December-4th-09. Suffix Trees - Suffix Extension Rules and Suffix Links. Haiku. The difficult theorem.
If you pay no attention to stars,
If you don't struggle hard,
An aura of strength encircled him,
I asked... He was..
Earning is joyful learning,
learning is gleeful yearning,
...turning me on!!
Difficult theorem today,
open up your mind,
let it blow!!
Go over it once more time,
try, you will make it,
Be proud of yourself. How? Learn
Learning is to open
(The treasure of yourself.)
Watch your laziness, the most
Now or never? Soon or late?
Promptly or idly?
Difficult words: Hand-waving /hændweiviŋ/, so called /səʊkɔːld/, sponge /spʌndʒ/, roughly /ˈrʌflɪ/, shortcut /shɜ:rtk.
December-7th-09. Suffix Trees - Creation of Suffix Links - The Difficult Theorem. E-VOID. I got a bit of mortification. I think pupils didn't grasp an important offspring that it was shown last class. I got a thing about making a proof too much compact and difficult to follow. I do it again to my pacification of mind and also th_irs. This offspring is crucial to grasp many parts of this algorithm. This clip it was caught profoundly. Possibly, or such it is my ridiculous simplicity, pupils did will to hit books. Possibly, pupils did study that important offspring from Friday to Monday. I know pupils will throw an inquiry (or many) about this offspring. I augur that it will occur. Just wait.
Second hour (NORMAL): This second class is about how to give talks. It is always one of my favoured lectures. I love speaking in public. I have a passion for convincing people, communicating ideas and emotions. I like very much to move among ideas and play with language, dike and springboard of our thought at the same time. First, I talked to them about the act of communication in general. The key, at least to me, is to speak always, always, always, and with no exception, about what you get excited with. Enthusiasm is of paramount importance when it comes to speaking in public. Without (sincere) enthusiasm there will never be effective communication. Then, I give some guides: say something, speak to somebody (not to yourself), repeat and do not repeat, make an outline, organize always, honesty is the best policy, look at your audience, introduce a dash of humour here and there, articulate your words, make a wise use of silence, summarize your points from time to time, and stop.
Difficult words: Thrill /θrɪl/,due /djuː/, overcome /ˌəʊvəˈkʌm/, decrease /dɪˈkriːs/, scheme /skiːm/, juggle /ˈdʒʌgəl/, show stopper /ʃəʊ ˈstɒpə/, wand /wɒnd/.
December-11th-09. Suffix Trees - Extension Algorithm/Edge-Label Compression - Talks. Today everybody will give their talks. Oh, yeah! Great acts of communication will laid in front of us! Marina Dazehair was the first one. She gave an interesting talk about her passion: engraving. She went a little too much into the technical details and less into the roots of her passion. Nevertheless, the way she talked about engraving greatly showed that passion. Next, Roger Jean Juro took the floor. He spoke about mental maps and creativity. His talk turned out to be very interesting. He had a clear idea of what creativity means and, unfortunately, how much formal educations, as is organized presently, has wasted his time... Yes, it was painfully blunt about that particular point. His talk, in spite of his problems with English, had a tight structure and presented with a sober excitement.
Then, it was Auralee Redfez's turn. This woman said something that perturbed me deeply. When I told them to pick a topic they would have a passion for, she told us she didn't have any. She said that life had been hard enough to her as for having a clear passion. After some thought, she said that kids would be her passion, especially his nephew. Her talk was about her nephew, but certainly I couldn't feel the passion. She mind went blank several times because of her English. The content was disorganized, presented in a stiff, ungraceful way. I know she is clever and very hard-working, but to me it is a mystery her apparent lack of passion (maybe, it is just latent, waiting to be fired up).
Leonine was next. He gave a funny speech about the virtues of virtual worlds. He told us how wonderful virtual worlds are, how many things you can build with a computer, how much fun you can get with it. However, he wasn't convincing at all. In fact, he didn't transmit the feeling his talk was prepared. He talked as he were in the pub with his buddies. It was entertaining, but there wasn't a true talk.
The conclusion is that giving a talk is not easy. First of all, one must be aware of the significance of such an act of communication. Secondly, it has to be prepared and it has to be rehearsed. There is some technique to be learnt. Bear this in mind.
December-14th-09. Suffix Trees - Phase Algorithm. Defying!
December-18th-09. Questions - Student's Presentations. Compassion. First casualty: Sigmund H. Noplace, pitiful creature, wasn't able to complete the last final project, the big project, the one trying out your real skills as programmers. I know this is hard. Unfortunately, you've never been taught what an algorithm actually is like; you, oh, poor students!, have never seen the proof of correctness of an algorithm; you, poor wretches, have seen very few times what a complexity analysis is; even worse, you, misfortunate victims of the damn flood of final projects raining these days, have not been accustomed to rigor and planning. I, poor me!, was really sad he, poor him!, couldn't make it. Again, I am quite sure Sigmund is clever to do this project, although under the adequate circumstances, which, somehow, they, educational system's scapegoats, don't know how to create. Because of the past projects and his class participation, he, as mortal as himself at least, has got a pass mark.
The rest of the class was in tears. Compassionately, with extreme sweetness, they made many questions to me. They all were the right questions. I, distrustful and unworthy mortal, was surprised by their general knowledge of Ukkonen's algorithm. However, I, contemptible being, had a foreboding about them completing the final project by next Monday. Humbly, I was seriously doubting it. Still, that wouldn't surprise me. Normally, I give them projects that are a little bit above his level so that I can reward effort properly. On the other hand, I have to do an enormous work to motivate them.
December-21th-09. Suffix Trees - More Student's Presentations. Today is presentation day. Three days ago they had barely started to programm suffix trees. Will they have been able to achieve it in such short period of time? The most optimistic part of myself insists on believing so. The pessimistic part of myself, or should I say the well-informed, realistic part of myself, murmurs ``they won't." I go to the classroom. None of my students are facing me and they all are crestfallen. I understand. They tell me they were unable to complete the project. They looked tired, even exhausted. I give them Christmas holidays to complete the project. I spent the rest of the class answering his questions about implementation, listening to their worries and chatting friendly. In spite of this, it is much what they have already accomplished.
Learning is a perpetual cycle of falling down and standing up (hopefully, higher). If you can find a path with no obstacles, it probably does not lead anywhere. I believe there are more urgent and important occupations than the ridiculous waste of time we call complaining.
I am looking forward to seeing what will happen January the 18th, the day they will finally do their presentations. Will they have been able to stand up?
January-18th-10. Pattern Recognition Course- Final Presentations. Today my students will make their final presentations. I can't wait to see them. They haven't asked me anything about the way to make the presentations or programming Ukkonen's algorithms. That surprised me quite a bit. It could be very bad news or really great news. I'll find out very soon.
Marina is the first speaker. Her English is awkward, unexpected, somewhat twisted, but I can see she has put some effort into it. She introduces the main definitions, but her presentation is following my notes to the letter. I can't see what she has assimilated, digested from the theory. She struggles against her English mistakes, which are more numerous as her presentation proceeds. She should have prepared her talk more. Surely, it is not well rehearsed (as I told them to do).
Then Auralee comes. Her English is awful. I cannot understand almost anything. Her grammar and pronunciation are confusing me and the presentation is quite sloppy. She is very nervous. Her mind goes blank very often during her talk. I can notice she is sweating. She finishes and walks along, head bowed.
Leonine is next. His accent is strong, but his grammar is acceptable and I can follow his ideas. He presents the core of Ukkonen's algorithm. I can tell he knows what he is talking about. He doesn't get lost in details and is able to answer my questions with aplomb. There were a couple of nasty surprises. The first was that they didn't implement one of the speed-ups as they should. It doesn't increase the time complexity drastically, but it should have been implemented correctly; it is question of principles. The second has to do with testing software. Their presentation ended without a single mention to how the program was tested. In view of my insisting questions, Leonine said that they had used 3 files to test their program (sic). I was speechless. After some thought, I asked how big those files were. Here my surprise was bigger. Leonine said that each file had 11 characters. There is no way on earth Ukkonen's algorithm can be tested with three 11-character files. Essentially, the message was that the program hadn't been tested. So sad, so disappointing. So much time and energy talking about how important is to test your program. Obviously, they lost precious points there.
That has been all. I have had very good moments with them. It has been really hard. If you are not determined, you better off not to teach a course like this. To me, in the long run it paid off.
See you next time.