Sunday, May 03, 2009

What's Next in IT?

Just over a month ago we learned that Microsoft had decided to bring its 'Encarta' encyclopaedia to an end. Like almost all Microsoft products, Encarta was widely popular; but eventually, it seems, it could not keep pace with the popularity (or rapid development) of Wikipedia -- nor indeed with the vast and constantly-developing encyclopaedic resources of Google: as Randall Stross wrote in yesterday's New York Times, 'The Google-indexed Web forms a virtual encyclopedia that Encarta never had a chance of competing against.'

I wonder if you, gentle reader, have ever made any systematic or substantial use of Encarta. If not, you might like to get a taste of the Encarta house style (while you still can) by clicking on their entry on Google itself.

In any case, Google (and Wikipedia) have had sure and steady ascendancies since their inceptions. I can remember a colleague, by about the year 2000, saying 'my home page is Google, and yours should be too.' Already by then, it seemed quite clear that it was leaving other search engines in the dust.

Very few people still remember the encyclopaedia project that was called Nupedia; fewer still will recall that I was on the board of that enterprise when it was active, as one of those responsible for the oversight of the Classics-related entries. Nupedia was founded, about a decade ago, on some of the same premises as Wikipedia, and on some fundamentally different ones; it came to an end not without some sadness and disappointment. But I mention it now because to no one involved in its early days did it occur that Nupedia might founder and fail.

I am guessing that the same could be said of Encarta. To be sure, they are not entirely commensurate examples, as Encarta has been in development since the 1980s, and (as Stross cites) claims to have been 'the No. 1 best-selling encyclopedia software brand for the past eight years.' So in terms of calendar years and of dollars earned, Encarta has had an honorable run -- whereas Nupedia was snuffed a-borning.

We could speculate on the problems entailed -- and the differences between, say, Wikipedia and Encarta. Two salient differences are that Wikipedia is both free and open-source. Free: you can make a donation to Wikipedia, as a discreet link at the top of the page indicates, but this is entirely optional. Open-source: this issue was at the heart of the Nupedia controversy, as some thought that encyclopaedia entries ought to be composed and vetted by experts. (Respect for Wikipedia's accuracy level rose considerably after a 2005 report in the British journal NATURE indicated that Wikipedia's overall accuracy, in matters scientific at least, rivaled that of the Encyclopaedia Britannica. (Britannica, of course, protested that the study was inaccurate, but NATURE has stood its ground. And indeed, 'accuracy' is a fluid notion when it comes to things like Wikipedia, because the very text of the encyclopaedia itself is continually changing -- probably literally by the minute. This is the peculiar strength of open-source content, particularly when it has such a vast and interested readership. Of course, the constant changing to which Wikipedia is susceptible also means that it runs the danger of constantly also having new error introduced; but if we take a macro-level view of things, over time the material is likely to be refined and refined to a point of extreme accuracy.

I am not privy to the inner workings of the behemoth that is Google, but given its runaway success (and corresponding gigantic monetary worth), it is safe to speculate that those in charge of it are overseeing a more or less constant process of updating and improvement in the technology. While it is not open-source software, it is so phenomenally valuable at this point that it would be absurd for them not to take every precaution to preserve their online supremacy.

The fledgling phenomenon that is the Internet may still be too young for us to pronounce upon long trajectories of development at this point. Still, it has been around long enough for us to observe some trends already, a few of which we have noted above. What I am interested in today is: what is it that makes some kinds of information technology succeed and others fail? I assume it is some combination of luck, shrewd timing, effective marketing, and intrinsic quality (the latter including, inter alia, efficacy and user-friendliness). A possible additional aspect (and Google and Wikipedia might both be cases in point here) is flexibility -- the capacity to change and adapt to the very march of progress.

Another question: is there a way to know, except in hindsight, which items are ultimately bound for success? This may never be possible -- or, again, it may just still be too soon in the overall history of IT to discern such signals reliably. But with the rapid recent proliferation of such items -- blogger.com, twitter.com, facebook.com, digg.com, and so forth -- I am sure that these are questions that weigh heavily on those with the most at stake in their development.

No comments: