Re: [xml-dev] a report on any xml file, what information is useful?

[ALT Mobile <dev@altmobile.com>]

Your list is a good start but confuses vocabulary specific analytics 
with generic XML issues. For example in your list you mention link 
analysis. WML links are different than XHTML links which are different 
from SVG links.


In the v6 release of our <alt> XML Studio, we implemented support for 
generic XML Analytics.

The technology is called Node Insight and it provides both interactive 
metrics as well as report generation of various XML statistics. As our 
tools are XML centric rather than text centric (or in Infoset speak: 
DOM-based rather than serialization-based), we were able to weave 
analytics into our core XML visualization technologies.

The product documentation for Node Insight is found here:

http://altmobile.com/Products/XML%20Studio/Product%20Documentation/1147831669593.html


For interactive use, the most important analytic is insight into node 
type (that is, attribute node vs text node as this is the most common 
problem for beginning XML developers), namespace usage (which is 
probably the most common problem in XPath development), and node sizes 
(which is important when viewing server generated XML as you do not want 
to overwhelm the user interface). A quick sample for these types of 
metrics can be seen here:

http://altmobile.com/sample_shots/namespace_insight_dom_browser_1.png


For detailed XML analytics our Node Insight implements these items:

1. Document specific XML analytics: such as document size, location, and 
namespace usage. In our product documentation, we explain why these 
metrics are important and how they influence XML architectures and 
implementations.
2. Node specific XML analytics: such as namespace information, element 
sizes, attribute values, and other items.


A sample Node Insight report can be seen here:

http://altmobile.com/sample_shots/node_insight_report_document_1.png


And for us, we provide both interactive point-and-click access to the 
Node Insight capabilities as well as XML-RPC web service APIs enabling 
remote access from tools chains perhaps written in Applescript, Perl, 
Ruby, etc. A pure Java language API is available for embedding.

Higher-level XML features such as analytics and differencing are harder 
to implement in editing tools because of the text-centric nature of the 
user interface. This is because the user has to constantly change 
focus/views. For this reason, we implemented our XML tools on a 
visualization core and are able to leverage all of the innovations in 
direct manipulation and object oriented user interface technologies.


As we like to say in describing the importance of XML visualisation and 
XML analytics:

"Visualization fosters comprehension and analytics fosters insight"



Good luck with your project.

--Zaid

http://altmobile.com



bryan rasmussen wrote:
> Hi,
>
> If you were generating views of XML files for display in some editor
> type environment what would people be interested in seeing in sort of
> a side display, things I can think of offhand -
>
> 1. Number of namespaces used
> 2. number of namespaced elements
> 3. Does it use specific namespaces
> 4. Any possible links in the documents
> 5. IDS and IDREFs?
> 6. XML Schema references?
>
> Anything else?
>
> Cheers,
> Bryan Rasmussen