Re: [xml-dev] XML not ideal for Big Data

[Kurt Cagle <kurt.cagle@gmail.com>]

Liam,

I think pushing architectures that are more conducive to mediating XML and JSON is one means of education. I'll be the first to admit that JSON is better for use within AJAX application within a web browser than XML, if only because browsers are increasingly becoming very conservative technologies, and when you look at most of the FUD about XML, it comes primarily from component developers thinking primarily about web components, designing for limited use cases and usually bound to a specific web API. The problem with this from a broad, long term standpoint is that this sea of APIs is fragile and transient. It's also something that most large data providers (big enterprises and government) are generally uninterested in - they have their data in XML and are dealing with Big Data that also crosses the threshold into Big Documents.

I think that the Mark Logic and Oracle initiatives are actually good ones. An xml-to-json() function in XQuery makes perfect sense, especially in a RESTful environment, just as a json-to-xml() function does on inbound data. Ultimately the question is whether you can work with the XDM model and XQuery as an abstractor for data manipulation in a distributed streaming environment. My primary concern with JSON is that you get idiots who seem to feel that you can take a complex ontology like NIEM or XBRL and represent it in JSON (you can't without imposing a lot of arbitrary syntactics) ... and that they can convince IT managers that this is in fact makes JSON a superior tool in such environments. 

Kurt Cagle
Managing Editor
http://xmlToday.org

On Thu, Sep 3, 2009 at 5:02 PM, Liam Quin <liam@w3.org> wrote:

On Thu, Sep 03, 2009 at 06:02:51PM -0400, David Carver wrote:
> I agree, that XML Databases are the way to deal with large data, but
> most of the people out there doing the work, don't know about them.

How to change that?

I suspect that, even today after the initial heavy breathing and
excitement has died down about JSON, the idea of JSON-native
databases would be appealing to a lot of people, even though
it's as meaningless in its way as XML-native -- the XQuery
products are mostly XDM-native, strictly speaking, just as
a relational database is not really CSV-file-native.

The importance of the distinction is the resulting perception.

So do we just need to blog and comment using newer, younger
keywords?  I see MarkLogic and Oracle articles about generating
JSON with XQuery, for example.

Liam

--
Liam Quin, W3C XML Activity Lead, http://www.w3.org/People/Quin/
http://www.holoweb.net/~liam/ * http://www.fromoldbooks.org/

_______________________________________________________________________

XML-DEV is a publicly archived, unmoderated list hosted by OASIS
to support XML implementation and development. To minimize
spam in the archives, you must subscribe before posting.

[Un]Subscribe/change address: http://www.oasis-open.org/mlmanage/
Or unsubscribe: xml-dev-unsubscribe@lists.xml.org
subscribe: xml-dev-subscribe@lists.xml.org
List archive: http://lists.xml.org/archives/xml-dev/
List Guidelines: http://www.oasis-open.org/maillists/guidelines.php