Re: DOM and SAX parsing in Ada

[Nick Roberts <nick.roberts@acm.org>]

Martin Krischik <martin@krischik.com> wrote:

> > In fact, it's because of AdaCore's 'not invented here' attitude that I
> > considered writing an alternative XML parser in the first place.
> > 
> > My understanding is that, although XML/Ada is released under the GPL,
> > any changes I made (or anyone not an AdaCore employee) would not be
> > merged back into the AdaCore branch (because that would contaminate the
> > copyright purity of their code, apparently). If I were to develop a copy
> > of XML/Ada (on SourceForge, say), it would be a permanent fork.
> 
> Last time I send them a patch I got an answer that it would be merged in.
> But since XML/Ada does not have cvs access I don't know if its true.
> 
> (If you are interested in my XML/Ada ad ons: they are part of AdaCL).
> 
> My patches for PolyORB however have been merged in - PolyORB has cvs
> access so I know.
> 
> AdaCore is better then you thing.

"Marc A. Criley" <mcNOSPAM@mckae.com> wrote:

> Well, from personal experience I can attest (and others in this thread
> have done likewise) that XML/Ada bugs I've reported have been accepted by
> AdaCore and been fixed in the code base.
> 
> (I've also submitted patches for GtkAda's Glade and those have gone right
> in as well.)
> 
> Contributions to GNAT itself might need to undergo a more rigorous review
> process before getging in, since that's AdaCore's family jewels, but
> there's ample experience with most of the other AdaCore products (GtkAda,
> GPS, ASIS-for-GNAT, and XML/Ada) demonstraing that AdaCore developers
> readily accept bug reports and patches, take them seriously, and
> incorporate them into the products.

"Jeff C" <jcreem@yahoo.com> wrote:

> Of course you mean to say GMGPL there...

Yes.

> But I am not sure why you are making that statemetn
> 
> At http://libre.act-europe.fr/xmlada/
> 
> Under the contributing header it says:
> 
> "Contributions to this library are most appreciated, including bug 
> reports."
> 
> Now, I have only ever submitted really really small things to ACT projects
> (GtkAda) and have never been turned away.
> 
> I suspect that if it was a large contribution that perhaps they would want
> copyright papers signed in some manner (either granting copyright to ACT
> or at least affirming that no other party had an interest).
> 
> Have you ever asked about submitting something large (or better yet
> actually tried to submit something moderate and see what happens)
> 
> I also strongly suspect that since this is a product for them that any
> large scale changes would get a reasonable peer review along with a
> reasonable number of rejections/peer review comments that would be needed
> before the item was accepted. This is all quite reasonable.
> 
> Even on some of my small patches, the level of comments about the changes
> were nearly equal to the size of the patch in question (and all perfectly
> valid comments that were indicative of the fact they they are creating
> libraries for customers and long term support and I am (in this case)
> working on a patch at 11 PM for no apparent reason in a best effort type
> fashion that sometimes works out and sometimes looks like I worked on it
> at 11 PM)

My comments are based on an e-mail exchange I had with the AdaCore employee
who was (at the time) in charge of work on XML/Ada, a couple of months ago.
His comments came as a surprise to me, in part because of what is said on
the web site ("Contributions to this library are most appreciated, including
bug reports"), but I have it from the horse's mouth, so to speak.

When I suggested that I wanted to make changes to the code, he indicated
that those changes could not be merged back into the AdaCore CVS copy unless
I signed a copyright assignment, in favour of either AdaCore or the FSF, to
the latter of which I am agreeable. When I suggested that XML/Ada be moved
to SourceForge, he indicated that it would have to be a separate project (a
fork), whose changes could never, in practice, be merged back into the
AdaCore CVS, because of the difficulty of getting all the contributors to
sign the assignment. This is to the best of my recollection and
understanding.

So, I could see very little advantage to making any contributions to
XML/Ada, since it would create two XML/Adas (mine and AdaCore's). Would that
be an advantage to the Ada community?

On the subject of comments, I suggested that the amount of maintenance
documention (in the form of comments or in any other form) was insufficient
-- it was nearly nonexistent, in fact -- and the answer was, in essence,
that no maintenance documentation is required, since the code is
self-documenting. I'm afraid, to me, that attitude is unacceptable (and
doesn't seem very professional, frankly).

-- 
Nick Roberts