Re: GNAT 2.04 Install Problem!!!

[Jim Dorman <jimd@pcada.com>]

acuff@lancelot.gcs.redstone.army.mil (Phillip R. Acuff) wrote:
>
> In article <dewar.798206910@gnat>, dewar@cs.nyu.edu (Robert Dewar) writes:
> >all installation problems should be emailed to gnat-report@cs.nyu.edu.
> >
> >Time stamp problems are generally quite different on different systems.
> >The problems in DOS usually come from not following the instructions
> >about TZ (in 2.05, this TZ annoyance finally disappears). In general
> >so far, we have found that all these problems come from not following
> >the installation intructions to the letter. But if you think you have
> >found a problem that is NOT the result of this, be sure to let us know.
> >
> 
> 
> Here's what works in the DOS/DJGPP environment:
> 
> set TZ=EST5EDT
> set TZ=CST5CDT
> ....
> set TZ=PST5PDT
> 
> See the pattern!  What doesn't work is any legitimate TZ entry
> besides EST5EDT.
> 
> What impact this has, I do not know.  However, this is an annoyance
> when I have other applications that depend upon a valid TZ entry
> (i.e., MS-Windows news readers, mailers, etc.).
> 
> Phil Acuff
> 
> 
> 
> 

These types of things should be expected when using FREE software! Just
plan to spend a lot of EXTRA TIME working around "bugs" and other
limitations. If you don't have the time or energy to "deal with it",
then you should consider a commercial Ada compiler that has been
developed for the commercial marketplace.

Commercial software developers are "on the clock" and working on a
deadline to complete a project. I know of many  software engineers
who almost failed their assignment because they tried to save a few
$$$ and used inferior development tools. Frankly, I have seen few
educational institutes produce commercially viable products that are
in daily use by the "common man". If there are out there, let me know.

I think, however, that the opposite is probably true. A potential Ada
convert tries GNAT, finds the above type of problem (and others) and
quickly decides to return to Microsoft or Borland C, C++ products
where they always install correctly, have documentation to hhelp you
with questions, and generally can be expected to "work as advvertised".

Commercial Ada products fall into that catagory. Yes, they are expensive
because the total # of actual sales are not as large as their C
counterparts -- and C compilers don't have to pay a team of people to
watch them run a series of tests. If you want ccheap Ada compilers,
then everyone should buy a commercial Ada compiler. Volume hand
competition have been proven, time and again, to reduce prices. If
everyone wastes time trying to save a  few hours wages, you will
eventually force more Ada companies to look to other sources for
income. When that happens, the remaiining Ada companies will have you
over the proverbial barrel when the FREE stuff no longer can do the
job because our tax dollars quit flowing. THEN WHAT????

If you want information on Ada95 commercial Ada compilers that can be
used as an alternative to GNAT and allow you to be MORE PRODUCTIVE,
then feel free to contact me or other commercial Ada vendors. I know 
of several available on various platforms that can probably feel your 
needs and SAVE YOU $$$ in the long run.

If you want to reach me, I can be contacted at (619)793-0245, or FAX
(619)793-5374, or email to ada_info@pcada.com. Contact information
for other Ada vendors is available from the Ada Information 
Clearinghouse.

Call us,you will be suprised at your overall savings! If you are a
student, ask for student (academic) prices. I have them, so do many 
others.

Jim Dorman
President
Active Engineering