Re: [Miase-discuss] miase file content - manifest?

[Frank B. <fbergman@u.washington.edu>]

Hello Richard,

i've been thinking about manifests. In my first implementation i  
decided against it, the task of actually reading the experiment  
description seemed negligible as compared to the updating effort that  
would have to be done in order to change a given miase file. Reading  
the notes from the sed ml, should be simple enough.

The one thing that a manifest should provide is a unique  
identification of *which* XML file the experiment description is.

I would obstain from putting the Model Names into the description as  
even something as simple as Model.getName implies an understanding of  
the API's needed to correctly interpret the model description language  
for encoded models. Which is not guaranteed at the moment.

> to parse all the files - what does anyone think of a manifest to be
> included in the archive?

as said above, I think if we have a manifest it should be very  
rudimentary, it should contain

- the name of the experiment desciption file
- the types of models encoded, (i.e: SBML/CellML...), this would allow  
tools at a glance to refuse files (or warn users)

we never talked about the format of experimental data, but that would  
be the things to be described here, be it CVS, SBRML or whatever

I think of manifests similar as to jar files, it tells me precisely  
what i need to know in order to run (the experiment description)  
successfully

>
> If we agreed that the manifest was the 'first' file in the archive
> then tools wouldn't even have to unzip all the files to find it.

thats actually harder to do (in some api's) than to agree on a  
predefined fileName. If we need a manifest then 'MANIFEST' seems as  
good a name as any. Which begs the next question, which format should  
the manifest take? For all those reasons  decided not to include a  
manifest in the first place.

>
>
> Other than that, the format Frank proposed seems quite effective -
> Frank I hope to be able to test out some .miase files generated by the
> java library on your website soon (at least before the Biomodels
> meeting).

That would be great,
thanks

best
Frank

>
>
>
> Thanks
>  Richard
>
>
>
>
>
>
>
>
>
> -- 
> Dr Richard Adams
> Senior Software Developer,
> Computational Systems Biology Group,
> University of Edinburgh
> Tel: 0131 650 8281/8285
> email : ric...@ed...
>
> -- 
> The University of Edinburgh is a charitable body, registered in
> Scotland, with registration number SC005336.
>
>
>
> ------------------------------------------------------------------------------
> Open Source Business Conference (OSBC), March 24-25, 2009, San  
> Francisco, CA
> -OSBC tackles the biggest issue in open source: Open Sourcing the  
> Enterprise
> -Strategies to boost innovation and cut costs with open source  
> participation
> -Receive a $600 discount off the registration fee with the source  
> code: SFAD
> http://p.sf.net/sfu/XcvMzF8H
> _______________________________________________
> Miase-discuss mailing list
> Mia...@li...
> https://lists.sourceforge.net/lists/listinfo/miase-discuss