Re: Maximum String size in Java?

From: Programmer Dude (Chris_at_Sonnack.com)
Date: 03/12/05


Date: Fri, 11 Mar 2005 17:51:34 -0600

Randy Howard writes:

>> (e.g. In twenty some years
>> of professional, corporate programming portability has *never*, I
>> repeat *never*, been a concern. The one time it even raised its
>> head was when I was working on unix but writing and testing code at
>> home on my MS-DOS box.).
>
> What's interesting about this, is that I have had almost the exact
> opposite experience, for roughly the same time period. My work
> has often (usually) involved supporting multiple (sometimes as many
> as 8 or 9 quite often) OS and hardware platforms from the same code
> base.

Your work has always sounded quite interesting and challenging to me!

> Obviously there is a huge expanse of programming work out there,..

Obviously! And just in case it's not clear, when I say portability has
never been an issue, that speaks ONLY for me and my work (well, and a
good deal of the work of most other programmers in my company). I'd
never presume to imagine (let alone dictate) the work of others (altho
I am trying to get certain co-workers to understand the value of writing
for clarity :-).

It's a big world and there's a LOT of different modes.

> ..what is most interesting is that C has been used to do a very large
> portion of that incredibly varied work over the years, both for
> those that are concerned about portability and those that are not.

I spent about 8 years developing solely on unix (HP-UX mostly), which
is where awk, sed, vi, grep, ksh and many others became my dear friends.
(Unix IS user friendly--just picky about it's friends. :-) I'd already
spent many years writing C by then, so most of app development was in
C, although I began exploring C++ about that time.

My C pride and joy at the time was a web-based app that pre-dated the
dot com boom (put it this way: Netscape was the hot NEW broswer at the
time). I couldn't count on cookies, so had to be very clever with
hidden inputs in forms. But the thing was fully stateful (where it
needed to be) and very robust. It remained in use until they migrated
to another backend system.

That app was about as portable as a large application could be. Needed
some #ifdefs for some of the more OS-y things, but it ran as fine on my
old 386 as it did on the HP-UX boxes.

> If K&R have something to be really proud of wrt to C, it's buried
> in those comments somewhere.

C will forever have a fond place in my heart as my first major (non-asm,
non-BASIC) production language. It's been years since I wrote any, tho.
I've post-incremented-by-one. :-)



Relevant Pages

  • Re: No .NET framework
    ... the Unix box and move your app over - Portable...No? ... <Just Kidding> ... To get most unix programs to run on ... > Portability and .NET do not go int eh same sentence... ...
    (microsoft.public.dotnet.languages.csharp)
  • Re: Advanced c++ debugging techniques on Unix
    ... seems you are python addict. ... The most advanced C++ debuggin technique on Unix? ... Writing your app in Python. ...
    (comp.os.linux.development.apps)
  • Re: Garbage Collection - Stop Making Trash
    ... Try moving Unix headers to Windows, ... attempt to sabotage portability -- we achieve limited portability ... except that no standards committee has endorsed an interface ... Microsoft loves this mentality. ...
    (comp.lang.cpp)
  • RE: Wonderful things happen to an OS when it has an internal champion
    ... rather culture issues which are the biggest impediment to App stacking. ... Now, in the UNIX environment, in order for an App to take advantage of ... are doing with all kinds of servers such as web servers, ftp servers, e-mail ...
    (comp.os.vms)
  • RE: Wonderful things happen to an OS when it has an internal champion
    ... rather culture issues which are the biggest impediment to App stacking. ... Now, in the UNIX environment, in order for an App to take advantage of the ... local OS platform clustering technology, it needs to follow that platforms ... portability is a a much bigger deal than it is on OpenVMS. ...
    (comp.os.vms)