xml-general mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Edwin Goei <Edwin.G...@eng.sun.com>
Subject Re: Mixing Two Terminologies
Date Tue, 06 Mar 2001 03:01:24 GMT
Kimbro Staken wrote:
> Let me give you a hypothetical example of what I'm talking about. Pretend
> you're a developer with some mid-size company who was just given the task of
> building an XML application by your boss. You read an article in your
> favorite Java magazine that said to build XML applications you need a
> parser. The article also said that xml.apache.org is the place to go for
> some great XML tools. So you fire up your browser and head on over to the
> site to download a parser. When you get there you're taken aback a bit by
> the long list of tools with funny names but it isn't too big of a deal. So
> you start looking down the list and see xerces is an XML parser but there is
> no link to find more info on xerces so you keep reading the list. When you
> get to the bottom of the list you find crimson, hmm funny that is also an
> XML parser. Still no link though so you fumble around a bit more and
> discover the link for crimson on the toolbar and go to the project page.
> When you get there you read a bit about the various specs it supports and
> then get to the roadmap section which mentions that crimson is going to be
> part of Xerces 2. So now you're thinking, hmm guess I should look at xerces
> so you find your way to the project page for xerces. Again you start
> reading, uh oh now it says that the xerces API is not stable yet. Hmm so
> what do I do, crimson said it wasn't ready because it is going to become
> xerces 2 and xerces 1 says that it isn't stable either. Ugh, I'm just
> confused I guess I'll just go see what Microsoft has to offer since the boss
> said VB was ok to use too.

Actually, crimson will not be part of Xerces 2.

As I understand it there are actually 3 (or maybe 2.5) parsers.  Someone
asked why there are two parsers.  Here's some history... the answer is
that it is because the ASF accepted two different code bases, one from
IBM (Xerces 1) and one from Sun (Crimson).  The original plan was to
merge them, but that doesn't really work.  Sometime last fall, some
developers of Xerces and Crimson (me) and others met and agreed that
there were problems with both parsers and so the plan was to migrate to
Xerces 2 the core of which is different from Xerces 1.  So really Xerces
2 is a third, different parser at the core.  However, Xerces 2 is still
not production quality, yet.  So that is where we are today.

> [more comments about website]

I agree that the website is confusing.  It would be nice for someone to
go to the site and find the _one_ Apache XML parser.  However, that is
where there is a significant difference between a company's website and
Apache's.  A company must decide which parser to commit to.  At Apache
this creates political problems, so there are currently two.  There are
other projects at Apache where there is more than one project that do
the same thing, sometimes they have the same name, sometimes not.  For
example, Cocoon 1 and Cocoon 2 or there are two different regular
expression projects.


View raw message