On Sun, Jul 06, 2003 at 11:21:26PM +0200, Alexander Maryanovsky wrote: > > >> >Linux is probably just a requirement so that the teacher could have a > >> >common language with the students and be able to check what the students > > > >It's also a practical requirement - we can't install whatever bizzare > >platform anyone will want, and anyway - to stress your point - I think > >one of the points in this workshop and in FS/OS in general is to be as > >portable and standards-compliant as possible. > > Point taken. Still not sure why it has to be a *requirement* though - if I > already have the bizarre *development* platform set up at home and my > software is portable (i.e. will run on Linux, so you can test it without > setting anything new up), why not allow it? >
I do not think anyone even has intentions to tell you what to do at home. Sadly, most students (both at home and here) write in VisualStudio, then port to Linux. I do not think that's the intention of at least this one workshop (and as far as I am concerned, not of any other course, but that's not up to me). > > >> Are we talking free or open source here? Also, I assume you know that > >> ~commercial != open/free :-) > >> Sorry, done nitpicking now :-) > > > >I guess you do not know Nadav very well. google a bit. > > I didn't imply Nadav doesn't know what OS or FS is. I was just nitpicking > about the use of the word "commercial" as opposed to "non-free" > (http://www.gnu.org/philosophy/words-to-avoid.html). > OK :-) > > >> That's not what I meant and I disagree - you can teach (and grade) the > >> spirit of mathematics, computer science, physics, biology etc. etc. > >> A good project in one of the above fields is one that achieves some > >> objective goals (proving a theorem, solving a problem). > >> A good project in open source is one that makes its developer happy about > >> it - you can't (objectively) grade that. > > > >If you agree with CATB (I mostly do), he says that most _good_ FS/OS > >is written because the author needed it. It was also probably fun to > >write, but no big, good FS/OS, is written _only_ for fun. Because life > >sux, and a lot of the work towards a real, long-lived, used often, > >program, is not that fun. > > By your definition of "good", which probably corresponds to the general > definition of "good software" (i.e. clean, useful, efficient and whatever > other qualifiers are usually considered good for software). But I am > currently writing a Scheme interpreter in Java - it's very cleanly and > neatly coded, but completely useless except for my own enjoyment and you > can only imagine how slow it's going to be :-) Does that make it a bad OS > project? > Sorry to say, but at least in some ways, yes. I am subscribed to freshmeat news, with the whatsnewfm filter (filters recently-seen projects away), and I can tell you there are, on-average, 1-2 interesting projects a day. For me. And I am not sure if you take all the software users in the world, there will be more than 5-10. From about 40-70 announcements per day. So: If you wrote it for yourself, without intentions to make a lot of noise with it, fine. Even if you announce it once, and get a few users, fine. But, if your project becomes "noise", that might be bad for the FS/OS world. As RMS said when he was here, 15 years ago he was afraid it would be too hard to get to a minimal working system. Today we are already far from not having enough FS - we actually have too much, in some areas. I will stress further: One of the practical points, for me, in favour of FS/OS, is efficiency. I personally used to think that if most of the software in the world was OS/FS, a lot more, and a lot better, software, could have been written. That's what was Nadav's original intention - to try and think of projects that will continue to live after the workshop. What happened in practice, is that there are tons of projects noone is interested in, and only maybe 1-2% of the entire "list of OS/FS projects" (list, as opposed to, say, work, effort, lines of code, etc.) are really used and alive. Now, I did not intend to offend you. Go ahead and continue with your scheme-in-java. I am really for it, and I am sure it will be a lot of fun, and will make you a better programmer. But if you also think about what I said, and look how many scheme interpreters (and even compilers, which are probably much harder to write) exist, and how few of them are actually used, you will get my point. > > >> >> Alexander Maryanovsky, a first year TAU student, btw :-) > >> > > >> >Maybe you should join that workshop - it looks like you already have > >> >an idea for free software! > >> > >> I'll probably fail on account of the the organizational requirements :-) > >> > > > >You can always try. Eddie doesn't eat people. If you are a good > >programmer, and have a good idea for a project, you are probably > >already more ready for the workshop than most of the students here. > > Maybe some other year, if I really need the grade ;-) No problem with me :-) You simply seem to feel strongly about the subject. -- Didi > > > Alexander Maryanovsky. > ================================================================= To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
