Google

The New York Times reports this morning that Google has unveiled a system to automatically prevent copyrighted videos being uploaded to YouTube. This system looks to be an olive branch in what’s turned into a bitter battle between Google and major media companies, including everyone from Disney, CBS, NBC to Viacom and Time Warner.

The system uses a fingerprinting technology, where media creators would upload content to Google’s servers. Google’s software, in turn, creates a fingerprint which is checked against any files that a user attempts to upload. If the file too closely matches the fingerprint, the upload is blocked.

The opinions of media representatives about the technology range from the lukewarm to the dubious, and for good reason. While from a copyright-owner perspective it’s nice to see Google taking a serious move towards blocking free uploads of copyrighted video clips, their position has long been that their position as a sharing-service is protected by the 1998 DMCA. Google is happy to remove copyrighted clips when alerted to their presence, but the lack of a cohesive, thorough means of blocking these uploads has frustrated media companies and lead to accusations of Google’s profiting off this copyrighted material (by having a service that is more popular because of the unauthorized content).

While the record-companies continue to wage war against file-sharing even as the future of DRM grows more and more dubious, it’s clear that the copyright battle is just beginning for video media. And Google, as much as they can within legal limits, is standing on the side of more open sharing and proliferation than the media companies that risk losing a lot more with this emerging technology.

Google, in their never ending mission to expand their sphere of influence and develop new revenue streams, has launched an innovative new spin on video content embedding they’re calling Video Units.

The new program takes the limitless content from YouTube and offers it to AdSense publishers, who can embed it in their site with a lean code snippet. The content of the videos can be hand-picked by the publisher, chosen from a variety of categories, or determined by Google with its complicated relevancy algorithms. The videos display in a very customizable players and feature contextual ads at the top of the video and as text overlays in the video content (similar to the contentious ad overlays YouTube added to their website a few months ago). Not only do publishers get free, easily customizable video content for their site, but they get a cut of the ad revenue, which is also split between Google and the video’s creator.

The program is an absolutely brilliant move by Google, as it has the potential of expanding the YouTube brand massively, as well pushes forward embedded video content on sites internet-wide and will be a great testing ground for how these kinds of ads are accepted by users. And while I’m sure the content providers are carefully screened at this point for quality control (as well they should be!), it may end up becoming a great method for independent filmmakers to get additional revenue for their work.

While pretty helpful as far as search engines go (think: all the gems from Matt Cutts, Webmaster Tools, Analytics, etc.), Google still has a vested interest in keeping its search algorithm and ranking methodology top secret, and while they occasionally do comment on issues raised by SEOs and webmasters, they’re typically pretty taciturn about what does and doesn’t work ranking wise. Thus, it was pretty refreshing to have Google’s Webmaster Blog talk about META Descriptions.

According to Google, “Accurate meta descriptions can improve clickthrough, [but] won’t affect your ranking within search results.” Instead, the point of a properly written META description is to attract people to your website by accurately displaying the contents of the page. They use the example of a dynamically generated eCommerce META description that repeats itself and provides vague information as the kind of entry to avoid. Instead, choose simple language that quickly and accurately defines the content of the page.

All of this reinforces what most of us in the SEO trade have believed for a while, but there were a few interesting things on best practices that are worth noting. First was that displaying the META description is a preference, but by no means a rule. Google explicitly states that a META description will only be used in the search results if it is of “high enough quality.” Failing this, expect to see excerpted text from your website or maybe even a DMOZ entry as the descriptive line for your website.

Another interesting thing to note is that Google sanctions programmatically created descriptions, so long as the results are not “spammy.” This is good news for eCommerce sites, and helps reinforce our position, which is that the challenges eCommerce sites face in organic indexing is only a limitation only to the creativity and SEO savviness of the web developer.

Finally, another “no brainer” that is still often ignored by many web designers, is the need for META descriptions to be different on every page — simple boilerplate descriptions copy/pasted sitewide just aren’t going to do it, and Google will be more likely to ignore your description, or worse, put your site in the supplemental index (unless, of course, you believe it doesn’t exist anymore).

Google neglected touching the topic of length with META descriptions, though 150-160 characters still seems to be the rule of thumb, based on how much text actually appears in the results.

While Google’s post didn’t throw any huge curveballs our way, it’s great to see Google talk shop and give a definite answer to questions SEOs have debated about in the blogosphere for years now.

An interesting article appeared in the New York Times last week concerning people who appeared in news stories that were wrong or incomplete coming back to haunt them later in life. The story points out the stories of several people for whom articles with inaccuracies appear at the top of Google — things like mismatched credentials, charges that were dropped later, stories phrased from one point of view without a counterpoint. While in the offline world these articles are long buried in stocks of microfilm no one will care to review, the supreme authority of nytimes.com keeps them at the top of Google ad infinitum.

It’s a very particular problem and I applaud the Times for taking it on, though the solution to the problem is a rather muddy one. Who’s to blame — SEO or Google? To what extent is an entity responsible for removing articles that may be inaccurate? And who polices all this information?

Of course, this is just one example of the bigger issue of what people can find about you on the ‘net determining the outcome of your job opportunities, reputation, and even criminal record — what with Facebook commonly scrutinized by job recruiters, MySpace pictures leading to criminal charges and the possibly damning result of Googling: “Insert your name here.” To some extent, this is the same question posed by the controversy around Google street view — how much privacy can you reasonably expect in our technology heavy times?

The glaring obvious answer is, of course, don’t say anything stupid on the internet, but it’s a fairly insufficient one. After all, while you may hope that no one sees comment #143 on an obscure message board or blog out there in cyberspace, the New York Times brings an inherent trust that is good enough for Google, and good enough for the person checking out your good name.