Print

Print


Birnbaum, David J wrote:
> Dear Viv (cc TEI-L),
>
> Thank you for your response. One of my concerns has been precisely how to keep the metadata associated with the image, since although I can link them easily enough on my own site (by, for example, embedding both metadata and image in a wrapper page), someone can easily copy just the image file and lose the metadata in the process. Embedding the metadata into the image file itself would ensure their integrity. I understand that digital cameras often associate EXIF metadata with the image automatically, although I don't know how reliably that is preserved when one converts an image from the camera's native format to something else. I hadn't known about XMP until I read your posting, and I immediately followed up on the Adobe site, but it sounds as if (as you note) XMP is currently somewhat difficult to use because of lack of tools. I wonder also whether this means that XMP is not sufficiently recognized as best practice for it to be a good choice for embedded metadata. Are there alternatives that might have a better chance of being supported/recognized across applications? Is the idea of embedding metadata too new for best practices to have emerged?
>   

I don't know whether the idea is too new, but certainly support of 
embedded media related metadata across applications is an open issue. 
See e.g. the just starting W3C Media Annotations Working Group
http://www.w3.org/2008/01/media-annotations-wg.html

Felix


> Sincerely,
>
> David
> [log in to unmask]
>
>
>   
>> -----Original Message-----
>> From: viv [mailto:[log in to unmask]]
>> Sent: Friday, August 22, 2008 9:33 AM
>> To: Birnbaum, David J
>> Cc: [log in to unmask]
>> Subject: Re: images and image metadata
>>
>> It is great to see this discussion.
>>
>> I don't disagree with any of the valuable comments made but would say
>> from my own perspective being concerned with digital preservation that,
>> like the TEI, I lean towards "openness" and closely coupling metadata
>> (such as in the TEI header) with its referent.  Also I distinguish
>> between the preservation format and any publication or dissemination
>> format.
>>
>> Hence in respect of images I am implementing an XMP(1) based solution
>> with a PNG(2) image preservation format.  The XMP allows me to embed
>> metadata concerning the individual image in the image file.
>>
>> Sadly as yet there is a shortage of tools to help so work is at proof
>> of
>> concept demonstration level using Perl hacks.
>>
>> Since I also promote TEI for preserving text where there is a need for
>> processing in future (as opposed to mere rendering) then I would be
>> delighted to see solutions for text/image integration.
>>
>>
>> Viv Cothey
>>
>> (1) XMP - extensible metadata platform, see adobe.com
>> (2) PNG - portable network graphics
>>
>>
>>     
>>> Dear TEI-L,
>>>
>>>
>>>
>>> I am organizing a small image-based project, involving the digital
>>> publication of a Russian festal icon with commentary. The icon
>>> contains 27 small images of church events (Annunciation, Nativity,
>>> etc.), arranged on the board much like a comic strip (an example of
>>> this type of icon is available at
>>> http://www.dormitionskete.org/DsWebStore/images/DsProducts/MS-FD-
>>>       
>> Cretan_org.gif). I would like the electronic edition to include
>> photographs of the icon as a whole and of each of the small images,
>> with textual information attached at both levels.
>>     
>>>
>>> I don’t have much experience with descriptive metadata for image
>>> files, or, for that matter, with image resources in general, and I’d
>>> be grateful for any information anyone might provide about 1) image
>>> formats (I envision something high quality for archival purposes and
>>> lower-resolution for web delivery--but I don’t know what resolutions,
>>> color depths, file formats, etc. are consistent with what
>>> professionals regard as best practice) and 2) descriptive metadata
>>> (whether art historical, photographic, or other--again, in a way that
>>> is consistent with professional best practice). I’d also welcome
>>> pointers to similar projects others might have developed, exposure to
>>> which might broaden my understanding of what’s possible and what’s
>>> desirable.
>>>
>>>
>>>
>>> The TEI guidelines don’t seem to deal with this type of project
>>> explicitly, which is reasonable, since it is not fundamentally a
>>>       
>> “text
>>     
>>> encoding” activity. Nonetheless, I suspect that many readers of TEI-L
>>> have dealt with these issues, and I’d be grateful for any advice or
>>> suggestions. I have poked around on the web in search of metadata
>>> standards for images, but those resources don’t give me a good sense
>>> of what people are doing in actual projects.
>>>
>>>
>>>
>>> Sincerely,
>>>
>>>
>>>
>>> David
>>>
>>> [log in to unmask]
>>>
>>>
>>>
>>>
>>>       
>> --
>> Scanned by iCritical.
>>