Emulator Coder, Site Developer, Site Owner, Expert player
(3572)
Joined: 11/3/2004
Posts: 4754
Location: Tennessee
So here is my tiered proposal.
So, I've been wanting to re-organize the site for quite some time now. In January I envisioned making this post to rally the troops and have an end of the year goal for this. Oops, it's almost Novemeber now >.<
Fortunately Ilari has been doing some heavy lifting on the site code to do this during this year. (Much thanks!). So we are still well on pace.
There is two aspects to this change I want to do. One is a game based organizational scheme (for the oldies here, it is basically "Tub's idea"). And a complimentary "Tiered publication" system to go along with it. This thread is geared towards discussing the latter. As I do want audience involvement with this idea.
The idea is to have publications grouped in "tiers" that have their own requirements. In long run, the backend will now allow for changing/removing/adding tiers to accommodate future needs. In the short run, we need to decide specifically the tiers and policy of each.
As you will see, the two main consequences of this change is that game choice would no longer be a rejection reason, as we have a tier to publish just about every game to. The other is allowing more diverse and numerous categories for a given game.
Since I can only have one poll, and the categories is more of a consequence of a better site organization scheme. I have made the poll about game choice. Because this change is largely motivated by it, I want to know what you think specifically about this issue. Is game choice rejections a good or bad thing?
I think the important thing the current "game choice" rejections accomplished is ensuring that people would not stumble across boring movies while browsing the site. Unfortunately this has the negative corollary that people sometimes can't find content that exists even though they're looking specifically for it -- because that content was rejected.
In other words, I see basically a user interface issue here -- there's little way to tell the good runs from the bad except for via the stars/moons system, which is pretty coarse. Perhaps you should be able to filter the runs that are displayed based on their ratings (e.g. only show runs with ratings of 7+). I guess your tiers are trying to accomplish something similar, by handing out moons to every run that might possibly be interesting to someone who has no prior experience with the game in question (i.e. good game choices), but it still seems kind of coarse to me.
Still, your suggestion is an improvement over the current behavior. I do think that the ability to filter results based on their rating, with a default filter at some sensible value (e.g. "Currently showing runs with ratings above 7.5 (showing 284, hiding 46)"), would accomplish much the same thing while avoiding the drama of judges "consigning runs to the Vault" or whatever. But if that's not on the table then I'd be happy enough with your suggestion.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Emulator Coder, Site Developer, Site Owner, Expert player
(3572)
Joined: 11/3/2004
Posts: 4754
Location: Tennessee
We are actually developing a better filtering system with this. For instance, looking by game, by tier, by category, or any combination of those. I haven't considered rating in there though. But currently you can filter by rating, and rating is rather arguable in its validity <_< But it wouldn't be hard to add that ability eventually.
Joined: 3/9/2004
Posts: 4588
Location: In his lab studying psychology to find new ways to torture TASers and forumers
Realize that this allows for a dozen different Super Metroid runs. Which is probably a good thing.
On the other hand, it also allows for a dozen different Monopoly runs.
Edit:
I'd also say that stars and moons currently have a specific meaning to them, and that meaning should be carried over as needed to the new tier system as well, which the proposal AFAICT doesn't account for.
Warning: Opinions expressed by Nach or others in this post do not necessarily reflect the views, opinions, or position of Nach himself on the matter(s) being discussed therein.
Emulator Coder, Site Developer, Site Owner, Expert player
(3572)
Joined: 11/3/2004
Posts: 4754
Location: Tennessee
The proposal doesn't allow for dozens of Monopoly runs if you notice. The vault is limited to two categories (and actually I'm on the fence about it including full game runs, I'd love some feedback there).
As such, at least 5 of those 7 monopoly runs are subject to "moon" requirements which means audience support, and entertainment value. Those 5 movies (which you are implying are garbage) would be rejected for their entertainment value and audience reaction.
Joined: 3/31/2010
Posts: 1466
Location: Not playing Puyo Tetris
Some games just don't make a good TAS. Fisher Price games for example. Though some "endless" games like Animal Crossing would be interesting to see get some sort of TAS (even if it doesn't have a "end").
I like the idea of having Tiers. Some runs are so "WTF?!" that they need more attention.
When TAS does Quake 1, SDA will declare war.
The Prince doth arrive he doth please.
Joined: 3/2/2010
Posts: 2178
Location: A little to the left of nowhere (Sweden)
I have some concerns about the tier names, because in the current system we would be handing out some 500-1000 moons if I understood correctly. Displaying a moon on that many TASes may cause people to wonder why we bother marking them at all, I am worried this can make the credibility of the runs take a hit.
Regarding the vault tier, I do think that 100% runs should be included, some games become very boring when they are completed to 100% because it is mainly grinding, with such a spot in the vault it is still possible for someone to make a good TAS of that, even if no one really sees the point to work on such a thing.
Also, does this mean that we will un-reject every well made "bad game choice" TAS that has been rejected in the past?
I will echo henke37's names for another tier. Those names would be perfect for a tier dedicated to the obsoleted runs.
I have voted yes, but I hope to receive some answers to my concerns.
I have some concerns about the tier names, because in the current system we would be handing out some 500-1000 moons if I understood correctly. Displaying a moon on that many TASes may cause people to wonder why we bother marking them at all, I am worried this can make the credibility of the runs take a hit.
Yeah, that tier could use a better name to avoid clashing with existing meaning of "moon".
Warepire wrote:
Also, does this mean that we will un-reject every well made "bad game choice" TAS that has been rejected in the past?
Yes (of course there's also improvements since to look at).
Emulator Coder, Site Developer, Site Owner, Expert player
(3572)
Joined: 11/3/2004
Posts: 4754
Location: Tennessee
Should we adopt a tiered publication, then yes, there would be a lot of un-rejections to do. In addition, we would probably want to rethink some of the currently published runs as Vault movies.
As for diluting the moon, I infer that you are implying concerns towards the meaning behind moon. Right now I'm just using moon as the name for this category, with an associated icon. Maybe you share the same concerns as Nach that you want a system to allow for a "noteable publication"?
All hail Neo-TASVideos!
The tier names could be different to differentiate them from current stars and moons. Perhaps the top tier could be something like Spotlight, Pedestal, Showcase, or Greatest Movies, the moon tier could be something like Bookshelf, Counter, or Good Movies. Vault or Archive sounds good for the bottom tier.
Edit: if the moon tier uses a different name and icon, the existing moons can be easily used in the new system.
Emulator Coder, Site Developer, Site Owner, Expert player
(3572)
Joined: 11/3/2004
Posts: 4754
Location: Tennessee
p4wn3r asked some question son IRC that I want to clarify here:
"Would users be voting on tiers"?
"Seems like this is a way to add more content but give the site staff the decision on which to highlight"
No, I think in this context the question "Should this movie be published?" is no longer a meaningful question. Instead we should go back to "Did you find this movie entertaining?".
Audience response will be the driving factor between the vault and moon tiers. (Or moon vs rejection in the case of non-vaultable submissions). The vault is largely non-subjective, and stars will eventually be decided by a dedicated user(s). Users will still be stating that they liked the movie, or not. For the most part there would be no change in the content of the submission discussions.
A dedicated user (Starman) will be deciding stars, but again the response from the audience will be a driving factor. I would say a submission with 100+ yes votes is a serious contender (for instance). So voting is about entertainment value. And audience response will be a significant factor in how to "Highlight" a movie.
Should game choice be abolished as a rejection reason?
Then judges/voters will probably use {lack of) entertainment as a reason for rejection. Some games mentioned above (Fisher Price, Endless, etc) are almost impossible to make an entertaining run out of it, so you might as well say they are bad game choices.
Emulator Coder, Site Developer, Site Owner, Expert player
(3572)
Joined: 11/3/2004
Posts: 4754
Location: Tennessee
Did you read the vault requirements? Entertainment value is not a significant consideration. When I mean "bad game choice", I also mean lack of entertainment value.
Did you read the vault requirements? Entertainment value is not a significant consideration. When I mean "bad game choice", I also mean lack of entertainment value.
Oops, sorry, didn't noticed the link. Brb.
Edit: Ok, for the vault, will games that don't really require "TAS-like" elements (Warioware, simple deterministic games, etc)be accepted into the vault as long as it meets the requirements?
Thanks for answering my questions, adelikat. For me the best would be to simply lower the publication requirements and then filter the runs based on a revised rating system. Assigning tiers like this would make drama go from accept/reject to tierA/tierB. But since you mentioned on IRC that you want to preserve "people bickering, and flinging poo at the judges" it's at least consistent. So, in short, I vote Yes, because publishing on a Vault is better than rejecting.
Emulator Coder, Site Developer, Site Owner, Expert player
(3572)
Joined: 11/3/2004
Posts: 4754
Location: Tennessee
jlun2 wrote:
Edit: Ok, for the vault, will games that don't really require "TAS-like" elements (Warioware, simple deterministic games, etc)be accepted into the vault as long as it meets the requirements?
The current wording of the vault would allow for these games. They are still records afterall.
Do we want to try to enforce some minimum standard for TAS-like elements? (this question is directed at everyone, not just jlun2)
Joined: 3/2/2010
Posts: 2178
Location: A little to the left of nowhere (Sweden)
adelikat wrote:
As for diluting the moon, I infer that you are implying concerns towards the meaning behind moon. Right now I'm just using moon as the name for this category, with an associated icon. Maybe you share the same concerns as Nach that you want a system to allow for a "noteable publication"?
In this situation I thought of the moon as any icon, the problem the way I see it is that so many runs will be given the same icon that the icon may lose it's credibility. But perhaps my mind is unconsciously poisoned by the current system.
----------------------
For standards I assume a good guideline would be that if it's impossible to complete the game faster by TASing it compared to playing it casually, it should not be allowed. Most (all?) of the WarioWare games fall into this category in my opinion.
When it comes to something being a tool-assisted speedrun, the speed and technical achievement should be the only criteria that matter, entertainment be damned.
When I first came to this site, I thought it was all about tool-assisted speedruns. But soon this naive presupposition faded as everyone started talking about how entertainment was the king, and seeing how well-made runs didn't always make it because they weren't entertaining enough to a hypothetical casual viewer.
But considering all submissions to be tool-assisted superplays will eventually run into problems. The relatively objective measuring stick of fast completion time loses its meaning and in its stead comes an endless plateau of subjectivity. Submissions of obscure and not as fast-paced or long games become underdogs that can be rejected on at worst arbitrary basis... throwing away decent runs and, to me, disrespecting the authors' hard work that's gone into making them.
And that's why I welcome this idea of a "vault", where even the most boring, obscure submissions have their place and can thrive (as long as they aren't completely worthless as far as technical achievement goes), for purely the purpose of precise cataloguing and for the few, rare people who can actually use the obscure data contained therein.
So kudos for this reasonable idea, that, if implemented, would be a step in the right direction.
To throw in an example, I'd like to see some tool assisted speedruns of fighting games. Those currently aren't welcome in here because of the rules.
Yes, these would be two steps in the right direction:
1. Emphasize the Tool-Assisted Superplay aspect of the movies by making the entertainment value decide the tier.
2. Let the non-superplay TASes (which are still a record) be published in the Vault tier.
But there shouldn't be more than two tiers. The Starts should keep their meaning as a very limited selection of the best movies (the showcase of the site, which must be a single webpage that can be browsed in less than several minutes, of course not counting the time possibly spent for watching videos). Stars should not become a third tier, or else there will be much more severe drama about cool movies not making it into the highest tier.
Warepire wrote:
For standards I assume a good guideline would be that if it's impossible to complete the game faster by TASing it compared to playing it casually, it should not be allowed. Most (all?) of the WarioWare games fall into this category in my opinion.
But if it's really impossible then such TAS naturally cannot be considered a record, so it doesn't fit into the Vault. However, it may suddenly fit into Moons if TASer manages to make something ingenious while waiting/playing.
Emulator Coder, Site Developer, Site Owner, Expert player
(3572)
Joined: 11/3/2004
Posts: 4754
Location: Tennessee
By the current wording, entries in the vault say it must MATCH or beat all existing records. Perhaps I should reword that to say it must beat all known unassisted records, and at least match all known tool-assisted records?
By the current wording, entries in the vault say it must MATCH or beat all existing records. Perhaps I should reword that to say it must beat all known unassisted records, and at least match all known tool-assisted records?
Indeed, what would be the point of TASes that only match unassisted record?
Didn't see anything to criticize there. However, it's a bit unclear to me whether it's intended for every single run to be categorized into those three tiers, or whether just some runs (those that match the descriptions) are given a tier, and then there's "everything else" (as a fourth, implied "tier").
After all, there are non-speed-oriented runs that nevertheless cannot be considered star- or even moon-worthy (and wouldn't match the requirement of completing the game as fast as possible of the "vault" tier either.)
Emulator Coder, Site Developer, Site Owner, Expert player
(3572)
Joined: 11/3/2004
Posts: 4754
Location: Tennessee
The intent is that every movie is a tier of some kind. The non-speed oriented runs fall into the moon category. Meaning they have to be entertaining enough to warrant it.
Also, I could be convinced that the moon tier is infact 2 tiers of sorts, where the moons are significant in some way, and the 3rd tier is more of an average/borderline level publication
