Hello Kate, personally, I don't think you need these attributes on <dao>
because I think the "show" and "actuate" attributes already define which
one is the first traversal (i.e. onLoad/embed) and which one is the
hyperlink. I do think you need to define a "localType" on <daoset> and
then define what that means for your institution and how it should be
encoded. For instance, you could define "dao_thumbnail_link" as "single
embedded thumbnail that hyperlinks to a larger representation of the
thumbnail image" and then define which elements/attributes this includes:
First <dao> with show/actuate set to "onLoad/embed". This is the
embedded "thumbnail" image that displays when the page loads.
Second <dao> set to show/actuate "onRequest/new". This is the target
link that the user clicks on.
You might also want to further refine the data type such as
"image/jpeg", "application/pdf", but this would be optional. I haven't
looked at the available attributes on <dao> to see where this
information might go.
I hope this helps.
Mark
On 4/21/2014 3:01 PM, Bowers, Kate A. wrote:
> Mark:
>
> Can you explain the need to attribute 'directionality' (that is 'start' and 'end'?) Who uses this and to what purpose? Why directionality ('start' and 'end') rather than, for example, 'displayImage' and 'targetObject'?
>
> I went into the xlink weeds a bit trying to figure out what arcrole and the like were for and ended up worse off.
>
> Your wisdom is much appreciated!
>
> Kate
>
> -----Original Message-----
> From: Encoded Archival Description List [mailto:[log in to unmask]] On Behalf Of Mark R. Carlson
> Sent: Monday, April 21, 2014 5:26 PM
> To: [log in to unmask]
> Subject: Re: [EAD] EAD3: <dao> or <daoset> for links with embedded thumbnails?
>
> Hello Kate, here is what I was proposing for EAD3:
>
> <daoset localType="dao_thumbnail_link">
> <dao localType="start" href="{url}" daotype="{value}" actuate="onLoad"
> show="embed"/>
> <dao localType="end" href="{url}" daotype="{value}" actuate="onRequest"
> show="new">
> <descriptivenote><p>Caption of image or hyperlink text</p></descriptivenote> </dao> </daoset>
>
> I was proposing that you create a "localType" attribute on <daoset> for this type of link.
> I think "localType" on <dao> is probably optional unless you wanted to define which <dao> is the thumbnail by attribute value rather than encoding practice.
>
> Mark
>
> On 4/21/2014 2:00 PM, Bowers, Kate A. wrote:
>> Mark:
>>
>> Thanks for responding... you're commenting on the current EAD, I think, though. Did you have any opinions on the EAD3, which is the bit I need to work out to make specifications for the programmer?
>>
>> Kate
>>
>> -----Original Message-----
>> From: Encoded Archival Description List [mailto:[log in to unmask]]
>> On Behalf Of Mark R. Carlson
>> Sent: Monday, April 21, 2014 4:36 PM
>> To: [log in to unmask]
>> Subject: Re: [EAD] EAD3: <dao> or <daoset> for links with embedded thumbnails?
>>
>> Hello Kate,
>>
>> I personally don't think what you've proposed is unorthodox in terms of encoding, use or practice.
>>
>> In xlink, the <arc> element spells out the direction that links are
>> to be traversed. I would have said that the first <arc> element goes
>> from "start" to "resource-1" and that the second <arc> element goes
>> from "resource-1" to "resource-2". Although there is no such mechanism
>> in EAD3, I would assign a "localType" attribute on <daoset> (for
>> instance
>> dao_thumbnail_link) and then define how that "localType" link should be encoded. Since the thumbnail has its attributes set to "embed/onLoad", that already implies that it's the thumbnail image. The second <dao> is used to encode the second "traversal" whose attributes are set "onRequest/new|replace". The only thing I'd change is to put the caption information in <descriptivenote><p> either in the first or second <dao> (you choose, but be consistent). If you don't think encoding practice alone is enough to define this use, use the "localType" attributes in <dao> to further clarify.
>>
>> Mark Carlson
>> University of Washington
>> Seattle, WA
>>
>> On 4/21/2014 9:56 AM, Bowers, Kate A. wrote:
>>> In order to pull together some specifications for a programmer, I need to come to grips, soon, with how one might encode a links to a digital archival objects via an embedded thumbnail in EAD3.
>>>
>>> I at least need to figure out the number of variables we need to allow for... so, for example, it would be really, really helpful to know how many attribute values overall might have variable values when encoding with a the goal of thumbnail + link as the html display.
>>>
>>> I do not know if TSEAD was thinking about embedded thumbnails as links to digital archival objects when developing <daoset> but there seems to be no other way to introduce two links that relate to a single digital object.
>>>
>>> In current EAD, we do this to embed a thumbnail:
>>>
>>> <daogrp xlink:type="extended">
>>> <resource xlink:label="start" xlink:type="resource" /> <daoloc
>>> xlink:href="[URN OF EMBEDDED THUMBNAIL]" xlink:label="resource-1"
>>> xlink:type="locator" /> <arc xlink:from="start" xlink:show="embed"
>>> xlink:to="resource-1" xlink:type="arc" /> <daoloc xlink:href="[URN TO
>>> DIGITAL ARCHIVAL OBJECT]" xlink:label="resource-2"
>>> xlink:type="locator"> </daoloc> <arc xlink:from="start"
>>> xlink:show="new" xlink:to="resource-2" xlink:type="arc" /> </daogrp>
>>> The HTML result can be seen in finding aids like this one (scroll
>>> down to "Inner crater of El Misti, Arequipa, Peru") :
>>> http://nrs.harvard.edu/urn-3:HUL.ARCH:hua03008
>>>
>>>
>>> In EAD3, I'm thinking it would look something like this, which is blessedly more straightforward than current, but it still strikes me as a kluge that subverts the intention of daoset:
>>> <daoset coverage="whole">
>>> <dao daotype="other" linktype="simple" href="[URN OF EMBEDDED
>>> THUMBNAIL]" actuate="onLoad" show="embed" otherdaotype="thumbnail"/>
>>> <dao daotype="derived" linktype="simple" linktitle="LINK TEXT"
>>> actuate="onRequest" href="[URN TO DIGITAL ARCHIVAL OBJECT]"
>>> show="new"/> </daoset>
>>>
>>> Am I right that this is subverting the intention of <daoset>? Can anyone suggest on a more orthodox option?
>>>
>>> Any help would be appreciated!
>>>
>>> Thanks,
>>>
>>> Kate
|