Re: backward compatability of tools

[John Burger <john at mitre dot org>]

> From: "Dr. David Kirkby" <drkirkby at ntlworld dot com>
>
> If the developers insist on dropping backward compatibility, would it
> not be better to at least do a test for old hardware/software (i.e.
> determine a machine is running SunOs 4.1.4), print an informative
> error message then exit, rather than produce the messages I received
> about /tmp/foo1 /tmp/foo2 etc not being found ? If it was not too much
> hassle, a suggestion to rebuild the application with autoconf version
> x and automake version y would be even more useful. So the old
> hardware/software is not supported fully, but at least it gives
> someone a hint on how to get it working.

Later:

> if I
> tried to install a piece of software that only run on Windoze 2000 and 
> XP,
> on a Windoze 95 machine, I would expect the software to have the sense 
> to
> tell me that it is not supported on Windoze 95. Perhaps that is a 
> better
> analogy.

Keeping track of every combination of hardware and OS would be bad 
enough - those predecessors to Autoconf that attempted to do this 
eventually died the death of a thousand cuts.  Cross-referencing this 
against working version numbers of Autoconf seems completely infeasible 
to me.

It seems to be explicitly against the philosophy of Autoconf to do 
anything in response to particular hardware or OS versions.  Rather, 
Autoconf tests whether particular =features= are present on a 
particular machine.  It could be that somewhere out there there is a 
SunOS 4.1.4 machine with some patch or another that does not have the 
problem you've run into.  So, detecting 4.1.4 and failing would be 
wrong, it seems to me.

That said, I must agree with your basic sentiment: Ideally, the 
configure script should die more intelligently, and earlier, than you 
describe, after having detected the absence of some necessary feature.  
This reminds me of the intermittent talk about making Autoconf use 
shell functions, or some other feature of the post-paleolithic OS.  My 
own take is that such backward-incompatible modernizations might be 
fine if and only if the resulting scripts check for the presence of the 
modern feature, and die with an intelligent error message if it isn't 
found.

- John Burger
  MITRE