## Saturday 12 October 2013

### Plagiarism

Ripping off someone else's ideas and presenting them as your own is a Bad Thing.  At The Institute we have a very strong policy on this.  That policy, as far as this New Boy can work out, is implemented in its entirety by printing out 30,000 Plagiarism Declaration Forms and insisting that all students sign, date and stick a copy in the front of every lab book they use.  I read, comment on and mark in excess of 100 of these books every week and I've been following policy and asking again and again for these forms to appear  . . . should crack the last few by the end of next week.

In my 1st year biology classes we've been getting to use microscopes to look at cells scraped off the inside of our cheeks, which is pretty standard practice.  Last week I wrote of the sense of wonder this exercise evoked in one of my older students.  To force some quantification into this exercise, I've been getting them to estimate the size of their own cells.  If you know that the width of the field-of-view is 0.4mm and you estimate that 14 cells could stand shoulder-to-shoulder across, then the average width of a cell is 0.028mm.  The nucleus of a cheek epithelial cell is maybe 1/8th of the diameter, so its width is, say, 0.0036mm.  But these are both estimates: 15 cells across the view and the average width is 0.027, which isn't the same as 0.028 but is close enough.  Probably the best estimate of the width is a much less "accurate"  0.03mm which is the same as 30microns (which is what the text-books all say is the width of a typical animal cell - so qualified Bingo!).

Marking the scripts early yesterday morning I came across two consecutive books with no plagiarism declaration, so I chid them both "tsk! get - this - sorted" for that.  Then I noticed that they both had
Quote:
Cell diameter = 0.0286mm
Nucleus diameter = 0.0038mm
UnQuote
That is clearly plagiarism or charitably, because I know they are lab-partners, co-operative action.  The chance of getting that same 3rd significant figure on an independent calculation from an independent count is small.  Lucky they didn't have the PlagDec in their books because then I'd have had them cold and it would be off to the Gulag.

Plagiarism is a Bad Thing but there are far worse things that can happen in a laboratory and my moral compass on the matter is not squeaky clean.  So I won't be reddening their books with their sin (permanent record) but explaining what the trouble is next week.  They're only beginners.  But it did remind me of the very first teaching assignment I had when I came back to Ireland 20 years ago.  I was asked to carry a course on Practical Computer Programming for my boss (about whom I was, by coincidence, bloggin' only yesterday) and the assessment was to write a program to count the different codon triplets in DNA sequence, check that it worked, print it out, sign and submit it.  The students for that course had been in college for more than two years and been competitively accepted into the premier department in the country, so they were smart.  Rather than all independently write such a program (boring and "peripheral"), most of them embarked on one of three creative solutions.
a) subcontract the task to a a math-major friend
b) form groups and thrash out a solution together
c) copy a colleague's program
All of these were streng verboten and the smart students were fully aware of this.  So they changed the code to individualise it, but they were insufficiently experienced ("ah, bless!") to realise that there is an infinity of different ways to crack that problem and merely changing the names of the variables doesn't make it an original enterprise:
 READ A,B,C READ X,Y,Z N = A*16 + B*4 + C M = Z + 4*Y + X*16 CODON(N)=CODON(N)+1 TRIPL(M)=TRIPL(M)+1
I dutifully read through all these parallel attempts and saw that I could draw a phylogenetic tree of their relationships. I also had to gaze blankly at a lot of really hairy code that I couldn't follow and clearly had been written by a nerd under contract.  I went to see the boss and he suggested that I haul each student  into a small room for a ten minute viva on 'their' program. That worked a treat.  I'd point to a random line on the print-out and ask "What's going on here?".  The competent students would say something like "I'm iterating the central module and tallying up the count for each codon" while the passengers would stutter.  Those youngsters have all grown up now and several of them have written Nature papers and are running their own research groups.  Clearly being incompetent in FORTRAN isn't much handicap to a successful life in science and co-operative FORTRAN coding is probably a positive asset: batting ideas off each other is a key builder of creativity.

Tom Lehrer expressed it rather well:
I am never forget the day I first meet the great Lobachevsky.
In one word he told me secret of success in mathematics:
Plagiarize!
Plagiarize,