Yahoo's desperate search times call for open source
Yahoo has opened its search for developers to customize in its latest step in what it calls its Open Strategy, a plan to make Yahoo a platform.
The effort is dubbed Yahoo Search BOSS, which allows "allows developers and companies to create and launch web-scale search products by utilizing the same infrastructure and technology that powers Yahoo! Search."
In a nutshell, Yahoo is opening its search kimono to spur innovation, build a community of developers and with any luck gain on Google and/or thwart Microsoft's desire to buy its search business. Yahoo began talking about its open strategy in January and has been rewiring the place ever since.
Yahoo says:
As anyone who follows the search industry knows, the barriers to successfully building a high quality, web-scale search engine are incredibly high. Doing so requires hundreds of millions of dollars of investment in engineering, sciences and core infrastructure -- from crawling and indexing technology to relevancy and machine learning algorithms, to stuff as mundane as data centers, servers and power. Because competing successfully in web search requires an investment of this scale, new players have effectively been prohibited from delivering credible alternatives to Yahoo! and Google. We believe the BOSS platform will begin to change that.
Very true, but why would you open all that investment to folks? Because you're not winning. For developers, the effort could be a boon. All these sites have to do is support Yahoo's ad business for the technology. To me it's a fine trade-off. Yahoo explains:
It's really quite simple. First, we believe that being open is core to Yahoo!'s future success -- opening our network, opening our own search experience via SearchMonkey, and now opening our search infrastructure via BOSS -- will lead to innovation both on Yahoo! and powered by Yahoo!. For BOSS, we see a virtuous circle in which partners deliver innovative search experiences, and as they grow their audiences and usage we have more data that can be used to improve our own Yahoo! Search experience and as a result, improve the quality of results our BOSS partners and their users get. Second, we do see new revenue streams from BOSS. In the coming months, we'll be launching a monetization platform for BOSS that will enable Yahoo! to expand its ad network and enable BOSS partners to jointly participate in the compelling economics of search.
- What happens to this effort if Microsoft buys the company or its search business?
- Is this effort about defense or offense, a distinction that matters if I'm a developer?
- Can I really build a search effort on Yahoo's APIs with the corporate uncertainty with Carl Icahn and Microsoft?
Despite those questions, BOSS is a good idea. Yahoo's infrastructure can be used to rank search results and tweak against the company's results, create new services, mash up applications and target verticals. Meanwhile, APIs will be expanded.
History isn't kind
History is another item that puts me in the wait-and-see camp regarding Yahoo's BOSS effort. Here's my working theory: Open source is a fine model for companies as long as they start that way. When you go proprietary to open source it's usually because you're losing the war and it's too late. In other words, open strategies have to be in your DNA from the beginning. It's very hard to switch streams.To wit:
- In 1998, Netscape created Mozilla.org to try to keep its Communicator browser going. Netscape was losing share to Microsoft's Internet Explorer and the company wanted to woo developers. Today, it's clear that Mozilla is a huge success, but it didn't do much for Netscape at the time. Mozilla at the time resembled a Hail Mary pass. And it was.
- RealNetworks shared code in an effort called Helix--but only after it was clear Windows Media Player had won the fight. RealNetworks later chastised Apple for having a closed system with the iPod, but we know how that turned out too. RealNetwork's efforts were commendable and it worked hard to recruit the open source crowd, but it never regained its late 1990s momentum.
- Sun has also gone hog wild for open source, buying MySQL, launching Open Solaris and contributing code to the open source community repeatedly over the last decade. However, if Sun had its old gravy train going--selling expensive proprietary servers with its own chips--I'd bet this open source thing wouldn't have happened. Admittedly, that's cynical. Today, Sun is an open source poster child, but it's trying to stir up an ecosystem to move software and hardware too. However, you can argue Sun is using open source as a Hail Mary pass too.
The chart (via Google Finance) tells the open source as Hail Mary pass tale:
Simply put, everyone loves open source when they are getting the snot kicked out of them.
And that brings us to Yahoo. The company invested billions in its search, spent a ton of engineering time and is getting lapped by Google. Meanwhile, Microsoft is hovering. In that context, Yahoo's open strategy makes a lot of sense. But let's not kid ourselves, Yahoo's open strategy could be characterized as a Hail Mary pass too. It may work. BOSS may turn out to be brilliant. But let's reserve judgment until we see some results--on the business and technology fronts.
Update: There are a few differing opinions on this one. Ed Burnette says I have it all wrong and one talkbacker agrees. Can you feel the love today on ZDNet? The crux of Ed's argument is that what Yahoo search plan isn't technically open source. I'll concede that to a degree, but the overall point is that when a company suddenly gets "open" religion it's because things are going so well. Ed also holds up Mozilla as a success--a point I acknowledge in my post. That didn't turn out so swell for Netscape, which was already in decline.
Dana Blankenhorn also has an interesting post on the Yahoo move that qualifies as a must read.