Slashdot Mirror
How Heraclitus would Design a Programming Language
CowboyRobot writes "Developer of Smalltalk Alan Kay has an interview on ACM Queue where he describes the history of computing and his approach to designing languages. Kay has an impressive resume (PARC, ARPAnet, Atari, Apple, Alan Turing Award winner) and has an endless supply of memorable quotes: 'Perl is another example of filling a tiny, short-term need, and then being a real problem in the longer term,' 'Once you have something that grows faster than education grows, you're always going to get a pop culture,' 'most undergraduate degrees in computer science these days are basically Java vocational training,' 'All creativity is an extended form of a joke,' and 'nobody really knows how to design a good language.'"
Notwithstanding the grammar mistake--I hope editors will have corrected it before I finish writing this comment--it is not "How [Would] Heraclitus ... Design a Programming Language" but rather "What Programming Language Would Heraclitus Design." On the other hand if you are wondering how should you desing and implement a programming language, I would suggest targetting Parrot which makes implementing compilers 1000 times easier than ever before, not to even mention future interoperability and e.g. access to the entire CPAN from the level of your own brand new language, effectively solving the most important problem of new languages: there are no libraries so people don't write anything, and people don't write anything so there are no libraries. Good luck.
Sincerely,
Pan Tarhei Hosé, PhD.
"Homo sum et cogito ergo odi profanum vulgus et libido."
Perl is a very powerful language to write small tools in the UNIX philosophy. It works very well. It requires some level of competence in the user, but so does every advanced tool. And it is not a general-purpose language.
Frankly, without Perl my work would be far harder.
I also like Eiffel. Guess I am one of those scary people that actually use different tools for different tasks and do not spam the world with simplistic, general and irrelevant statements. about
Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
It was in Dobbs' Lightning that I first read Tux Sneezed, which I still think is a rip-roaring good yarn. The scene where Atlanta Hope sees Niklaus Wirth and it's her old *ahem* "boyfriend" with the gaunt cheeks, and he said "I am Bob Dobbs", man, that's writing. The 103-page long speech afterwards, explaining the importance of strong typing and showing why all the anti-Heracletians are destroying civilization by destroying strong typing, certainly is persuasive, especially to me who's got three (going on four) contracts, each of which share the same include files. "Without strong typing there can be no civilization."
Her nonfiction book, "Antitrust: The Unknown Ideal for the New Heracletian" is, I think, a distinct letdown, but the Dobbs' Lightning bumper stickers sure give people the creeps.
I met Atlanta Hope at the time of the IEEE Committee Riots. I was in the thick of it (you have no idea how bizarre civil war gets when one side uses nerf weapons as a large part of its arsenal), and met Atlanta herself where the last stand was being made. She grabbed my right arm and howled something like "War is the Health of the State! Conflict is the creator of all things!"
Seeing as how she was on a heavy Heraclitus wavelength, I quoted with great passion, "Men should fight for the Laws as they would for the walls of the city!" That won her, and I was Atlanta's personal lieutenant for the rest of the battle.
(Epilogue: Heraclitus -- He was apt to say odd things. Once he even wrote that "Religious ceremonies are unholy." A strange duck.)
I'd disagree that there aren't people who can design decent languages. The problem is that they can't market them,
No, the problem is that the people who know a lot about languages know little about application domains, and the people who know a lot about application domains know little about how to design languages (or at least don't spend much time on it).
That's why languages like MATLAB dominate scientific computing and languages like Perl, PHP, and Java dominate web computing, and why languages like CAML, Haskell, Lisp, and Smalltalk have never ended up being good general purpose languages.
The problem isn't language designers its us developers, we don't want to spend a week learning a new syntax for a loop, we want to use what we used before. In other words we are luddites.
Programmers contribute to the problem. But while many people have syntactic hangups, even more of them just "don't get" a different approach to programming at all.
I'm not a luddite, I just don't feel like typing an essay to make the computer do something simple.
... until ... loop ... end? Why do I need all that just to specify a loop? Why is for( init ; condition ; increment ){ } harder?
I used Eiffel in college, and didn't it find it to be all that great. To make your compiled program run at a decent speed, you had to 'freeze' it (melting IIRC was the quick & dirty no opimization compile) which would take hours. Hours if find if you are compling a huge project (say a kernel) but for a simple graphical poker game..ugh.
I know terminology is probably something that held it back...freeze, pebbles, etc.
It was wordy and too strict style-wise. Why the hell should i have to type a : when doing an assignment? From
Having to type all that extra stuff does add up, and while typing isn't the largest task, when it does finally come time to code, I don't wnat to spend my day writing extra, unnessary crap to make the program work.
Please explain why C# or Java is rubbish compared to Eiffel? I'm all for learning a new syntax if it buys you something new...but Eiffel's syntax doesn't, I just takes more typing and requires you to remember even more keywords.
You don't understand what his criticism for Perl is all about. You'll understand it once you try to implement Perl in Perl (as an exercise you could start to implement the Perl RegExp in Perl without using Perl RegExp).
(Non-ANSI, but original) Lisp is a language that is implemented in itself in about half a letter page.
Even Smalltalk can be implemented in Smalltalk in a quite compact manner.
Perl needs a quite large compiler to turn Perl code into the runtime code. And every new feature in Perl is done in a way, that the compiler has to be changed and a formerly syntactically wrong line gets a semantic put on.
Perl is fine for what it was thought out for: To crunch large amounts of similarily structured data into human readable reports. That's what most CGIs or admin scripts are about: filtering database entries or long logfiles and converting the results into webpages or other human accessible documents.
(Or getting huge amounts of data and turning them into log file entries and data base lines.)
It is still a Practical Extraction and Report Language. Alan Kay didn't dispute it. But it gets most of it's power as a language from what Alan Kay has called "an egyptian pyramid", a huge number of bricks and buildingblocks put together by brute force and thousands of slaves, but not from its inherent design qualities.
But that's not a "good" program, that's a bad one. A good program would teach algorithms, design, and team working and at least three languages, one of which should be Lisp, just because it lets them know there are other paradigms out there. That will teach flexibility and allow students to cope with legacy and oddball systems. Learning Java is simply not good enough. Indeed, I would say that learning C and Smalltalk would give anyone the tools to walk into a Java post with a week's notice, and many other posts too.
Get them all doing assembler, that what I say. Assemble some sense into 'em!
TWW
"Encyclopedia" is to "Wikipedia" what "Library" is to "Some people at a bus stop"
Well, I was trying to be funny, but obviously it didn't come off too well. But shush, don't tell anyone: I really love the ubercomplex discussions going on.