[Bioperl-pipeline] DataMonger

[kiran]

> it's ok... won't it lead to the problems we identified in the first place
> with that particular design?
i am just thinking aloud..
the SQL is usually overloaded and is designed for database implementation
efficiency and we would want to remove that overloading in XML (which serves
as a configurational fiile now) and also as much as in the object layer
too.. if we map the SQL exactly into XML, the effort for the users to
understand the XML would be equivalent to understanding the SQL schema
itself, which we wanted to overcome..  it is the like case between object
design and database design which are usually different for similar reason..
is there anything i missed or wrong conceptually ?

kiran

----- Original Message -----
From: "Elia Stupka" <elia@fugu-sg.org>
To: "kiran" <kiran@fugu-sg.org>
Cc: "Shawn" <shawnh@fugu-sg.org>; "bioperl-pipeline mailing list"
<bioperl-pipeline@bioperl.org>
Sent: Wednesday, September 11, 2002 8:21 PM
Subject: Re: [Bioperl-pipeline] DataMonger


> > This is because, though the datamonger is implemented as a runnable
> > for other reasons, this structure of encaspulating it as a
> > preprocesser seemed better.
>
> I am always worried when I read these things... just curious... apart from
> user-friendliness (which I totally understand), why would design decisions
> vary between XML and SQL? In other words, go through some design
> discussion and decide "it can't be represented like this" and then in XML
> it's ok... won't it lead to the problems we identified in the first place
> with that particular design?
>
> It's not particuarly data-monger, it's a general issue we need to be
> careful of, since we now have perl object layer design, SQL design, XML
> design, and sool java object layer design... which could lead to headaches
> unless they are tightly kept in sync...
>
> Elia
>
> ********************************
> * http://www.fugu-sg.org/~elia *
> * tel:    +65 6874 1467        *
> * mobile: +65 9030 7613        *
> * fax:    +65 6779 1117        *
> ********************************
>
>
> _______________________________________________
> bioperl-pipeline mailing list
> bioperl-pipeline@bioperl.org
> http://bioperl.org/mailman/listinfo/bioperl-pipeline
>