Re: gnat-3.10

[dewar@merv.cs.nyu.edu (Robert Dewar)]

Ronald says

<<You yourself said that 3.10 is your "current production release".  Do
your customers know that you consider that it's not even ready for
public release?  And why would I want to provide free software to the
world under a licensing agreement that makes it less-than-freely-
available to everyone?
>>

Right, when a production release is first made, it obviously has not been
as widely used as a production release that has been made some time ago
and widely used.

When we first make a prerelease, those customers that want to go to the
new version, to get the benefit of bug releases and new features can.
Some choose to, some choose not to. Those who choose to move to the new
release know that we will provide immediate help to them if there are 
any problems. This does not guarantee freedom from problems, but it
helps to reduce the risk.

For example, if there is a slight installation glitch that causes trouble
on some particular hardware configuration, we can easily help anyone
who runs into it, and let people know how to get around it. 

But a public release is a much different thing. Thousands will use it
who have no access to support from us or from any one else. It is therefore
much more important that public releases be glitch free, particularly
when it comes to installation problems.

That is why we delay making public releases until we think that all
the glitches have been ironed out. Some of our customers wait until
well after the public release to switch versions, which seems quite
reasonable, others want to move immediately to the new release.

Ronald is all upset over "what we have done to him", but all we have done
to him is to tell him to wait until we think that the version is ready for
public distribution. As we all know, Ronald had trouble with the 3.09
public release, perhaps that says we should have waited LONGER rathern
than SHORTER to distribute 3.09, though one cannot be sure that all
problems will be solved at any time.

Incidentally Richard Stallman is definitely supportive of the notion of
not releasing free software before it is in an appropriate state. The
Linux folks have often argued that the Linux distributes should contain
the latest GCC snapshots, since obviously they fix bugs, but such an
attitude is not a good idea in the long run.

Sure, everyone would like to be able to get the latest and greatest GNAT
as early as possible, but we do not think it will help the community to
have very frequent releases which have not been thoroughly user tested.

We are willing to let customers have versions earlier, precisely because
we know we are there to help if there are any problems. 

Actually Ronald's experience in failing to build from sources is a really
good illustration of the problem. If Ronald had been a customer, then of
course we would be spending time with him to get around the problem (whether
the problem is our problem or the customer's problem is not the issue in
this situation, getting around the problem *is* the only issue). On the
other hand, Ronald's frustration with 3.09 is a good example of what we
want to avoid happening by making sure that public releases are in good
shape before they occur.

This is simply a matter of quality control, which seems in our judgment
to be as important for public releases of GNAT as it is for any other
customer releases that we make.

Robert Dewar
Ada Core Technologies