Re: Ada book recommendations

["Genro Kane Gupta" <genro@niestu.com>]

In article <9kdsua$no$1@news.huji.ac.il>, "Ehud Lamm"
<mslamm@mscc.huji.ac.il> wrote:

> <tmoran@acm.org> wrote in message
> news:YBna7.24659$Kd7.15196504@news1.rdc1.sfba.home.com...
>> If you already know Ada pretty well, I'd add "Concurrency in Ada" 2d
>> ed, Burns and Wellings, ISBN 0 521 62911X
> 
> This is indeed essential reading if you use tasking - and important and
> interesting reading, even if you don't.

I've "used" tasking in the sense that I've compiled and understood several
examples of it (the philosophers come readily to mind), and given some
time and a sufficiently non-critical audience, I could probably churn out
a credible tasking application from scratch.  But it's certainly an area
in which I feel "soft", so the recommendation is much appreciated.

> I also second the recommendation for Cohen's book ...

As have several others.  It has moved to the top of my list.

> (I'll just add that someone told me that it is too dense. I, however,
> don't think so myself).

For me, dense is good. <g>

> What we don't realy have is a "techniques" book - and this is something
> important. I suggest reading Heaney's pattern work (avialable online),
> hand in hand with the GoF pattern book. Trying to follow and extend the
> examples in the Rationale can also be rewarding.

Ahh, design patterns.  Another area I've brushed against but haven't
explored in depth.  I have so far viewed patterns in the same sense as I
viewed the introduction of TDSP and OO--touted by many as a silver bullet,
full of useful concepts that I'll adopt myself once I get around to it,
but a bit over-hyped in general.  Am I wrong?

I want to learn more about them--can you give me specific references to
the above works, or any others you deem valuable?

> Do you have specific areas of interest?

I am coming from two angles:  first, and to me foremost, I want to become
more proficient in Ada95 for my own use.  I have several personal projects
in the design stage that I'd like to see move forward into implementation
and release, and would like them to use the language as it was intended,
not as my own prejudices and somewhat superficial exposure would force me
to.  In that area, I am interested in everything--tasking, generics,
controlled types--all the OO and advanced features this modern language
has to offer.

Secondly, I'm interested in doing an honest and worthy job on my company's
conversion effort.  We have a quarter of a million lines of
proprietary-dialect Pascal comprising our software development tools:
compiler (one of my jobs is maintaining our in-house P4-based Pascal
compiler, as well as all these other tools), assembler, linker, loader,
symbolic debugger, and a buttload of other goofy tools needed to produce
our product.

Our old platform went away, so we're moving to a new platform (which will
probably go away around the time this project is complete, if history is
any guide), and I somehow convinced management that porting it to yet
another proprietary Pascal, or to C, would be a bad decision.  So that
project probably won't use tasking, but a comfortable knowledge of the
breadth of the target language will undoubtedly make the job much easier.

Thanks for your reply, Ehud.  I'll move the Concurrency book up a notch on
my list.

-- 
Not Genro, but I play him on the net

______________________________________________________________________
Posted Via Uncensored-News.Com - Still Only $9.95 - http://www.uncensored-news.com
   With Seven Servers In California And Texas - The Worlds Uncensored News Source