OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.


Help: OASIS Mailing Lists Help | MarkMail Help



   Re: [xml-dev] xml schema best practices question.. one big xml or lots o

[ Lists Home | Date Index | Thread Index ]

[got caught by the reply/reply-all plague again -- resend of original
I just sent Leevi]

You may want to take a look at cross over solutions such as eXist

> http://exist.sourceforge.net/

This is a native XML database.  In essence it uses XML as the
foundation and the optimizes the storage and retrieval, allowing you
to utilize up and coming XML DB-centric Query technologies such as
XQuery to retrieve your data.

There are others.  In fact SQL Server, Oracle, etc... already allow
storage of native XML and while I can't speak to Oracle (although I
understand both their here and now as well as their coming XML support
is nothing less than marvelous so its worth a gander for sure) but as
far as the Beta releases of SQL Server and the personalized MSDE both
support a subset of XQuery support that you can play with right away.

See the blogs of Mike Champion and Michael Rys for more details,
http://blogs.msdn.com/mikechampion/ and
http://blogs.msdn.com/mrys/default.aspx respectively and if I have my
link correctly you can access the download for MSDE (actually its now
called SQL Server 2005 Express) via >

> so now i am armed with this new information and clarity.. i have to ask
> what is xml good for??.

XML was never designed as a database replacement but instead as a
cross-platform data exchange format.  When you are speaking in terms
of individual transactions the benefits of working with XML become
very apparent.  No open and closing of database connections, a human
readable format that can be quickly and easily understood if designed
correctly.  A simple way to express structure and function that can be
passed from platform to platform (think SOAP) without concern for how
each platform handles the processing.... etc...

Theres obviously a lot more than this but in my opinion the
combination of a high powered DB backend in which can be easily
queried (XQuery), the returned XML transformed (XSLT) into the proper
context/format for the application in questions, using a path based
references (XPath) to quickly and easily navigate through the XML
infoset/dataset/whateveristhecorrectandpropertermfortoday, XML Schema
and RelaxNG for data validation based on a predefined set of rules,
etc...  Huge benefit to using XML with the power of a heavy load
bearing DB is a lethal combination of flexibility, speed, power, and
ease of creating kick butt software like never before...

Cheers :)


On 4/16/05, Leevi Graham <leevi@theleviathanproject.com> wrote:
> thanks for your response michael,
> it seems that after all my study and research into xml it is not the
> right form of storgae for the application i wish to create. The info
> would hopefully be accessed numerous times and as more people add their
> business details the xml file will become bigger.. So i guess that a
> relational database od the key afterall.
> so now i am armed with this new information and clarity.. i have to ask
> what is xml good for??. If files sizes become a problem when accessing
> the information repeatedly what is a real world situation where xml is
> ideal? possibly storing seperate bits of data for occassional retrieval?
> i understand that xml is browser, computer, etc independent and that one
> of its key bebefits in extenability but what good is that if you cant
> share large amounts of data quickly and easily.. will native xml
> databases solve this problem?
> as always one answer leads to a 1000 questions...
> cheers
> Leevi Graham
> Michael Kay wrote:
> >There's no right answer to this: it depends how you want to use the data.
> >Remember that an XML document is not a database. If the document is 1Gb big,
> >then opening it (parsing it) every time you want to find out a single phone
> >number is going to be very expensive. On the other hand, if it's 1Mb and the
> >number of transactions is 1/hr rather than 1000/sec, then it may well be
> >viable.
> >
> >Generally I would implement hierarchic ("contains") relationships like the
> >one you describe by using the XML hierarchy, unless this makes the document
> >impossibly big.
> >
> >If you decide to store the menus separately, either because two pubs can
> >have the same menu, or for physical design reasons, there are many ways you
> >can represent the relationships. If each menu is in a separate file then
> >it's probably most flexible to use a URI. However, where objects have
> >natural identifiers like part numbers or personnel numbers, then I prefer
> >just to use that identifier, and leave the query logic to work out how to
> >find the XML document for a given personnel number or part number. There are
> >some people though who insist that the only respectable way to identify
> >objects is by means of URIs.
> >
> >Michael Kay
> >http://www.saxonica.com/
> >
> >
> >
> >
> >
> >>-----Original Message-----
> >>From: Leevi Graham [mailto:leevi@theleviathanproject.com]
> >>Sent: 16 April 2005 03:41
> >>To: xml-dev@lists.xml.org
> >>Subject: [xml-dev] xml schema best practices question.. one
> >>big xml or lots of little components?
> >>
> >>hi guys this is my first post to this dev-list... hopefully
> >>it wont be
> >>the last...
> >>
> >>i have a aquestion that has been bugging me for quite some time..
> >>
> >>i am attempting to create a business directory with xml as
> >>data storage.
> >>
> >>i wish to create business elements that contain the obvious
> >>information.. address, contact location etc...
> >>
> >>however the business type may change from retail to professional to
> >>hosplitality....mainly hospitality like pubs and restaraunts..
> >>
> >>so finally here is my question:
> >>
> >>*lets say i have a business element that is a restaraunt and that
> >>restauraunt obviously has a menu.. should i include the menu
> >>as a child
> >>element of the business or should i create a seperate xml file called
> >>menus. Also if I do create a menus.xml file that contains the menus
> >>seperatley how will i be able to tell which menu belongs to which
> >>restaraunt..*
> >>
> >>things to consider my businesses.xml file may eventually contain many
> >>businesses with many menus....
> >>
> >>hopefully someone can help
> >>cheers
> >>
> >>-----------------------------------------------------------------
> >>The xml-dev list is sponsored by XML.org <http://www.xml.org>, an
> >>initiative of OASIS <http://www.oasis-open.org>
> >>
> >>The list archives are at http://lists.xml.org/archives/xml-dev/
> >>
> >>To subscribe or unsubscribe from this list use the subscription
> >>manager: <http://www.oasis-open.org/mlmanage/index.php>
> >>
> >>
> >>
> >>
> >
> >
> >
> >-----------------------------------------------------------------
> >The xml-dev list is sponsored by XML.org <http://www.xml.org>, an
> >initiative of OASIS <http://www.oasis-open.org>
> >
> >The list archives are at http://lists.xml.org/archives/xml-dev/
> >
> >To subscribe or unsubscribe from this list use the subscription
> >manager: <http://www.oasis-open.org/mlmanage/index.php>
> >
> >
> >
> -----------------------------------------------------------------
> The xml-dev list is sponsored by XML.org <http://www.xml.org>, an
> initiative of OASIS <http://www.oasis-open.org>
> The list archives are at http://lists.xml.org/archives/xml-dev/
> To subscribe or unsubscribe from this list use the subscription
> manager: <http://www.oasis-open.org/mlmanage/index.php>


:: M. David Peterson ::
XML & XML Transformations, C#, .NET, and Functional Languages Specialist


News | XML in Industry | Calendar | XML Registry
Marketplace | Resources | MyXML.org | Sponsors | Privacy Statement

Copyright 2001 XML.org. This site is hosted by OASIS