[XML-DEV Mailing List Archive Home] [By Thread] [By Date] [Recent Entries] [Reply To This Message]

Re: Jim Gray article on Next Generation Databases

• To: xml-dev@l...,bob@w...
• Subject: Re: Jim Gray article on Next Generation Databases
• From: David Lyon <david.lyon@c...>
• Date: Thu, 5 May 2005 13:14:33 +1000
• Cc: "'Bullard, Claude L \(Len\)'" <len.bullard@i...>,"'Ken North'" <kennorth@s...>
• In-reply-to: <200505050148.DCD05878@m...>
• References: <200505050148.DCD05878@m...>
• User-agent: KMail/1.7.1

On Thu, 5 May 2005 11:48 am, Bob Wyman wrote:
>  This is a "development"? Doesn't anyone remember VAX/VMS? It had RMS
> (Records Management System) which presented the user with a choice of file
> types: sequential, block I/O, record oriented, indexed, etc. All built-in.
> Having indexed ISAM files as a core function of the operating system was a
> wonderful thing...
>  I remember meetings in the mid-80s when we were seriously
> considering bundling a stripped down version of RDB (our relational
> database) as a component of the VMS operating system as well. The thinking
> was that since we could see how all of our application products would be
> enhanced if they used RDB for data storage, it might make it easier for us
> to build and ship products if developers could simply assume that there was
> an RDBMS available on every platform.

I think we truly at this point now. Be it ADO on the Windows platform or
mysql on Linux, every machine now seems to have at least one 
half-decent sql database built in.  

Yet, from an XML perspective, none of these databases are really
tapped. And I think the 'brand' of database is now becoming
irrelevent.

I keep on getting rejected from the powers that be, but if there
was a group of people that would be prepared to get together
in Europe (at a time other than when ceBit Sydney is on) to
bash this out, I think it would be a very useful excercise.

The proposal is simply to define a simple set of tables
that make the basis for an xml<->xml transfer system
with rudimentory inbuilt indexing capabilities, ie
document_type,document_date, document_reference
etc...

This should be able to be implemented on any o/s, using
any vendor database.

I'll even provide the coffee for free....

If we already have all the parts in existance, I really
don't understand what are the obstacles to doing
this. 

It really seems that there really is a standoff of the big
guns. And they have everybody hypnotised like
stunned mullets with patent threats.. Do anything
and we sue you... yada yada

I just take these as afronts to freedom of speach...

but after a while you realise that there is work to
be done and customers to be served. 

After all we we shouldn't let the wheels of xml grind 
to a halt just because the big guns are in stalemate..

there's many small holes in the big picture that
still need to be filled in.

David

-- 
Computergrid : The ones with the most connections win.

• References:
  • RE: Jim Gray article on Next Generation Databases
    • From: "Bob Wyman" <bob@w...>

• Prev by Date: Re: Jim Gray article on Next Generation Databases
• Next by Date: Re: Jim Gray article on Next Generation Databases
• Previous by thread: RE: Jim Gray article on Next Generation Databases
• Next by thread: RE: Jim Gray article on Next Generation Databases
• Index(es):
  • Date
  • Thread