After Lotusphere 2008, we wanted to collect all the session presentations in one place for our internal knowledge management.
There are two sources available to get both the information about the LS sessions and the PDF files of the speakers:
Genii Software's 8th Annual Lotusphere Agenda Database and
Lotusphere Presentation Batch Download from Stephan Wissel
So how do you combine the two data sources and get a nice visualization like the following?
(you can see the whole picture when you click on the image)
We added a simple Lotusscript agent in the Genii database, that reads all the session documents from the view "Session ID" and produces MPXML, the MindPlan XML format, out of them.
The agent took about an hour or two from coding to importing the Lotusphere data.
The MindPlan XML format is described in the MindPlan SDK. In short, with MPXML you may not only create new nodes (like the track 1-4 nodes and the root node "Lotusphere 2008 - Sessions by Track" you can see above), but you can also import already existing Notes documents as MindPlan nodes.
<mpnode>
<title contenttype='text/plain' contentencoding='xml'> Lotusphere 2008 -
Sessions by Track
</title>
<mpnode>
<title contenttype='text/plain' contentencoding='xml'>Track 2 - Application Development</title>
<importnotesdoc dbid='852573AA0016C628'
unid='3AFC6E3E4ADFB6AB852573AA0017D564' dbpath='LSESS08.nsf' server='' importas='copy'>
<type value='Branch'/>
<title contenttype='text/plain' contentencoding='xml'> AD101 Building Social Software Solutions for Your Applications</title>
</importnotesdoc>
</mpnode>
</mpnode>
</MindPlanXML>
That's what we did with the session documents, which preserves the body field containing the session details and all the other fields.
Ok, we automatically define our own MindPlan structure and put together new nodes and import Notes documents. But what about the attached PDF files in the picture?
Well, we used Stephan Wissel's batch file (thanks for that!) and downloaded all of the files to the locale machine. In our Lotusscript agent, we then search for files like "-sessionid-.pdf", e.g. "AD101.pdf".
If a file exists, the agent adds an < attachments > block in the MPXML code, that tells MindPlan to add an attachment after creating/importing the nodes.
<importnotesdoc dbid='852573AA0016C628'
unid='3AFC6E3E4ADFB6AB852573AA0017D564' dbpath='LSESS08.nsf' server='' importas='copy'>
<attachments>
<attachment uri="file:/c:/temp/AD101.pdf" />
</attachments>
</importnotesdoc>
...
With the contents of the MindPlan SDK, you can also create your own node documents without MPXML, even in a background agent, because we describe the field values and design elements of the MindPlan database in detail.
A use case I would think of is an automatic MindPlan-based company portal map, that aggregates information from several sources, like
- RSS feeds
- recently added documents in Lotus Notes databases
- some sort of text mining tools, that cluster text (e.g. from spidering the company intranet) and use MindMap technique to visualize the results
- and of course static contents, e.g. a bulletin board with shared web and Notes favorites or upcoming company events
MindPlan XML is an easy way to import any data into MindPlan and export node information to third party applications. The MindPlan SDK lets you automatically create MindPlans via code and also add your own forms and views in the MindPlan database.
P.S.:
Just a short hint for those who try to adapt the MPXML above and enter some valid unid/dbid/dbpath, to test the Notes document import:
You do not need to write the code in a MPXML file and import it via "File/Import/Import from file...". Just copy the adapted code into the clipboard, select a MindPlan node and paste the code into the MindPlan application. That invokes the same import mechanism.