An evening with the Dijkstras

I think it wise, and only honest, to warn you that my goal is immodest. It is not my purpose to "transfer knowledge" to you that, subsequently, you can forget again. My purpose is no less than to effectuate in each of you a noticeable, irreversible change. I want you to see and absorb calculational arguments so effective that you will never be able to forget that exposure. I want you to gain, for the rest of your lives, the insight that beautiful proofs are not "found" by trial and error but are the result of a consciously applied design discipline. I want to inspire you to raise your quality standards. I mean, if 10 years from now, when you are doing something quick and dirty, you suddenly visualize that I am looking over your shoulders and say to yourself "Dijkstra would not have liked this.", well, that would be enough immortality for me. Your obligation is that of active participation. You should not act as knowledge-absorbing sponges, but as whetstones on which we can all sharpen our wits. If you don't understand me, ask for clarification; if I am going too fast, slow me down. (If I am going to[o] slow, you may try to speed me up, e.g. by yawning?) Finally, don't expect me to motivate you, but during "office hours" I would like to get personally acquainted with you.
- Edsger Dijkstra in an address to his students (EWD1213)


Preface

Those who do not know much about Edsger W. Dijkstra (henceforth called EWD) will not be able to appreciate the contents of this post. So here is some information which might help you:
Dijkstra Eulogy by J Strother Moore
How can we explain Edsger W. Dijkstra to those who didn't know him? by David Gries
Remarks at the funeral of Edsger W. Dijkstra by J Strother Moore

Amongst Dijkstra's accomplishments is the 1972 ACM Turing Award (widely regarded as the Computing Science equivalent of the Nobel Prize). He was the first person to organize operating systems as a set of mutually co-operating sequential processes, a model which is used by all modern operating systems. He built the first multiprogramming operating system (The THE system), and the first implementation of a compiler for a high-level programming language (AlGOL60). He formulated many fundamental concepts in computing science like the semaphore, the notion of deadlock, the notion of mutual exclusion, the self-stabilizing systems etc. All of these spawned entire fields of research and are part of the common computing science vocabulary today. As was said in one of the above articles "In Computer Science, we are all Dijkstra's children".

I hope the foregoing will give you a flavour of the kind of person Edsger Dijkstra was. Hopefully you will be able to understand the great esteem in which I hold him. If you do, you will know why this visit and this story is so exciting for me.

The Visit

Ria Dijkstra (EWD's wife of 45 years) said she would pick Jeremy and me up from my Space box at 1430. She arrived at 1425 and was surprised that we were waiting downstairs. She was planning to come up and knock on my door. We reached the Dijkstra residence within 10 minutes. Its located in the neighbouring village of Nuenen and the drive to it was very scenic. There were farmlands and orchards on either side of the road. Mrs. Dijkstra said that it is really pretty in spring when all the trees have leaves of different colours. Right now, they are in the process of shedding their leaves. But it was very pretty nonetheless.

Those who read the manuscripts of EWD (the manuscripts are also called "EWD's") would notice that around half of them sign off with the address "Plataanstraat 5, 5671 AL Nuenen, The Netherlands". Dijkstra worked out of this house for more than 12 years while he was on the payroll of Burroughs Corporation. There was even a wooden, painted logo of the shape "B" outside the front door, which is an imitation of the Burroughs logo. EWD made it himself.

EWD's son, Rutger Dijkstra arrived just moments after we did. He works for a software company in Groningen, two and a half hours from Nuenen by car (Speaking of cars, he was driving a gorgeous black Jaguar sports coupe). According to Wim Feijen, EWD agreed that Rutger was a better mathematician than himself. This tremendous compliment, combined with the fact that he was a programmer in the software industry, meant that I was keen to know his opinion on how he used his father's methods.

It was a wonderful three storey house. Dijkstra's famous study is on the first floor. We sat in the dining room and talked for a while. Mrs. Dijkstra fed us well with coffee and two Dutch sweets she said were only available around Christmas time. I forgot their names. Afterward, we went for a walk around Nuenen. I walked side by side with Rutger while Jeremy and Mrs. Dijkstra walked behind us. It was a beautiful walk for two reasons. First, because the surrounding area was beautiful. Green fields, meadows, cattle, farm houses. All of these made for a very picturesque country scene. Second, because it was a very instructive half an hour conversation with Rutger. The details shall follow in due course, be patient!

We then sat for a long while in the dining room. Mrs. Dijkstra was always bustling in and out with food and tea. We had loads of snacks, including the very Dutch one of smoked eel on biscuits. One entire wall of the living room is lined with books from floor to ceiling. It is an impressive array. It appears that EWD liked history very much. There were numerous history books from volumes on the fall of the Roman Empire, to Dutch history, to American history (he was particularly interested in this subject). He also liked biographies. There were plenty of them. Finally his favourite novelist was Dorothy Sayers. I had never heard of her before, so I borrowed one of her novels. I also borrowed the biography of Alan Turing titled "Turing: The Enigma".

We then settled down for dinner. It was a fabulous Dutch meal of soup, boiled potatoes, boiled french beans, mushroom broth, and chicken broth. I ate a lot. I can't remember eating this much in a long while. It was accompanied with some nice wine (which Jeremy and I had taken with us). It was followed by a lavish dessert of chocolate mouse, vanilla ice cream, apple tart (which was also a contribution from Jeremy and me), and strawberries. I don't know how I managed to fit it all in. (As an aside: I must really start my jogging regime this week. Some quarters say I am already looking chubbier. But I think that has more to do with the optics of my camera.)

After this we settled down in some lovely arm chairs in the living room and listened to some music from Tom Lehrer. It is vocal, satirical music on the state of affairs in America. One song was about modern American Math education to which we all laughed heartily. Mrs. Dijkstra let me borrow the CD's and I am enjoying them as I write this.

After that Rutger said he had to leave (it was getting to around 2045). I took the opportunity to take my only picture of the day. Here are Rutger Dijkstra, Ria Dijkstra, and Jeremy in the Dijkstra living room. In the background there is a hand embroidered quilt which has a Texan theme. It was stitched for the Dijkstras by their Texan friends during their time in Austin.


After Rutger's departure, Mrs. Dijkstra took us on a tour of the house. We saw EWD's study where he wrote more than 600 of his manuscripts and also his seminal work "A Discipline of Programming" which layed the foundation for the formal derivation of programs. It was filled with books on physics, mathematics, and of course computing science. I saw books on modern languages like Haskell as well. I saw his famous Mont-Blanc Meisterstuck fountain pens with which all his handwritten manuscripts were written. He used to have six of the them, but gave many of them away toward the end of his life. It was a nice to see them.

Also in the study were the coffee mugs of the Austin Tuesday Afternoon Club. It is the sister of the ETAC and was run by EWD while he taught at UT Austin. Mrs. Dijkstra said that Jeremy and I could have one each. They have the words : "Rule 0: Don't make a mess of it" printed on them in EWD's handwriting. It was a wonderful gift. My coffee will taste better now! Here is a picture of the mug, taken once I reached home:

We saw plenty of pictures of EWD in complete cowboy attire, from the bolo tie, checked shirt, cowboy hat and glares. We saw pictures of him giving lectures, accepting awards, standing at famous landmarks etc.. Finally in the living room I saw the 1972 ACM Turing Award which, as I mentioned earlier, is the computing science equivalent of the Nobel prize. It is a simple silver bowl with an inscription on it. Mrs. Dijkstra said that the Dutch customs people wanted to charge a tax on the prize when EWD was returning to The Netherlands after receiving it in New York. He managed to avoid paying the tax. Phew!

We also saw EWD's bedroom and the bedrooms of his children. By then it was 2130 and Mrs. Dijkstra said that we could spend the night there if we wished. Jeremy and I looked at each other in a slightly stunned silence. She was extremely hospitable and motherly all through the evening, but nothing prepared us for this! We never managed to utter a word though I think Jeremy managed to shake his head. So she said she would drive us home. And that was that!

Impressions

There are many lasting impressions that I carry away from this visit. I think the most eloquent summary of the change which this visit has effected in me is thus : It drove home the point, beyond any doubt and with crystal like clarity, that the Dijkstras are perfectly normal people. I was very excited about the prospect of visiting "Dijkstra's house". I couldn't believe that I would be actually seeing a study where so much history was made. I thought I would be awed.

Nothing of the sort happened.

From the get-go it all appeared most normal. The house was a normal Dutch house. The study was a normal room. Mrs. Dijkstra was a normal old lady. Rutger Dijkstra was a normal middle-aged man. Yes, they were both very sharp, but normal nonetheless. Flesh and blood. Willing to listen to you. Most happy to talk. I was feeling "at home" within an hour of being there. It is the most comfortable as I have ever been. I was even calling EWD "Edsger" toward the end!

Mrs. Dijkstra is adorable. She is so sweet and so hospitable. She was constantly getting us things to eat but she never forced anything. She never let us clear the plates or set anything up. She insisted on doing it herself. She knew a lot about the mathematics that was discussed. She had an opinion on a lot of the topics. She related anecdotes about EWD's experiences in relation to some of the things that were being talked about.

She asked me several "motherly" questions like "Is your bed all right? Because, the night you arrived, Wim and I made it for you. But we did not remove the plastic covering on the mattress under the sheets. If it is irritating you, you can remove it". Or, "I was going to take you'll out for dinner, but then I realized that you would appreciate a home made meal, so I decided to cook for you". Or, "Are you managing your laundry properly?". Or, while we were in the car on the way to my space box, "Oh, I should have had you call your family from my house, you would have liked to talk to them". The gestures speak for themselves. She is a gem.

Rutger Dijkstra is an accomplished mathematician. His master's thesis is "One of the most beautiful works I have ever read" in Wim Feijen's words. As I mentioned earlier, we had a half an hour conversation while walking around Nuenen. I asked him about the use of formal methods in his work. He says that he does not use it directly at all, but his training enables him to be in full grip of the software he writes which cannot be said for the other programmers on his team He said that he has been employing "Extreme Programming" techniques for the last few years. When I asked him about EWD's comment that "Object Oriented methods do not live up to my standards of elegance" he replied "My father was often guilty of quickly making generalizations based on only one or two examples. There are many bad things in the object oriented methodology, but looking at them alone is dangerous". When I asked about EWD's notion of structured programming he replied "It does not scale for larger software. In my father's day, it was a huge step forward, but the world has moved on. Toward the end of his life, my father was un-willing to accept the realities of the modern software industry."

I also learned a bit about extreme programming and Rutger's explanation has inspired me to explore it further. I shared my own reasons for studying EWD's methods and how I thought they would help me as a programmer in the future. Rutger agreed that I was on the right track and that in essence EWD had the right ideas. That said, there were some places where his detractors had a point, like the examples in the preceding paragraph suggest. It was a nice talk and it served to confirm my ideas on Dijkstra's work and how it can help me. That was satisfying.

Dreams do come true

I remember that it was a year ago to the month when I said something like the following to my good friend Dhruv back in good old Mumbai: "It would be a dream come true to go to The Netherlands and work with Wim Feijen. It would be a dream to see Dijkstra's study and also his grave." To which he replied "Some dream you got, but it may happen!".

Well, sometimes dreams do come true. You just got to want them badly enough to do something about it.

I should now get back to working on "improving my quality standards". Bye!