odf-adoption message
[Date Prev]
| [Thread Prev]
| [Thread Next]
| [Date Next]
--
[Date Index]
| [Thread Index]
| [List Home]
Subject: Interop camp example
- From: robert_weir@us.ibm.com
- To: odf-adoption@lists.oasis-open.org
- Date: Wed, 20 Jun 2007 15:13:50 -0400
A few words to explain what I was proposing
for an interoperability camp.
I was thinking we could concentrate
on word processor format interoperability, since we're more likely to get
a critical mass of participants if we focus. The idea would be to
get those who knew the internals of their word process and came with the
source code and the ability to prototype changes during the camp.
So we want the coders.
In advance of the camp, we would send
out PDF files illustrating some sample documents. In advance of the
camp, each vendor would attempt to replicate that document to the best
of their application's abilities. They would then save the document
in ODF format.
An example document might be to replicate
the first page of this PDF: https://www.socialsecurity.be/foreign/en/employer_limosa/infos/documents/pdf/act_mb_28122006.pdf
This is an example of Belgian statute
format, where the text is given in parallel translation. To get the
proper alignment would require probably the use of tables with hidden borders.
There are also some challenging aspects of getting the numbered lists
and list levels correct .
When we send out the PDF we would also
provide some annotations to explain some finer points of the document that
might otherwise be overlooked, like the use of em-dahes and en-dashes in
the page header.
Depending on the complexity, we might
have 5 or so documents that we do in this way, each one emphasizing different
practical issues concerning interoperability.
We would collect the ODF's from each
vendor in advance of the camp, and distribute them to everyone at the camp,
or a few days in advance. So if we have 5 vendors, then everyone
would have 5 different ODF versions of each document.
And what do we do at the camp? Say
we have 5 vendors and 5 target documents. That gives each vendor
20 documents to try loading, giving 100 different interoperability tests
we can perform. How do we handle this? One way would be to
set aside an hour for each document and have each vendor take a turn first
explaining any issues they had in creating the document originally, and
then they could load each other vendor's version of that document, and
the group could note what new problems, if any, show up. If each
vendor can be projecting their application on the screen while doing this,
we can all look and observe together, and do this efficiently.
We would then want to reserve some time for coding, and some time at the
end for vendors to demonstrate any fixes, and some time to discuss next
steps.
As you can see, this is going to be
time consuming. So maybe we would only be able to do 2-3 test documents?
-Rob
___________________________
Rob Weir
Software Architect
Workplace, Portal and Collaboration Software
IBM Software Group
email: robert_weir@us.ibm.com
phone: 1-978-399-7122
blog: http://www.robweir.com/blog/
[Date Prev]
| [Thread Prev]
| [Thread Next]
| [Date Next]
--
[Date Index]
| [Thread Index]
| [List Home]