tm-pubsubj message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]

Subject: Re: [tm-pubsubj] Requirements oddity

Warning: longish!

Peter,

Finished up some more admin stuff this morning so back to the draft! 
Going slower than I hoped. :-(

Peter Flynn wrote:
> On Sat, 2003-09-13 at 14:35, Patrick Durusau wrote:
> [...]
> 
>>"A Published Subject Indicator may provide human-readable metadata about 
>>itself."
>>
>>But the "metadata" isn't really about "itself" is it?
> 
> 
> I think what we were trying to say is that the PSind can have metadata
> about itself *as well as* about the PSident (as you say).
> 
> 
Well, that makes sense in a way, but may illustrate a lack of clarity in 
our thinking or the prose.

Consider that I have the PSident:

"http://psi.fruits.org/apple.html"; // note change from committee spec

Now, in order for that to be a Published Subject, it must "must resolve 
to an human-interpretable Published Subject Indicator." (Requirement 2)

So, lets construct the Published Subject Indicator, apple.html:

<head>
<meta name="keywords" content="apple" />
<meta name="location" content="http://psi.fruits.org/apple.html"; />
<meta name="type" content="published subject indicator" />
<meta name="publisher" content="Johnny Appleseed" />
<meta name="date" content="20030914" />
</head>
<body>
<p>Apple: Round firm fleshy fruit of a rosaceous tree</p>
</body>

OK, do these two together meet the requirements and recommendations?:

Requirement 1: URI? yes

Requirement 2: Resolvable? yes

Requirement 3: State its URI? yes

Recommendation 1: human-readable metadata about itself? yes

Recommendation 2: machine-interpretable metadata about itself? yes

Recommendation 3. 1 and 2 are consistent? yes

Recommentation 4. Declares itself a PSI? yes

Recommendation 5: Declares its publisher? yes

Question 1: What in this HTML page is about the Published Subject 
Indicator and which is about the Published Subject Identifier? Is there 
a difference?

Question 2: More specifically, is the date metadata about the Published 
Subject Indicator or about the Published Subject Identifier?

Would it make a difference if it read:

<body>
<p>Apple: Round firm fleshy fruit of a rosaceous tree</p>
<p>Date: 20030914</p>
</body>

Question 3: Declare itself to be a PSI? Shouldn't it say Published 
Subject Indicator? It is obviously not a Published Subject Identifier.

One problem I am wrestling with is how do machines discover information 
about Published Subject Identifiers?

If the semantics of metadata in a Published Subject Indicator were 
declared to be "about" the Published Subject Identifier, we could have:

<rdf:RDF
     xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#";
     xmlns:dc="http://purl.org/dc/elements/1.1/"/>
    <rdf:Description
        rdf:about="http://psi.fruits.org/apple.html";
        dc:publisher="Johnny Appleseed"
        dc:description="Apple: Round firm fleshy fruit of a rosaceous tree"
        dc:type="Published Subject Indicator"
        dc:date="2003-09-14" />
</rdf:RDF>

(Does not have to use RDF, could declare a <link> element in XHTML and 
prepend dc. to all the same names inside meta elements.)

Which not only meets all the requirements and recommendations but is 
also machine-processable metadata "about" the Published Subject 
Identifier. (Actually since we suggested consistency between 
human/machine metadata, we could simply say it applies to both?)

While it is true that topic map processors will only "match" the 
Published Subject Identifiers for subject identity purposes, making it 
easy for other software to seek out and store metadata from Published 
Subject Indicators "about" Published Subject Identifiers, looks like a 
good strategy.

Comments?

Hope everyone is having a great day!

Patrick

-- 
Patrick Durusau
Director of Research and Development
Society of Biblical Literature
Patrick.Durusau@sbl-site.org
Chair, V1 - Text Processing: Office and Publishing Systems Interface
Co-Editor, ISO 13250, Topic Maps -- Reference Model

Topic Maps: Human, not artificial, intelligence at work!

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]