[Zope3-dev] renaming MementoBags..conclusions?

Jeffrey P Shell jeffrey@cuemedia.com
Mon, 13 May 2002 16:33:04 -0600


On 5/12/02 4:17 PM, "Guido van Rossum" <guido@python.org> wrote:

>> But does a memento-bag only hold meta-data?
>> 
>> I think it can be used to hold a lot more kinds of data besides
>> meta-data of the object it is associated with. In fact, I think it will
>> very rarely be used to hold meta-data of an object it is associated with.
>> 
>> Or, perhaps there's a less-strict use of the term meta-data that I'm not
>> aware of.
>> 
>>    http://www.dictionary.com/search?q=Metadata
>> 
>> I would not like to confuse the ideas of "associated with an object" and
>> "information about an object or its structure".
>> 
>>> (Jim and Guido both seemed to favor this to my proposal, IAssociatedData)
>> 
>> I prefer this spelling because to my ear it more accurately describes
>> the purpose of the thing.
> 
> Depends on the ear :-).  To me it's just more verbose.  Unless you
> restrict the definiton of metadata to the Dublin Core, the concept of
> metadata is vague enough that anything you'd want to be associated
> with an object can be considered metadata.
> 
> What to call this is not a new issue, by the way.  MacOS calls it
> "resources" (its "resource fork" is pretty similar to what's in the
> memento bag).  AtheOS calls it "attributes".  I think both those terms
> are too general.

And speaking of resources - I have seen some interesting debates flare up
over whether Resource Fork data is metadata or not.  I guess the debates
over IMetaData have already been resolved out, but I thought I'd share this
anyways: 

> In general, the resource fork did not contain metadata. It certainly did not
> contain any of the pieces of metadata discussed so far (name, size, type,
> creator, etc.) In the case of applications, the resource fork did contain data
> about what types of files the application could read, and so on, which
> qualifies as a type of metadata specific to applications. But the vast
> majority of the contents of a given file's resource fork (if it had one) was
> data, not metadata.
-- http://arstechnica.com/reviews/01q3/metadata/metadata-6.html

For me,

+.5 for IAnnotation
+1 for IAttachedData

-.4 for 'IAttachData' - it reads a little too 'pronoun-verb-noun'-ish, and
every time I read it I want to say "oh you DO now, do you?  Well, IPFreely,
how 'bout THAT!"

-- 
Jeffrey P Shell 
www.cuemedia.com