Scott Tavares wrote:
>
> >
> > Let's be blunt. I'm writing the XML book for O'Reilly. That should
> > mean something to you (I'm sorry, I don't usually do this folks, but I'm
> > trying to steer Turbine from a big mistake). That should mean I know
> > what I'm talking about. O'Reilly, and soon half the computer based
> > development groups in the world (O'Reilly's distibution) stake their
> > dollars and reputation on people like me and Richard Monson-Haefel and
> > Jason Hunter not only being caught up, but being ahead of the curve. So
> > I'm telling you that you would be making a big mistake using DTDs here,
> > when there is no compelling reason not to use XML Schema. In fact, XML
> > Schema will allow expansion that DTDs don't. I for one think that while
> > Scott Ambler knew what he is doing, there is more to what OPaL can be
> > than just doing what he outlined and stopping. There is tremendous room
> > for growth. So don't start out the XML portion (which will be very
> > important) by using an old, rusty technology.
> >
> > And if your time was wasted, or even "not productive" in your mind by
> > reading the XML Schema spec, then you are going to sorely left behind,
> > as XML is quickly going to change everything... believe me, I get paid
> > every day to read these specs, to wade through 100+ mails at xml-dev and
> > xsl-list, etc.,etc.
> >
> > Scott, this isn't meant as a flame. But you made some very poor
> > assumptions about XML Schema, none of which were well-founded, and then
> > you questioned the viability of an option anyone in the XML community
> > would stand up for. So I'm responding strongly, not towards you, but
> > towards your mistaken assumptions. Don't take it personally, as it is
> > not at all meant to be.
>
> OK, lets be blunt. i'm not taking it personally, but I believe you are! We
> could have avoided this entire conversation if you simply posted a
> comparable XML Schema to the DTD and explained why and how we should use it.
> But no, you choose drop names and use a peacock like stance and proclaim
> your dominance in your knowledge of XLM technology. i for one, could care
> less about who you are, who you know and what books you write. none of which
> have anything to do with the issue at hand... boy, talk about an ego. Now,
> how does the saying go ...put up or shut up... post a comparable XML Schema
> if you feel so strongly about it. in the mean time i'm going to move forward
> with what i have, if and when someone creates a XML Schema that trumps what
> i do then good, replace it, all the better.
It's premature, which was my original point. I was happy to see George
say the same thing. You are hung up on implementing what Scott Ambler
suggested, and nothing more (someone who you obviously do care about,
interestingly enough _because_ of what he wrote, sic!). George I was
happy to see was willing to look beyond just "a class map, as described
in Scott Ambler's paper [insert URL here]. I don't understand why you
are content to leave it at that. As soon as you put that DTD in place,
you cut off significant imagination and creativity channels, because
people start to build XML around that.
Push the envelope, for crying out loud. Be a developer, a designer, not
just an implementer. That's my point. XML Schema is about pushing the
envelope, not "providing a comparable solution." I could care less
about being comparable to anyone. I want to be _ahead_ of _everyone_.
So I never said use XML Schema because it does the same thing as DTDs
better, _only_. I said use it because it is a better data mechanism.
Because it will force you to stretch, as well as force OPaL to stretch.
To be based on Ambler's paper, but far more. No one is going to care
about Scott Ambler, they are going to care about object mapping and
object persistence, so when they ask why isn't such and such there, and
why isn't that feature implemented, your answer of "It's not in the
paper" is going to be embarassing.
-1 on constraining the XML for class maps there. I tried to get around
this without plain vetoing it, but there it is. If you want to
personally build your own DTD and conform to it, fine. But don't commit
it. And I would prefer you not post it to the list, other than if
someone personally asks you for it, and then send it to them in private
e-mail. Let those of us who think there is more to these class maps
than you currently intend not be confined to them, or even think about
them _yet_. George, who you do obviously respect, who you said
understands OPaL's purpose and code better than you in many instances in
an earlier e-mail, thinks that it is premature. If you won't believe
me, believe him.
And to address it being personal, I find it embarrassing when someone, a
committer even, posts to a general list that learning about a new
technology, even if it is _totally_ unrelated (which this was not), was
a waste of their "valuable" time. That is an embarassment to Open
Source, and promotes other users to not push the sides of their box.
That's what riled me up, and keeps me riled up. That and that you still
have missed the point - a comparable XML Schema would _not_ have
addressed my point; and your suggestion of it makes me think you need to
spend more time on XML Schema, or at least re-read my e-mails.
>
> ...putting on my fire proof suit now,
Why? There's yet to be anything personal, merely about the manner in
which you characterize OPaL and this project, as something "constant"
and "unchanging" and easily "constrained." These I take issue with, and
anyone that suggests them, not you ;-)
-Brett
>
> -scott-
>
> : - | (me ignoring you)
>
> ------------------------------------------------------------
> To subscribe: [EMAIL PROTECTED]
> To unsubscribe: [EMAIL PROTECTED]
> Problems?: [EMAIL PROTECTED]
------------------------------------------------------------
To subscribe: [EMAIL PROTECTED]
To unsubscribe: [EMAIL PROTECTED]
Problems?: [EMAIL PROTECTED]
