I wrote this to someone in PM, but I decided that it's a good idea to post it here.
Many people have expressed disapproval on several occasions, when it seems that I'm going to reject a submission that is better than existing publication, leaving a non-optimal publication stay there.
This applies to situations where a submission is pointed out to be slightly improveable, and a judge rejects it in favor of a new submission that does not exist yet.
Here is my justification for it.
I think what we publish (i.e. the publication events) is more important than the list of movies in the archive (i.e. what is found on movie pages).
-- Making insignificant new publications (such as 32-frame improvements) decreases the signal-to-noise ratio, causing that the people who want to keep up-to-date with publications and to follow the front page, will lose interest in doing so. I want new publications to be radical.
-- Keeping old movies on the movie pages is a lesser evil; people don't care about what's the newest record NOW; they just want to see an entertaining movie. And if it was published, it is entertaining.
This why I don't see keeping the current movie there even if a better has been made, as a bad thing. Of course it only works if a new movie is coming; otherwise an excellent publication opportunity is lost, which also is a turn-off.
Examples where I have applied or wanted to apply this rationale, include Hitler no Fukkatsu, Super Demo World, Super Mario Bros. 2, etc.
I don't have any problem with submissions that are better than the relevant publication being rejected, I just feel that when this happens, it should be taken as a sign that the relevant publication no longer need be maintained.
With exceptions, of course, for "improvements" that are faster but significantly less entertaining than the publication they are attempting to obsolete.
Addendum: I think that a statement like "if this game's movie isn't brought up to par we'll remove the publication" could provide a significant motivation for fans of the game to try harder to make a good TAS.
How fleeting are all human passions compared with the massive continuity of ducks.
In general, rejecting an obsoletion often goes against two of the things mentioned on JudgeGuidelines:
1) "Hard work should have a reasonable chance of being published."
2) "Don’t demand them do tedious laboursome optimizations that can’t actually be noticed in the viewed video."
So unless the drop in entertainment is extremely noticeable, I think it's a bad idea.
<Swordless> Go hug a tree, you vegetarian (I bet you really are one)
Is your point that the movie is slower, but was accepted anyway?
I think for the pursuit of perfection to be true, even tiny improvements must be accepted.
If there's a possible improvement somewhere in a movie, I don't think it's unreasonable to ask the author to redo. Sometimes he will definitely want to fix whatever is wrong with his movie, I'm guessing. If he doesn't, though, because it's too much work or because it's lame or because he's moved on from that game or whatever reason he might have, I think rejecting the movie is a significant mistake. In that scenario you're left with the choice of having either the good or the worse movie on the site, and choosing the worse movie is just kinda stupid imo.
I would like to bring your attention again to my suggestion at http://tasvideos.org/forum/viewtopic.php?p=144974#144974
If there was a semi-automatic "world record" category (alongside the other categories) then there would be less controversy about this subject.
But should you still be required to do the most interesting stuff you can throughout the run, especially during the slow parts, despite it still being a fastest completion? Because I think you should.
put yourself in my rocketpack if that poochie is one outrageous dude
Perhaps a redesign of the front page could help this out. Have separate lists for improvements and new movies. The improvements list would probably move a bit faster than the new movies one. Maybe have a way to put radical improvements to a movie in another list or on the new movies list. This way movies that are a big deal can be shown more prominently than minor improvements, which are still good to have.
This may be slightly off-topic, but I somewhat dislike the "movies posted on these three days", because the number of movies on the front page fluctuates wildly sometimes. There's days when I've seen just 3 movies and days with a whole bunch.