Hacker News new | comments | show | ask | jobs | submit login
Youtube retires video responses (youtubecreator.blogspot.com)
89 points by mxfh 1304 days ago | hide | past | web | 67 comments | favorite

The thing that bugs me about the way Google retires something is that they do this:

1) Retire a feature

2) Offer a half-arsed work-around

3) Wait an indeterminable period of time

4) Release a new feature that covers the original use-case

That's crap, as users are stuck at #2 for ages and some of them migrate away to use other tools that cover this use-case. All this method achieves is to annoy every user that used the feature (even if they seldom used it).

Surely it would be better to:

1) Release a new feature that covers the original use-case

2) Wait a declared period of time (short or long)

3) Retire a feature

Which means people can immediately start using the new way. And if the new feature fully covers the use-case of the old one, then #2 could be skipped altogether.

> 1) Retire a feature

> 2) Offer a half-arsed work-around

Oh man, so many times.

They recently did the same thing to /my_subscriptions, which was the only proper way to keep up with a lot of subscriptions. It was a grid view where watched videos would disappear. The only option now is a "feed" view where it's a long list and watched videos are just grayed out.

It has produced quite a lot of anger[0] and two kinds of solutions have emerged - one is userscripts that restyle the new "feed" view into a grid and another one is what I'm currently working on[1] - a proper replacement page that just uses the YouTube API, stores your history locally and lets you hide videos that you don't want to watch (this was why I wanted to build this, even before they removed the original grid view).

[0] https://productforums.google.com/forum/#!topic/youtube/uVzkG... [1] http://daviddeutsch.github.io/yt-sanegrid/

This was a feature actively being abused by certain users(google "Reply Girls", potentially NSFW) to piggyback off of popular videos and sometimes even enrage communities associated.

The new planned ability of adding videos to comments doesn't address this issue at all...

But sometimes they don't care for the original use case, so there's no point in replacing it, just retiring

The original use-case was an engagement mechanism.

Their statement is:

> So, on September 12 we’re going to retire this little-used feature as we work to develop more effective fan engagement tools for creators.

They are going to replace it with a new thing to cover the engagement mechanism.

It is true that with things like Google Reader it was not being replaced. But in this case their statement declares an intention to replace the retired feature with something else that helps solve the engagement scenario.

'video responses' feature was not used a lot imo.

but for all startups this is a point of concern, after experimentation, which features to digg further and which ones to keep. user analytics definitely is a must.

Really? We must surf different parts of YouTube. I saw it quite frequently, especially around music, video-game related things, and a few different vloggers.

Edit: that said, I didn't actually click them. I just saw them.

I thought Hacker News was the home of pivoting, launch-'n-iterate, and all that.

Listen, I agree with you on this with things like Google Reader.

But 4 in a million people clicked on the video responses. On a mathematical level, it just wasn't worth it. This isn't a big deal, I don't know why everyone is getting so worked up about it. It's an extremely marginal feature, not a flagship component being eliminated tomorrow without warning.

People aren't mathematical entities. On average, you're a fool, because most people are fools. But if I start out treating you that way, you're right to be upset.

In evaluating a feature, it's important to have a global view. But it's also important to have a local view. If you choose features at random, they're generally used by a small number of people, but they get made because they're important to somebody.

In this case, you can't think just about the people trying to view the video responses. You also have to think about the people who have been making them. Content creators are a vital part of the YouTube ecosystem, and getting non-creators to cross over and become creators is an important engine for content growth.

To somebody who has been using this feature, this is a fuck-you twice over. The first time is shutting down a feature they were using. The second time is the recognition that YouTube was having them do something that didn't work well, but that YouTube doesn't really care.

Can YouTube get away with offending a bunch of people? Sure. But treating people like they only matter when they are currently useful to you? That's for sociopaths and supervillians. When you're building things for people, it's important to treat them as people. They'll remember when you don't.

>People aren't mathematical entities. On average, you're a fool, because most people are fools. But if I start out treating you that way, you're right to be upset.

That's true, but the two cases can't really be compared. Groups of people are, very frequently, in nearly every discipline and line of work, treated as mathematical entities. It's how governments determine how to allocate resources and funding in social programs, and how advertizing and marketing work.

In a public company, one of the main goals is to make as much of a profit as possible. Reducing costs is paramount to keeping profits in the green, and when you're operating on the same scale as YouTube, even a tiny optimization like this can translate to huge savings in the long run.

I agree that people are treated as mathematical entities. And I even said that was fine if that's not all you do. The mathematician's solution to the problem of overpopulation is just to shoot people until you get below the required number.

You are also wrong about the main goals of a public company, and wrong again about how to get there. Those are central dogmas of the MBA worldview, but they're just dogmas.

A great example is cars. American car companies minimized costs and focused on profits. Toyota maximized user value and minimized waste in creating that value. Toyota has been kicking the asses of American car makers for decades. Even when Toyota teaches American car-makers their secrets, they can't adapt, because those and other items of dogma make it impossible to change: http://www.thisamericanlife.org/radio-archives/episode/403/n...

Your point is valid, but your opinion is out of context.

  The team is focused on enabling you to share video links in comments.
They're not removing the feature because people don't use it, but because they're planning to replace the feature with one that improves the original use-case.

Although it's true that the execution wasn't smooth, they should have removed the feature when the replacement was ready.

Right. And your last sentence shows that you are thinking of people as humans, not abstract mathematical entities who it ok to offend.

That's all nice and well, but my point is, no matter what change you make, you're going to alienate users. Youtube can't make everyone happy. It isn't feasible. This change, that affects the view-flow of 0.0004% of users, is an intelligent one. The change Youtube made to subscribed videos (mentioned elsewhere in this thread) might not have been as intelligent.

Youtube has to optimize for the greatest number of users it can. When you're working with a business at scale and there are millions of users (or even a billion), you don't get a bellyache over features that have to be scrapped because they're not used enough. It's a company, not a charity. That's how it works.

No, it does not affect the view-flow of 0.0004% of users. The actual number of users affected is not stated and cannot be inferred from the data given.

My point is that sure, sometimes you must alienate users. But good product management requires you to minimize that number, and the size of the pain. That's just good business.

it just wasn't worth it. Fully agree with that, but was it worth it to remove it before its replacement is ready? Since it was more or less unused it isn't like it was causing a big burden.

Nobody used that feature.

I used this feature. It was a great way to connect videos in the community I belong to. Just because popular videos like "Gangnam Style" or whatever don't use this feature doesn't mean that it isn't useful for other users.

4 out of every 1 million.

Which isn't many, but it might have been critical to those 4.

And given the numbers involved when we talk about YouTube, it's not like it's really as low as 4 people, it's 4 out of 1 million views on a site that generates 1 billion unique visits each month ( https://www.youtube.com/yt/press/en-GB/statistics.html ). It's actually a large number of users then.

> Which isn't many, but it might have been critical to those 4.


I understand it as 4 in a million are clicking on video responses. People creating video responses as self-gratification might care a lot about it..

I think the 90-9-1 rule applies.

The reach of those 1% is pretty substantial.

I have never used the feature in question, but I'd certainly think twice about removing any feature that 1% used.

If the click through rate is 0.0004%, I don't think the power users will care all that much. They're probably not getting any significant source of views from video responses. The more powerful tactic on YouTube seems to be linking to other videos at the end of your video.

Content creators on Youtube have a bootstrap problem. Once their videos become popular enough, Youtube's content discovery mechanism will direct potentially interested viewers to said videos. The problem is that if no one has seen the video, then the algorithm has no idea what its quality is, or what type of viewer would be interested. Depending on your target audience, even getting a small number of views by piggybacking a popular channel may be enough to get you onto the algorithms radar, which generates a lot more views that the click through rate would suggest.

Linking within the video works, but it requires that the creator of the original video does more work, and if it was not planned when the video was created then, while doable with annotations, it appears scammy.

I was referring to linking in the video, but you make a good point about wanting to link to videos made after the fact.

While piggybacking may be an option for garnering initial views, using it for this purpose encourages response spam. Using good 'ol search optimization should really be the first option for content creators looking to get off the ground.

'Which isn't many, but it might have been critical to those 4.'

Well yeah. But you can say that for nearly any feature of program. I don't see this as a big deal.

It's such a small number that it makes complete sense to remove that feature. Removing feature that may significantly lower user experience on 0.0004% visits while slightly improving the user experience on 99.9996% visits? Quite obviously good decision.

I wonder what the reverse click-through rate was.

When you watched a video that was a response to another video, YouTube showed "This is a video response to ______"

I personally used that a few times, unlike video responses which I don't remember ever using.

I had a small video on my channel a while ago which originally procured a few thousand hits but when someone made a much more popular video response (made popular due to virality through reddit/digg) my viewings for that video shot up due to, as you described, "reverse click-through".

Yeah, is it any surprise that the click-through rate would be abysmal? People want to watch the video, not responses to the video. But as you say, when you are on the response, you may actually want to watch the original video and so the reverse click-through rate should be much higher.

At first I thought this meant video comments had been retired. For a split second there I thought the world had been suddenly healed of a great evil and had great joy. Alas, it passed.

Exactly. Video is so bloody inefficient for anything other than how-to. Video is lazy journalism for those unwilling to subject themselves to the rigors of concision or clarity demanded by the written word. When the video/text ratio of a site rises, I stop visiting.

When the video/text ratio of a site rises above zero, I stop visiting.

Fixed that for you.


Hacker News has featured videos on its home page. A lot.

I stopped clicking on video responses when I realised that almost all of them were spam videos trying to get some views after the original video went viral.

Youtube Video Response spam was pretty nasty. It did attract some attention for unique spam detection algorithms for video responses as well [0,1].

[0] Fabricio Benevenuto, Tiago Rodrigues, Virgilio Almeida, Jussara Almeida, Chao Zhang, and Keith Ross. 2008. Identifying video spammers in online social networks. In Proceedings of the 4th international workshop on Adversarial information retrieval on the web (AIRWeb '08), Carlos Castillo, Kumar Chellapilla, and Dennis Fetterly (Eds.). ACM, New York, NY, USA, 45-52. DOI=10.1145/1451983.1451996 http://doi.acm.org/10.1145/1451983.1451996

[1] http://www.iiitd.edu.in/~ashish/PST-2013.pdf

They were abused by spammers, but it was a nice feature, and had a lot of legitimate users as well. Maybe it wasn't used a lot on mainstream youtube, but there were places where people commonly made videos to respond to the arguments made in another video, for example. I don't see the point in getting rid of it.

They suggest people should search for responses to the video instead. Few will and even if you do, you don't have a good chance of finding them necessarily.

I've only ever seen one useful thing done with video responses: a kind of pseudo-playlist, where the next video is a response to the previous one, making it easy to click from one to the next. With YouTube's new playlist interface, even that's no longer useful. (And with the ten-minute length limitations lifted, the need for multi-part videos has gone down significantly, so only long-running series need them anymore, and those can easily create playlists.)

Now if only they'd bring back the ability to subscribe to a playlist rather than a channel. I still have some playlist subscriptions, but you no longer have the option of following a playlist without seeing all the other videos from that channel.

Video responses are often used by content creators that want people to submit content for mashups. Wonder how they'll do it now.

Ideally, an existing or a new competitor will fill the void. I really like YouTube and I realize there are viable alternatives, but web video often feels like more of a monoculture than it should be.

You can't effectively compete against YouTube for this particular space (which is why even though there are alternatives it still feels like a monoculture): the reason that website exists is because Google has a sufficient monopoly in search advertising that they can afford to dominate any other market they want by operating well below cost. Supposedly they finally got YouTube profitable (at which point you can just view YouTube as a legitimate investment), but this was after years of running the website at a loss of hundreds of millions of dollars a year.

Youtube has been doing nothing but fuck content creators lately.

Videos of interest would be tagged accordingly.

There are lots of guitar competitions hosted on YouTube. The way to enter this competition is by replying to the original video.

Maybe by using the title, description, etc. the competition creator can aggregate them and create a playlist but that's more work for them. Just browsing through the video responses is a much quicker way currently for both the competition creator and participant.

Perhaps a hashtag would be in order? ;)

Sure. But what I don't understand about the spam video responses meme in this thread is that the video owner can currently choose to automatically accept video responses or not.


Features need to be trimmed now and then for a piece of software to survive in the long term - otherwise it evolves towards infinite complexity and feature bloat, and subsequently ossifies.

If you're one of the 1 in 100,000 youtube users that enjoyed this feature, you're upset that it got removed.

But for the greater good you should just accept you were unlucky.

The alternative is a system (and UI) that gets more and more complex over time to the detriment of everyone


There can be features with VERY poor CTR overall that are still valuable for very specific and valuable segments of your population of users. I wonder how much they dug into this and explored the potential systemic impact? I assume they did... but I can't help but have some doubt.

It is right and good to kill things nobody uses. Come on with the bellyaching, on this one. Google Reader this wasn't.

I used it

Why are people upvoting this submission? What's interesting about the fact that YT removed some insignificant and unused feature?

I find it notable that something that seemed so prominent on a very popular website like youtube only had .0004% click-through rate.

I find it interesting that basically 100% of videos were from spammers (spare me the rounding error anecdote) so the CTR for video spam is "about" two orders of magnitude lower than the supposed (although highly gamed and faked) CTR for banner ads which are also by definition 100% unwanted spam.

So are there really two orders of mag of fraud in the bigger industry of banner ads, or despite popular belief is video two orders of mag less compelling than static banner ads or email spam... Something doesn't add up.

Because Google. Similar thing happened to me when I could not publish a paper for a long time due to no interest in a subject... Until somebody published a similar thing with a Google affiliation. Then I couldn't publish it cause it was not original any more.

I support removing video responses. They were useless. Upvote!

Because it wasn't insignificant, and is actually used?

It was actually used, but by a fraction of users.

A fraction of users upload videos on Youtube, guess this feature also needs to be phased out.

.0004% of the respondees or .0004% of visitors?

A lot of vitriol in the comments of that post.

I wonder how many people will stop using Google products as a result of this design decision.

About two and a half, probably.


NO! I'm afraid of change!

Not surprising. There's really not much justification for getting worked up over this. They're not sunsetting a critical component of Youtube. This isn't a Google Reader...come on.

4 people in a million? You'd get rid of the feature too. That's what you should do - it might alienate four people, but honestly, your business has to pick and choose which customers to piss off. You really can't please everyone in the world with your decisions, and this frankly wasn't useful to Youtube from a mathematical perspective.

The math pretty much proves it - this was a smart decision. People will adapt and get over it. This is barely news.

Four clicks in a million is not 4 people in a million. Unless all of your users show up exactly once per month, click one thing, and leave. Which is probably not the YouTube model.

Further, who these people are matters. The number of active editors on Wikipedia is about 0.0004% of total users. But if you kill something even a small percentage of them use, god help you.

Most importantly, people aren't mathematical entities. You may sometimes have to piss some customers off. But most of the time you don't have to piss anybody off. And that's a good choice, because unlike, say, your average integer, people have memories, and they talk with one another.

Product decisions aren't mathematical decisions; they're human decisions. Ones often helped by good math.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | DMCA | Apply to YC | Contact