'The mess they are in has been cased by too much unscientific thought.' - E.W. Dijkstra(1930-2002)
Today i came across a interesting book when i was looking for research-related materials in the library. It's a selected writings from Dijkstra, one of the most influential computer scientist. To my surprise, he had wroted not only papers, but also more than one thousand insightful manuscripts( known collectively as 'EWDs') on computing, mathmetics, research, science, and even on life.
I really benifit a lot from reading EWD447: On the Role of Scientific Thought. Most frequently, we made mistakes because we did not pay as much attentions on those simple, yet very important principles as we should. Among those is 'the separation of the concerns', which Dijskstra emphazied a s 'the only available technique for effectively ordering one's thoughts'.
Yes, the general acceptance of our approach is a big concern, but it can be ignored temporarily 'when the significance of your messeage can justify the care you give it to its presentation'. Scalibility is another critical issue, but it only becomes important when the problem has been formulated and your model functions correctly.
Sometimes, our problem looks so complicated just because we consider two many aspects at the same time. Accuracy/Efficiency/Privacy/Implementation...oh, pls, one thing at a time!
'A scientific discipline separates a fraction of human knowledge from the rest: we have to do so, because, compared with what could be known, we have very, very small heads. It also separates a fraction of the numan abilities from the rest; again, we have to do so, because the maintenance of our non-trivial abilities requires that they be exercised daily and a day, regretfully enough, has only 24hours.'
