Wikipedia talk:MIT Media Lab survey

I have moved this to the Wikipedia talk space as discussions on a survey don't really constitute an article. Discussions should be here on the talk page with info on what it is on the meta page. Angela 23:35, Sep 11, 2003 (UTC)

It was nice. Kind of like a first date, only with less petting. -戴&#30505sv 00:30, Sep 12, 2003 (UTC)

Was it a really good survey or do you just have really boring dates?

It was fun, but my date was really english-centred in his questions...

The survey is indeed interesting. Thanks to Nanobug for finding it and Angela for organizing the discussion. Has anybody contacted MIT to find out:

What they intend to do with the information?
If we can get a copy of the raw or their assembled results?
Who are the authors, and how do we contact them?

As implied, I enjoyed the survey. I wish them well, and I hope we can too derive some good from their results. Lou I 03:08, 12 Sep 2003 (UTC)

Actually, Nanobug gets the credit for both those things. All I did was move it to the Wikipedia namespace. The only information I have is that there was this post on the mailing list. Angela 03:15, Sep 12, 2003 (UTC)

Thanks Angela, I've located the author by email, if I get answers on the above questions, I'll post them. Lou I 04:32, 12 Sep 2003 (UTC)

Actually I didn't find it, Jimbo Wales put it on the front page for the MIT Media Lab survey people (see page history on Main page on 14:25 11 Sep), and I just created a page where we could discuss it. Angela then moved the discussion to a more appropriate page (thanks Angela). Looks like an interesting survey, but I'm sure Wikipedians could together come up with a better one. I particularly think that some of the essay type questions (like "If you were building a visualization tool...") would have been much better being placed in Wikipedia itself, so we could all see each others answers, and brainstorm properly. Nanobug 11:17, 12 Sep 2003 (UTC)

I thought that the survey was kind of irritating. It felt like it was over analysing what is basically just a tool.

How often do you use the Hammer?
Have you ever built a Hammer?
Which end of the Hammer do you usually use?

and so on.

-Anon

Which indeed reads like a transcript of my typical first date. -戴&#30505sv 04:20, Sep 12, 2003 (UTC)

Do your dates make you fill out surveys as well? --Delirium 04:36, Sep 12, 2003 (UTC)

On a more serious note, I agree mostly. It seems like it'd be an interesting undergraduate project, but it just doesn't really seem to be something I'd expect professional researchers to be doing. If you're going to analyze wikipedia, the stuff the IBM researchers were doing with tracking article histories seems a little more substantive. But I could be missing something here. --Delirium 04:38, Sep 12, 2003 (UTC)

One of the IBM researchers is the author of this survey: http://researchweb.watson.ibm.com/history/contact.htm. Alex756 04:52, 12 Sep 2003 (UTC)

My take is that he is trying to find some aspect of updates, interractions, etc. where he can use a visulization or graphic technique to help understand things like the IBM data. Lou I

The homepage for the person doing the survey is here-- and now I remember his note sent out a while ago I feel bad for not giving it a second thought. It seemed to have caused no ripples, but then depending on the time it was sent, the ripples of new pebbles are easily drowned out by bigger boulders that happen to be falling into the river -- or is WP a vast lake? I forget. And its all related to this -- a collaborative, graphical way to look at data in a 3D environment here . Very nice, actually.--戴&#30505sv 05:09, Sep 12, 2003 (UTC) http://researchweb.watson.ibm.com/history/ http://researchweb.watson.ibm.com/history/gallery.htm

I think Fernanda is a she, not a he. See her IBM contact information [1]. - Alex756 04:56, 12 Sep 2003 (UTC)

Oh, I never was good with genders. -戴&#30505sv

Maybe this is your chance. Alex756 05:16, 12 Sep 2003 (UTC)

Dont jinx it. -SV

Is that better(See above)?

Sorry -- I see white (light) text on black (maybe grey) background only --much much easier on the eyes. Everybody knows that;)-SV

I wish I had known about the above links yesterday, when I filled out the survey. I could have provided more intelligent input, rather than puzzled responses (e.g., no, I don't engage in vandalism on Wikipedia, & don't understand why anyone would). -- llywrch 05:19, 13 Sep 2003 (UTC)

Will the graphical Wikipedia edits viewer be available to the public directly from the "page history" page, or just for reserachers at IBM? LDan 11:32, 12 Sep 2003 (UTC)

I think its an IBM-funded research project, and as such may be subject to thier managerial controls-if it were ever to be released as open-source.-戴&#30505sv 22:15, Sep 12, 2003 (UTC)

Richer online discussions

Have you seen the survey guy's web page? He's interested in "creating a richer environment for online discussions".

Maybe he can help us improve the WAY this website works, so we can get better articles written faster!! --Uncle Ed 14:08, 12 Sep 2003 (UTC)

It would be nice, but its not clear--judging that those involved are choosing to remain somewhat detached from the WP --as an objectively-viewed reserarch subject--I cant tell where the desire for patent/proprietary tech merges with the open. At some point no doubt this may be cleared up.-戴&#30505sv 22:15, Sep 12, 2003 (UTC)

Words rather than sentences

They have interesting history flow charts on here (as already linked to above). I see one problem with their methodology: since they consider changes to sentences, rather than characters or words, the program interprets typo fixes and single word changes as changes to the whole sentence - it appears on a lot of the pages in the gallery as if contributions made by people early on the life of a page are almost entirely wiped out over time, and this is just not true. In reality a significant portion of the changes are words here and there. It would be interesting to see the charts using algorithms which compared characters or even words rather than sentences. Nanobug 14:18, 12 Sep 2003 (UTC)

Interesting.-SV

To me, it looks like it was done by words. In one of the examples, it broke up a sentence. It wouldn't be bad if they switched to characters, though, but it might be more computationally intensive to do that. LDan 14:20, 13 Sep 2003 (UTC)

Survey results

I've been in contact with Viegas, the survey sponsor. She promised that when they produce a summary, it will be posted to Wikipedia, the meta-page for this one. If I can get her to post it directly, we may gain a contributor. :). Lou I 16:41, 12 Sep 2003 (UTC)

It would be nice. Im glad Jimbo stuck it up there.-SV

Main page?

I don't like seeing this on the main page - it would have been better placed on announcements or the village pump, in my opinion. Martin 18:03, 12 Sep 2003 (UTC)

Its bringing the communities' help to legitimate and perhaps vastly useful research. I couldnt disagree with you more. Im glad Jim set this precedent and hope that we see this kind of thing more often. We may have to debate the particular merits of each research item, (if there are more) -but who can deny that its a high honor for the WP to be the subject/object of unique and serious academic research? -戴&#30505sv 22:15, Sep 12, 2003 (UTC)

Fair enough sv, we can agree to disagree. I just felt that as essentially free advertising for a commercial endeavour it is somewhat unwelcome. Martin 01:59, 14 Sep 2003 (UTC)

Is it free advertising? I thought MIT was a not-for-profit corporation like Wikimedia? I know that the survey sponsor is also involved with the IBM research project, but she is doing the survey as part of her graduate work at MIT. Won't it be published at some point as academic research if it is under the balliwick of MIT? Alex756 08:58, 14 Sep 2003 (UTC)

MIT is somewhat notorious for keeping their work closed, and funnelling it to spin-off companies owned by their professors. This is a large part of what led Richard Stallman in disgust to found the Free Software movement. --Delirium 19:22, Sep 14, 2003 (UTC)

Nice generalization. There is no point in keeping little research projects like these "closed". Better to publish it, as it helps the student compete for scholarships, and profs need to show papers to prove their doing something. dave 14:11, 15 Sep 2003 (UTC)

Fine, as long as they don't horde their results... -- Cimon Avaro on a pogo-stick 07:28, Sep 13, 2003 (UTC)

I would not have seen it if it wasn't on the main page. Fernkes 20:41, Sep 13, 2003 (UTC)