Metatagger and Datadeploy

wmh

Good day all,
We have just begun to really start our implementation of Metatagger here and I would love to hear some tips, suggestions, and/or war stories to help us implement this right.

This is requested more from the application setup than the Taxonomy/vocabulary perspective.

Some questions I have are:
1) is Datadeploy to a DB (oracle) the right direction, or has more success been had by pulling from the EA files into XML directly. (or other methods).
2) What other IW applications can be used to further add functionality to this (opendeploy?)

Thank you and look forward to the responses.

-Marty

Adam Stoller

Can you provide more context regarding what you want to do with the metadata?

I think that might have a big bearing on what applications can be used to assist you and/or how to use them in conjunction with one another.

--fish
(Interwoven Senior Technical Consultant)

wmh

Our overall goal is to build a Dal/Dam with search capability on items such as document/page owner, Overall subject matter and specific document topic.

Most of the metadata needs to be kept outside of the document, with in a DB or the EA data.

Adam Stoller

I'm not an expert in such matters - but I think that using DataDeploy to transfer the metadata / extended attributes to a DB will probably be the right direction to go in here. Others who are more familiar with what you are trying to accomplish will hopefully pipe in and contribute additional ideas.


--fish
(Interwoven Senior Technical Consultant)

Migrateduser

I think it will depend on the underlying search technology of your DAM application. If it is RDBMS backed then DD with DAS is probably the way to go. If you have some other search technology in mind like Verity then export of EAs to XML could work better.

Migrateduser

In addition to leveraging DD, an alternative approach to building a digital asset library is to leverage the newest product in Interwoven's portfolio: our own DAM system from our recent acquisition of MediaBin.

You can find out more about MediaBin in our upcoming webcast, which you can register for at our www site. If you have complex DAL requirements, a MediaBin approach may be prove a better long-term solution.

Kevin Cochrane
Interwoven

wmh

Our current plan was to use Verity K2 for the search functionality.

I will also investigate the new IW product MediaBin.

Any other tips, suggestions, lessons hard learned?

Migrateduser

What about not using MetaTagger and instead use TeamSite templating? User would go to a template and upload the asset and fill out any needed meta data fields. When done, a workflow could read the generated DCR and create an XML file that Verity can index.

The advantage might be more flexibility over the user interface - the MetaTagger GUI won't allow you to do JavaScript tricks such as having the contents of one form drop-down be based on what was selected in another drop down.

What do people think of this approach? Integrating a vocabulary created by MTStudio might be tricky, but I guess possible.

I'm not suggesting this approach, but was curious on feedback.

David

Adam Stoller

The general limitation / cost of this approach is that you need to create "wrapper" DCRs around all your non-HTML assets. If you use the extended attribute approach you can tag any kind of asset without the need for this "wrapper".

As with most tasks - there's definitely more than one way to approach it - but each method has it advantages and disadvantages - I believe that we are working to reducing the discrepancy between the UI features of Templating and MetaTagger so hopefully the approach of using "wrappers" will become less necessary when trying to present a seamless interface to the process as a whole....

--fish
(Interwoven Senior Technical Consultant)

NathanW

In our installation we use DD to send our EAs to an Oracle DB for 2 reasons:
1) We use Oracle Intermedia indexing for site search
2) We have custom scripts executed in some cases from scheduled tasks that read the db and generate XML files in specific formats

The main benefit of point 2) for us is we can capture the metadata once - deploy it to a central repository - then repackage it however we want.

Nathan Wall
Department of Transport and Regional Services (Australia)

Adam Stoller

That's fine - the piece you miss with setting EA's with MetaTagger is the ability to have automated and 'intelligent' metadata (guided based on automated processing and controlled vocabularies) set.

Again - a lot depends on the needs of the organization - the larger the site, the more efficient ways of collecting metadata and ensuring that it is consistent from asset to asset becomes more important

--fish
(Interwoven Senior Technical Consultant)