FW: FPODBC - follow-up

[Brian K. White]

Jay R. Ashworth wrote:
> On Thu, Dec 18, 2003 at 03:04:58PM -0500, Kenneth Brody wrote:
>> Terry Slater wrote:
>> [...]
>>> Isn't backward compatibility one of the keys to making a good piece
>>> of software? IMHO, the new License model is flawed if it doesn't
>>> allow for backward compatibility with older OS's. Just a thought.
>>
>> Should filePro 5.0 be made to work on an MS-DOS 3.1 system with a
>> 286?
>>
>> How far back should the application support?  (See the other
>> subthread regarding Micosoft no longer supporting those O/S's.)
>
> Should an application be able to be used on a workstation (where it
> clearly will actaully *run*) whose operating system likely comprises
> 50% or more of the installed base, and fairly clearly will continue
> to, regardless of the fact that the vendor is dropping support for
> that version of the OS (in favor of an OS whose EULA *requires* you
> to grant Microsoft root access at their discretion for no good
> reason)?
>
> It should be pretty clear which side of the value judgement *I* come
> down on, and I'm not even deploying the product.

I think any company should support whatever it's users want to buy a lot of.
If there are a lot of 98 boxes out there, and a lot of users want to use
them, it's only the height of stupidity (ahem :)  for a vendor to tell those
users no we won't sell to you or we won't support that platform.

If on the other hand most users are not asking for that, and supporting that
minority would drain resources enough to adversely affect the majority, then
it's perfectly understandable and fair for the vendor to stop or at least
seriously limit support for that minority.

However, that assumes that other factors are roughly equivalent. The logical
defensibility of their stance changes in the presence of factors such as,
lets say they have statistics (even if only anecdotal and in their heads and
not written down in black & white somewhere) that indicate that it's harder
or impossible for them to produce a product that's as robust as they'd like
on one platform vs another. Even if customers are asking for it, you can't
fault a person or organization for being unwilling to put out an inferior
product with their name on it. Heck, that's why we mostly don't even offer a
windows version of our app anymore. Having not sold a new windows install in
probably 2 years our customer-in-distress support calls have dropped 90% and
the average time to fix the average support problem is also much less. Even
if it's only what they think, it's a valid thing to base policy on whether
it turns out to be true or not. Maybe they simply don't want to be
responsible for putting out a good product that they know can not actually
be any better than the not very good windows98?

OTOH... what with the known ntfs bug that ms has still not fixed and waits
there to kill any xp/nt/2k box at any time for no known reason (and thus no
known way to avoid it) yet... It's hard to call xp unilatteraly better. It
is definitely better at many things but having the filesystem go corrupt
without warning or provocation is nasty enough to counteract a whole lot of
other good points. It's one of those things that's kind of just plain a show
stopper and it doesn't matter how much else is good or how much better it
is.

Brian K. White  --  brian at aljex.com  --  http://www.aljex.com/bkw/
+++++[>+++[>+++++>+++++++<<-]<-]>>+.>.+++++.+++++++.-.[>+<---]>++.
filePro BBx  Linux SCO  Prosper/FACTS AutoCAD  #callahans Satriani