Posts Tagged: youtube

Noticed / Thoughts — 3 Comments
7
Jul 08

Ongoing Analysis of YouTube-Viacom

News has moved quickly since Wednesday’s ruling by Judge Louis Stanton in Viacom et. al. vs YouTube et. al., the landmark ruling ordering the transfer of all YouTube user histories. Foremost, Google has indicated it will not appeal the ruling, choosing instead to fight the battle in the court of public opinion. To that extent, Google lawyers have reached out to Viacom, offering to anonymize the transferred logs. Viacom attorneys seem to be open to the option, but have not agreed to anything binding.

Viacom attorneys have stated that they won’t be able to follow the RIAA model and suing individual users. In an article posted today, Saul Hansell of the New York Times disagrees, stating: “Viacom says that it isn’t going to use the information from Google to sue individual YouTube users for copyright infringement, but there is nothing under the law to stop it from doing so.” This wealth of information, tied with a ribbon and presented to Viacom, will present intriguing, appealing options. Why not sue YouTube users, demolishing trust in the net’s eminent video-distribution brand?

What role does Google play in this mess? While not a viable option for a public company, Google could have settled the lawsuit in lieu of turning over our information. Additionally, Google’s practices of storing information for 18 months – far longer than necessary – compounds the snakebite here. If Google regularly expunged or anonymized our records, damage could have been minimized.

As Google rolls over, it is hard not to be angry about the situation. Why does Viacom get a record of every legal video I’ve watched? What right do they have? Wendy Seltzer writes about the dangerous precedent being set: “I worry that this discovery demand is just the first of a wave, as more litigants recognize the data gold mines that online service providers have been gathering: search terms, blog readership and posting habits, video viewing, and browsing might all “lead to the discovery of admissible evidence” — if the privacy barriers are as low as Judge Stanton indicates, won’t others follow Viacom’s lead? A gold mine for litigants becomes a tar pit for online services’ user.”

Furthermore, this class of data – one generated in a seemingly private transaction between one’s self and a server – should be recognized and protected as unique. Not only for the particularly private nature of the information, but the scope of the information that comes with these log transfers. It is one thing to subpoena phone records, it is another thing to get a digital recording of every phone call one has made. This transfer is both content and history; that the information Viacom is receiving is federally protected only adds to the terrible irony.

Noticed — 2 Comments
3
Jul 08

YouTube’s Privacy Catastrophe

In a decision catastrophic to digital privacy rights, a federal judge has ordered that Google turn over the video-consumption histories of all YouTube users. The order, which represents the decision of Judge Louis Stanton in the ongoing YouTube et. al. v. Viacom et. al., stipulates that Google must turn over the following:

  • A record of every video watched on YouTube or embedded in a third-party website via YouTube
  • The viewing histories of all IP’s in YouTube’s database
  • The viewing histories of all users in YouTube’s database

The only videos that are granted protected under Judge Stanton’s decisions are those explicitly private; Google will be compelled to turn over all of the records of every other public video you’ve watched. The EFF states that the order is in violation of the Video Privacy Protection Act, a law enacted following the disclosure of Supreme Court nominee Robert Bork’s video rental records.

That the court failed to understand the privacy implications of such a disclosure is astounding. YouTube’s records capture the viewing habits of a wide swath of the web’s users. Such data should be considered personal and private, such as search data, log data, and telephone records. Such disclosure, public or private, is a clear violation of privacy rights, and sets a dangerous precedent going forth.

In a moment of hubris for the behemoth Google, Judge Stanton cited Google’s own public policy blog, that foolishly argued that IP’s are not personally identifiable information. Indeed, IP’s, usernames and histories are identifiable, certainly in corpora the size of YouTube’s.

I’m certainly not a lawyer, but there are a few procedural questions I’d like answered: What processes or procedures exist for the judge to amend this opinion? Even if the case can’t be made against the transfer of data, can he amend his ruling to allow for anonymizing the data? Additionally, what protections follow the data, post transfer? In the worst case, do the logs become public record? Can Viacom et. al. analyze the logs, using them for business or future legal action?

And what can we, the internet, do to convince the judge of his error? I know that I wave my arms about Google and privacy a lot, but this is not the time to gloat. This is a dangerous precedent, and I hope that we can harness the collective to see if we can put it right.

Other smart people writing about the issue:

Update: Via Siva Vaidhyanathan, this quote from Michael Froomkin:

While it may be the case that some of these videos are trying to share copyright protected materials under the radar, it is undoubtedly the case that many of these videos are (1) truly private and of very limited distribution and (2) the author would be identifiable from the associated information ordered to be disclosed. (The order also is opaque as to what sort of precautions if any Viacom would be required to take to prevent leakage of this data.)

There are some procedural obstacles to getting an immediate interlocutory appeal of this decision, but assuming they can be surmounted I think there’s a strong chance of reversal before the 2nd Circuit.

Update 2: The NYT has quotes from Viacom:

“We are investigating techniques, including anonymization, to enhance the security of information that will be produced,” said Michael D. Fricklas, Viacom’s general counsel.

Mr. Fricklas said Viacom would not have direct access to the data, and that its use would be strictly limited by the court order. Viacom would not, for example, chase down users who had illegally posted clips from “The Colbert Report.”

“The information that is produced by Google is going to be limited to outside advisers who can use it solely for the purpose of enforcing our rights against YouTube and Google,” Mr. Fricklas said.

Thoughts — 1 Comment
2
Aug 07

Why doesn’t YouTube care about relevance?

As we all know, YouTube is a Google property, and perhaps Google’s hottest property going. In the time since Google has acquired the company, I’ve seen lots of feature additions – playlists, “Who’s watching now”, even a Google advertising channel. With all of these feature additions, what I haven’t noticed is meaningful improvements to search and relevance, anti-spam efforts, or efforts to prevent gaming. Granted, a lot of this is likely going on behind-the-scenes, but I’ve noticed YouTube getting less and less of my time as I grow increasingly frustrated with the service.

It strikes me that so many of the problems with YouTube could be fixed by some very simple modifications to the way they present the video thumbnail. The first option would be to not just grab a predictable frame, and instead pick a random frame. Seems simple enough, right? If YouTube wanted to get a little bit more tricky they could offer a multiview thumbnail, which would show a few random frames from the movie. Frankly, any strategy that prevents me opening videos of teenagers talking into their webcams would make me happy.

At the same time, it seems that YouTube hasn’t put much effort into making search useful at all; the first result by views for Minneapolis bridge is some XXX Paris Hilton spam. I mean, if my email filter can catch these, why can’t YouTube? And is YouTube doing anything to prevent the botted results that have made the Most Popular feature almost useless?

Conspiracy theorists will say that YouTube/Google actually don’t care much about developing spam filters or screeing fraudsters. The more video the merrier, for Google to stress-test systems and accumulate the world’s content. However, YouTube being Google, and Google being Google, I have to believe that a number of geeks behind the Google wall of silence also secretly pine to make YouTube better. So do it! While YouTube will always serve a purpose (uploading videos), there’s plenty of competition in this market. Google bought YouTube for its community, and unless Google invests in tools that make the community experience more sane (and less spammy), the community will start to dry up. Let’s hope that this point is not lost on the YouTube/Google team.

Hacks / Software — 1 Comment
19
Feb 07

YouTube RSS Feeds

A few weeks ago, I needed to set up an RSS feed of a YouTube tag. When I searched YouTube’s help index, I couldn’t find anything about RSS, which was very surprising. However, a Google search quickly returned a YouTube page explaining how to set up RSS feeds for users, tags, and a whole host of other options.

I’m not sure if this is just an oversight on YouTube’s part, or if YouTube is trying to hide RSS (hoping that you’ll keep your subscriptions inside the service). Nevertheless, YouTube RSS works, and I’ve found it is a huge timesaver when trying to keep track of videos. Here’s the ten-second tutorial.

  • To set up an rss feed for the tag “monkey”, use the following link (modify accordingly): feed://www.youtube.com/rss/tag/monkey.rss
  • To set up an rss feed for the user “bob”, use the following link (modify accordingly): feed://www.youtube.com/rss/user/bob/videos.rss
Thoughts — 3 Comments
17
Oct 06

Object Oriented Conversation

The acquisition of YouTube by Google provides a very interesting glimpse into the value of community and conversation in social applications. And while this acquisition has been over-hyped, over-analyzed and blogged to death, I figured it might be valuable to explore how YouTube created conversation in their social application. Indeed, that extant conversation created tremendous value – value that contributed to the price paid by Google.

It might be useful to first characterize YouTube as one of the more perfect examples of a long-tail site. Indeed, even the story of the site’s creation is long tail: The founders wished to create a site that would allow them to easily share personal videos with their friends. From inception, the site was designed to capitalize on micro-audiences – one friend sharing a video with her small, real-world social network – exactly the audience comprising the long tail.

Let’s analyze this a little bit. Sure, we all know that YouTube is a great place to go and look for stuff like Daily Show clips and music videos, but what exists beyond the surface? In a site like YouTube, Daily Show clips, viral video, etc. represent the head of the long tail. Indeed, there is a lot of traffic there, but it is transitive traffic. This traffic is comprised of people who get links mailed to them, follow blog links and the like.

However, if you conduct a random search for a term on YouTube, you’re likely to not find the Daily Show or viral videos with 500,000 views. In fact, you’re likely to find tons of content that has been viewed 1 or 3 or 10 or 15 times – and a lot of times that content doesn’t seem all that valuable to us. However, it is this content that drove YouTube’s valuation, and it is a very effective illustration of the value of the long-tail.

As we fill out a social network profile, or post pictures on Flickr, we are illustrating our identity and creating content for others. When we use social applications to create a representation of our identity, we are doing it for our social networks – people we want to impress and entertain. Each time we post a video on YouTube, we are creating content for our social network, the small group of people who benefit from our content creation.

Because video became cheap – our cellphones took videos and digital cameras recorded movies – it is only natural that we would want to take advantage of the channel to use it to communicate with friends. However, this required non-traditional thinking. This required us to stop thinking about video as a one-to-many (and very many at that) and start thinking of video as one-to-one or one-to-very few. Just as the average blog post has an audience of a just a few people, the average video posted to YouTube is just for a few people. And just as that average blog post is quite valuable to that small audience, that average YouTube video is extremely valuable to its small audience. The long-tail scale-free economics that support the blogosphere (and its A-list) are precisely at play in YouTube, creating an invaluable ecosystem.

This ecosystem, the one purchased by Google, came about because of innovative thinking and social architecture. For YouTube to go long-tail, a large percentage of users had to become content-creators. Of course, just as it is tough to find topics to blog about (requisite to being a “good blogger”), it is equally tough to find inspiration to create new video content. Therein lies the beauty of conversation. YouTube, just like the blogosphere, became a place for conversation – conversation that drove content production and flooded the system with interesting stuff for micro-audiences. And since YouTube was a walled garden, for conversation to be valuable it had to occur within the system – an individual on Revver conversing with a person on YouTube just didn’t work for that long tail of users.

The social architecture that enabled conversation in YouTube was built in, perhaps subconsciously, from the beginning. The founders built a site so they could share party videos with friends. The founders, while they probably have more friends now, likely had a relatively small social network. It was the millions of users like the founders, using the service in a similar fashion, that drove the value of YouTube. The fact the site also became the perfect home for viral videos and pirated video was completely secondary – they simply had the infrastructure to support the long-tail, hence the capacity to support non-long-tail uses. Other video sites that aren’t targeting the long tail are missing out on the social forces that drove YouTube – while people like viral videos, it is the long-tail of peer-produced content that keeps people coming back. It is the peer-production that enables conversation, and the iterative process that drives value back into the site. Without this value, a video sharing site is just expensive infrastructure built on a house of cards.

If you followed my blogosphere analogy, it seems to make sense that similar, Technorati-like services could emerge to break the walled-garden aspects of video sharing sites. If I like Revver or Vimeo or Clipshack I should be able to use these services and have conversations with YouTube users, right? That would be sort of like if Wordpress.com users weren’t able to converse with Typepad users. Of course, with video thrown into play, things get a lot more challenging – but it would be useful to define some open standards on video conversation that will allow users on multiple sites to talk back and forth. While the Technorati of video likely wouldn’t index the video content, they would be able to examine standardized info (links, backlinks and trackback, etc) to find relationships between videos on different sites. Of course, this would require the video sites being willing to agree on a common format, and for YouTube to be willing to join the conversation. There’s a tremendous amount of potential value here.

YouTube, and most social sites, are about conversation. It is the conversation between users that drives the creation of content, and the network effects that make the network extremely valuable. YouTube serves as a highly illustrative and useful case study in designing a social architecture – hopefully you will be able to put it to work in your applications.