On Fri, Mar 15, 2002, guy keren wrote about "Re: pthreads question":
> or online tutorials. i'll offer my usual plug (what took me so long? ;)  )
> for lupg, http://www.actcom.co.il/~choo/lupg/tutorials/ , and click on the
> 'Multi-Threaded Programming With The Pthreads Library' link.

I know it's a matter of taste, but I usually find learning a non-trivial
subject (like Posix threads) taking hours (at the very least), and it's not
very convenient to sit by a terminal for hours upon hours reading from the
screen. A pile of dead-tree sheets (i.e., a book or printout) is more
convenient for me.

And if you don't happen to work/study in a place that lets you print hundreds
of pages for free (or if you work in a place where it's just as easy to order
a book as it is to print one out), the alternative is to get hold of a real
book.

> (some people
> beleife in free software. i believe in free documentation).

Anybody still remember the term "library"? It used to be a place where you
go, lend a book for a couple of weeks without paying (or paying a very small
sum), and then return it without needed to buy it. I read a few good books
in the Technion that way, and a few other good books in AT&T's library.

For some reason the great concept of a public library has been seriously
decaying in the last decade - not only it's harder to find the books you want,
but you usually don't have things like software, musics and videos in a
library. I remember taking out music CDs in a public library in NJ - how come
it doesn't exist in Israel? Why should the RIAA (or the Israeli equivalent)
get rights that book publishers never had? Why does taking out a book in a
library costs pennies and you can take it for a week or two, but taking out
a video in a video rental place costs 10 shekels and you get it for 2 days?
O tempora, o mores :(

Free documentation is nice, but I'm much less fanatic about it than about
free software. Remember that even a relatively-expensive book costs $50
(compare this to hundreds of dollars for software), and you also get some
added convenience over the free alternative (if, like me, you like to read
physical books).

> btw, the best way to learn a subject, is to write a tutorial about it, or
> prepare a _thorought_ lecture. you'd be surprised how effective these
> methods can be - provided you are a profficient programmer, with a good

Even in this case, you'll need a good source of information to base your
tutorial on. If you read a couple of manuals and didn't figure out why/when
pthread_detach() is needed, you're not likely to be able to explain it well
to the class (or you'll explain something but do it wrong or miss the
imporant things).

> theoretical background. when that's the case - the concepts are usually
> not realy new - if you did enough multi-process programming, switching
> into multi-threading is much easier then if you have never done any
> multi-process programming. (multi process is NOT client-server - that's a
> very simplistic model, which teaches only a small part of the IPC world).

This is very true. The book I recommended is specifically about Posix Threads,
and not parallel/distributed programming in general. When I read that book
I was already quite experienced in these things and all the relevant concepts
(semaphores, deadlocks, shared memory etc.) - just not using the Posix API.
But I think you can read that book without previous experience, and it will
explain everything you need to know.

-- 
Nadav Har'El                        |        Friday, Mar 15 2002, 2 Nisan 5762
[EMAIL PROTECTED]             |-----------------------------------------
Phone: +972-53-245868, ICQ 13349191 |Guarantee: this email is 100% free of
http://nadav.harel.org.il           |magnetic monopoles, or your money back!

=================================================================
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]

Reply via email to