Deitra-- In the Jackson finding aid referenced below, I decided to use the
EAD <table> element for positioning the images within the flow of the
text, but if you don't want to do that, you can just use <linkgrp>
directly within <p>.  Since the EAD <table> element doesn't have an
attribute where you can specify table alignment, I am using the
"altrender" attribute which is mapped to the "style" attribute in HTML:

<table style="float: right;">

The use of the child elements of <linkgrp> is based on a short article I
wrote on the use of XLink in EAD:

http://staff.washington.edu/carlsonm/XLinkToEAD.pdf

I am also using some javascript code in the "altrender" attribute of <arc>
to open the larger image in a popup window.  The HTML code this produces
is:

<a href="..."
onclick="javascript:window.open(this.href,'','width=362,height=297');
return false;><img src.......>

<table altrender="float: right;">
<tgroup cols="1">
<tbody>
<row>
<entry>
<linkgrp linktype="extended">
<resource label="start" linktype="resource"></resource>
<extptrloc label="small"
href="http://www.lib.washington.edu/specialcoll/manuscripts/finding_aids/jackson/ead/jackson1sm.jpg" linktype="locator"/>
<extptrloc label="large"
href="http://www.lib.washington.edu/specialcoll/manuscripts/finding_aids/jackson/ead/jackson1lg.jpg" linktype="locator"/>
<arc from="start" to="small" show="embed" actuate="onload" title="Click on image to enlarge" linktype="arc"/>
<arc from="small" to="large" show="new" actuate="onrequest" altrender="javascript:window:width=350,height=275" linktype="arc"/>
</linkgrp>
</entry>
</row>
</tbody>
</tgroup>
</table>



On Tue, 15 Feb 2005, Dietra Stanley wrote:

> Hello all!
>
> I am doing the same as Mark has suggested....i have run into so much
> difficulty using <dao> , that I just use the <extref>.  I am looking to
> place thumbnails of the actual larger image directly into the text so if
> anyone out there has examples of how they coded thumbnails directly beside
> the descriptive text please HELP :)) .  I will for now take a look at this
> link you provided below.
> ----- Original Message -----
> From: "Mark Carlson" <[log in to unmask]>
> To: <[log in to unmask]>
> Sent: Monday, February 14, 2005 4:59 PM
> Subject: Re: use of <dao>
>
>
>> <dao> is not a valid EAD element within <p> and since it is not a PCDATA
>> element so you can't use it to create a textual link as you describe
>> below.  The subelement <daodesc> is not meant to be used in this fashion
>> either.  I brought this up several months ago (the fact that you can't
>> link to a DAO within elements like <scopecontent><p> like you can with
>> <extref><extptr>.  But the fact that you are linking to a DAO is only
>> important for portability.  You could use any linking element as long
>> as you could flag it somehow so that you could get that information out
>> later should you need to do so.  If it were me, and I wanted to do what
>> you describe, I would just use <extref> instead and then use the term
>> "dao" in the "role" attribute of <extref>:
>>
>> [[If you were actually embedding DAO images within the text, I would
>> suggest something else.  See our Henry M. Jackson finding aid as an
>> example:
>>
>>
> http://www.lib.washington.edu/specialcoll/findaids/docs/papersrecords/JacksonHenry3560.xml
>> ]]
>>
>> <scopecontent><p><extref href="..." show="new" actuate="onreqest"
>> role="dao">Coloring books</extref>, <extref href="..." show="new"
>> actuate="onrequest" role="dao">puzzles</extref>, ... </p>
>> </scopecontent>
>>
>> Other than this, there is no easy way (IMO) to do this within elements
>> like <biohist> and <scopecontent>.
>>
>> P.S. Don't forget to use the "behavior" attributes "show" and "actuate"
>> when using linking elements.
>>
>> Mark
>>
>> On Mon, 14 Feb 2005, Amy McCrory wrote:
>>
>>> I'm processing a collection that contains many examples of various types
> of
>>> merchandise.  We're going to create digital images of representative
>>> objects, rather than show every item.  My question has to do with how
> we're
>>> going to link to these image files.
>>>
>>> In the past, I have always linked to a <dao> from a component-level
> record
>>> describing one named thing.  But in this collection, for certain groups
> of
>>> materials, I prefer to write collective descriptions, such as:
>>>
>>> [lots of tags omitted]
>>>
>>> <unittitle>Walt Scott merchandise</unittitle>
>>> <unitdate>1954-1964</unitdate>
>>> <scopecontent><p>Coloring books, puzzles, squeeze toys, party
>>> decorations.</p></scopecontent>
>>>
>>> --and put the links into the <scopecontent> text:
>>> <scopecontent><p><dao href="-------">Coloring books, </dao><dao
>>> href="--------">puzzles, </dao><dao href="--------">squeeze toys,
> </dao>,
>>> <dao href="--------">party decorations.</dao></p></scopecontent>
>>>
>>> Does anyone see a potential problem in doing this, rather than creating
> a
>>> separate <c> record for each type of item?  It's nice to have a
> <unittitle>
>>> that matches the title or caption in your image file metadata, but is it
>>> necessary?  Any comments--or even better, experience with this--would be
>>> most welcome.
>>>
>>> With thanks,
>>>
>>> Amy McCrory
>>> Archivist, Cartoon Research Library
>>> 023L Wexner Center
>>> The Ohio State University
>>> Columbus, OH
>>> (614) 292-0538
>>>
>>
>