[DAS] LDAS vs Dazzle

Thomas Down td2@sanger.ac.uk
Thu, 3 Jan 2002 22:27:35 +0000

On Thu, Jan 03, 2002 at 02:02:56PM -0800, Simon Chan wrote:
> Hi Everyone,
> Right now I'm trying to determine which to use:  "Dazzle"
> or "LDAS" I found one of Lincoln's old email responses in the
> archive:
> Subject: [DAS] System requirements
> "I think the choice boils down to whether you feel more
> comfortable with Java or with Perl.  Also whether you
> prefer the benefits of the full-featured Ensembl data
> model or the simplicity of Bio::DB::GFF."
> ***QUESTIONS*** 
> 1.  In what way is Dazzle more full-featured than
> LDAS?  
> 2. Can anyone give me some specific examples?


Dazzle isn't actually tied to the Ensembl data model
(although it's [relatively] easy to set up a Dazzle
installation backed by and ensembl-format database).
The main difference (other than language) between 
the two is that LDAS seems to talk directly to a relational
database (using it's own, reasonably simple, schema)
whereas Dazzle uses plugin modules to feed data into
the core server, using the BioJava object model.

If you want to run a DAS server on top of an existing
database (and don't mind doing a little bit of coding),
Dazzle might well be the neater option.  Otherwise,
it probably depends which server-side framework
(mod_perl vs. Java servlets) you're happier with.