Mailing-List: contact general-help@xml.apache.org; run by ezmlm
Message-ID: <01d801bfeb75$5c8e64c0$397c9081@eng.sun.com>
From: "Edwin Goei" <Edwin.Goei@eng.sun.com>
To: <general@xml.apache.org>
References: <20000711060007.66863.qmail@hotmail.com>
 <019801bfeb70$3406f940$397c9081@eng.sun.com> <396B80BF.377A5387@lutris.com>
Subject: Re: [spinnaker] Announce
Date: Tue, 11 Jul 2000 13:19:47 -0700
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: 7bit

> I think this is the key - there is nothing wrong with using any and all
> we can from Xerces. But if we say this is Xerces 2, or even assume that,
> we have to justify /not/ using Xerces code. I think this is the exact
> reverse - I think all code, from day one, needs to be justified to /go
> into/ the tree - that keeps a lean, mean, clean API (excuse the Mr. T
> type string of adjectives).

Yes, I agree 100%.  I think we need to justify that code should go into an
implementation rather than the reverse.

> I know at JDOM we have been looking at going to Project X or a newer,
> tiny little parser called Spitfire - not because we are unhappy with
> Xerces, but because of (a) its size and (b) its complexity to work
> within [the code, not the usability]. I would love to see Spinnaker
> prosper (and plan to help it do so), and if it becomes Xerces 2, that's
> great! But I think it is foolish to set this as an initial requirement.

Spitfire, haven't heard of it.  Do you have a pointer?

-Edwin