2009-06-05 03:45 am (UTC)
Nothing. This is why people have editors and partners. They are the people who say "enough, already!", albeit for different reasons.
Incidentally, how's your dissertation coming along?
2009-06-05 03:57 am (UTC)
A bad situation to be in is when you and everyone except your advisor is saying "enough already". In that case, you have to suck it up and burn a summer making them happy with the dissertation.
2009-06-05 07:10 am (UTC)
deadline. It's always a deadline.
(Famous last words:) Then maybe I need more deadlines in my life!
It seems to me that asking for the "True Answer" on Stack Overflow, especially while invoking the "gods", is Bad Idea Jeans, because it'll especially attract those responders who consider themselves godlike and all-knowing.
For mere mortals like us, I recommend starting on PLT. That's how I started, and I now get to hack Scheme for a living. Well, sort of a living.
I wouldn't worry too much about picking The Right Language And Implementation For All Time. All languages and implementations have their problems.
What's so super bad idea jeans about it? It's not like I have to listen to them. Also, I know funtional programming, and I get paid (well) to do top-tier developing :) With Lisp's in particular, the fragmentation really retards the language growth and development, which is what the real Bad Plan is.
And I know that all implementations have their problems. I want to find one to follow that has fewer problems than others, since I have limited time and energy.
Is it that hard to see the humility and politeness of asking questions in such a way? Also, it's been fun baiting that Rayne d-bag (he's 15!) :)
(and seriously, I know languages. I've worked in R, Python, C, Haskell, Erlang, and others :) )
You think it's fun to bait 15-year-olds; I think it's fun to bait extremely intelligent men with whom I went on bike rides several years ago who suddenly seem to be acting oddly defensive about their programming ability. I guess we all gotta do what we find fun.
Edited at 2009-06-09 04:48 am (UTC)
I think that you're conflating two different kinds of work. See, to me, there is definitely a part of a song that is as you said - a living, breathing beast. To me, that is the live performance part of the song, and I really cherish and love that. That is the part of the song that comes out each time you play it, whether it's in private to yourself, in practice with your mates, or in public in front of a crowd. It's the song evolving, as you said.
But that "photograph", as you describe it - that's a different thing, and photographs of different kinds serve different purposes. Some photographs are documentary and may become out of date - it sounds like The Reasoned Schemer is somewhat like that. A classic text like SICP may be less like that, a book like David Copperfield even less so, and a religious text even less than that. It's a matter of intent, a matter of culture, and a matter of execution as to how stale these things get. Whatever it is, once the work is done, it's done and distinct from whatever thought processes it documents.
I guess my point is that in my view an author has to have a different attitude towards a work in progress versus a finished work, because they are different things. This is why we get upset when Steven Spielberg turns the guns into walky-talkies in E.T. - sure, his ideas about the work have continued to evolve since he made it, as has the culture surrounding the work, but to everyone else the ship has left the port. It's not his anymore. It's going to get outdated, it will be left behind. This applies to technical works just as it does to artistic works.
Thankfully, because of this, finishing something is a burden lifted as well. Once you hit that deadline and release your work into the world your mind will actually be free to think about something else if you want! New beasts are free to come to life and feed off of the fruit of your invention. This is a wonderful and healthy thing that is to be encouraged.
And now I have to stop writing and revising and post this message...
Shamefully, I don't remember that moment of E.T. (it's been a long, long time), so I don't get the reference.
I think that the "snapshot" and the "beast" are both kinds of unfinished work. That's part of my point, I guess -- that actual finished work is very unusual, but that unfinished things get packaged as "finished" all the time by people other than the creator, when the creator would not necessarily be ready to declare them finished, or maybe never even intended there to be a "finished" version.
It was a reference to a revision Spielberg made in E.T. - apparently, in a rerelease of the movie he edited guns out of some soldiers' hands and edited walky talkies in instead.
My instinct here is to disagree with you for some reason, because, well - it is a big relief to be able to let go of something I'm working on and work on something else, and I still think that's an important thing. But that's me letting go of it, when I decide to do that. I don't know that I would like it at all if someone made that decision for me.
Working with other people, though, I don't always get what I want....
The Reasoned Schemer, which I just finally finished reading a few days ago, and I see that even it, which is probably the least out-of-date thing on my bookshelf, has gone somewhat stale in the last seven months. Well, that's great! It's a sign of progress!
Today I finished reading The Elements of Programming Style by Kernighan and Plauger. It's from 1974! And all the examples are in FORTRAN and refer to card punching! Programming has come a looooooooong way. Talk about progress; reading that book has taught me that we programmers today have it good. (Though, of course, it could still be better.)
Strangely, the book is only slightly stale. It has a timeless, though dusty, quality about it. I wonder what it is that imparts timeless quality onto a subject matter so ephemeral as code?
Well, with TRS, I don't think I would be able to detect any staleness if I weren't actively working on the book's subject matter. Because I work on that stuff, I happen to know that the book is out of date. That's kind of my point -- that I have lots of books that don't smell stale to me, but if I were as intimately familiar with the subjects as the authors are, I'd be in a better position to judge.
Of course, some books are obviously dated, even to the casual reader, and others are quite timeless. I think the timeless quality you're talking about comes from the fact that the book is making incisive points that are not really about code, but rather about things that haven't changed much since 1974 and aren't liable to change soon, such as the nature of computation and the nature of being human.
Edited at 2009-06-06 03:35 am (UTC)