Hacker News Reader

Comment by danhak | Hacker News Reader

I don't mean to be too discouraging, but I really don't understand the point of this as a business. It seems like something that anybody could code in about five minutes, and YouTube itself has started to add playlist functionality.

View on HN

I pursued this idea with randomwalker for about 6 months before realizing there's absolutely no business value here. My advice to Nowmov: choose another market before it's too late. We did, and it's been going much better for us. Fortunately, we didn't come across a strong celebrity endorsement like Ashton Kutcher's, so there was little cost to switching gears.

The problem with video recommendations is content ownership. YouTube has a monopoly on user-generated content so they immediately became our number one risk. Our business was henceforth bound to their ToS. Shortly after we switched gears, the Totlol debacle surfaced on Hacker News, confirming our suspicions that the risk was too great for a business to tolerate.

The main obstacle with the YouTube ToS is that they don't allow advertising on pages where the video content is the focal point of the page. Nowmov clearly focuses on videos, so the 'get-big-and-strike-it-rich' model is not an option for them. Moving beyond YouTube -- say, using licensed content -- cripples their product, because there simply isn't enough content to create a compelling recommendation engine. Even Hulu, being backed by industry giants, is bleeding content providers.

I can't emphasize this enough: this is a cool product, but not a good business. We had cool algorithms that provided fantastic recommendations and kept users on our alpha site for an average of 13 minutes before adding any type of social features. "Pandora for videos" was dead simple to pitch, too. I wish NowMov the best of luck and I hope they can succeed where we failed. The world needs a product like this.

Look at it from the other direction. In 5-10 years, there will be a lot of really successful businesses made as the line between TV and internet blurs. Who knows what those businesses will look like, but I'd bet that a few smart hackers who are able to build and iterate quickly will figure it out before Comcast - and maybe even before Google.

Recommendations is an 11-figure business. Seriously. This is a pretty reasonable angle of approach, so I can understand that part.

What I have trouble with is what chaosmachine points out: the recommendations are not that good. At least from a cold start.

How does this have 18 votes in 16 minutes? At a time when the site is pretty quiet? That is very suspicious.

From other YCers one would guess ? - maybe someone did a shout out at the scribd party or something to get upvotes.

I get that recommendations are a big deal. But YouTube itself already DOES have recommendations, on the home page if you're logged in!

If a service like this catches on, YouTube could replace it overnight, especially as they have data on what you (and the people you correspond with, via Google) enjoy watching.

The Pandora analogy has been thrown around a fair deal, and it is flawed. Pandora does not play and recommend content that is owned and managed by exactly one company.

>How does this have 18 votes in 16 minutes? At a time when the site is pretty quiet? That is very suspicious.

I'm sure it's nothing nefarious. But a good number of people who use this site have strong affinity for YC startups either because they're YC alums or just they're people who really like YC startups and feel like part of the YC community due to being members on this site and having watched YC grow.

I would guess that a subset of said people will basically upvote anything about a YC startup. And launches are especially prone to this treatment because people think that users of this site will find any YC launch interesting (which many users will).

I upvoted it because I worked on a similar idea with abossy and randomwalker. I agree with abossy's comment above: this is a great idea, I wish them all the best and I hope they kick ass, but the business side is difficult. So long as you rely on YouTube for videos, you are at the mercy of changes to the YouTube terms of service. You can look for other sources of videos besides YouTube, but this is challenging.

So I'm hoping they succeed where I failed. Hence the upvote.

Can you name one recommendation service that is an 11-figure business? The only company I can think of whose core business is mostly recommendation is Pandora, and they're not even that large; nine figures at best.

One could make a pretty strong argument that Amazon is at its core a recommendation service (with really good logistics that take care of the rest of the whole "oh, I just bought this thing I didn't know I wanted" process).

I also can't figure out why in the world would anybody use this - neither me, nor anybody I know, regardless of age/sex/other characteristics.

But then again, I said the same thing about Twitter when it came out and I was wrong.

YouTube could also easily replicate this idea and probably do much better recommendations since they have more resources, more data on the users and videos and more people that are experts in ranking algorithms.

Their business model might be being bought by Google :)

What business model?

Reply on news.ycombinator.com