Archive for April 2008

Unexpected results

4 April 2008
Categories: Uncategorized

Google reports the following results for the term “search engine”:

    1. AltaVista
    2. Dogpile
    3. Wikipedia’s article on “Web search engine”
    4. Yahoo
    5. An optimization company
    6. A CBC Radio show called “Search Engine”
    7—11. More optimization companies
    13. Live Search
    14. Mamma Metasearch
    15. Google… UK!

What’s going on here? Are the optimization companies starting to figure out Google’s arcane secrets, or is it that Google’s expanding horizons are distancing it from its search engine roots?

Tough love for Moo U

2 April 2008
Categories: Research, Wikiality

Over the past couple of years, ever since I first heard about it in Walter Adams’ excellent memoir The Test, I have been fascinated with the Michigan State University Group, a technical assistance project that MSU provided to South Vietnam from 1955 to 1962. Part of the reason it intrigued me was the fact that up until two years ago, I had never heard of the project—and I’d even worked for one of the participating MSU departments for seven years (albeit thirty years after the project ended). I know I’m not alone in this; MSU seems perfectly happy to forget it ever happened.

The thing is, all I could find online on the subject was a reprint of the Ramparts article (“The University on the Make”) that had generated such controversy and campus uproar. In reading John Ernst’s even-handed 1998 history Forging a Fateful Alliance, and more recently Scigliano and Fox’s “official” 1965 overview Technical Assistance in Vietnam, I could see how one-sided the Ramparts article was. Yes, it raised some valid questions about the project, its motivations, and its CIA connection. But it did so by deliberately ignoring any positive benefit that MSUG might have had for the people of Vietnam.

This hardly seemed fair to me. After all, the people who had initiated and participated in MSUG had done so with (mostly) good intentions. That the results were less than stellar was pretty much par for the course in that era of overseas technical projects run by American universities, as Adams and Garraty so deftly illustrated in their book Is the World Our Campus? (1960).

Most people are unlikely to do as I did and take out interlibrary loans of dusty, seldom-used, sun-faded volumes from such far-flung locales as Southern Illinois University and the former Northeastern Illinois State College (as one book’s stamp reads). For them, the sum total of MSUG history on the Internet was a possibly spurious, certainly unsubstantiated tale of cloak-and-dagger nefariousness. History is written by the victors, and in this case the victors were the conspiracy theorists.

I’m not trying to defend the project. Diem was a complete bastard, the U.S. was wrong for backing him at all (much less as long as we did), and MSU managed to piss away a lot of its intellectual capital and new-found respectability by playing along. But the University’s motives were not 100% craven, and MSUG was not merely a CIA front.

So, to provide a more even rendition of the history to a wider audience, I wrote an article for Wikipedia. (My other, more personal, motivation was that it allowed me to finally get the story out of my head, where it had been bouncing around without a proper audience for months.) Before posting it, I ran it past my wife, who called it “tough love” for my alma mater; and one of my closest friends, who suggested I run it past my dad, who remains active in the University community.

Dad was very supportive: not only did he purchase and send me another book on the subject, but he also contacted a fellow retired professor who had run the MSU international programs office for many years. (This professor begged off the question by reminding Dad that he had started in that role several years after the end of MSUG, belying the fact that he had of course worked closely with nearly everyone involved. This may be seen as an indication of how the subject remains a sore spot with the MSU administration.) The professor read the article and offered only minor copy edits and a few vague suggestions, and said he thought I should submit the article to a magazine or scholarly journal for publication. That was a pretty clear sign that I had nailed it.

And now, a digression into Wikipedia minutiae…

With no red flags waving, I posted the article just after noon UTC on March 26. I added a few redirect pages and also incorporated links to it in several articles that mentioned (or should have mentioned) the project: Ngo Dinh Diem, John A. Hannah, and Operation Passage to Freedom among them.

It soon was noticed by a user with a vast Wikipedia résumé on South Vietnam, Blnguyen, aka YellowMonkey, who praised it and kindly offered assistance in polishing it up for Featured Article status. He (I assume it’s a he) made some minor improvements but otherwise left it well enough alone. Unfortunately, he doesn’t have any sources for related photos either, so I’m still trying to figure out if the image of Fishel and Diem I scanned from Ernst’s book can be used.

The other thing I did that day was to self-nominate the article for Wikipedia’s “Did You Know?” feature. DYK has only a five-day window for eligibility, meaning that nominally the hook could go up on the main page any time through the 31st of March, or maybe the 1st of April (or, not at all). For the next several days, I watched it closely.

The DYK template has a minimum refresh interval of six hours, but as I watched it over the past week, it generally had a refresh interval of closer to 7½ or 8 hours, or about three new lists per day. It was pretty easy to follow, even though checking on some updates meant stopping by the computer after a 4 a.m. trip to the head. Thing was, the DYK nomination list is usually backlogged, so the stuff that appears is almost always the five-day-old hooks. This meant I should have expected to see my DYK, if it was chosen, around the 1st.

Except of course the 1st was April Fools’ Day, and all the DYK hooks were jokey twists on reality. So everything got delayed by a day. The first DYK update went up promptly at 00:01 UTC, 2 April, signifying the official end of April Fools’ Day (not that that abated the attempts to edit the Oldsmobile page, but that’s a different story).

The next DYK went up at 06:02, exactly 1 minute after the refresh window opened. And lo and behold, there was my hook.

Or, a facsimile of my hook. Although it exceeded the 200-character recommended limit by a handful, I had offered up:

…that the Michigan State University Group, which gave technical assistance in public and police administration to the government of South Vietnam from 1955 to 1962, provided cover for the CIA?

What appeared was:

…that Michigan State University, which gave technical assistance to South Vietnam from 1955 to 1962, provided cover for the CIA?

In other words,

…that the Michigan State University Group, which gave technical assistance in public and police administration to the government of South Vietnam from 1955 to 1962, provided cover for the CIA?

By the way, the edits and inclusion of my hook in DYK were done by my new acquaintance, Blnguyen. I suppose “in public and police administration” was simultaneously too detailed and not explicative enough. Pulling “the government of” was a good bit of clean-up, since the contract was with the government but the assistance was, at least in intent, given to the country—i.e., not just the government, but the people of Vietnam.

Piping the article title into “Michigan State University,” however, shocked me a bit. At first I thought my trouble with the piping is that it implies, at first glance, that “Michigan State University” is a new article, which of course it is not. But the truth is, it was shocking to me because it pulls no punches, and gets to the heart of the matter: MSU itself provided the assistance—and the cover. MSUG was just the instrument of providing. This edit made me face the fact that I’m still a bit uncomfortable about my school’s integrity having been risked on this enterprise, and perhaps subconsciously I was trying to distance MSU from MSUG.

What chagrins me most, though, is the fact that the next DYK update occurred at 12:14, or 6 hours and 12 minutes after the previous. On average, the DYK refresh intervals today (including a relatively laggard one at 18:34) have been merely 11 minutes longer than the required 6-hour minimum. That’s a considerable change from last week: from 3/25 to 3/28 the average interval was an even 8 hours. Clearly they’re working to eliminate some of the backlog by updating as often as possible. (And yet, by 03:00 UTC on 3 April, they were back to an 8 hour interval.)

In short, if we go by Michigan State time (that is, EDT), the DYK hook appeared from 2:02 AM to 8:14 AM on a Wednesday morning. So much for getting it noticed. I’ll have to check tomorrow to see if the article received any bump in traffic at all.

Follow-up: Well, that was at least a bit impressive. 1 April: 11 reads. 2 April: 1172! (Also bumped up views of my user page on that day to 25, from a pre-posting YTD average of 2.34.) Nevertheless, I still feel that the article didn’t get the audience it deserved from DYK.