I have been visiting the Helsinki Ateneum museum last weekend. It is wonderful to be able to spot places where our MOLTO work in cultural heritage could be used. One issue is as usual the spelling mistakes. I read only one or two of these paragraphs written on walls, and in one of them, right in the middle, there is a spelling mistake. One thing is to find a spelling mistake in a paper, but on a wall, for every visitor to see?

Public amend, namely, I just fetched 133 emails stuck in my account at Chalmers: those of you who answered my emails in the past week, know that I am only reading it now. I should have remembered to change the password also in the mail reader, that is. In any case, I am working on the backlog now.

This week is one of those in which I seem to spend my time in chasing deliverables and other data that has to go into the progress report. It is not that easy to keep all threads in place and coordinated, one feels like one sends the email and the recipient simply disregards it - just as we do when our conscience calls.

Recently I did a major improvement in the efficiency of the robust GF parser and now I am eager to try it out on the patent corpus. The first thing to notice is that although the parser is now usable even for nontrivial sentences, for some long sentences it still fails with "out of memory" error. The borderline is somewhere around sentence length of 20 tokens. There is still room for improvement and I know what must be done but before that I want to do a pilot experiment in translation.