Google News: Where's the context?

People who gripe about free services (especially beta services) are one of my pet peeves. So Google, please don't take this the wrong way. Your basic services are like oxygen to me. But I don't turn blue if I can't get Google News. Today it occurred to me to ask: why am I not yet hooked on your new drug?

At this moment, the top health story concerns the lumpectomy/mastectomy debate. It's an Associated Press story about two long-term studies published today in The New England Journal of Medicine. (The abstracts are here and here, and you can get the full texts for $10 each here and here.)

Google News then list 98 related stories. It is mildly interesting to watch how these NEJM studies flow through the news channels: the syndication of the AP story, the various bylined interpretations of the same event. But I happen to be curious about the context of this story, and this audit trail yields no context.

Reflexively, because it's a habit of mine, I started to follow the threads. Monica Morrow is widely quoted. She apparently wrote an editorial wrapper around these two studies, but there's no abstract, only the paid fulltext version, in today's NEJM. What does Google know about her? The first result gives me some background:

In a study published in the March 1998 issue of the Journal of American College of Surgeons, Dr. Monica Morrow, a breast cancer expert and professor at Northwestern University Medical School, lamented that thousands of women could be spared mastectomies every year, but many were not given an alternative.

The second result leads me to a bio of Dr. Morrow. Were I researching this topic in earnest, I'd be off and running. The NEJM abstracts provide one source of clues. Those authors have probably left trails that can be followed. Dr. Morrow's appearance in other news stories around 1998 would be a fruitful avenue as well. What I'd be looking for is context. How did the consensus in favor of radical treatment evolve? Who were its advocates? When and how did the pendulum begin to swing the other way? Under what circumstances is mastectomy still recommended? Where is the data?

The tool I'd use for this research would, of course, be Google. It would be awesome if the "related stories" feature of Google News could take a first cut at doing the correlation -- finding the NEJM links (which none of the news reports bother to cite, though in fairness the URLs may have been unavailable to them), digging up background on Monica Morrow, ferreting out the history of breast cancer surgery.

I'm sure Google will get there. You don't need artifical intelligence to do the first level of correlation, just brute force -- a technique that looks better and better the more Google applies it. And maybe, if the first-level correlation were automated, the writers of those 98 related stories could deliver more context to their readers.


Former URL: http://weblog.infoworld.com/udell/2002/10/17.html#a472