Re: RDBMS vers new data model

On Thu, 22 Apr 2004 [email protected] wrote:

> Hello Guenther,
> 
> > Hi Hari,
> >
>
> That's exactly my opinion too! It less than optimal to adapt the special 
> properties of biological data to a RDBMS. Instead a global view will be 
> necessary to develop a new kind of a data model combined with an adapted 
> Information Retrieval or Data mining, Storage layer,...  to fulfil the 
> biological needs. 
> 
> Do you have already an idea, where to go?


This is a very interesting debate, which may turn out to be purly
academic, as many nice data warhousing systems have died, and people tend 
use whatever is commonly available good or bad.

IMHO, I think no one datamodel can / should / would or could be set in
place, rather every biologist makes their own data model either from
scratch, by combining other datamodels in to larger models and/or with
available ontologies.

All 'published' data should be available on the internet via the model.

As time goes by standards / basic / core models will emerge.

I.e. We will not do away with big, well funded, centralised resources, but
we will maintain the flexibility to express our own data acording to our
own beleifs about the nature of that data.

Our own (biological) understanding of data is what drives model
development. 


However, this may be like saying "if we all share poverty would be a thing
of the past". Nice idea, probably never going to happen.

 
> Regards,
> 
> Thorsten
> _______________________________________________
> Biodevelopers mailing list
> [email protected]
> https://bioinformatics.org/mailman/listinfo/biodevelopers
>