Stupid Old Pencils

As Bertie Ahern famously described our manual method of voting in Ireland.

It turned out our manual method was far superior to a proposed e-voting system. This came to mind again last week as I saw students in some pain from the effort of three hours of non-stop handwriting. Clearly exam time is the only time that prolonged handwriting is required anymore.

e-voting and e-learning would, at first glance, appear to have little in common.

At first glance both appear to be “easy” activities to revolutionise using technology. Computers can count (extremely quickly and accurately), hence e-voting should be simple - no need for hundreds of people manually counting votes. Modern methods of communications mean that the very best educational material can be widely disseminated with ease via the internet - no need for the classroom!

And that's what they have in common - in fact they are not easy problems to solve at all! On the face of it technology should be very relevant to these processes but in fact its very frustrating to get it to work effectively. E-voting involves all sorts of subtleties around security, and preventing things like vote buying, and coercion, not to mention the integrity of the poll and trust in the result. With e-learning the problems are mostly around accreditation. In both cases it proves very hard to come up an alternative to our “stupid old pencils” for voting, and old fashioned in-class lecturing and the handwritten exam for measuring learning.

Consider for a moment the problem of setting and marking an exam over the internet. It doesn't take long to see what the problems are. How do you prevent collusion and cheating? How do you know who is actually doing the exam? How can you prevent candidates communicating?

Getting the worlds best teachers to record lectures which anyone can access is the easy part! But the economies of scale which should therefore arise are pretty pointless if each student still needs to be individually tutored, assessed and examined.

We still vote by making a mark on a piece of paper. We still mark exam papers by red-penning hand-written scripts. Both activities are in fact stubbornly resistant to technology.

So are we to conclude that there are activities for which technology in just naturally inappropriate?
No. These problems are probably not insuperable. Cryptographers have already come up with very cunning techniques to support e-voting, even internet voting. For example see
http://eprint.iacr.org/2011/568

Perhaps in time some-one with come up with a high-tech solution for e-learning/e-accreditation. But we are not there yet.

DIY University Ranking

Rank your own!

Worried about the methodology (and agenda) of existing university ranking organisations?

Consider this institutional citation count. Note that a variant of this proposed measure is also used, and heavily weighted, by those other ranking organisations. Indeed it is probably the biggest, and least subjective component in the mix. It is based on the number of citations to research papers written by staff members.

Its easy – go to this Web page
http://scholar.google.com/citations?hl=en&view_op=search_authors&mauthors

(I know you get an error message) and input the name of a university in the box at the top and click on “Search Authors”. Then, starting from the top count down (1,2,3..n) through the staff until your count n is greater than the citation count for the next staff member. Lets call this the institutional or i-index. For example type “Dublin Institute of Technology” without the quotes, click on Search Authors, count down through the list, not forgetting to click Next to move to the next page. You get a score of 13. Not bad! (Note all these counts were observed on January 10^th, and are liable to change in the meantime)

Now let’s try and calibrate the method using some world famous universities. In major ranking exercises Harvard and Stanford often comes out ranked number 1 or 2, and Oxford typically 4 or 5. According to the proposed method Harvard scores 233, Stanford scores 159 and Oxford 126. Looks about right.

For Irish Universities (and some prominent ITs) the ranking is (drum roll please!)

1. DCU – 40
2. UCD – 38
3. UCG – 34
4. TCD – 26
5. QUB – 26
6. UCC – 22
7. Maynooth – 15
8. DIT – 13
9. Ulster– 11
10. Limerick- 11
11. Waterford IT – 5

Not quite what you expected? Well you can hardly complain that Google’s algorithms have an axe to grind when it comes to Irish Universities…

Lots of interesting observations can be made. Obviously DCU comes out on top – so no surprise there . Drilling down a little deeper, and from a purely parochial point of view, observe that Computing is still far and away the dominant research school within DCU.

Colleagues from other universities (and departments!) might complain that their staff have not “signed up” with Google Scholar. My response is – if you choose to hide your light under a bushel, who is to blame? You or the bushel?

But of course I accept that as more academics sign up with Google Scholar, the more accurate the measure will become. I have observed that academic applicants for jobs now often attach their Google Scholar citation profile, so its use is rapidly becoming more widespread. Certainly if I am to interview an academic for a research post, it’s the first place I look, and I know that internationally colleagues are doing the same.

Lets fall in Love

Everyone is doing it (with apologies to Cole Porter)

Irish kids are doing it..

http://coderdojo.com/

Regular guys are doing it..

http://codeyear.com/

Girls are doing it…

http://girldevelopit.com/

The Mayor of New York is doing it...

http://www.bbc.co.uk/news/technology-16440126

They are all coding! Also known as Computer Programming.

It would be great to be able to say that Irish Schools were doing it (by and large they aren’t ). But the pressure for coding to become part of the secondary curriculum is building up all over the developed world, and hopefully we will not be too far behind.

Of course I am old enough to have seen all this before. For me coding is still a hobby that I love.

The early 80’s was the era of the coding hobbyist magazine, with titles like Personal Computer World (PCW), Practical Computing (UK) and Dr. Dobbs Journal (US). PCW closed in 2009 having long morphed into a rather dull trade magazine, Practical Computing closed in the 1990’s, and even Dr. Dobbs disappeared as a print magazine, but lingered on as a Website. However interestingly in March last year Dr. Dobbs commenced a resurrected PDF version of the magazine. In fact it would seem like the perfect time for some entrepreneurial type to launch a new hobbyist magazine to ride the current wave of popularity.

In the 80’s inexpensive home computers first became available, and since they didn’t do much of anything unless you programmed them yourself, learning to program was an absolute requirement. I still recall my pleasure in having an article accepted for publication in Practical Computing, featuring a program to solve alphametic puzzles.

Then in the late 1990s personal computers became things you played games on rather than programmed yourself, and the hobbyist scene died away. But now its back with a vengeance, and our discipline will certainly benefit from the renewed surge in interest. Exposure to coding at a young age brings huge benefits to all.

So maybe its time to fall in love with coding all over again.

And the bottom line is - coding is a skill that gets you a well paid job.