How come there is such a strong streak of Puritanism in the developer
community, as though it was immoral in some way to write code that
works, but which fails to conform to some ideal of perfection.

Our clients have written hundreds of queries which are saved and re-
used as part of their in-built reporting systems. Many of these will
have so-called ambiguities, even though they have been running
correctly for a year or more. How do I explain to them that their
report-writing is morally defective and is no longer to be tolerated?

You can imagine how difficult it is going to be for us to move people
from IB to FB.

--- In ib-support@y..., lester@l... wrote:
> > Why can't RC1 do its best to understand the query, and maybe give
a
> > warning if there is ambiguity, and run it anyway. Leave it up to
the
> > developer to remove ambiguity.
>
> Perhaps we could get it write the queries whilst it's at it
> <g>
>
> > This is IMPORTANT. Is there anyone out there who agrees with me?
>
> I have been caught on a number of ocassions by 'leaving out
> the flexible bits' and not picking up on them when making
> changes later. So anything that forces tidier coding and
> complains when I'm updating things is fine by me.
>
> Old systems stay with old versions of libraries and
> software. When you start to change to a new base, everthing
> should be checked anyway, so the oportunity to review code
> in the light of 'new bugs' is more of a help than a
> hindrance.
>
> --
> Lester Caine
> -----------------------------
> L.S.Caine Electronic Services