Subject: Re: XML and GILS
Carl Hage (carl@chage.com)
Date: Fri, 27 Mar 1998 10:59:01 -0800
Message-Id: <199803271919.NAA10290@rgate.ricochet.net> From: "Carl Hage" <carl@chage.com> To: gils@cni.org Date: Fri, 27 Mar 1998 10:59:01 -0800 Subject: Re: XML and GILS In-Reply-To: <351AC95A.B483179A@fgdclearhs.er.usgs.gov>
Mark Kelly <mark@markkelly.com> wrote:
>
> A new W3C submission
> <see http://www.microsoft.com/standards/xml/xmldata-f.htm> called XML-Data
> is being proposed as way of getting any information from any data source
> to any client - browsers, applications or applets.
Neither XML or SGML is the solution that allows information to be
exchanged between _any_ source and _any_ client, even though that
seems to be a popular claim. However, the new XML-data is a major
advance because it changes XML from being a record-level formatting
standard to a database-level format.
Previously XML only provided a means to construct data records, e.g.
a little better than the Tab/Comma Separated Values format, but no
way to define the fields and semantics. Now it allows a data schema
to be defined, so it is possible to load a database and do something
with the data without having to write a custom program.
Now it's possible to write a GILS equivalent schema in XML, which
allow future XML-data applications to manipulate GILS data. It also
fits in nicely with the BSR work, since both use a more object
oriented model of data.
> XML is touted as working well within or beside the Resource
> Description Framework (RDF) that we should watch.
RDF has had the same problem as XML-- lack of semantics for
attributes. If you ask me, 99% of the value of GILS is in the
definitions of the attributes and controlled vocabularies. Previously
RDF/XML was like taking GILS/Z39.50 and deleting all the attribute
definitions. You can exchange arbitrary attributes, but so what--
applications need to manipulate the same attributes. Whether the
attributes are exchanged with Z39.50, XML, SMGL, SUTRS, HTML, or
ASCII is irrelevant since it's easy to write a front/back-end to
reformat records. If the attributes aren't standardized or aren't
precisely defined.
Data exchange between applications only works if one application
interprets semantics of the data in exactly the same way as another.
The EDI standards for business transactions have been hampered
because the standards are too vague (purposely), and each pair of
trading partners negotiates the details in thier data exchange and
customizes the software at each end. Only a few big companies can
participate because it costs thousands of dollars to setup each
trading partner.
> ...The mapping or negotiation between XML for GILS and GEO entries
> will be an interesting and necessary exercise, and one I hope will also
> be facilitated through software. To assure some semantic crosswalk we
> probably ought to use the same XML tags, though we could with greater
> difficulty provide external mapping.
Yes, I think this would be useful, and should be done in conjunction
with a BSR/BSI model. Each . in the BSR name matches nested object in
XML, except the last name which is sometimes just a data type.
To make data exchange as well as external mapping feasable, it will
be crucial to have the schema definition (e.g. XML-data) as well as
all code values (Controlled Vocabulary terms) online, machine-readable,
and network accessable without copyright restrictions. Paper-only
standards and copyright restrictions are a major problem with many
current data excchange standards.
--------------------------------------------------------------------------
Carl Hage C. Hage Associates
<mailto:carl@chage.com> Voice/Fax: 1-408-244-8410 1180 Reed Ave #51
<http://www.chage.com/chage/> Sunnyvale, CA 94086
This archive was generated by hypermail 2a16 : Tue Mar 23 1999 - 03:55:42 EST