Computer Science: Trying Mitra’s Experiment

In the last post about computer science, I posted a link to one of Sugata Mitra’s videos, in which he reports giving a bunch of Italian kids in Turin a bunch of questions in English, and asking them to find the answer without the aid of a human translator.  To his surprise and pleasure, the students were able to answer the questions he posed them by translating them in Google Translate, and then finding Italian answers to the questions, and then translating them back into English so Dr. Mitra could understand them.

I’m enthralled by this idea.  I’ve been interested by the problem of Search Fu for a long time.  And I wanted to give my students a greater appreciation of how much information was available to them, constantly.  And how much they could learn by getting curious about something, and learning to manage and harness their digital skills to their curiosity.

And frankly, I was curious.

So with one of my computer science classes today, I tried it.  My methodology was … well, dubious, I suppose you could say.  I took a bunch of questions, and I used Google Translate to put them into various Western languages and non-Western languages (since I don’t speak those languages myself).

Here are the questions I asked in this way:

  1. ¿Quién fue Pitágoras y lo que hizo?
  2. Ποιος είναι Bartholemew Dias?
  3. মুম্বাই কোথায়?
  4. Qual è la seconda città più grande in Albania?
  5. Che cosa significa la teoria della relatività dichiarare?
  6. איפה הוא נוביה Zemlya?
  7. Wer entdeckte Desoxyribonukleinsäure, und wann? Was ist zu tun?
  8. มาจากการประชุม TED และสิ่งที่มีสามที่แตกต่างกันเข้าร่วมพูดคุยเกี่ยวกับ?

I gave students a few guidelines, which in reality I should have thought about more carefully.

  • Each group  gets to use only one computer
  • Each group can send people to look over the shoulders of other groups
  • Each group can use what they learn from other groups’ screens.
  • I forgot to specify that any person in any group can switch which group they’re attached to at any time,
  • I forgot to specify that any group can only have a maximum of four people at one computer.

The first unexpected result was that, between all the groups, they were able to find at least one answer close to what I had in mind for each question.   Each group was able to determine the language of origin of each of the questions, and learned how to replicate that question’s language in Google Translate, and derive an approximate answer to the question (in English, at least — we didn’t try the latter half of the Turin Experiment).

The second unexpected result was that it got treated like a contest.  Each group of students believed that their goal was to complete the eight questions before any of the other groups.  Each group believed it was possible to ‘win’ by being ‘first’ with the answers.  Each group sought to shield their results from all of the other groups, and each group became angry when other groups tried to use  their discoveries to advance their own learning.  In other words, rather than treat the effort as a playful game, the groups treated the questions and their answers as dangerously serious.

Moreover, in the guise of competition, an relatively large amount of time was lost from educational activity in favor of social activity.  The students wanted to make it a competitive activity — although I didn’t necessarily intend it to occur in that way or in that spirit.

This is a complicated awareness to come to.  There’s the old saw about how Information wants to be free, and maybe it’s so — but under the alleged rules of the game, there was a genuine effort to shield the groups’ individual results from one another as part of an overall effort to win.

From this, I think I’ve learned a couple of critical things about being a Computer Science and Informational Technology teacher:

  • The social and emotional curriculum of “Internet Citizenship” cannot be forgotten: it must be front and center.
  • The power of group learning has been bypassed in some ways in favor of individual or paired learning, at least here.
  • The access to information doesn’t necessarily mean that it will be used.
  • I have to design the human elements of my classes with as much care — or more care! — than the uses of software and hardware.

So, for a first effort at replicating this work as a classroom activity, I give myself a D — I wish I’d intervened earlier and given them greater mobility between groups to harness their group effort together.  I give the kids a B+ for their ingenuity and skill in translating the questions out of languages they didn’t understand, and finding reasonable answers to them.  But in terms of the A+ for all, I think it eluded them and me.  The structure of competitiveness between groups really had to be broken down for them to have a complete success.  And I think that means that I need to bring this activity back to the drawing board for the next time.

Because there will be a next time.

Liked it? Take a second to support Andrew on Patreon!

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.