Haskell, guarantees and pusillanimity
Many times in the course of my Haskell apprenticeship, I've had conversations of the following form:
Me: I want to do X in Haskell, but the obvious thing doesn't work. How do I do it?
Sensei (usually Duncan or totherme): Oh no, you can't do that! If you could do that, then you could have a transperambic Papadopolous twisted asiatic sine-curve ( Read more... )
Me: I want to do X in Haskell, but the obvious thing doesn't work. How do I do it?
Sensei (usually Duncan or totherme): Oh no, you can't do that! If you could do that, then you could have a transperambic Papadopolous twisted asiatic sine-curve ( Read more... )
Except, in all the above cases, it screws up. It turns out that you really need dynamic memory a lot of the time...
I think it's worth trying to imagine what it would mean to not screw up. Is there an underlying assumption here, that there exists a holy-grail language, which can precisely and elegantly describe everything in the universe? Is it possible that the languages mentioned work well enough in their own domains, only to require hacks and/or extensions when you try to use them for something unexpected?
TBH - most of the time, I write pure, not extended regular expressions (I use macros, like \w \W and \s - but these fall well within the theory - being equivalent to pure regexps like [a..zA..Z0..9], etc). When I want to do more complex things, I tend to use different tools...
If we resign ourselves for a moment to working with a language that comes short of being the holy grail, then perhaps the interesting question is about the shape of the domain of that language?
Of course, most "general purpose" programming languages have quite funny shapes... I certainly think that there's a huge number of problems which can be more elegantly solved in haskell (using the wonderful guarantees it gives us) than in any other common current language...
Reply
That's an interesting question! And to be honest, I don't really know the answer. I guess by "screw up" I meant "the restrictions you enforce condemn the language to an excessively narrow niche, and make things that ought to be easy hard". So yeah, you could argue (as Duncan does, in fact) that Fortran is a great domain language for very fast numeric algorithms that don't need dynamic memory, and that we should only use it as an embedded DSL in some general-purpose language. But that's not how it actually gets used, most of the time. My comments about regexps are a bit dodgy, in that the only proof I know that Perl regexps have been extended is that you can embed arbitrary Perl code into a regexp :-) Do lookahead/lookbehind assertions fit the framework? The case I really wanted to talk about is Prolog, where you spend half your time second-guessing the optimizer and the other half the time writing Yet Another meta-interpreter (from my admittedly limited experience). Or think about SQL - surely you've had that experience where, no matter how far round you twist your brain, you can't write the query you're thinking of, but you can describe it in a simple sentence, and could write it straightforwardly in a more conventional language?
I certainly think that there's a huge number of problems which can be more elegantly solved in haskell (using the wonderful guarantees it gives us) than in any other common current language...
Quite possibly - I'm still trying to decide that :-) It seems that in many cases (Laziness? Referential transparency?) "guarantees over expressivity" gives the right (or at least a good) choice - the leverage you get outweighs the restrictions. But I'm really just trying to isolate the general design principle. Because I'm not convinced that it's such a great idea as a general principle, if your goal is to design a usable language. In the limit case, of course, it leads you to prefer the guarantee of program termination to the expressiveness of Turing-equivalence :-)
Reply
Peronally, the problems I think I've had most with SQL don't stem from its lack of expressivity as from its poor integration with the rest of the programming system... Getting no compile time checking over SQL which was all in strings in a java programme really sucked. Also, dumping java data into BLOBs, which are totally inaccessible to SQL queries was annoying....
The usual complaint about SQL is that you can't search for a substring, or a regex within a value. This is annoying, it's true - I don't understand what the reason for those original restrictions were and it's probably the case by now that a database could provide this functionality and still provide the guarantees that make the database worth using - but let's pretend for a moment that they couldn't. In that case, you have to pull everything out of the DB, and do the substring check in your general purpose langauge. Would you be better off without the database? If so, why are you using a database in the first place? It seems likely to me that you're actually gaining a lot from having the DB there, probably as a result of some guarantees it can give you (about concurrency, or guarantees allowing performance optimisations or whatever) - otherwise you wouldn't be using it. If that's the case, then you're not complaining that SQL restricts what you can do (relative to your general purpose language), you're complaining that it doesn't extend what you can do far enough. If the language can be made more expressive without sacrificing those guarantees, then that's great. But if allowing a substring match meant that concurrency didn't work any more, and the database was as slow and memory hungry as your general purpose language - what would be the point?
In the limit case, of course, it leads you to prefer the guarantee of program termination to the expressiveness of Turing-equivalence :-)
If the scale we're thinking of is the one that stretches from 0 guarantees to infinite guarantees (presumably some kind of constant statement, or something?), then I don't think limit case analysis is particularly useful. If guarantees were all I cared about, I'd bury the computer in my back garden - possibly after burning all the bits that would burn. Then I'd be absolutely sure that no bad things were going to happen. If I wanted no guarantees at all, I'd programme with "cat /dev/random > a.out"
The interesting thing about language design is finding sweet spots, where you have the guarantees you want, and the expressiveness you want (and it's not necessarily always a simple trade off between the two).
Reply
I've been having a look, and can't find the specific problem that I'm thinking of - it was some feature that Mat wanted to add to Moblog (I think it was "Comments posted on (posts that you've commented on) since you last logged in") that we found we simply couldn't do in SQL. He had to set up a cron job to create an intermediate table. Possibly this was just a problem with MySQL (which at the time didn't support nested subqueries), but it was pretty lame.
Reply
IIRC one of Richard Gabriel's complaints about Common Lisp is that because the standard allowed programmers so much flexibility, it was very difficult to write an optimizing Lisp implementation--which meant that novice-to-intermediate Lisp programmer would get lousy performance until they learned the optimization tricks of their particular implementation.
So I think that in general, constraints on the language end permit optimizations on the implementation end, and when a language user is doing things that don't continually bump into those constraints, that leads to happiness on the user end.
The proper way to accommodate users that do run into those constraints, IMHO, is to provide escape hatches to other languages: SCRIPT tags, the Haskell foreign function interface, PostgreSQL "createlang", etc.
Reply
And welcome! How did you find me?
Reply
The lack of such a language is (as I believe you know), a source of continuing annoyance to me. Although you could argue that it could only exist in a universe where humans were capable of describing things precisely and elegantly in the first place. I expect it's all quantum.
Reply
- While we can't program in the language they'll use in 100 years, we can try to predict its features, and this provides a useful heuristic for choosing languages to use or designing new languages
- More powerful computers will demand languages that can provide a greater range of efficiencies: highly optimized for crypto/image rendering/etc, highly expressive but possibly highly inefficient for most stuff
- "The desire for speed is so deeply engrained in us, with our puny computers, that it will take a conscious effort to overcome it. In language design, we should be consciously seeking out situations where we can trade efficiency for even the smallest increase in convenience."
... and a bunch of other stuff, most of it interesting.Basically, he thinks that in a hundred years time, most people will be using a descendent of Lisp :-) But it's an interesting idea: if you're trying to design languages now, you should aim for the holy-grail language: even if you don't succeed, you're likely to get something better than if you'd just tried to write a slightly better Fortran.
Reply
Leave a comment