If anyone does want to work on it, I'd be happy to help. Maybe I'll 
contact clay.

The most immediate and clear need I see is for application/marc+xml and 
application/mods. 

MADS could be useful, I dunno. Not sure if a seperate one would be 
needed for MFHD?

With all the effort on making web-friendly APIs for library 
bibliographic control systems (DLF task force, jangle, etc.), having 
MIME types for these formats will make everything flow much more 
smoothly and clearly.

Of course, even without them being registered, we can use 
application/x-marc+xml and application/x-mods right away, which is 
probably what I'll do.

Jonathan

Ross Singer wrote:
> His point, though, is that you can't tell the format being used until
> you open the document and try to negotiate it that way.
>
> So if you think in terms of content-negotiation and a particular
> resource is available in EAD, MARC XML and Dubin Core, you have no way
> of expressing that.
>
> Jonathan, this has come up before.  Ed Summers and I kicked around the
> idea of registering these but never got anywhere (mainly because
> neither one of us was really interested in writing the RFCs).  Clay
> Redding might be doing something, as I recall...
>
> -Ross.
>
> On Wed, Feb 4, 2009 at 10:52 AM, Ethan Gruber <[log in to unmask]> wrote:
>   
>> Correct me if I'm wrong, but wouldn't the mime type for MARC-XML and MODS be
>> application/xml, like every other xml file?  As for MARC-binary, I can't
>> say.  I don't have any of those files handy.
>>
>> Ethan
>>
>> On Wed, Feb 4, 2009 at 10:47 AM, Jonathan Rochkind <[log in to unmask]> wrote:
>>
>>     
>>> I am actually rather shocked that it seems that MARC-XML, MODS,
>>> MARC21-binary, do not have registered Internet Content Types (aka MIME
>>> types).
>>>
>>> Am I missing something, or is this really so?
>>>
>>> Anyone know what the process is for registering such?  Anyone want to help
>>> try to do that? I guess we'd probably have to talk to the standards
>>> organizations for each of those types, rather than doing it independently?
>>>
>>> Jonathan
>>>
>>>       
>
>