 |
C++Talk.NET
C++ language newsgroups
|
| View previous topic :: View next topic |
| Author |
Message |
Michael Jørgensen
Guest
|
 Posted: Thu Sep 25, 2003 1:35 pm Post subject: Necessity in comp.lang.c++.portability group? |
 |
|
Hi friends,
Writing portable code gets increasingly important. Most developers - myself
including - are familiar with only one platform. I've tried a few times
porting some of my software to different platforms, only to be faced with
great problems.
Instead of having to deal with porting problems on a case-by-case basis, I
seek advice on how to write code that is portable from the start.
This requires knowledge of several different platforms. Additionally, it is
nice to know in which direction the various platforms are heading, e.g. is
POSIX becoming a de-facto standard? Which platforms and/or compilers are
known to be seriously non-conforming? To what extent is it necessary to
cater for these systems? Other questions that could be asked are: "Which
header files should I include if I want a timer with resolution of one
second?". "What about 1 milli-, 1 micro-, or 1 nano-second?" There should be
(if not already) a "Right Way" to achieve this in a portable manner.
On-topic for such a newsgroup would also be questions about installation of
the created software, e.g. Windows Installers versus RedHat RPM. What
information do these install utilities require, besides the compiled binary
and/or source code itself?
So I seek a newsgroup that specifically addresses portability issues.
Instead of going to several different platform-specific newsgroups and
asking the same question (and getting completely unrelated answers), I want
to concentrate these specific issues and ideas in a single newsgroup. The
idea is to "Do It Right The First Time", at least to the greatest extent
possible.
What are your comments?
-Michael.
[ See http://www.gotw.ca/resources/clcm.htm for info about ]
[ comp.lang.c++.moderated. First time posters: Do this! ]
|
|
| Back to top |
|
 |
Lorenzo Bettini
Guest
|
| Posted: Thu Sep 25, 2003 8:44 pm Post subject: Re: Necessity in comp.lang.c++.portability group? |
|
|
I think it would be GREAT! Especially for the open source movement,
where portability is MUST :-)
Lorenzo
--
+-----------------------------------------------------+
[ See http://www.gotw.ca/resources/clcm.htm for info about ]
[ comp.lang.c++.moderated. First time posters: Do this! ]
|
|
| Back to top |
|
|
WW
Guest
|
| Posted: Fri Sep 26, 2003 10:41 am Post subject: Re: Necessity in comp.lang.c++.portability group? |
|
|
Michael Jørgensen wrote:
| Quote: | Hi friends,
Writing portable code gets increasingly important. Most developers -
myself including - are familiar with only one platform. I've tried a
few times porting some of my software to different platforms, only to
be faced with great problems.
[SNIP] |
I think it is a good idea. I also think that if you want to keep it topical
it should be moderated. I was lately thinking about a "fast moderation"
thing. You must post to an address with the email you are going to use.
You get the FAQ. Before that: you cannot post. Then when you start to post
it will just go through. This can be done with a simple, automated SW on
top the the current way of moderation. Now when someone starts to be an S
shaped hole (think about how to do you call an I shaped beam) the moderators
go in, cancel his offending posts and put him on blacklist for some time.
And of course it could any time be swicthed back to full moderation. It
would also be possible to keep newcomers on the hand-moderated list until it
is seen they are not trolls (so trolls cannot come back from another email
address).
Why do I mention this here? Because the allowed topic of that group would
be vague, more vague than clc++m and also portability is one thing which
does not attract newbies (and hopefully trolls), so it seems to be possible
to prefer automatization and speed of moderation over total moderation.
===
Then the C++ related. You have listed in your post two out of the 3
portabilities:
- platform
- compiler
There is a third, which people often forget. The Deathstation 9000++:
- any future platform allowed by the standard
Why do we have long long int for 64 bits on the 32bit platforms? Because
people thought long == 32 bits in the 16bit era. And vendors are forced to
make (limiting) workarounds in the future (now current) languages, because
people did not know what the language really is/allows. I honestly believe
that this is a portability issue and IMO just as important and just as
subtle as the others. And even more dangerous. You have a chance to get
compiler x on platform y (even if just buying CPU time somewhere) and can
test if your code works. You can do it *now*. If not, there are people who
can tell you the gotchas *now*. But there are no such things for future
platforms.
In addition to his all IMO the portability newsgroup could serve as a
"union" of programmers, which forces QoI on the implementors. For example
to make them come up with a common inline assembler syntax for Intel
plaforms (I dunno if it is important or not, I just have discussed that in
another thread as a QoI issue). So that community coould gather people and
opinions and have an affect QoI of current compilers - regarding
portability.
--
WW aka Attila
[ See http://www.gotw.ca/resources/clcm.htm for info about ]
[ comp.lang.c++.moderated. First time posters: Do this! ]
|
|
| Back to top |
|
|
|
|
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
|
|
FAQ
Search
Memberlist
Usergroups
Register
Profile
Log in to check your private messages
Log in
