Hi Andy,
> I guess the problem I have with it is that, to me, a sub-
> project should be related to a parent project which has
> some physical code.
What's the "physical code" in the XML project to which all the Xerces-*,
Xalan-* et al are related to?
> But if you make Xerces-J, Xerces-C,
> etc. sub-projects, what is their parent?
Pretty much the same kind of thing that is their parent now. Except under
this proposal there's the commonality of "intimately related to XML
parsing" binding the subprojects together; in XML, it's some vague
association with XML or its applications.
I would go so far as to say that it's a myth that there's a Xerces
subproject today. There's Xerces-J, Xerces-C, and Xerces-P, each with
their separate dev lists, PMC reps, and communities. They're related in
name and because they descend from code that was originally architected
similarly. You could argue that this proposal, spawned as it is by a Board
requirement that PMC's need to be creatures intimately involved with the
code they're overseeing, in fact gives us an opportunity to build more of a
Xerces community, not less of one.
> And if the Xerces sub-projects are defined as a parser
> in a particular language, where is the room to add real
> sub-projects related to a particular parser codebase?
I wonder if you had a chance to glance at my proposal on how to modify the
charter to permit "closely related" technologies. I'd actually prefer to
keep us doing parsers, and have components like HTML parsers live in
XML-commons (or some other more common place), but I thought this might
address your need.
> Anyone in dis/agreement? What are your thoughts?
I'm also really curious to see what other folks in the Xerces community
think. If this isn't the right way forward, then it's not obvious to me
how we come to a place where the Board can feel confident that the PMC's
responsible for the Xerces-* code bases are actually capable of discharging
their responsibilities.
Cheers!
Neil
Neil Graham
XML Parser Development
IBM Toronto Lab
Phone: 905-413-3519, T/L 969-3519
E-mail: [EMAIL PROTECTED]
Andy Clark
<[EMAIL PROTECTED] To: [EMAIL PROTECTED]
> cc: [EMAIL PROTECTED], [EMAIL
PROTECTED],
[EMAIL PROTECTED]
03/31/2004 09:28 Subject: Re: [VOTE]: motion to
transform Xerces into a top-level project as a
PM member of the "federation" of XML
projects
Please respond to
xerces-j-dev
Neil Graham wrote:
> But they do seem to have quite a number of the characteristics of
> subprojects: their architectures are quite vastly different (especially
> Xerces-C and Xerces-J 2, the most active); with a few exceptions their
> committer bases are disjoint--and in the two cases I'm aware of in which
> [...]
> Does that meet any of your concerns?
Not really.
I guess the problem I have with it is that, to me, a sub-
project should be related to a parent project which has
some physical code. But if you make Xerces-J, Xerces-C,
etc. sub-projects, what is their parent? An HTML document
with links to the respective sets of documentation? That
doesn't really seem like a project.
And if the Xerces sub-projects are defined as a parser
in a particular language, where is the room to add real
sub-projects related to a particular parser codebase? As
written, this charter doesn't allow for that -- donations
would have to be fully adopted into the codebase for a
parser code in a particular language.
Disjoint developer communities for parser implementations
does not imply that they should be separate sub-projects.
So I don't think that I can support this draft of the new
charter.
Anyone in dis/agreement? What are your thoughts?
--
Andy Clark * [EMAIL PROTECTED]
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
