[MOBY-l] MobyDataType 'Object'

Ola Spjuth ola.spjuth at farmbio.uu.se
Fri Oct 19 16:36:24 UTC 2007


Thanks for the help, I understand a lot better now.

Cheers,

    .../Ola

On Oct 19, 2007, at 15:29 , groscurt at mpiz-koeln.mpg.de wrote:

> Hi,
>
> In the moby world the basic datatype is an object. Any object can be
> identified by its id and a namespace. Both values guarantee to  
> describe
> the object explicitly.
>
> As an example:
>
> 00492215062447
>
> does not tell you anything for what this number / string stands  
> for. With
> the namespace "telephone number" it becomes clearly what its for.
>
> The same id can be used differently - e.g. with the namespace "social
> number" it is a different object than the one with the namespace
> "telephone number" - so with the same content (the id), they are
> semantically different (because of the namespace).
>
> If a MobyService has as input such kind of object - e.g. a database
> identifier - the MobyObject has to be used. There is no more  
> information
> one needs, because its all defined with these object.
>
> The DataType String is semantically incorrect because not all  
> identifiers
> are Strings (see telephone examples).
>
> Therefore dont use MobyDataString if an Object is needed.
>
>
> Thats my point of view - please correct me or add to that
> andreas
>
>
>
> On Friday 19 October 2007 14:41, Ola Spjuth wrote:
>> Hi,
>>
>> How should the primitive MobyDataType 'Object' be treated? If a
>> service only has a input with a primitive 'Object', what does it  
>> expect?
>>
>> Example (implemented in jMoby):
>> The service 'getKeggPathwayAsGif' has primary input name='pathway'
>> with MobyDataType 'object', hence instantiating from this
>> datatypeName creates a MobyDataObject, which cannot contain a value
>> (the static method createInstanceFromString(typename, value) sets the
>> 'value' parameter as ID for the MobyDataObject). Is this intentional
>> and in that case what is the intention?
>>
>> Why does not the service use the DataType 'String' if that's what it
>> expects? Is the dirty workaround to always supply a MobyDataString if
>> a DataType is set to Object?
>>
>> Cheers,
>>
>>     .../Ola
>> _______________________________________________
>> moby-l mailing list
>> moby-l at lists.open-bio.org
>> http://lists.open-bio.org/mailman/listinfo/moby-l
>
> -- 
> Andreas Groscurth
> Diplom Bioinformatik - PhD Student
> Max Planck Institute for Plant Breeding Research
> Carl-von-Linné-Weg 10
> 50829 Cologne
> Germany
> E-mail:    groscurt at mpiz-koeln.mpg.de
> Phone:    +49(0)221-5062-447
>





More information about the moby-l mailing list