Slashdot Mirror
Edsger Wybe Dijkstra: 1930-2002
Order writes "Edsger Wybe Dijkstra, one of the founding fathers of computer science and the author of the famous "Go To Considered Harmful", has died on Aug. 6, 2002 after a long struggle with cancer."
← Back to Stories (view on slashdot.org)
After reading this article, I think we all need to pause for a minute, and consider the insight of this simple observation.
Add his definition of things human minds are geared to list: static relationships. It's perfectly in line with Dawkins statement that human minds are designed to comprehend things roughly human-sized moving at roughly human-speeds.
I keep forgetting how long people have been programming. Think about how many people using GOTO there were back in 1968. Probably only a few thousand. Crazy.
https://www.accountkiller.com/removal-requested
GOTO Heaven
I remember when Tulsa U. brought in Dr. Dijkstra back in the spring of 1984. He spoke at length about software design principles, and how design was the lynchpin of good systems. He was there only for a day, and had insisted on taking time out to talk to anyone interested in hearing him. I'm very glad that TU invited the Computer Science students from ORU over to hear him.
The Computer Science profession has lost another giant.
A pity he's gone.
Any service that uses pathfinding algorithms (such as MapQuest) should pay their respest.
Some links from my article that slashdot rejected some hours ago: the University of Texas announcement has a list of his awards and discoveries. (He taught at UT.) A brief paper (in PDF, it's scanned from a handwritten paper for CACM if I recall) shows his brilliant, clear, and concise methods of thought and writing.
If you ever used an application that made use of shortest-path searching -- say, any real-time strategy game -- then you owe this man a debt of gratitude.
You cannot apply a technological solution to a sociological problem. (Edwards' Law)
Dijkstra was very good at producing quotable remarks; in addition to his comment about computers, thought, submarines, and swimming (RTFA), he made the following remark about computer science:
"Computer science is as much about computers as astronomy is about telescopes."
Tarsnap: Online backups for the truly paranoid
I found the quotes here: http://www.cse.iitb.ac.in:8000/~rkj/dijkstraquotes .html I paste them here in full to counter the slashdot effect.
:-) -->
Some Quotes of Edsger Dijkstra
"Always design your programs as a member of a whole family of programs, including those that are likely to succeed it"
"Separate Concerns"
"A Programming Language is a tool that has profound influence on our thinking habits"
"The competent programmer is fully aware of the strictly limited size of his own skull; therefore he approaches the programming task in full humility, and among other things he avoids clever tricks like the plague" (from 1972 Turing Award Lecture)
"Progress is possible only if we train ourselves to think about programs without thinking of them as pieces of executable code"
"Program testing can best show the presence of errors but never their absence"
"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"
And then my quote
Use Adsense for Charity
I first learned that Dijkstra was ill back in February at a conference. Apparently he was sick with cancer and returned home to the Netherlands to live out his remaining days. Since that time, I periodically checked with my source, but they heard nothing new of his condition. I'm shocked to learn he lasted this long, considering what I heard back then.
I was fortunate to be introduced to Dr. Dijkstra at SIGCSE 2000 in Austin by my advisor. Its unfortunate that our field is so young that its pioneers are just now starting to pass on (compared to other sciences such as Physics, Chemistry, etc.).
...he is not going to need his forks anymore and the other guys are finally getting to eat?
seriously though, i think dijkstra will be remembered as long as there is the need to prevent race conditions... which in my eyes is quite an accomplishment.
-strangeloop
Two articles posted in a row to depress me.
-Pete
Soccer Goal Plans
It's a shame that /. seems to think "Go To Considered Harmful" is Dijkstra's signature achievement. He was profoundly influential in developing the theory of operating systems. He was one of the first proponents of layered design. He also did pioneering work in mutual exclusion (IIRC, he invented semaphores) and deadlock. In short, he is responsible for a lot of the fundamental concepts that we use to build complex systems today.
Well, I was finally able to get the page discussing the evils of 'Goto' to come up. I think I understand his gripes about it, although I'm still puzzled as to why it needed to be 'abolished'.
I can certainly understand pleading with people to make more sensible code, but I didn't strike me as being that urgent. I don't think I have a full picture of what's going on here. Could somebody enlighten me? I'm really curious.
In today's computer world, dominated more by marketing folks more than the technicians, I wonder how many people have heard of this man. It is sad that in the last decade of so, CEOs like Bill Gates and Jeff Bezos have gained so much public recognition while people Dijkstra languish in relative anonymity.
A few weeks ago, there was a post in /. about Knuth. I was surprised to see many ask who he was!
All your favorite sites in one place!
For those of us who have chosen the fields of computer science & engineering as our professions, this is a time to reflect and realize just how lucky we are.
We're getting in on the ground floor. The folks who were there in the VERY BEGINNING of our field are still around to teach us something. We need to remember just how privileged we are to have these fantastic people with us to "pass the torch" so to speak.
Look at how far the medical field has come in its history. Or chemistry. Or physics. And these are just scientific professions.
Think about other things, like teaching or agriculture.
We're the next group to advance CS/E. We've got to adopt these folks as our mentors and learn all we can from them.
Not just _how_ their stuff works, but _why_ they did it. Fundamental practices 30 years ago are as fundamental today as they were then.
"Those who fail to learn from their past tend to repeat it."
RIP, Mr. Dijkstra. And thanks for being such a great mentor.
--NBVB
I like spaghetti code. I grew up on AppleSoft Basic and GW-Basic (thank you microsoft).
I read books I picked up from the library for free which showed Basic programs threaded back and forth in sequence, for no apparent reason, and like this sentence, confusing the heck out of me. I saw it as a challenge. I also loved condition gotos'. They were evil.
Gosub? Bah. They ran out of memory too much. Because I hadn't the discipline to Return before I Goto'd out of the subroutine. So I used Goto's to simulate procedures. I also eventually used Goto's in a way that I would eventually learn is like structured programming. Set some variables, goto here, do stuff, goto back, set the same variables something else, goto here, do stuff, maybe goto back. Or it would be the end of the program.
Then I got my first C book. I still haven't got the hang of this language. Before the book even mentions "goto" it gives me a lecture on how awful goto's are and that they can produce spaghetti code. But I *like* spaghetti code. And whats with these labels? Line numbers were so much cooler. But I took the man's advice, I used functions.
But Basic spoiled me. I was never an effective programmer since. It wasn't long after I learned of structured programming that I got my first book on C++ and was introduced to object-oriented programming. Now, for someone using structured techniques for a couple years, the need for objects seemed to make sense. But I was lost in a sea of hierarchial classes and virtual methods.
When I first went on the internet, I started learning all kinds of crazy languages, hoping some of them would be simpler. And there were many. Except for forth and common lisp. Except for ML and Smalltalk. So I am still toying with scheme as I speak, still trying to figure out what exactly the difference between a recursive and iterative process is.
Eventually, I'll figure out how to write spaghetti code in this otherwise clean and elegant language too. Continuations sound promising, from what it sounds like.
I wish the best of Dijkstra--hope he rests in peace. Honestly, I've never heard of him until this post to slashdot.
But maybe it is slightly better for him not to know that some of us never learn.
Most CS (and by this I mean the academic field "computer science", not the engineering field "programming") is really just a subset of applied mathematics.
10 PRINT CHR$(205.5+RND(1)); : GOTO 10
Non-Linux Penguins ?
I assume "dike-stra", I could be wrong though. The "Wybe" part... now you've got me there.
Learn to Play Go
Two chapters from one of Dijkstra's books improved my program correctness by an order of magnitude, and this was after I had fully digested Bertrand Meyer on programming by contract. His notion of guards is the number one item on my top ten list of everything I know about writing correct code.
Just looking at his U texaspublication list is an awesome (pre-1990s meaning) experience. Let your eyes scan it, as they would the Grand Canyon. Then wander around the UTexas site, where many publications are online, and start reading. You'll be a better person for it. And you may experience a thrill of understanding, when you see that his hands hold up so much of today's code, as Shakespeare's hands hold up so much of the language and common experience of the English world.
To get a feel for the span of his life's work, consider his thesis title, "Communications with an automatic computer." The word "automatic" was necessary then, to distinguish it from a person with a calculator. The machine he used in his thesis? It had a 32K memory unit. He divided this into what he called "living" and "dead" memory.
Let's hope that his memory will be of the living variety.
To a man I never shall meet, thank you.
His writings on semaphores gave me the key thoughts on scheduling real time OS's some 30 plus years ago and I went on to create and use those RTOS's in a variety of micro and mini computers over the years on a lot of consulting jobs. Most of the time, I neglected to tell management just how I did things, and the results were some amazing systems way back then. He nudged me to realize that "Simplicity is Elegance" when it comes to software design, which is really a matter of efficient resource management. If you can understand your own designs, they might even work. He will be missed.
I am frankly not convinced that he found that nested blocks are *objectively* better than goto's. His description is not really rock-solid reasoning in that paper that I can ascertain.
Nested blocks are "better" because they are more consistent from programmer-to-programmer I have tentatively concluded.
More about my GOTO ramblings at:
http://geocities.com/tablizer/goals.htm#goto
There is yet to be a "killer proof". I heard that when that paper came out there was a lot of contraversy. Goto fans rightly claimed that it was just an opinion. Regardless, most programmers now prefer nested blocks for the most part, whether they know why or not.
I can't find any GOTO fans to interview, so their preference reasoning is unfortunately lost to history it seems.
Table-ized A.I.
I've never seen anything that could be accomplished with a goto that couldn't be accomplished by simply calling functions.
Then again, now that I think of it, if gotos are bad, recursion could be considered bad for the same reason - because it can be difficult to tell sometimes where code is being executed. Try following the code in a recursive descent parser and you'll see what I mean.
No, Thursday's out. How about never - is never good for you?
actually, yes. in c#, the following is illegal:
/* FOLLOW THROUGH */
/* FOLLOW THROUGH */
switch (value)
{
case 0:
DoSomething();
case 1:
DoSomethingElse();
break;
}
you need to do:
switch (value)
{
case 0:
DoSomething();
goto case 1;
case 1:
DoSomethingElse();
break;
}
the reasoning being that it's better (more readable) to have explicitly defined follow throughs.
This man contributed many great ideas to our field. The sad thing is how many programmers are still in ignorance of them, even now. You did great things, Mr. Dijkstra, and will be sorely missed. I just hope we're still allowed to have generic computing devices in ten years' time, so we can continue to refine and develop the revolutionary ideas you left us with.
Remember that? Only algorithm on my CS course I ever put into practical use. aka "No bracket required", (for Phil Collins fans).
-- Free software on every PC on every desk
Its unfortunate that our field is so young that its pioneers are just now starting to pass on (compared to other sciences such as Physics, Chemistry, etc.).
Yes. Computer science is indeed in its infancy. Dijkstra cleaned up algorithms by eliminating spaghetti code and introducing structured programming. In my opinion, we are still mired deep in the dark ages of computing. If only someone would clean up software engineering by eliminating the algorithm as the basis of software construction.
Do a search on Google for 'synchronous reactive systems' and find out about the next big advance in software engineering.
Project COSA
I heard reference to his algorithm (only way I heard about him) and just pegged him as another Renaissance man with too much free time on his hands (like Fourier).
Learn what he taught. Avoid GOTO. Learn about structured programming and CSP. Strive for elegance and simplicity in your programs. I can think of no better testament to his work than to show that we really were listening.
Slashdot - News for Herds. Stuff that Splatters.
find . -type f -exec grep goto {} \;
/usr/src?) or a mix of source code, binaries and grocery lists which just happen to have the string 'goto' somewhere, and then seeing how many times that happens with wc? Dijkstra would be ashamed of you.
So you're running this without any indication that the files you're grepping are really source code (are you running it on a clean
I understand that this could be construed as funny, but it strikes me as rather distasteful.
(drivers/scsi/NCR5380.c, lines 1466-1468)
We are informed when great men die. We are never informed when great men are born. Here's hoping whoever he comes back as can live up to his potential.
[o]_O
When I was just fresh out of college back in 1978, a collegue of mine who had been on Dijstras circulation list gave me a large stack of photocopied papers from Dijstra...all written in his own handwriting because he liked to invent his own symbols and found typewriters too limiting. I was working for Philips Research at the time - and I suppose Dijstra was working at Philip's "Math Center" in Eindhoven, Holland.
:-)
I've kept a whole boxful of his papers over the years - just because they are so fascinating to browse.
He invented his own programming language for expressing algorithms - but doesn't seem ever to have written a compiler for it. He refers to algorithms his mother came up with...almost every document has something interesting like that.
The notes are written in the most perfect handwriting you've ever seen.
They could have been printed - they are that precise. Then, one of them out of the blue seems to have been written in someone else's handwriting - it's just as amazingly neat though and when you get to the end of it, it says something like: Apologies for the poor handwriting in this note, but my left hand could use some practice.
These cannot be stored as text files without losing most of their historical interest. Maybe I should spend an evening or two to scan them and put them online. There could be no more fitting tribute to the man.
www.sjbaker.org
somebody in the internet advertising business like me? or are you a publisher?
"Old man yells at systemd"
The classic example is breaking out of deeply nested looping constructs. Note: This does not apply to languages that have exceptions.
General Relativity: Space-time tells matter where to go; Matter tells space-time what shape to be.
Although Edsger is remembered for the article on the
goto, his development of the stack model was an
evolutionary leap in the development of computers.
Every computer made today embodies his model.
Interrupt handling, recursion, reentrant
programming, multi-programming, multi-processing,
virtual memory all come out of Edsger's model.
I had the great fortune to work on a Burroughs B5500
and later the first B6500 that made it out of
manufacturing. This entire series of computers
was based on Edsger's model and his Algol 60 compiler.
Tony Hoare may have put it best when he quipped
"Algol is an improvement over all its successors".
Certainly Edsger was an improvement over most of
his successors.
Jim Tarvid
I took it in 1992. He was an amazing professor--as eccentric as you'd imagine him to be. I remember him once asking us if any of us knew of a good place to get his worn Birkenstocks repaired.
He was the person who first made me realize--at a visceral level--that clear thought is as important in a program as clear prose is in writing a novel.
After the final exam (conducted verbally, one-on-one, at his home), he asked me what one thing I would most take away from his class. He seemed to consider my answer to this question more important than my performance on the test itself. I told him what the above about literature and programming. He nodded, thought for a bit, and said 'Very good. Can I offer you some tea before you go?'.
I got an A, so I guess he liked my answer.
Honestly, I think the biggest crisis in software development is the "function follows form" problem. People are choosing implementation parameters before the problem is even designed!
:) These are just my personal feelings about software development today ........
;)
It's the "I don't know what it is we're going to do exactly, but I know it'll be done in Java" problem.
The second-biggest crisis in software development today is the bloat problem, IMO. The fact that hardware speeds & memory capacities are following Moore's Law is no reason for us to bloat the code so badly.
Re-writing things that used to function just fine in a new paradigm just for the sake of rewriting it is asinine!
The concept of "Webifying" everything is just silly. Whoever thinks that stateful tasks should be done with a stateless protocol (HTTP) is insane!
Anyway, enough ranting.
And you are completely correct. We need to learn not only the lessons of our mentors, but their mistakes too. Mistakes like C shouldn't ever be repeated...
Oh, did I just say that?
Sorry, I'm letting my personal feelings out again
--NBVB
It's bad to think of Dijstra only in terms of *one* memorable thing he did...but since we're doing that anyway...
...processLeafNode... ; ;
When I learned to program in the early '70s, our lecturer told us that if we used even *one* 'goto' in our work, we'd score zero for the entire assignment.
I've been programming for ~30 years now and never felt the slightest need to use one in a high level language since that day.
I strongly disagree about recursion though. Used properly, it's *very* readable:
void binaryTree::walk_binary_tree ( void )
{
if ( isALeafNode () )
else
{
leftBranch -> walk_binary_tree ()
rightBranch -> walk_binary_tree ()
}
}
Try writing that more cleanly without recursion!
(NOTE: You may be able to make it faster or more memory efficient without recursion though).
The art of reading and writing recursive programs is to try to forget that they are recursive.
I think to myself:
"In order to walk this tree, I walk the two child branches
- hmmm - I have a function to walk trees - I'll just use it
and assume it'll do what it's told."
The fact that the routine you are calling is also the routine you are currently writing just doesn't matter in most cases.
www.sjbaker.org
At the UTexas EWD archive.
Put the inner loop in a function and return if disaster.
Hmmm... somebody give me a short description of why 'goto' needed to be 'abolished'?
:-)
870 if j go to 800,900,400
Any other questions?
A dingo ate my sig...
You should really learn how to use find if you don't know what {} does. Silly xargs user!
Put the inner loop in a function and return if disaster.
And watch your code run like molasses.
Moderators: This is one of those posts where I say screw karma. Mod me to redundant hell if you wish, it just doesn't matter.
This is an extremely sad day for computer science. There is hardly a field in CS that Dijkstra's work didn't touch. His work can be seen everywhere we use computers.
Personally, this is an extremely sad day for me as well. Although I never met the man or saw him speak (now one of my greatest regrets), being in college, he's my equivalent of a Joe DiMaggio or a Ted Williams. This man was a hero and an inspiration to me.
Sometimes it really pisses me off that we show such public sorrow for sports figures who pass away like Ted Williams who for the most part didn't do a damn thing to really and truly improve our lives (granted Ted Williams was a marine and fighter pilot but that's not why most people were mourning him). This man greatly and directly contributed to a vast improvement of our quality of life as human beings. His obituary will be a foot note and page Z-42 of the NY Times and Washington Post but when celebrities die, they're front and center on page 1. It makes me sick.
That's my 2 cents and I'm not giving any damn change. >:o
If you know him for nothing else, Dijkstra's Semaphores (aka P and V operators) are a fundamental construct in pretty much any parallel processing environment. Many CPU's have Dijkstra's semephores implemented at the hardware level. It's hard to think of anyone else who has such a fundamental construct named after them. OK, maybe 'Booleans' and the long defunct "Hollerith String".
www.sjbaker.org
He wrote "goto considered harmful" in 1968
autopr0n is like, down and stuff.
This is speculation, but I would guess that he's speaking of the first couple chapters of "A Discipline of Programming".
Dijkstra's mindset is not for everyone. It's the mindset of a computer scientist who wants to have confidence in his code, confidence that does not come from ego (I am aswome, therefor my code never stinks).
I think the ugly hack has it's place. After all, breaking a window is usualy a bad way to accomplish something, but if the context is you're trapped in a burning house, it's more likely a good move. Bodgeing it out has a similar context, but 99% of the people who don't value clean, well considered code are not in that context. And of the 1% who are, I would bet many of them are in that burning house because someone before them didn't value clean and well considered code.
I'll bite.
What's wrong with the algorithm?
Predictable runtime, finite outcome.
I wholeheartedly agree.
I have always looked at the default fall-through behavior of the case in C as a design error - a complete disaster waiting to happen.
Why is it so?
Except I thought good compilers these days are able to recognize a sitation like this an inline the function?
Anyone have good profiling tools available for the really good compilers?
Maybe, but probably not. After all, the inner loop is all in one function. And with a decent compiler it will probably be in-lined and unrolled a few times anyway.
this is the slashdot crowd. The same one that gets all crazy about MS or DRM or RMS. The rabble cannot "clean up" for this important post. Be kinder.
There is an old zen saying:
Show a swordsman your sword
Show a poet your poem.
Slashdot is just slashdot.
Why is it so?
Because Original C is almost completely a block-structured language. With very rare exceptions (that tend to involve comma-delemited lists of statements), all control structures control a single statement -- the beauty of C, however, is that a block is acceptable in any case a single statement is allowed.
As this relates to the switch/case statement, it becomes a question of what exactly each component can control. The switch statement has to control a list of something, obviously, because it has to contain the cases. What the cases control is a more interesting question.
If the case controlled a statement/block, such as what happens in Pascal, then we end up with code that would look something like this:
switch foo {
....
....
case bar {
}
case baz {
}
}
With structure like that, there is no mechanism for fall through, even explicitly -- using a goto would involve gotos between a block and its sibling (instead of its ancestor), which creates spaghetti-code problems with gotos all over again. The only way that the language could implement fall-through would be through a new explicit statement to do so, and taking that route too many times leads to PL/I or APL.
Kernighan and Ritchie, the designers of C, were expressly designing the language for systems programming -- operating systems and compilers. These people want (and even need) a language that is both high-performance and flexible. To them, a "weak switch," one that allows fall-through, was potentially useful and came at neither a performance nor complexity (of the language) price.
Therefore, the optimal decision was to implement switch controlling a block of statements, with the innovative implementation of cases as labels within those statements. Programmers who used C were supposed to understand what this meant for fall-through, and although I'm sure they made the mistake of leaving out a break occasionally (just as often as Us Normal People leave out semicolons or the like), the error it caused wouldn't be impossible to find or fix.
"Evil company X is threatening to restrict our rights! Let's all get together to stop--OOOH! SHINEY!!!" -- AC
I had Dijkstra for a graduate CS class in the Fall of 1996. It was an exploration of elegance in the process of quantitative reasoning. I must say that he taught me the virtue of careful thinking more so than any other instructor during my formal education. Check out this link starting around manuscript 1237 to see the course notes. As an example, he showed us an algorithm for calculating increasing cubes (x^3 for x=1 to N) of integers that reduces to 2 C statements and uses only integer addition and initial assignment as operators. E-mail me if you want the code. Hint: It would only be a 2 statement algorithm for any arbitrary polynomial function.
k u r t AT s p a c e s h i p . c o m
Dijkstra's Law (of Programming Inertia):
If you don't know what your program is supposed to do, you'd better not start writing it.
I'm very sorry to hear he's died, even though I never met the good Doctor. In fact, each time I'm led off into the weeds by some dumbass project manager who misinterprets XP or RAD or ??? into contradicting this law, I quote Dijkstra's Law to anyone nearby. Along with quoting from Yourdon's "Death March", it's my favorite self-help therapy method.
Let's not forget this bit of fun. We can banish goto forever now that someone finally invented
comefrom.
dike-stra is much like it, not totally, but the 'ij' sound is very typical for the dutch language, so there is no similar sound in English.
:)
'Wybe' is pronounced as 'Wee-buh' where the 'wee' part is a short sound, not a long weeeeeee
Never underestimate the relief of true separation of Religion and State.
Those who actually read the linux kernel source codem probably already knew Dijkstra and his god-like powers in the computer-sciences.
But for those who put their nose in there and juts read the comments, there are some references
Fr example: drivers/scsi/NCR5380.c
echo '[q]sa[ln0=aln80~Psnlbx]16isb572CCB9AE9DB03273snlbxq' |dc
It is sad that in the last decade of so, CEOs like Bill Gates and Jeff Bezos have gained so much public recognition while people Dijkstra languish in relative anonymity.
and in his time, whom do you think was more famous, Newton or his King/Queen ? Lagrange or whatever Louie ruled then ?
True metal survives the acid test of time. The ornamentations, the hype-sellers, the gates'es and Bezos'es, will be forgotten by everyone (except historians) by the next century.
Dijkstra's shortest-path algorithm and other works will be remembered in centuries to come.
Working for necessity's mother.
C is a perfectly good cross-platform assembly language.
C++ is an inconsistant, bloated, pseudo high-level, partially object-oriented, Frankenstein somewhat C-compatible language.
My other first post is car post.
Hmm... This might be nice, The problem is that Mr Dijkstra's name comes from the northern part of the netherlands. This 'state' or province as we call it has an own language, with different pronouncement. My go at it:
Normal Dutch:
D'ii'kstra: the ij is pronounced a bit strange, it is hard to explain......
Fries:(northern language/accent)
D'ee'kstroa : the ee is pronounced as in english the seperate e but longer.
Make a man a fire and he will be warm for a day, set a man on fire and he will be warm for the rest of his life
I hope I had moderation points for this !
:)
So true
I believe the Internet core routing protocols use Dijkstra's shortest path algorithm, whereas RTS games probably use the A* algorithm to find approximate shortest paths. So everyone who accesses slashdot remotely uses his algorithm... :) IIRC, Dijkstra also developed semaphores and mutexes, according to our old friend Andy Tannenbaum, which are an absolute requirement for any multitasking, multithreaded OS. Gosh, the man was a legend...
Qu'on me donne six lignes écrites de la main du plus honnête homme, j'y trouverai de quoi le faire pendre.
That's why open source is such great thing, everybody want to improve your code ;)
It is practically impossible to teach good programming style to students that have had prior exposure to BASIC; as potential programmers they are mentally mutilated beyond hope of regeneration.
I've always had great respect for the man, but after reading that quote it has increased.
Also to note: More than half the college books I've had to read have some reference to Dijkstra in them. He is one of the greatest contributers to computer science, not only in the theoretical level, but on the practical design level as well.
The Moore-Murphy Law: The number of things that will go wrong will double every 2 years.
I had a friend working for the computer center at UT. She emails me that she just done such-and-such a thing for Professor Dijkstra. Upon reading this, I of course send some reply about all the stuff he has done, and how it must be cool to have met him, etc, etc, and she replies, "I didn't know he was famous. I just thought he was a nice old man!"
So, here's to a great computer scientist and a "nice old man". May he rest in peace.
I took a course he taught at UT Austin. The final exam was oral, one-on-one, for about two hours or so. I walked into his office at the end of the hall, which was larger than the classroom was. Actually, I don't walk into the office; I walk into the waiting room leading to the office.
So there's this huge room, with two walls covered in bookshelves, filled with books, periodicals, publications, a picture of Dijkstra in his graduation robes, awards, etc., all neatly arranged. I get the feeling the Doctor has written half of what's shelved there. (Knuth wrote half of the rest, I reckon.)
Dr. Dijkstra sits me down, and after a quick chat, launches into the first problem. It's a proof, fairly simple. After presenting the problem, he sits down in the chair across from me, and waits, quietly and patiently. On me.
I got so flustered I ended up with a B. One of my great regrets.
Lately democracy seems to be based on the skybox, the Happy Meal box, the X-box, and the idiot box.
Other people missing from the list include Whitt Diffie and Ron Rivest. I am suprised to see Eric Raymond but not Linus Torvald or any of the Apache people.
Given some of the names that are on the list I don't see any reason to complain about TBL. If Tim was lucky then so was Denis Richie, C was after all merely an incremental development of CPL, BCPL and B. It is very strange to have Ritchie on the list and not Hoare whose work on Algol came long before.
More interesting however than a pioneer's list is a contemporary list. I would much rather be on a list of people currently at the forefront of research than a has been's list.
Looking for an Information Security student project suggestion?
Try http://dotcrimeManifesto.com/
For an older view, try:
A Discipline of Programming
for a more refined, later view, read:
Predicate Calculus and Program Semantics
Many people are showing their lack of knowledge by endlessly citing the GOTO thing--which was a letter, not even a paper. But Dijkstra's main work was in what he called the calculation of programs: developing provably correct programs. Yes, it isn't everyone's cup of tea, and no it isn't completely realistic for most purposes, but it is some brilliant, pioneering work.
There are three classes of programmers out there:
First, those who say, "Dijkstra was a genius; if only every programmer followed his dictates."
Second, those who say, "Dijkstra was a complete iconoclast; he had a lot of good ideas, but his approach wasn't practical in the real world."
Third, those who say, "Dijkstra who?"
Sadly, there are a lot of third class programmers out there.
Stupid job ads, weird spam, occasional insight at
The Dijkstra Algorythm is the mechanism that determines path selection in an OSPF routing domain.
RIP determines best path based solely on hop count, irrespective of the bandwidth of the links along the path.
OSPF determines path without considering hop count. Instead, it utilizes metrics derived from the bandwidth of the links between the source and destination as the determining factor in deciding the best path.
I'd rather send my packets across two DS-3 connections than one 9.6kbps connection in reaching the destination.
Couple that with it's support for VLSM and Classless operation, and it's super efficient update "flooding" mechanism, and you have the basis for the Mac Daddy of interior routing protocols.
This was a brilliant man. I'm sorry to hear of his passing.
For those that would die defending it, Freedom
has a sweet taste that the protected will never know.
Cute. He died of cancer, so you get to kludge in the same metaphor to plug your megalomaniac, ah, "theory" (when I can find evidence of sound CS or even engineering principles on that page, I'll remove the quotes).
You should be ashamed of yourself.
I've finally had it: until slashdot gets article moderation, I am not coming back.
I received my computer science degree from the University of Texas, where Dr. Dijkstra taught before retiring. I never took the undergraduate class he offered (I was kind of intimidated at the time), but the professor who taught my Software Engineering class had him come in to lecture one day.
This software engineering class was very pragmatic, emphasizing methodical design, implentation, and testing. As I recall, Dr. Dijkstra gave his lecture near the end of our semester, by which time we had been heavily involved in something resembling a team development evironment for a few months.. There was a very corporate feeling to our regimen of meetings and reports.
So one day we all go to the faculty lounge to hear the esteemed professor speak. He comes in the door of the lounge appearing to me most unlike the kind of man who could write so forcefully about programming, dressed in shorts and a tee-shirt with a distinctly old-grandfather look on his face.
In his very soft-spoken manner, he told us that he beleived that the main problem with programmers was a lack of rigor. People were so concerned with coding and testing that they never learned how to write something correctly the first time. He asked us to prove the correctness of the code for a binary search and spent the next half-hour proceeding glumly as we slowly worked through the process with him.
I got the impression we were a vaguely dissapointing group of students who he could tell were not convinced of the validity of his approach. It wasn't even a bitter dissapointment, though. I felt as though he was someone who had totally convinced himself that he knew how to make the world a better place, but that noone was listening.
He answered our questions about "gotos considered harmful" (it was his editor's idea to give it the cute title) with what I considered obvious patience. He talked about how he really only was able to keep up on the research that people referred to him these days. And then the lecture was over.
Our professor and Dr. Dijkstra were good friends, and I hung around after class talking with them about computer science and Dijksta's past. I ended up in his office after a while and we chatted about the current state of the industry as he saw it, why he really liked Texas, and so on. He was so intelligent in his conversation--asked so many probling questions--that by the time I was done I felt both touched and exhausted. He put on his cowboy hat and walked out of the office with me and headed off to his next appointment.
That was the last time I saw Esdgar Dijkstra--the only real time I ever talked to him. But I feel that the world has lost a quiet crusader, and I feel a tug in my heart thinking about this old dean of computer science with his cowboy hat.
-- Have you ever imagined a world with no hypothetical situations?
It is practically impossible to teach good programming style to students that have had prior exposure to BASIC; as potential programmers they are mentally mutilated beyond hope of regeneration.
APL is a mistake, carried through to perfection. It is the language of the future for the programming techniques of the past: it creates a new generation of coding bums.
The use of COBOL cripples the mind; its teaching should, therefore, be regarded as a criminal offence.
When FORTRAN has been called an infantile disorder, PL/I, with its growth characteristics of a dangerous tumor, could turn out to be a fatal disease.
COBOL is for morons.
With respect to COBOL you can really do only one of two things: fight the disease or pretend that it does not exist.
The question of whether computers can think is like the question of whether submarines can swim.
Stupid job ads, weird spam, occasional insight at
It just happens to be able to find all shortest paths in the same amount of time it takes to find one, hence the running out of memory comment.
I hate to respond to a troll, but I'll bite.
Sanity is a sandbox. I prefer the swings.
This makes sense, but I don't understand why they didn't make the default behavior equivalent the break, and provide a "fallthrough" keyword or something similar that does the same as the default requirements.
What I mean is, why did they make the default the way they did? It seems like an error to do so, and there is no technical reason it couldn't be done the other way.
He was the Euclid of our times. Reading his writings I get the same feeling of economy, conciseness, clarity, effectivity that I do when reading the Elements. His calculational style of proof is, to me, entirely on par with Euclid's use of the Theory of Proporions.
I hope it won't have to take two thousand years before we recognize his fundamental brilliance.
I'll admit -- what you describe is very technically possible. However, implementing this would make the language more complex than necessary.
Firstly, implementing something like this would put an 'implicit break' at the end of each case-list of statements. Doing this would crack the program sequence of C-code, in that (unless specificially altered by the deliberate use of a control statement by the programmer) the lexically next statement is always the chronologically next statement. Admittedly, this specific case wouldn't have _too_ much of an impact on code writing, but for the language in general the statement 'program instructions are executed in order' becomes 'program instructions are executed in order EXCEPT when it's part of a switch statement.' It's inelegant, and doing something like this to solve this particular problem could lead to similar stopgap solutions for other preceived problems, and before long we may as well reintroduce the unlimited goto.
Secondly, introducing a specific keyword for this function means that all C programmers would have to learn YAK (Yet Another Keyword). No existing keyword serves a similar purpose as fallthough (continue being reserved only for loops, and 'switch' is most definitely not a looping construct), and the addition of any new keyword in a language should come only after much debate -- with too many keywords, programmers learn only a subset of the language, and you get 'dialects' of the language based on coding style.
In my opinion, modern compilers should catch this kind of thing (when possible) and, on suitably detailed warning levels (such as -Wall or -Wstudent [yes this one's made up]) should emit a warning. Although a prime source of small errors in code, constructs like these (and assignment-in-ifs) have their uses. Expressly dealing with them makes the language more complex than it needs to be, IMO, and forbidding them takes power from the programmer -- better to recognize and warn (if told to do so), but otherwise compile just fine.
"Evil company X is threatening to restrict our rights! Let's all get together to stop--OOOH! SHINEY!!!" -- AC
RIP.
As he also said "OOP is an exceptionally bad idea which could only have originated in California," you becoming proficient in OOP after starting with BASIC is unlikely to convince him that his assessment of people who start with BASIC was wrong. =]
10 PRINT CHR$(205.5+RND(1)); : GOTO 10
"You are old, Father Edsger," the young man said,
"All your papers these days look the same;
Those EWD's would be better unread --
Do these facts never fill you with shame?"
"In my youth," Father Edsger replied to his son,
"I wrote wonderful papers galore;
But the great reputation I found that I'd won,
Made it pointless to think any more."
"You are old," said the youth, "as I mentioned before,
And make errors few people could bear;
You complain about everyone's English but yours --
Do you really think this is quite fair?"
"I make lots of mistakes," Father Edsger declared,
"But my stature these days is so great
That no critic can hurt me -- I've got them all scared,
And to stop me it's now far too late."
"You are old," said the youth, "and your programs don't run,
And there isn't one language you like;
Yet of useful suggestions for help you have none --
Have you thought about taking a hike?"
"Since I never write programs," his father replied,
"Every language looks equally bad;
Yet the people keep paying to read all my books
And don't realize that they've been had."
"You are old," said the youth, "and I'm told by my peers
That your lectures bore people to death.
Yet you talk at one hundred conventions per year --
Don't you think that you should save your breath?"
"I have answered three questions, and that is enough,"
Said his father. "Don't give yourself airs!
Do you think I can listen all day to such stuff?
Be off, or I'll kick you down stairs!"
"I'm a ChickenHawk, and I only eat Chickens!" heh.
The classic example is breaking out of deeply nested looping constructs. Note: This does not apply to languages that have exceptions.
I can't beat Knuth's paper; I have to refer you to it for a complete discussion of why and where gotos are useful.
I just wanted to say that exceptions do not, can not, and MUST not be used as a tool to avoid GOTOs. Exceptions are dynamic (runtime) events; GOTOs are static, constant effects. An uncaught exception results in a crash; an uncaught GOTO results in uncompilable code.
-Billy
I'd support renaming the Ted Williams Tunnel here in Boston to the Edsger Dijkstra Tunnel but then it would never let me GOTO anywhere!
*load groan*
It is by the juice of the coffee bean that thoughts acquire speed, the teeth acquire stains. The stains become a warning