Different Formats of XML????

[Eric (25)]

Hi Folks,
Can someone point me in the direction of some info on the differences between the following two formats of XML? I work in an ORG that utilizes both so I'd like to start gathering some data on which one is better so that we may consolidate.

Any help would be much appreciated. Thanks!

<DATASET>
<DATAROW FNAME="Eric" LNAME="Langland" />
<DATAROW FNAME="Larry" LNAME="Ellison" />
</DATASET>

<DATASET>
<DATAROW>
<FNAME>Eric</FNAME>
<LNAME>Langland</LNAME>
</DATAROW>
<DATAROW>
<FNAME>Larry</FNAME>
<LNAME>Ellison</LNAME>
</DATAROW>
</DATASET>

 

[Andy Dingley]

Can someone point me in the direction of some info on the differences between the following two formats of XML?

You're standing in it. Do some Google on subject lines of "element vs. attribute"

 

[Mike McCarthy]

I was just looking at this today - W3 schools suggests the second option, ie. using child elements.

Check - http://www.w3schools.com/xml/xml_attributes.asp

 

[Stefan Ram]

Can someone point me in the direction of some info on the differences between the following two formats of XML? <DATAROW FNAME="Eric" LNAME="Langland" /> <DATAROW> <FNAME>Eric</FNAME> <LNAME>Langland</LNAME> </DATAROW>

In XML, there is a discrepancy between the semantics and syntactical restrictions. The result is that today no one can answer this question with certainity.

Usually, there are two kinds of predicates: Unary predicates P, written as Px, and binary relations R, written as xRy.

The assertion you want to notate is:

x is a data row and
The FNAME of x is "Eric" and
The LNAME of x is "Langland" .

In a more formal notation, using unary predicates and binary relations this would mean:

data_row(x) ^
FNAME(x,"Eric") ^
LNAME(x,"Langland") .

Using the guideline that it enhances readability and
usability when syntactic structures are related to
semantics, this suggest to use element types for unary
predicates and attributes for binary relations, so:

<DATAROW FNAME="Eric" LNAME="Langland" />

But - this is not always possible in XML, because type and attributes can not always be repeated as needed and the values of attributes may not be structured.

So, some suggest to always use subelements and to ignore attributes, while others flip wildly between attributes and subelements with out any clear rules or even use attributes whenever posssible just because it is easier to enter them with their current XML-editor.

My own solution is to use "Unotal" instead of XML. Unotal allows repeated types and attributes and it allows structured attributes. So I can tell between subelements and attributes by their meaning and not by arbitrary syntactical restrictions. Thus, I have a "clean source notation". If needed, I than can convert that to XML or whatever else is required. See also:

http://www.purl.org/stefan_ram/pub/unotal_en

 

[Eric (25)]

Great. Thank you both.

Cheers

Eric

I was just looking at this today - W3 schools suggests the second option, ie. using child elements.

Check - http://www.w3schools.com/xml/xml_attributes.asp