point where we can't have enough staff to handle things with our current standards. This is because it isn't just submission volume that's increasing, it's also publication volume. We're getting more submissions and we're also publishing more of what we get.
Just for some statistics:
In 2023, we received 885 submissions, and published 712 of them, for an 80% publication rate. It was had 74% in 2022, 70% in 2021, 70% again in 2020, and 60% in 2019.
We published more movies in 2023 than we did in 2019, 2020, and 2021... Combined. 701 total in those three years, 11 shy of 2023 by itself.
We had more publications in 2023 than we had submissions in any other year.
Samsara wrote:
Oh, and on top of that, we've historically always been understaffed, particularly in publishing.
We had roughly 10 active Judges and 4 active Publishers in 2023. 10 Judges splitting up 885 submissions, 4 Publishers splitting up 712 publications. 10 Judges scrutinizing everything that comes in to see if it's up to our optimization standards, tech standards, and current movie rules, and 4 Publishers processing the 80% of acceptances up to our current encoding standards.
The thing is, Judging and Publishing aren't necessarily difficult. I've done both, and I'm a complete heckin' idiot and frickin' moron! The problem is that they can be very time-consuming. A Judge or Publisher being inactive already means that everyone else has to work more, but even active ones can get hard stuck on complicated submissions, requiring other people to step in and away from what they're working on, slowing down multiple people with technically 0 loss in actual staff activity. We've genuinely had submissions that no staff could handle, leading to us needing to reach out to the community for help.
Not every run is over a month long like Desert Bus or a sync nightmare like (INSERT HIGH PROFILE GAMECUBE GAME HERE), obviously, but the number of runs like that coming in is increasing over time as we refine our rules and start accepting more and more complicated systems as they become available to us. For example, Flash has been a pretty popular platform as of late, which is lovely! It also means that our majority Windows-using staff all need to set up Linux VMs in order to process those libTAS submissions, and not all of us are able to do that, meaning more work goes to the people who can.
So... Where does that leave me? It leaves me precisely... here:
ikuyo wrote:
Now, if you ask me "What should TASVideos look like if it were to abandon its submission pipeline and rules" the answer is... I don't know. I have some ideas, for sure. I think archival of movie files and knowledge bases for both the general TASing process and specific games would definitely be part of that, and these elements already exist in the currently standing TASVideos site. But I can't say for certain, and even if I did, I don't think figuring such a thing out should be my work alone. More than ever, we need people, and people are there. We ought to listen to you, to what you think the best way to achieve the site's goals is, and how we can work towards that goal.
I've spent the last four years helping TASVideos to get to the point it's currently at, both in terms of spearheading major changes to submissions and encouraging discussions about various other aspects of the site. Everything I've done here has been with the intent of helping the site progress. I don't think I've completely succeeded, but I'm damn proud of the work I've done and even prouder of what the site has become thanks in part to my work.
This, above all else, is going to be the hardest discussion the site has ever had.
I, too, have no idea how to proceed from here, and I feel we're a very long way away from even starting to reach that point. TASVideos needs to change to be sustainable, but the way it needs to change involves fundamentally changing the site in a way that not everybody is going to be able to agree on. As easy as it may sound for us to drop our standards and make it easier on the staff, inevitably there's going to be culture shock as a result. Not just from us, but from the community at large. The site as a whole will have to adapt to a brand new way of doing things. We have to shift away from core principles we've all seen and followed religiously since the beginning, core principles that were never and should never have been challenged before... but they're core principles that were made for a smaller, more niche TASVideos. That's not us anymore, and that's definitely not what we're going to be with the direction we're heading.
In theory, the easiest short term bandage is appointing more Judges and Publishers, but that requires people to be interested in stepping up, and given this entire thread is about the site's increasingly unsustainable workload... I don't exactly see anyone volunteering their time right now ._. I guess the best question to ask everyone right now is what can we do to change that? What things can we change, or what possible misconceptions can we clear up, that would make people more likely to volunteer to help us? Is it going to take that large, sweeping reform we'll eventually need, or is there something smaller we can do in the meantime? I'll give it some thought, myself, but the less brainpower that I use outside of my priority of taking care of my family, the better.
As long as I'm above ground, though, I'm going to do everything I can to ensure TASVideos lives. It's a damn good thing the problem is "oh no, we're getting too big" instead of the opposite, which could have easily been the direction we went without everything we've done so far.
This ain't RHDN, we're not giving up on this.
TASvideos Admin and acting Senior Judge 💙 Currently unable to dedicate a lot of time to the site, taking care of family.
Now infrequently posting on BlueskywarmCabin wrote:
You shouldn't need a degree in computer science to get into this hobby.
Hi, I haven't been around enough lately, sorta been burned out. Anyways, I had some ideas on what we can do to change things to make things potentially easier. I'm not going to bother quoting people cause tired and lazy.
First of all, what do we really want out of tasvideos? I assume a big thing is that we want a place where people can reasonably find the tases they're looking for. Whether that be the fastest way to beat Mario 1 or some funny tases that they haven't seen before. So at the very least we need people to ensure things are some sort of organized. Also presumably we want things to promote collaboration and transparency. I doubt any of us want to start seeing "TAS cheating scandal!!!!!!!!!!" in our youtube recommendations.
Sync verification is a major issue and leads to a lot of frustration. Save file verification has some similar complaints. Instead of it being pass/fail, we could just mark tas entries as "verified emulator sync" or "unconfirmed emulator sync" or "confirmed no sync" and not reject over it. People can decide for themselves whether or not they want to trust that.
Instead of requiring a tasvideos staff created encode before an entry for a tas is created, we could simply create an entry with the temp encode linked and then update later with an "official" encode later if so desired. Or maybe the community doesn't see the need for such encodes anymore. I don't know! I am open to listening however.
[16:36:31] <Mothrayas> I have to say this argument about robot drug usage is a lot more fun than whatever else we have been doing in the past two+ hours
[16:08:10] <BenLubar> a TAS is just the limit of a segmented speedrun as the segment length approaches zero
I personally believe easier tools for laymen to help with publishing would be the most immediate endeavor you could use to multiply the number of publishers, as they're the fewest.
I want all good TAS inside TASvideos, it's my motto.
TAS i'm interested:
Megaman series, specially the RPGs! Where is the mmbn1 all chips TAS we deserve? Where is the Command Mission TAS?
i'm slowly moving away from TASing fighting games for speed, maybe it's time to start finding some entertainment value in TASing.
The title of this topic was so goofy that it made it hard to take seriously, since shutting down the site itself isn't really a serious goal that can be acted upon. But reading through the thread, it seems like the topic has evolved into being more about ideas to streamline the site to make it more sustainable and better equipped to handle long-term growth. So I'll give my ideas on a way to restructure the site to support that.
The first thing is I think Playground submissions should be more visible than they currently are. I think this has been a general goal for a while and I'll go into a few more specifics later in this post, but it's important to mention because of my next point...
Second, I think non-standard goal submissions should bypass the judging process entirely and go directly to Playground. This could be accompanied by basic filtering/moderation for spam or incomplete WIP TASes, but I feel like this would take some of the workload off judges, especially around the beginning of April when intentionally Playground-worthy submissions dramatically increase. And with this system comes the next thing I'd propose...
Third, I think such TASes that are submitted directly to Playground should have the ability to be moved back into the queue and judged for the Alternate class if there's enough positive feedback. This could be based on a certain number of nominations by TASVideos members and could possibly be done based on a quota system in peak submission times (although I'd hope that this would be relatively rare and reserved for times when staff is especially short-handed). If you look at submissions right now then you'd see that most of the Alternate-eligible TASes are aiming for publication, but if Playground is better established and more visible as a "baseline" for TASes on the site then Alternate could slot in as more of a "curated" set of non-standard TASes, ultimately publishing fewer to the main Youtube channel but highlighting more on the site.
I think the above structure also plays nicely with a few things others have already mentioned:
TiKevin83 mentioned having a second channel for community-provided encodes. Perhaps this could be a space for encodes supplied by the submitter for TASes in Playground. This has the side benefit of allowing Youtube viewers to watch brand-new submissions as they come in rather than 4 months after the fact like with publications, and it could also drive more engagement on the submission threads when feedback is most important.
Along with the previous point, KusogeMan mentioned streamlined tools to help with publication. Although most submissions are submitted with temp encodes, some of them are not. If there was a streamlined and mostly-automated process on most platforms to generate some more-or-less standardized good quality encode then the previous point would be much more effective.
Memory mentioned the ability to mark entries based on whether they've been confirmed to sync. I think this would fit in nicely with this system. And it doesn't take a judge to check whether a TAS syncs, so this could be performed by a much larger group of people. It might not be a good idea to make it available to literally everyone on the site due to the potential for people abusing it, but it could be available to published authors, for example, or if outside community members want to help then they could be given a specific permission just for verifying sync.
Now, what I've said so far does leave some gaps, most notably Standard goals. I think the current queue system for Standard goals should remain in place. The main mechanism for increasing sustainability for the site under this proposal would be to drastically cut down on the number of what are currently Alternate submissions, giving judges and publishers more time to spend on Standard submissions. And since this will be primarily focused on the technical aspects of a TAS, expanding the number of people who can perform that sort of evaluation would certainly help, especially speedrunning community members who don't currently engage with TASVideos.
It would certainly take some getting used to if this were to be implemented, just like the shift to the current Class system took some getting used to. I remember people getting upset when their TASes were reclassified from Moons to Standard because Moons used to be the "better" class. And it also seems pretty counterproductive against the goal of showing off more TASes to respond by publishing fewer TASes, but that's exactly why it needs to be accompanied with better visibility of Playground submissions. And that leads to my last point...
Alongside the system I proposed, I suggest renaming "Playground" to "Self-published" or something similar. The name "Playground" sounds like the kids' table of submissions, and if this were to be the starting point of all non-standard goals then I think a rename would be in order. By calling it "Self-published", it would better reflect the vibes of what it would mean and would intuitively delineate this from the site-published Standard or Alternate classes. And this name would also better support the optional self-publishing of Standard submissions as well.
I know this is only one of many ideas that have been proposed over the years, and it almost certainly won't be implemented verbatim, but I hope that something conceptually similar to this would be what the site moves toward eventually.
(As a side point, "Self-published" would also be a great way to be able to publish and highlight showcase runs like the Triforce% ACE Showcase Submission #7726S. I know there's something more specific in the works, but showcase runs attract a lot of attention immediately after they happen and so having a fast self-publishing process would certainly help.)
So in summary, what I'm proposing is the following:
Playground should be renamed to Self-published. Non-standard goal submissions can only be Self-published and can't be submitted to the queue, while standard goal submissions must be submitted to the queue but can optionally be self-published as well.
Self-published submissions should be more visible on the site in some way, and submitters should optionally be able to submit a standardized-format self-made video encoding for publication to a second TASVideos Youtube channel (with light moderation to prevent spam/undesired content etc).
Self-published non-standard submissions should have the ability to be nominated by community members based on their subjective evaluations of quality and entertainment (made easier to evaluate if it's self-published on the Youtube channel), and submissions should go to the queue and judged for the Alternate class once they pass certain thresholds for these nominations. This decreases the number of TASVideos-published Alternate runs and it offloads the current evaluation of entertainment away from the judges and onto the general community of viewers. Community members could also verify sync, further offloading some of the workload from the judge.
I think we can all agree that TASVideos has pretty much become a victim of its own success.
...But I'm extremely not sure if getting rid of the submission and publication process is the right approach, especially its implication. To me, and I hope I read it wrong, it reads off as getting rid of the publisher role altogether, which is not something I personally can support. And in fairness, I'm also very not sure about publishing the authors' temp encodes/having an alternate channel for users to upload their published TASes. If that were to happen, then firstly, IMHO, TVC will have failed at its primary purpose: an archive of every published TAS on this site. Secondly, what would be the point of having publishers in the first place?
I'll be frank, as it is, both proposals really don't sit well with me, for the reasons I mentioned above.
Joined: 10/12/2011
Posts: 6449
Location: The land down under.
It's worth also noting that Publishers were waiting for a Judge to pick up something to be Accepted for a good couple of months, where there were 0 Accepted TASes on the bench even when there was 50+ Submissions on the bench unclaimed/being judged.
Now as of current typing. Publishers have 11 Submissions on the bench that's Accepted (not claimed), and this topic is a bit of a killer for some of us, even prior to the creation of this topic, some of us are having other struggles outside the site.
And we only have 4 Active Publishers that held the fort-- well let's be fair 3 Active Publishers.
removed themselves from that count
We were unable to move forward without a TAS being accepted in order for it to be Published.
Even when something was "New" for a period of time, some of us knew the eventual end result to a Submission.
To create an idea that works without overlap for a Publisher/Encoders is to create a system that prevents us from stepping on each others toes in the process.
The method is to have an actual integrated system to which we can Publish in 4 dedicated slots (4 hour splits, this is what we've had) where the account is most active for viewership, and for Reencodes 4 dedicated slots (2 hour splits, this is what we've had) where the account is less active.
We know that for viewership (in GMT+0) 12p, 4p, 8p, 12a is when it's most active. And at 4a, 6a, 8a, 10a is when it's at its least.
But, the site does not support a scheduling system to be hooked in.
If we had such a system, that's 28 Publications and 28 Reencodes a week if Encoders/Publishers decided to do extra.
We can work, without stepping on each other, without reminding one another when one of us are going to publish, and so we don't have to create a burnout when we go too hard on something, cause something is set.
In the current state of how the site operates we can use YouTube's scheduling system for Reencodes specifically.
At a period of time I was uploading Reencodes at 8a (GMT+0), because that's when I knew that not many people were watching the account, because it was considered a lower viewership count.
I had the descriptions pre-done, and the only thing that I would have to do after those Reencodes went live, was to switch the YouTube links later.
This wouldn't really work for a Publication due to well we would need a feature to actually do something about it.
There's no point in a scorched earth type policy that doesn't work out, when there's many ways to prevent that, but it's dependent on site advancements which we don't have the manpower to complete well other things that are also required for Submissions, such as "Event Entires".
And there's also no point for a second channel. It's something that's really easy to prevent.
And if a second channel was made, then as Stella wrote. We will have failed.
I didn't know where to put this, so I'm putting it here, below the break.
And yeah. Publishing isn't necessarily difficult, depending on the system target. But the importance of something can be very time-consuming when something hits out and requires our attention.
I archive the important scripts when I can, because I know the importance of a later generation where this sort of information is practically vital, and having this is better for the future.
And sometimes I remember to ask other Publishers for their more time-consuming ones, and sometimes, their scripts are lost if they aren't backed up.
Then again, who would actually decide to backup 3TB of avi Desert Bus footage?
Me, the answer is me, since I published from the lack of action, whilst getting more demotivated by the day behind that publication.
For Judging, and the Judges have seen this from me.
I put something in a table, and mark it down in a colour based on how the rules are set for something to be visibly optimal versus technically.
It's an aid to help them with the 50+ Submissions on the bench that they have to go through and I won't lie, I started doing that system when there was/is 0 Accepted TASes.
WebNations/Sabih wrote:
+fsvgm777 never censoring anything.
Disables Comments and Ratings for the YouTube account.Something better for yourself and also others.
As a simple TASer, I am interested in popularization of my work. TASVideos has a YouTube channel with a lot of subscribers. I want my TAS to be recorded and published on that channel. I want views, likes, comments and a storage of my input files, in case published videos are deleted/banned/whatever.
Give me your requirements, I will try to satisfy them!
I don't TAS shovelware games, I use TAStudio all the time (no slow-motion recording), I don't enter cheat-codes.
If you trust me, feel free to be lenient regarding my submissions.
If not or if you prefer to take your hobby-job seriously, here is my input file with ROM hashes. Check everything suspicious in my movie.
For now I have nothing to complain about, including rejected submissions.
Regarding possible contribution to the community.
- I am not very skilled at describing movies. That's why I don't ask to be a reviewer.
- Judgement? No, I don't have time to perform deep investigations.
- I have a computer (GTX 1060, Ryzen 7 3800X), that's not utilized that often. My internet is fast.
If you provide a step-by-step instruction for complete dummies (because I previously didn't get lucky with TASVideos AviSynth package), I may encode and upload some TASes.
TASing is like making a film: only the best takes are shown in the final movie.
I agree that there should be some degree of curation or moderation at least. Just I do not think it needs to be a process that blocks a submission from being on the site. Or even that everything needs to be curated. Like if we have playground type stuff that could just be there on its own like SFII single fight playarounds for example. Just need to be upfront with what the movie is, and where it is in the process with all flags set (whatever we could decide they are). About publication...honestly I would be fine if people just linked their own youtube videos on it.
Here is an idea for a process.
Submission. User selects game and game region. Video link inclusion possibly required.
Verify correct game. Submission goes to temporary area where a judge would verify it is the right game. But NOT check goal and sync at this point.
Set flags. Set "to be curated", "video not verified" "sync not checked" "goal not verified" and then it would be archived to the correct game.
Judge checks Sync OR Goal. Judges can select which ones to check goal or sync.
After either of those are done then mark that flag.
When both are complete the rest of the flags like fastest completion could be set if it satisfies those criteria if someone wishes.
Publication.
----------------------
Publication has had some ideas. Publication needs to be manageable and I am not sure what that would look like. However personally I would advocate for something that is not so dependent on staff to get a more streamlined approach.
Publication process ideas summary abbreviation. Think I got the gist of the ideas here.
Include Alternate tasvideos channel where the submitter or other community member provides the encode. main channel could host for example the fastest character "any%"
Be more restrictive on what could be published so publishers can handle the workload. Such as only publishing "Staff picks".
Still have publishers continue to try to publish everything
Feature Ideas to assist publication
Streamlined tools to help with publication for members.
This thread has a lot to process so I don't quite have fully thought out statements to make, though I do have two things I wanted to at least do.
Dimon12321 wrote:
Regarding possible contribution to the community.
- I am not very skilled at describing movies. That's why I don't ask to be a reviewer.
- Judgement? No, I don't have time to perform deep investigations.
- I have a computer (GTX 1060, Ryzen 7 3800X), that's not utilized that often. My internet is fast.
If you provide a step-by-step instruction for complete dummies (because I previously didn't get lucky with TASVideos AviSynth package), I may encode and upload some TASes.
I can use this post to help with my first thing, since I can also respond to each part. From my experience as a staff member, I've definitely seen folks that think being a judge is some sort of difficult task. In reality it's usually not all that hard since my general process is to watch the TAS in the emulator for sync and note if anything feels visually wrong in terms of optimization. Authors usually have good author notes that explain those concerns. I usually don't go in to full investigations unless it's something that feels wrong in a major way and there's no justification for bits to feel off. If it's something minor ("maybe a few frames and that's it") but otherwise feels optimized when watching it then it's just something I bring up in judging notes.
As for the points in Dimon12321's posts:
A review doesn't even need to be comprehensive. It can be as simple as "the submission's input file syncs from start to finish", which at minimum means that it's sync stable. Even something as small as that can helpful in situations where a judge doesn't have access to something that can play the TAS themselves for whatever reason e.g. being in an area with limited internet access or having to deal with computer repairs.
I've already said my piece that the judgment process isn't actually all that daunting. Even if you have questions, you can always talk with other judges, reviewers, as well as the community for feedback to help with judging.
Joined: 4/17/2010
Posts: 11495
Location: Lake Chargoggagoggmanchauggagoggchaubunagungamaugg
Speaking from direct personal experience has always helped me improve things, and I happen to have a lot of personal experience with matters described in this thread...
The first TASes I saw in my life, of the games I played myself, I didn't realize they were done with tools, so you can imagine my absolute insane fascination with what those gods could pull off in real time! It was one of the strongest positive emotions in my life.
Fortunately it didn't really fade out after I've learned that TASing is not a real-time thing at all. Instead while discovering every aspect of this hobby in 2010, I got fascinated over and over: by the expertise and creativity needed to design those movies, by HD encodes, by judges and their high-level work, by emulator coders and all the unique tools, and to top it off, by the warm and kind overall mood of the forum (especially after the places I've been before TASVideos).
Fascination with judges persisted for years, and I was dreaming of becoming one, which happened in 2013. The first few months were very interesting, but eventually it became boring and there were a lot of submissions to judge, so I cut a whole bunch of corners to speed myself up. Which led to my demotion in 2015, and in 2016 I developed a new approach to judging (maximal thoroughness) which made it interesting for me again... for a few more years.
Then in 2018 I became a senior judge and got to organize people as well as the rules. It was hard but rewarding. But I slowly started feeling a certain problem: I stopped having any time on TASing. The only submissions with my name after 2016 were things others did on top of my old work, or my trivial tweaks to movies I judged. #8889: feos's Flash Samorost in 04:21.24 was the only bright exception where I had a chance to be creative again.
I was unable to finish the MAME core for bizhawk because I had to look after the submission queue. I stopped publishing because I stopped having any spare energy on that extra job. I don't have time to finish my current emucoding project - Amiga in bizhawk. Heck, I was only able to have this insane Wiki: Activity rate because I kept switching from job to job the second I got tired of doing the same thing. Combined with constantly working with people on people problems, this made it not unbearably boring.
But even though I don't have a real life besides my RL work, I still don't have enough energy to process the submission queue even if it's the only thing I do for like half a year (like it was in 2023). We never reliably have enough active judges for enough time to let them work in a relaxed way while still not drowning in submissions they have to process.
Due to insane amount of judgments I have to do, I stopped enjoying it many many years ago. The only bright side is when there's some movie that requires thinking out of the box or some research, and I accomplish it after fearing it for a bunch of weeks. That feels awesome. All the rest is empty duty, and the more new submissions I see these days, the more it kills me that I have to work more and more and more just so the queue is not 500 in size. I have to sacrifice everything I ever wanted to be doing, to this. And even then no matter how hard I crunch, the best thing we get is the queue not increasing for a while. Right now it's 83. In 2018 when I first started caring about it, it consistently stayed at 30 no matter how we crunched.
This summer I realized I do not want to overwork anymore, and I can't. I spent a few weeks working on AmigaHawk instead of judging, felt refreshed, claimed like 10 submissions to handle them quickly, and halfway though got overwhelmed again by a certain issue: no matter how much we overwork, there's no goal to possibly accomplish with judging. No matter how much you've overworked before, there will be infinity of new submissions to work on like nothing happened. It doesn't get easier, it doesn't get more efficient. Every edge case is like building everything from scratch, because reality forgot to give a fuck about our policies, so all we do is constantly hacking our policies to be less optimistic and more realistic.
That's right, we've been working on policy improvements for years like crazy, and they are not going to ever match what this hobby needs. It's always what we need from the hobby. But as it was rightfully noted in this thread, our rules limit creativity more than they help it.
Now I can't say I have a clear vision of what exactly needs to be fixed in our policies and how, to make everyone happy. And interestingly, I can't even come up with a way to improve TASVideos to be more sustainable. The staff team that we've had for the last few years is so awesome that after all the reforms, the overall current system does feel ideal in terms of achievements and rewards. The only fundamental problem truly is we don't have enough people to do this at the required volume.
Maybe we just need to be more demanding for new judges and publishers. Because really the site will die if everyone burns out, which will happen with the current ratio between queue size and the amount of people who manage it.
And either we manage to attract more staff members, or we'll have to reduce the work they do. There's no easy and enjoyable solution here. And the community has to understand it too.
I do feel that the current situation is deeply rooted in our site culture: admins and staff were so dictatorial that only minimal amount of things were outsourced to the community. Staff would manage everything according to whatever vision was prevalent at the time, and the feedback would be parsed only within that vision, not as a way to constantly improve that vision. The community got so used to it that now people don't seem to even think that judging and publishing are important and interesting jobs to do - they know staff magically does everything for them. So now if we try to outsource some things, barely anyone would volunteer. Or maybe it depends on how eagerly we ask? I'm not sure.
There's a high level-matter to all this as well.
I spent the last couple decades of my life pondering leadership and proper ways to inspire people, while also ensuring the job is being done. And on one hand I had real-life dictators that cling to their unlimited power like it's the only good thing in their life, and on the other hand I had my own first-hand experience in keeping things alive and well.
The more power a dictator collects, the more catastrophic the crisis will be, because complex systems can't function with only one point of failure: work needs to be distributed, or it won't be done. And since they never distribute the work, it's not being done. They just monopolize the spheres they don't understand, and then they ignore what needs to be done in them, because they're incompetent and dumb. So everything gradually degrades.
A true leader is not just one that gets things done, but also one who's built a system that works on its own without them. A true leader does not want unlimited power! Because they know how hard it is to successfully control things and ensure they work well! So they try to distribute the workload while also dreaming about not having to continue the manual routine.
Which has been my dream for a lot of years. I kept dreaming about having a team that would be able to do the grind without my direct involvement, so I could do things I like while only overlooking and resolving their questions.
And I can't even say there's no such staff team... they just have such an insane amount of work to do that I can't afford enjoying life while they're grinding. I'm dreaming about not having to judge anymore, but ever-increasing queue has been one of my nightmares for many many years. I don't know how to solve this, but this summer I decided to stop worrying about the queue size, while still doing whatever I can. The hard part is that I seem to have reached my internal judging limit. I may still do precedents or technical things, or maybe trivial things, but I really want to start judging less. I really want to do other things without failing to sleep at night over giant queue.
So once again I have no idea how to solve this in a neat way that everyone would love. We either need to magically get 2-3 times more people on staff, or we need to do 2-3 times less work.
Warning: When making decisions, I try to collect as much data as possible before actually deciding. I try to abstract away and see the principles behind real world events and people's opinions. I try to generalize them and turn into something clear and reusable. I hate depending on unpredictable and having to make lottery guesses. Any problem can be solved by systems thinking and acting.
Moderator, Senior Ambassador, Experienced player
(907)
Joined: 9/14/2008
Posts: 1014
I took some time to work out how best to respond but I still don't feel like I have solutions to everything. Still, I have long felt similarly to ikuyo, feos, and others on staff that our current volunteer pool is inadequate to handle the volume, is unlikely to ever be large enough to not be a limiting factor, and efforts to get more productivity out of the system will always lead to burnout. For the sake of everyone, I feel we need to reconsider how to change the conversation from killing the site (effective post title notwithstanding) to considering the curation focused rework TASVideos needs to continue existing as an ongoing concern.
On the one hand, we must find a way to allow the diverse TASing communities to find a home here or they'll find a home elsewhere as I've repeatedly reinforced every time a similar topic has come up, extending to the TASBot community where there are still barriers to event content publication on TASVideos. On the other hand, we are a curation site highlighting the "best" TAS content and always have been, and anything less than that would be devastating to many in the community. I believe it is possible to serve both needs simultaneously but doing so will require nontrivial site code changes.
Join me in a thought experiment about where speedruns and superplays of various types end up. On one extreme is SDA glitchless style human RTA runs with no tools allowances whatsoever. At the other extreme is classic TASVideos fastest goal completion at all costs TAS content. In the middle with no home is a long tail of users and communities neither extreme serves, including gamers with disabilities or accessibility needs who use tools openly in honorable ways. TASVideos is far more in a long term position to change to accept that long tail but doing so threatens the historical image of the site by eliminating the high standards our publications have been held to. So what do we do?
I feel strongly, and have felt for a long time, that we need ways for communities to be self-regulating with their own community categories that meet their needs, such as the fighting game folks or individual communities like Celeste. If we give them a home similar to a speedrun.com page where they can make categories and hash though their own requirements, they'll do a lot of the hard work without requiring a huge amount of staff oversight. Encoders would continue and expand as needed, publishers would continue to handle descriptions and other aspects of promoted runs, and judges would primarily move into curator roles focused on highlighting runs with far less frequent adjudication.
What makes a run valid in this new community-driven approach? How do we choose what to highlight and when? Some of these implementation questions will need to be hashed out, but I feel strongly after discussions with adelikat and others that the baseline requirement across the board for emulator-based submissions is a confirmation a run syncs for someone other than the submitter, preferably along with verification of hashes and other common header/metadata fields. Event or non-TAS content such as on-console superplays with tooling (think CZR playing SM64 on drums on a real N64) would need community and category defined requirements for marking it as synced. Instead of words like self-published, I'd simply call this organizational structure "Synced", and it would be the base level requirement to be added to a given community branch/category. (Imagine for a moment what it would be like if 20 years ago we'd made it a requirement that users uploaded movie files as as userfiles first and another user had to mark them as synced before that userfile could be sent through the submission system - that's effectively what I'm describing here).
Unlike our existing submission process that uses a common form, a runner would instead select the appropriate game/community hub page, select the category/branch they want the run to appear under, and add a submission there (new game pages and new categories could be created by sufficiently established users). Once the submission was successfully synced by someone else from that game/community it would be marked as Synced and be automatically assigned a numerical page similar to our existing ####S for submissions. They could, in fact, *be* submissions as they are implemented on the site today but adjusted as needed to be visible, filterable, and searchable site-wide.
It's fine for a Synced run to never be promoted to a Published run in this new model because it's no longer treated as "not accepted" by default. Individual game/community members could use their own decision process to decide what runs to promote up to Published. Perhaps fighting game communities have obscure character-specific categories that remain as Synced runs while new and improved any% category runs get promoted to Published runs with descriptions and ####M pages. Encoders would process Synced runs to archive them but would uncheck the "Publish to subscriptions feed and notify subscribers" box on YouTube (i.e. it wouldn't be promoted). If future volume exceeds encoder capacity, individual communities could set additional criteria for archival or encoding queue priority as needed.
We don't have to treat these stages using existing site naming and features but it may be helpful for the sake of discussion to think of it as various levels of searchability and visibility:
State
Visible
Searchable
Promoted
Hidden userfile
No
No
No
Visible userfile
Yes
No
No
Synced
Yes
Yes
No
Published
Yes
Yes
Yes
In essence, what I'm proposing is something similar to a userfile as the basis for what a user would add in a particular category/branch in a given game/community space where it would show up in a pending list prior to another user validating it as Synced at which point it would become searchable on the site (and likely eventually archived by an encoder subject to availability). Synced runs which the individual community deemed worthy of promotion would bubble up to Published runs, complete with descriptions and a publication and posted fully on YouTube. Judges would be much more hands-off, only stepping in for significant issues or corner cases and possibly not even called judges anymore. Perhaps their role would morph more into one of curation, selecting the best Publications to highlight on the front page.
In summary, I feel we need to move to a curation model as quickly as possible for the health of the site while simultaneously giving room for more TAS oriented communities to have their own spaces. My framework above is just here for the sake of discussing what's possible rather than the way it has to be but I'm pretty convinced the concept of Synced is sound and worth exploring further regardless of how it's ultimately implemented or coded. Thoughts?
It's fine for a Synced run to never be promoted to a Published run in this new model because it's no longer treated as "not accepted" by default. Individual game/community members could use their own decision process to decide what runs to promote up to Published. Perhaps fighting game communities have obscure character-specific categories that remain as Synced runs while new and improved any% category runs get promoted to Published runs with descriptions and ####M pages. Encoders would process Synced runs to archive them but would uncheck the "Publish to subscriptions feed and notify subscribers" box on YouTube (i.e. it wouldn't be promoted).
I personally cannot support this approach as written, because it invites the possibility that runs which would be guaranteed to be published under current site rules (Standard eligible runs) wouldn't be guaranteed to be published under this model if a particular game's community decides not to push the synced run up to the published/promoted status.
I don't want to support any changes to site policy that could effectively produce a situation where currently publishable runs wouldn't be guaranteed to be published after site policy changes.
If the above proposition had the caveat that runs which are currently acceptable for publication cannot be left only at Synced status, then I'd have less of a complaint.
Moderator, Senior Ambassador, Experienced player
(907)
Joined: 9/14/2008
Posts: 1014
DrD2k9 wrote:
If the above proposition had the caveat that runs which are currently acceptable for publication cannot be left only at Synced status, then I'd have less of a complaint.
Sure, that's reasonable - the goal here is to reduce the crushing workload on the judges while opening up the site to allow more diverse content to find a home, not to reduce the number of publications. I'd expect more publications overall rather than less. Having guidelines for individual communities to ensure categories we publish now would generally, likely with few exceptions, still be published is entirely sensible.
It's fine for a Synced run to never be promoted to a Published run in this new model because it's no longer treated as "not accepted" by default. Individual game/community members could use their own decision process to decide what runs to promote up to Published. Perhaps fighting game communities have obscure character-specific categories that remain as Synced runs while new and improved any% category runs get promoted to Published runs with descriptions and ####M pages. Encoders would process Synced runs to archive them but would uncheck the "Publish to subscriptions feed and notify subscribers" box on YouTube (i.e. it wouldn't be promoted). If future volume exceeds encoder capacity, individual communities could set additional criteria for archival or encoding queue priority as needed.
To add to what DrD2k9 said, I'm worried about the future submissions too.
In the last couple of years the forum audience has shrinked in 2-3 times. It's already common to see a TAS discussion thread where the first post is "This movie has been published". So concerns the polls too, like nobody is interested is what you have done! Veterans are gone. Most of active players visit the site just to submit TASes and respond to posts in their discussion threads.
By not having a big audience on the site, the role of subjectiveness and ignorance becomes very significant. If I got your suggestions rights, publishers will basically become "gods" since they will decide which synced runs to publish. That's a freaking wonderful perspective! I looked through my submissions and, seeing how little feedback I had got for various works, it would be no wonder if some of them would end up being just "synced". I've been spending months to make a TAS, frame advance all the time, replaying sections in case of discovered improvements, and all I would get is "Synced"? That's not what I was striving for, and I think people will agree with this.
TASing is like making a film: only the best takes are shown in the final movie.
Moderator, Senior Ambassador, Experienced player
(907)
Joined: 9/14/2008
Posts: 1014
Dimon12321 wrote:
In the last couple of years the forum audience has shrinked in 2-3 times. It's already common to see a TAS discussion thread where the first post is "This movie has been published". So concerns the polls too, like nobody is interested is what you have done! Veterans are gone. Most of active players visit the site just to submit TASes and respond to posts in their discussion threads.
Like most communities in the last few years, conversations have increasingly moved to realtime discussions in places like Discord. Still, as others have noted with extensive statistics, submissions themselves are way up. While the shifting location of conversations may feel uncomfortable, I don't see that it's an indication that veterans of the site are gone. You're bringing up valid points around the drop in engagement around things like commenting on runs and voting on them but that aspect is true no matter what solution we investigate and it's likely we can brainstorm ideas like integrations that may help there.
Dimon12321 wrote:
If I got your suggestions rights, publishers will basically become "gods" since they will decide which synced runs to publish.
No, as a site we'd have guidelines that game/community members would need to follow about category handling and most of them would publish the fastest run for time oriented categories. Nothing would change from the way things are handled today in the sense that, after a new, faster run was confirmed by someone else to sync and meets the goals of the category, it would first be marked Synced and subsequently published. (For more subjective or art oriented categories where speed isn't the primary goal of the TAS it would require a bit more nuance as you may have many runs with different artistic elements but that same challenge applies to event content.) Some communities may reasonably opt to not publish, say, every single combination of characters in a fighting game but may opt to designate one as the new best and promoted to a published run. To your earlier point, perhaps the conversation itself would end up happening in Discord but of course changes to any runs would naturally happen on the site itself.
Dimon12321 wrote:
I've been spending months to make a TAS, frame advance all the time, replaying sections in case of discovered improvements, and all I would get is "Synced"? That's not what I was striving for, and I think people will agree with this.
Heck, I agree with that - having said that, that's not what would happen here. Assuming you were creating an any% category run and your final time was faster, your run would first be reviewed and marked Synced by someone else in that game community (or a judge/curator for games without their own communities driving specialized category rules) and would be added to the publication queue automatically. It's already been validated as the new fastest run after all.
Runs deliberately created for alternative playaround categories or branches the game/community has set up, such as the different fighting game character example I gave earlier, would stay around as just Synced more likely than not. The significant change away from a centralized submission form to a model where you're submitting directly into a category will immediately make it clear to submitters if the run is likely to be published or not. Of course, community response could change if a particular run was found to be special and any run could be promoted to Published as warranted. The significant change here is the complete redefinition of what a judge does and a significant offloading of many judge functions to individual game communities, giving them the ability to perform the validation that a run can be marked Synced on their own but with judges/curators still around to help answer questions.
These are all good discussion topics and it's definitely worth hashing all this out - I look forward to more civil discourse and thanks. :)
Joined: 4/17/2010
Posts: 11495
Location: Lake Chargoggagoggmanchauggagoggchaubunagungamaugg
One important thing. The bigger the changes, the longer they take to implement. Currently we have about 2 active site code developers, and a bunch of tickets that are critical or important. If the reform we agree on is too big, it will take years to implement (nobody knows how many).
Warning: When making decisions, I try to collect as much data as possible before actually deciding. I try to abstract away and see the principles behind real world events and people's opinions. I try to generalize them and turn into something clear and reusable. I hate depending on unpredictable and having to make lottery guesses. Any problem can be solved by systems thinking and acting.
Moderator, Senior Ambassador, Experienced player
(907)
Joined: 9/14/2008
Posts: 1014
feos wrote:
One important thing. The bigger the changes, the longer they take to implement. Currently we have about 2 active site code developers, and a bunch of tickets that are critical or important. If the reform we agree on is too big, it will take years to implement (nobody knows how many).
That's one of the main reasons my proposal attempts to re-use as much of the existing site code as possible (using the existing submissions storage, etc). Still, many things about this thread are honestly about needing more volunteers in general and site code needs are just as important of a need. Any changes we make will require at least some coding even for the most minimal things we can do. For now, I'd propose starting with a flag for "Synced" based on adelikat's interpretation where it implies hashes and versions are correct. That feature on its own has plenty of uses even if my proposal isn't implemented in full or at all.
There are certainly ways to begin implementing some of this without any site changes at all. One idea is to repurpose the "Needs More Info" status to also include requests for sync verification. Right now a judge marks a submission as "Judging underway" to claim it. If the judge doesn't have the setup to confirm sync, they could instead mark the submission as "Needs More Info" and explain that the info needed is an independent verification of sync (and also a temp encode to judge from, if the submitted TAS didn't include one). This would also provide more transparency to the community for where sync verification is most needed which could bring people to volunteer when they wouldn't have even known there was a need otherwise.
I feel like the workload of actually needing to set up the specific game and emulator for each game that a judge looks at is part of what makes it difficult for new people to commit to volunteering as a judge. But if some of that workload is offloaded to other community members, it becomes a lot more approachable to contribute in a more effective way. As an example, an author might not really know if a judge doesn't have the appropriate setup to replicate sync, but if the submission is set to "Needs More Info" for a requested sync verification, the author would get a notification indicating such and could even help find someone who already has the correct setup to verify sync.
