I'm not really into 15, 10 might be a little too low, but I don't like 12 either.

How about 11? You already vote on your own video, so setting it at 11 would require 10 actual votes from other people for it to be sifted.. either that or some system where you wouldn't have to vote on the videos you submit, so that the 10 votes would all come from other people.

Why raise the amount of votes needed? Its hard enough as it is to get videos sifted...

*looks at own's personal Queue count currently at 245*

Might as well keep it at 10...

We don't want to open the floodgates so that the front page is a random collection of videos.

Why not?

It's difficult to get 10 upvotes. 12 is reasonable, but I'm voting for 10.

Comment hidden because you are ignoring dag. (show it anyway)

>> ^MrFisk:
We don't want to open the floodgates so that the front page is a random collection of videos.
Why not?
It's difficult to get 10 upvotes. 12 is reasonable, but I'm voting for 10.


Why not? Because that's not what this site is about. We could just develop a random YouTube post displayer otherwise.

How about we remove the ability to vote for your own video, at least?

I likes me some numbers. What percentage of submitted videos eventually get Sifted?
What percentage would we expect?

Are we at a point where too many videos are reaching the front page, turning VideoSift into little more than a masking layer over YouTube? i.e.: "YouTube without the insipid YouTubers!"

I agree that it makes sense to remove the ability to vote for your own video.

Comment hidden because you are ignoring dag. (show it anyway)

For the past few months it's moved around between 3K to 3.5K videos per month being published, depending on the number of days in the month.

I voted for 12, even though it's quite clear that 10 will win by a landslide. That's fine, I'm not that wrapped up in the issue. I do think 15 is too high, because raising the bar that much will make this place more like Digg or Reddit, in that only the most popular videos get rewarded (and most of us know that popularity doesn't always mean quality), while rarer gems are relegated to PQ oblivion. I like this place because it showcases both popular AND off-the-beaten-path videos, and if we lose either part of that equation the site will suffer, I believe.

However, the escape threshold has been at 10 votes with a 2-day queue for over a year now, and I'm sure the active membership has grown enough to warrant some sort of increase. Therefore I suggest you develop some sort of formula for raising the escape threshold automatically whenever there's enough of an increase in active members.

For example let's say that right now we have 1000 active members, and for the sake of argument let's define "active member" as anyone who has logged in at least once in the past week. And let's also say that escape threshold is set to 10 (which I'm sure it will be once the poll closes). That means that for every 100 active members you need 1 vote to get a video published. Now let's say in three months the number of active members has risen to 1100 members. Therefore the escape threshold should be 11 votes. Doesn't that make sense? And the threshold wouldn't have to be raised immediately once the requisite number of active members calls for it. It could be reviewed every three months, for example, and if it's going to go up you could announce it in Sift Talk maybe a week ahead of time, that way every one has plenty of notice. Show everyone the numbers too, that way there's no doubt that the increase is indeed warranted.

I say this because if you leave it to a poll every time you want to raise the escape threshold, chances are you will never be able to raise the escape threshold ever again, even if active membership numbers double, triple, quadruple, etc. And then this place will become like a random YouTube player.

People like to post videos not just because they want to share their tastes, but also to get rewarded when enough people appreciate what they've offered. Hence the existence of the star system. If you make it too hard to get stars, you'll be discouraging a lot of people from posting videos. However it's not unfair to raise the bar a little when the number of active members warrants it. So I hope you'll give my idea some serious consideration.

i like 12. how many "we need to something about the publish threshold" sift talks have we had over the months, and now were actually discussing doing something about it and now were crying about the change. pffty.

15 maybe too high, we all have tons of videos in the 10-14 range. but 10 is too low and alot of crap gets sifted. the more people there are in any given place, the lower the standards fall. im fine with raising the bar a bit. .... but this is coming from the wench with 1 video in her pq.

>> ^dag:
For the past few months it's moved around between 3K to 3.5K videos per month being published, depending on the number of days in the month.

Out of how many total submitted?

Why not? Because that's not what this site is about. We could just develop a random YouTube post displayer otherwise.

The quality caliber of people here insures that won't happen.

I prefer old 10, but 12 is OK. 15 is too high.

You're right about the two goals conflicting, but I don't really understand the intellectual backing of point one.

Quality means different things to different people. To me in the sites context a vote barrier of 15 for example means only highly viral videos will make it out of the queue, most of the time these videos always follow the same line of content. Most of the time that kind of content is already popular or viral across most of the internet. So how does that differentiate what VS does if most of the content can be mirrored on other sites?

As I said before its never really clear what kind of quality we are going for. Most of the time it feels like the site is catered for quality that appeals to mass mainstream audiences, a sort of billboard Top 15 of the internet. That kinda is strange because the billboard top 15 of the real world is filled with shit most of the time.

I'm undecided on this. The videos I sifted following this change would definitely have got at least 10 and so getting 15 wasn't a problem. The others are so-so submissions but with a few helpful votes and lots of promotes they would make it past 10. Now it's harder and in a way I don't see it as a bad thing. We all know what videos should probably stay in our PQ but yet get sifted.

The fact is raising the threshold to 15 does, on balance, prevent lesser videos getting through. On this site, this is not a bad thing.

And sure it will stop the good stuff as well. This is where we need solutions. One thing that occurred to me (and this is full of holes) would be to have a primary channel for each video (up to the submitter to show good judgement and channel owners to monitor). Each channel is weighted depending on the mode average votes (the number of votes that occurs most frequently) - a comedy video would then need more votes to sift than something in the Obscure channel.

Also you would effectively have Collectives back...

[edit] btw I say let's keep experimenting with 15 for a while more.

I've stayed quiet on this subject, waiting for this poll to come up. My observations about the 15-vote experiment:

- One good thing about it: Users are more likely to scan the field of Unsifted videos.

- With 4.0, star levels have been raised. If the 15-vote rule were to stay, it would make it even more difficult to achieve higher star levels, needing to get 25 videos sifted with 15+ votes.

- Not only that, but if the 15-vote rule were to stay, there would be many more videos sticking around in the Unsifted section. The further your video gets buried over time, the less likely you are to get additional votes. Casual users do not explore that far...

- In which case, if we are going to keep the 15-vote rule, you might as well do away with the "Hotness" factor (which I would banish anyway, but that's another story), considering any videos that get 15 votes within 24 hours would be on that list to begin with.

- Quality is subjective. We all know that. Considering that, if the 15-vote rule were kept in place, members will be more likely to use their Power Points for selfish begs, self-promotes, general brown-nosing, etc. Or misanthropists like me will have to pretend to be friendlier and likable in order to get more votes from people we don't really care about.

- People like me will stop wasting their time posting videos. I know you probably couldn't care less, but there are others out there like me (Mr. Fisk, mintbbb, Farhad) and I know you don't want to give QM more material to work with, you elitist scum.

I could easily use my time browsing for the latest TDS/Colbert/Bill Maher/News/Zero Punctuation/Break/The Hot New Thing On Any Stupid Site/Whatever, or spend my time looking up "cat" or "ukelele" or "baby" or "fat Christian redneck" or "funny animal unexpectedly interacting with another cute species" on YouTube... but that's not my thing. Everyone has their niche, and I'd rather leave those things to other people.

About one-third of the videos I've sifted are between 10-14 votes. While a 15-vote rule wouldn't stop me from submitting my random music (if only to annoy those quality-hawks), it would take the fun out of it.

It's pretty obvious at this point where the vote is going to go, but I wanted to say my peace.

I like the idea of not voting for your own videos and the tiered reward system that gwiz665 proposed. That would put more emphasis on finding the popular vids to sift.

12.

that's the way to go.

end communication.

12 and remove the ability to vote for you own videos.

10. Not really worth debating.

Optimally, queue escape would just be removed and PQ with it.

^that was so stupid I think you gave me cancer.

^Good.

Why not just do the queue escape algorithmically. Hottest video every ten minutes gets published, or something like that. It could also be tempered by number of users online, total number of videos in the queue, etc.

If we're looking to change the "quality" of content that gets published, I think we need to start by really analyzing what it is that people think shouldn't be there.

I know that I often just submit stuff that I find interesting, and hope it catches with other people. Sometimes I start out knowing it's unlikely to sift, like this DNC ad about healthcare. But sometimes I think I've found a gem, and it doesn't sift for reasons that escape me like FDR's Second Bill of Rights, or 8 Must See Documentaries.

On the other hand, stuff like Bill O'Reilly being a moron skyrockets onto the Top 15, presumably because schadenfreude trumps enlightenment.

There are probably a lot of different ways to go about shaping what's sifted to match our notions of quality, but just raising queue escape mostly just affects how quickly individuals get new ranks, not what gets on the front page or Top 15.

In my opinion the big issue with how VS works now is that timing plays too big a factor. If I sift something late at night, or early in the morning, it's a lot less likely to sift than something I post early in the evening.

I think that the number of votes to get a video sifted should be placed back at 10. I dislike having to post my videos at a time when the largest number of users is online just to have a shot at making it out of the queue.

Further, I have enough faith in fellow sifters to know that if 10 people like something, it is at least worth a viewing.

many here have already stated the fact that quality is subjective,which i agree.the majority of my sifts have barely made it out of the que when it was 10.i receive approx 10% votes in relation to views,discount the 5% pity votes and i would sift nada,zilch,zero.the ratio changes drastically though when the video escapes the que and that ultimately is what the sift is about yes? sharing things of interest?quality videos we may find interesting and wish to share? maybe in the future things may need to change but as it is right now many quality videos will be consigned to pque while viral vanilla sameness will dominate the front page and that is a shame.maybe i have it all wrong and im not getting the plot but i would hate to see those rare quality nuggets be doomed to pque obscurity.
thats my 2c anyways.

I'll sidestep the actually question of what the magic number is (though I voted for 15) and address something else for a moment.

Every time we have this discussion there's one argument that comes up, usually repeatedly:

"It's hard enough to get things sifted."

It's usually followed up with an anecdote about that person's difficulties. Nobody wants to accept that maybe their submissions don't deserve to escape the queue. Videosift isn't here to provide a daily affirmation to its users that, yes, everyone else likes what you like. If you need 100% positive feedback, log off and go play with a light switch.

^I understand your point, but this place thrives on positive feedback, i.e., star points for sifted videos. Now if we make it too easy to get star points then they'll have the approximate value of a Deutsche mark during the Weimar Republic. However, if we make it too hard for the more obscure gems to receive star points, then the majority of videos that people will sift be the ones that have a greater probability of success. Therefore you'll see more TDS, Colbert Report, and cat videos, and fewer small but worthy gems. In other words, this place will become like a Hollywood studio.

Now I like seeing cat videos sometimes, and I also like seeing obscure music videos too. That's why I like this site so much, because you can find virtually every variety of video under the sun here. So a balance needs to be struck, I believe.

I don't think 12 votes will be too high for good small-niche videos to be rewarded, but that is admittedly based on a hunch, because I don't have hard numbers to back my case. But raise the bar too high and like I said, this place will become more like Digg or Reddit. That is not meant as a slam against either site, because I have occasionally posted videos from both. Still, I don't want this place to be a clone of either of them, because why should we even exist if we're nothing more than an RSS feed from both sites? Then that would make VS one of the worst things I can imagine a web site to be: BORING.

Now if you'll excuse me, I've got a light switch to play with.

"The first over-arching goal of VideoSift is to extract really good videos from a larger pool of not so noteworthy videos, for display to a larger audience."

How does raising the vote threshold help achieve that? Just because there are more people voting doesn't mean everyone's perception of quality is weaker. The quality of a video is not inversely proportional to the number of users. Quality is objective in the sense that a video considered quality when there were 200 users is still quality when there are 2000. Unless you believe the opinions of those extra 1800 users are inferior to the first 200. But if that's the case, just don't let anyone else join.

My ego says 12, since I was the one who suggested it.

But my heart says 10. The front page is moving at a snail's pace this week.

I dunno, maybe the front page would move at a similar speed between 10 and 12. I suppose we'll find out if this poll equates a change.

I say 10 mostly because I remember when I first started sifting and repeatedly self promoted and * begged to no avail.

Put longer queue time if higher than 10 votes requirement.

I'm going back for 10 now. I thought that maybe you'd moved up the level because, well, lots more people were voting and it was 'too easy' to get a video sifted. But, dangit this video made it past the 10 vote mark on it's own, and I think it's a worthy video, and that it got to the release point on it's own nicely

Except that it hasn't here.

And to say that there are all these other tools to get a video out of the queue is a bit silly I think. To have to promote and beg and push and push for a video to get out is kinda of wrong I think. I think there should be no need to really do that, or maybe do it only every now and again when you think something has been overlooked.

But to have to do it as a rule if it's not a cute cat video or similar, I think, means that the system is broken.

I think put it back to 10... work out some ways of allowing videos to have more of a chance to get out of the queue with no external help... and THEN if too much seems to start getting through, then maybe bump up the limit to escape.

Oh yeah, and as said... 15 votes to get a video out, PLUS a new too high barrier to get any star... makes getting 'into the club' too damn hard.

You're going to slow your new memberships right the crap down and maybe kill the growth of the site.

>> ^spoco2:
But, dangit this video made it past the 10 vote mark on it's own, and I think it's a worthy video, and that it got to the release point on it's own nicely.

Apparently there's a third option... find a popular sift talk and link to your video.

I think the sift's stated purpose of finding the best of the best conflicts with everything everyone wants the sift to be. People want every video they submit to pass, primarily because they get a "reward" when this happens. The only way I think this site can ever truly be about finding/indexing the best of the best is to get people to do it just because they want to. So long as it's a game, a competition, people will continue to throw video after video at the wall in the hopes that some of them stick.

Hitting the queue escape mark says nothing about a video's quality, because they can float around in PQ indefinitely. If a video gets 10 (or 12, or 15) votes after sitting in PQ for 2 years, should it really be on equal standing as a video that rockets out of the queue in an hour? There are other factors, yes, but I think the point stands.

I don't really think it matters what the queue escape limit or expiration times are so long as PQ and Beggar's Canyon are around. You can get damn near anything sifted so long as you keep pushing it.

I still think the only way we can improve the quality without rewriting the system from scratch (and I'm not sure how to improve it, offhand) is to give votes value by limiting how many times you can vote in any given day. Eliminate the downvote (possibly crushing all of ant's hopes and dreams in the process) and give us all, say, 10-20 votes per day. I think people will be more likely to search for videos that really deserve the vote when they can't just vote for every video they watch every day.

this is exactly what i was about to say.
lets kiss.

>> ^xxovercastxx:
I'll sidestep the actually question of what the magic number is (though I voted for 15) and address something else for a moment.
Every time we have this discussion there's one argument that comes up, usually repeatedly:
"It's hard enough to get things sifted."
It's usually followed up with an anecdote about that person's difficulties. Nobody wants to accept that maybe their submissions don't deserve to escape the queue. Videosift isn't here to provide a daily affirmation to its users that, yes, everyone else likes what you like. If you need 100% positive feedback, log off and go play with a light switch.

10 always felt perfect to me.

The way my own videos always escaped, it seemed to really reflect their value in the grand scheme of things... the stuff that got close to 10 tended to get a lot of votes once they escaped, whereas the stuff that didn't or barely made it out didn't...

I like 10.

Does anyone know when the polls close on the vote?

^3 days. So this should close in about an hour.

This is a fundamentally flawed way of thinking.

The idea behind videosift is that something should be able to "sift" on its own merits - this is basically impossible with 15 votes to go to the frontpage, because few users actually go to the unsifted tab. You have to time your submission to the most visited time and get a perfect title etc. etc. - stars have to align for your relatively obscure video to get through easily.

The recent thinking with PQs and begging shifts the perspective from the sifts own merits to "how much effort the sifter wants to put in" the more power points you use on your own sift, the better it will likely do. This is also a bad idea, but there we are.

By raising the bar for something making it out of the queue, we only inflate the size of everyones PQ. This is a BAD THING. Very, very few people actually go into anyone's PQ, unless they're all huggy in their clique.

Moving it back to 10 is the only viable choice for now. Eventually, we should move it back down to 5, and just kill things that don't gather 5 in the 2-3 days they have in the queue. Maybe make a PRIVATE PQ, that no one else could go into, that only exists for people's own reference, for future sifting at a better time, or just for viewing it themselves.

>> ^xxovercastxx:
I'll sidestep the actually question of what the magic number is (though I voted for 15) and address something else for a moment.
Every time we have this discussion there's one argument that comes up, usually repeatedly:
"It's hard enough to get things sifted."
It's usually followed up with an anecdote about that person's difficulties. Nobody wants to accept that maybe their submissions don't deserve to escape the queue. Videosift isn't here to provide a daily affirmation to its users that, yes, everyone else likes what you like. If you need 100% positive feedback, log off and go play with a light switch.

And obviously later introduce the tiered rewards, so that a really popular video gives the sifter a lot, and a more eclectic one still gives SOMETHING back, and only a really bad one gives nothing back.

How about we do percentage wise? How does Digg do theirs for front page?

Voting for this poll ended with the majority of users voting Return it to 10.

How about this...

Eliminate the queue entirely. When someone submits a video, it's part of the site immediately. The only reason we currently need a threshold is for awarding star points. What if, rather than basing star points on the number of published videos, we based them on the number of votes accumulated?

It eliminates the threshold argument entirely. It simplifies the site (Beggar's Canyon and PQ would no longer serve a purpose).

I think we need some threshold to get sifted, but I disagree that having patience and eventually sifting an obscure video is a bad thing. If you focus on sifting more obscure videos, you should have the potential to sift them over time if you score points in ways that contribute to the site. Otherwise, we are left with the lowest common denominator - a far worse outcome.

I'm not saying working to get something obscure in is a bad thing either. I'm saying that with enough patience, time and effort, you can get anything in. We hold the publishing threshold to be indicative of quality and I'm arguing that it's not.

It used to be that videos were queued for 2 days or 3 days or whatever it was and if they didn't make it in that time, they were done for. In all honesty, I hated that system, because it required too much extra effort most of the time. You had to watch your submissions like a fucking ebay auction and save, promote, etc. That system suffered from one of the same problems as the current one except that it required a lot of intervention. That problem is that we have potentially infinite queue time. Before you had to keep saving, now it's automatic with the PQ.

I'm arguing that there's no difference between something sitting "unsifted" in PQ now and something "sifted" with 3 votes in a system with no queue. As it is, when you hit 15 votes we decide your contribution is of sufficient quality to give you a star point. It doesn't matter if it took 1 hour to get to 15 or 2 years... if 15 people on the internet were willing to click the up arrow, then it must be good. I'm arguing that we've got an overcomplicated system that contradicts itself by saying "You've got 2 days to make the cut!" and then if you don't, "Ok, ok. Take as long as you want."

Throwing out the queue will have zero effect on content compared to today. The videos that are languishing in PQ with 4 votes now will be languishing in a dark corner of the site with 4 votes without a queue. The only time sifted/unsifted currently matters is for rewarding stars, and I think that stars based on votes might actually be a better measure of contribution value than the current system.

As for the "lowest common denominator", that will never change. LCD doesn't mean lowest quality; it means broadest appeal. The top sifts will always be the LCD.

I prefer 10 but will settle for 12 - 15 seems way too high!

Edit - Am I too late to vote, It doesn't give me an option to cast a vote

Comment hidden because you are ignoring dag. (show it anyway)

Voting is over.

OK. I'm not going to do my usual guilt trip thing - let's go back to 10. That's the community's overwhelming opinion and I'm not that sure myself. I do get some feedback about declining quality of the front page, and perhaps I'm a bit knee-jerky in trying this out to correct it. Sorry. I feel like the Fed trying to control the economy by cranking the interest rate around. Quality is in fact a very subjective thing. Myself, I like a good mix of politics, geekery and funny pet videos- but that's me.

>> ^xxovercastxx:
I'm not saying working to get something obscure in is a bad thing either....


I've read that comment 5 times now and can't think of anything to contradict its conclusions. It would definitely simplify things, but it means a very big change in how VideoSift works. At the very least, how often you can submit a video would have to change.

It would also require an incredible act of humility. Dag, Lucky, and Campioni would have to be willing to abandon a tremendous amount of code that took who knows how many hours to build.

>> ^dag:
let's go back to 10.

What happens to pqueued videos with 10 or more votes?

The queue escape limit has been lowered back down to 10 - "return it to 10" requested by the community.

>> ^siftbot:
The queue escape limit has been lowered back down to 10 - "return it to 10" requested by the community.


Now that the queue limit is reduced back down to ten, can the ones that are stuck in pqueue of 10 - 14 be bumped into sifted status?

That was an amusing time to take some sift-vacation.