This blog is now closed. I have finised my MSc project and there is nothing left to say. My full report project can be downloaded here, if anyone is interested.
If you which to have a look at the SLinks prototype itself, please ask me by email.
I am now working on a PhD at the Ecole polytechnique fédérale de Lausanne with Prof. Martin Odersky. Please, have a look at my new weblog about my PhD.
Et voilà, I'm out of here, on Monday an aeroplane will fly me back across the channel to mountainous Switzerland. And I have one month of holiday before I start working again ...
Between the cleaning of my desk, last re-readings of my report etc. I discussed with Philip about whether there could be better, more formal ways to represent all theses optimisations I've worked on. For the projection optimisation we came up with a neat and nice way based on the type of variables. Re-implementing the optimisation with this new solution could be a great idea for whoever will continue to work on SLinks in the future. For the other optimisations, we wrote some basic transformation rules, but did not come up with any completely satisfactory solution. If you read on, you will find the notes we came up when thinking about these optimisations.
That's it, my report is finished. I had a little panic attack this morning when I realised that the entire section I had written about algorithm W and unification was garbage, but I rewrote it and everything is fine now. Jeremy has promised he'll proof-read it until tomorrow morning and I will re-read it too until then. But basically it is done. I even have a nice title page. The grand total of pages is 61. It is quite a pleasant feeling to have finished this report, and even well in time!
More or less everything is written now, even most of the introduction and conclusion. Tomorrow I will have to make a re-reading orgy and make sure all the layout is nice and neat. I might even finish my report in time (not that I have any choice about that anyway).