incubator-general mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Aleksander Slominski <>
Subject Re: XNode 1.1 API submission
Date Thu, 18 Sep 2003 22:38:42 GMT
Murray Altheim wrote:

> Hi all,
> This is a proposal for a metadata API for use in Xindice, though it can
> be used more generally for adding an XML metadata wrapper around an XML
> node. In short, it's SOAP-like, but *actually* simple.
> Brian suggested I join this list and make the proposal here. I've 
> included
> my initial message below, unedited. (I've still not heard at all from 
> Sun.)
> Anyway, I'm wondering what is the easiest, lowest-pain way to approach
> getting XNode 1.1 submitted to the Apache project. I'm under no IPR
> commitments at the Open University (I checked), so I can freely fill
> out whatever forms are necessary. If people want it, I can post the
> API javadocs or send them to interested parties. It's pretty small,
> 7 files, 9K compressed.

hi Murray,

please send it to me - if it is open source send me also source code or 
pointer where it is available (or previous version is available).

i think have similar API - one of its goal is also to make processing of 
SOAP headers described as infoset much easier - the API is translating 
XML Infoset constructs into Java interfaces (like XmlElement) and is 
available as part of MXP1/XPP3 (it as open source ).

i would like to better understand what problem you solve and how 
approaches are different.



> Thanks,
> -------- Original Message --------
> Subject: XNode 1.1 API submission
> Date: Tue, 19 Aug 2003 20:33:01 +0100
> From: Murray Altheim <>
> Organization: Knowledge Media Institute
> To: Brian Behlendorf <>
> References: <>
> <>
> Brian,
> I don't know if you're aware, but while I was still at Sun I wrote an
> API to wrap XML documents in a SOAP-like wrapper to enable the attachment
> of metadata. That API was com.sun.xnode.XNode. I'd originally written it
> as part of a non-Sun project, as I was at the time involved with Doug
> Englebart's group at SRI, where Lee Iverson (now at UBC) had floated
> his "Nodal" project. I thought XNode was a Nodal-like thing for XML. 
> Well,
> I was wrong, but that's where it started.
> So anyway, my boss was laid off and I found myself (and the rest of my
> office) under Java management. I suggested using XNode and dbXML as
> part of Sun's Registry project, as we needed an XML database and a means
> of storing metadata. Long story short, the API was intended to be 
> submitted
> to Apache but due to office reshuffling and my leaving off to school, it
> never happened. Sun still delivers com.sun.xnode.* as part of their
> Registry project (part of JWSDP), but has never published the API nor 
> does
> there seem to be any interest in it. Likely below the radar, really.
> I'd submitted the code to Xindice, but somebody pulled it off the server
> because there hadn't been the proper paperwork from Sun releasing the 
> code.
> (No criticism -- that's what should have happened) So I sent a message 
> off
> to Sun's "feedback" line, and one to Jeff Suttor, but after two weeks 
> I've
> not heard anything at all. Like I said, probably under the radar.
> Because XNode is a core API within my Ph.D. project Ceryle, I've gone
> ahead and rewritten the API as "XNode 1.1". It is not quite backward
> compatible with XNode 1.0, uses a different package designation (i.e.,
> it uses the intended package of org.apache.xnode.*) and a different
> XML namespace URI ("" rather than
> "", noting the www/xml difference as well
> as the version number). I've altered/improved the method names in a few
> places, added a method to allow embedding of a DOM Element as metadata
> rather than simply name-value pairs, and rewritten the documentation.
> Now... what would it take for me to submit XNode 1.1 to Apache? I've
> always intending to do this and perhaps now it can be done. I realize
> there is some muddiness as to IPR, though several things keep that from
> being a problem for you:
> 1. the original API was never published publicly, except by me after
> leaving Sun, hence in theory I own the copyright (but not the IPR
> since I was under contract while at Sun)
> 2. Sun has not patented com.sun.xnode.XNode, so org.apache.xnode.XNode
> is not an enfringement on copyright. You can't copyright ideas, only
> patent them, so the ideas in com.sun.xnode.XNode that resurface in
> org.apache.xnode.XNode are protected under Sun's IPR only insofar
> as they are Sun's ideas. In reality, XNode is a combination of Lee
> Iverson's NODAL and SOAP.
> 3. I've rewritten the API so you'd be receiving a *different* API.
> 4. They've never made any public splash about this, nor have they ever
> published anything about it either. They could potentially come
> after me for violating my IPR agreement (since in theory everything
> I've ever done or said is now owned by them, even going back to my
> birth), but there's no violation of copyright since they've not
> published anything (which is required of US copyright laws). As
> I mentioned previously, I actually beat them to it, so I in theory
> own the copyright on XNode 1.0.
> 5. Sun has no history of going after things like this, esp. ones that
> are way below the radar, esp. since I'm an ex-employee with no
> resources, not Microsoft. They would not spend resources bothering
> with such a thing, even if they could figure out where there was
> any infringement (since I've not given away anything demonstrably
> Sun's, having been away from them for over a year and a half).
> I've gussied the thing up, do you want it? :-)
> Murray
> ........................................................................... 
> Murray Altheim
> Knowledge Media Institute
> The Open University, Milton Keynes, Bucks, MK7 6AA, UK .
> "There have been ongoing problems with cockroaches getting on the
> floor... they can wreak havoc with 350,000 dominoes." -- Brian Kim
> ---------------------------------------------------------------------
> To unsubscribe, e-mail:
> For additional commands, e-mail:

If everything seems under control, you're just not going fast enough. —Mario Andretti

To unsubscribe, e-mail:
For additional commands, e-mail:

View raw message