[ome-devel] OME Server lite...

Ilya Goldberg igg at nih.gov
Tue Dec 14 21:41:47 GMT 2004


Hi Mike
Good to hear from you.

On Dec 13, 2004, at 2:33 PM, Wood, Mike wrote:

> Hello everyone. It's been a long time since I've posted anything but I 
> now have a few questions.
>
> Is it possible to extract from the OME project a simple database that 
> uses the OME data model and can provide simple insert and query 
> features using the OME XML data transport? In other words, how 
> difficult would it be to remove components and create a simple data 
> store?

It seems that the current release of OME is what you're talking about:  
Its an image and meta-data store.  It has the analysis engine, but you 
can just turn that stuff off so you don't have to look at it.

Things are pretty compartmentalized so taking things apart is 
definitely doable, but it really depends on what you have in mind 
wether or not it would be "easy".

> I'm am not interested in the analysis chains or the ability for the 
> server to create images. I want a DB that can store images with 
> metadata and allow me to query them both afterwards.

Not sure what you mean by creating images - you mean like generating 
thumbnails/jpegs/tiffs/etc?
If you can't get a jpeg/tiff/thumbnail as a result of an image query, 
you want to get back just the raw pixel data?
That may or may not be a good thing depending on what you have in mind 
for a client.

I guess I'm not sure what you mean by "lite".  A lot of the complexity 
(heaviness?) in OME is in dealing with the meta-meta-data required to 
support an extensible data model.  If you want to keep this, then I 
don't think there's a whole lot you can eliminate to lighten things up. 
  If you want to have a fixed data model (support only OME XML "core", 
let's say), then you  might be better off starting from scratch and 
reusing only things like our XML parser and possibly OMEIS.  If you 
eliminated everything having to do with analysis from the code-base and 
the DB, it would eliminate a bunch of stuff, but not as much as you 
might think.  Its the flexible data model that makes it what it is (and 
enables analysis to boot).

I think the best way to think about this though is in terms of the 
functionality you want to see in the client.  Or maybe figuring out 
what's undesirably "heavy" about what we have now.

-Ilya


>
> Thanks,
>
> Mike Wood
> Applied Precision
> _______________________________________________
> ome-devel mailing list
> ome-devel at lists.openmicroscopy.org.uk
> http://lists.openmicroscopy.org.uk/mailman/listinfo/ome-devel
>



More information about the ome-devel mailing list