Ars on Google at 10 years old

Ken Fisher on Google vs. Microsoft

Just 10 years ago, most tech analysts believed that the future was Microsoft's. Not only that, but on many editorial pages, Microsoft already owned the present. In the years surrounding Microsoft's battle with the US Department of Justice over its business practices, many believed that only the government could prevent Microsoft from eventually steamrolling all of technology. Past technology giants were just weak copies of their former selves, many would say. IBM, it would seem, was doomed as as a tech powerhouse once OS/2 was done asunder by Redmond; Apple, it was commonly argued, was on its deathbed (an argument that has been made yearly since Apple first released the Macintosh, of course). The 90's were a time when most people believed that whoever controlled the desktop operating system controlled the heart and soul of technological development and thereby, its promise.

Google has, with only a little help from the Department of Justice, proven that competitors can come out of nowhere and change the game. At a time when AltaVista was seemingly the best search on the planet, here came a search engine that appeared to know that you were searching for as if it could read your mind. Its pinpoint accuracy—largely lost since, in my opinion—was only outdone by its lack of ads. Google seemed, dare I say it, altruistic.

Of course, when Google first appeared on the scene, no one thought it could become a counter-weight to Microsoft's influence. But with a steady stream of new (beta) product releases and several critical acquisitions, Google became a giant. Its $103 million acquisition of Applied Semantics in 2003 was the turning point. The acquisition gave birth to AdSense and AdWords, the two revenue-side programs which are responsible for almost all of Google's largesse. Only a few years later, one could now see how the desktop was was necessary, but not sufficient in terms of owning the lion's share of promise. Now it was the interconnection of all of those desktops worldwide, in an open platform (i.e., the web), that offered the most promise. Google proved that point better than anyone.

For all of this, Microsoft hasn't been diminished. A look at the company's financials shows that it's strong, and the silly predictions that Windows is dead or irrelevant are easily dismissed by even a cursory understanding of the IT world—something not grasped by emo kids and some of the day dreamers in the Valley whose interest in the Fortune 500 ends at the salaries of their execs. But while Microsoft has experienced some strong linear growth, and has more pie today than it did 10 years go, that pie itself has gotten much larger, and its percentage share has clearly dimmed. It's amazing what Google has accomplished in 10 years, and even more amazing to imagine who will do to Google what it has done to Microsoft. Students of history know that nothing lasts forever, and as we celebrate 10 years of Google, we already hear the moans and complaints of those who find the company too large, too powerful, and too self-centered. But the honeymoon for Google hasn't really ended yet, over-inflated stock price corrections notwithstanding. Here's to another 10 years of Google.

Jon Stokes on Yahoo!, Google, and the post-scribal archive

Another major step [in the evolution of the written word] was the division of a stream of writing into standardized units of transmission. That is a fancy way of saying "chopping the text up into pages." Simple as it sounds, this step was slow in coming. Cutting up the scroll into uniform but arbitrary portions, and sewing them down the side to make a codex—a manuscript book—must at first have seemed a leap into arid technological abstraction. Once that leap is made, many other things are possible: page numbers, running heads and indices, for instance. The codex does for the text what the alphabet does for the language. It articulates it—just the way the joints articulate the hand.

Warren Chappell & Robert Bringhurst, A Short History of the Printed Word, p. 39

When Jerry Yang and David Filo began the part-time undergraduate project that would eventually turn into search giant Yahoo!, their aim was the create an index of the contents of the then-miniscule World Wide Web. The idea that the Web could and should be "indexed" like a library or a book was an obvious one at the time, and it was obvious because indexing was what people had done with suitably large collections of unstructured text since writing made the leap from public monuments and tombstones to paper in antiquity. Scrolls were indexed in library collections, and when the book form was invented its individual pages opened up new possibilities for detailed, fine-grained indexing. The early WWW, then, was a mass of largely unsorted pages and groups of (partially indexed) pages, and to anyone with eyes to see the whole thing was begging to be indexed. So Yang and Filo, eventually aided by an increasingly large number of employees, set out to do just that; they came up with an indexing scheme, and they got to work sorting.

Two years after Yang and Filo started Yahoo!, another pair of students embarked on a similar project using a radically different approach. Instead of throwing humans at a large and growing mass of unstructured data and asking those humans to sort it into a given index structure (i.e., the Yahoo! model), Sergey Brin and Larry Page threw algorithms at the data; and they asked those algorithms to discern the perpetually changing structure that was already there. In this respect, Google's most important contribution to the organization of knowledge was that it abandoned the "index" as traditionally defined and deductively realized (i.e., "index" as a top-down organizing rubric for a body of content), in favor of the "index" as inductively realized (i.e. "index" as a content- and meaning-agnostic form of organization that simply matches keys with return values).

Ultimately, the key difference between Google's index and the Index that was Yahoo! was this: Yahoo!'s (user-facing) index was a human-invented topic hierarchy that was imposed on the archive from without, exactly like the Dewey Decimal System that was its immediate precursor; in contrast, the semantic or topical structure of Google's (machine-facing) index reflected only the relationships that the search algorithms found in the archive, with the result that the index's structure was (usefully) a direct expression and artifact of the data at any given moment. Importantly, that structure was never visible to the user—and it wouldn't have been useful if it was—but it was accessible via the simple search box.

Of course, Larry and Sergey weren't the first to forsake the age-old model of sorting data by hand into a fixed topic hierarchy and instead turn algorithms loose on the unstructured textual mass, but the PageRank algorithm that Google gave us turned out to be the best tool for uncovering, from moment to moment, human-usable patterns of relationships between terms in our dynamically evolving archive, and for exposing those patterns to individual users.

Thus it was that sorting finally gave way to searching, and searching to finding, and the scribes ceased two millennia of organizing the archive to watch as the archive organizes itself.

Ryan Paul on Google and open source

Over the past 10 years, Google has become an influential force in the open source software landscape and has made many important contributions to furthering the evolution and adoption of open technologies. The company is widely recognized as an important ally in the quest to bring software freedom to end users, but it still has much to learn about transparency and inclusiveness.

Google's most celebrated gift to open source software is the annual Summer of Code program, through which the company pays students to develop useful new features for a large number of software projects. Some of the students who were introduced to open source software through the Summer of Code program have become valuable and respected members of the community.

Google has also provided valuable funding and developer resources to improve several components of the Linux desktop stack—most notably Wine, a Windows binary compatibility layer that helps users who are dependent on Windows software make the jump to Linux. Google also offers some infrastructure to help facilitate open source software development, including a code hosting service and code search tool.

But while Google has aided the advancement of open source software in many important ways, the company has also stumbled on several occasions. For instance, Google has faced criticism for locking enthusiast developers out of the Android development process despite touting the platform's openness. Google has also received criticism for controversially discouraging adoption of the Affero GPL.