On 27/03/2014 6:26 PM, Vincent Hennebert wrote:
(CC-ing to general@ as it’s the place where I believe the conversation
should take place.)

I too would be in favour of welcoming DITA-OT as a sub-project of XML
Graphics. I think it matches rather well the purpose of this project.

A few comments:

On 27/03/14 18:32, Ron Wheeler wrote:
I think that you are right about being a sister project to FOP within

The discussion from everyone has been very helpful even if the writer was
negative to the idea.

//I would be happy to see some interaction in a number of areas.
a) Clearly, the DITA group should be pushing for the FOP features that are
required to produce the documents that can be described in DITA.

Pushing for having some features implemented is one thing; actually
implementing them is another. Correct me if I’m wrong, but I don’t
believe the skill sets of a DITA developer and a FOP developer
intersect, do they?
The DITA-OT uses XSLT to transform DITA xml to other forms so there is a certain amount of XSLT expertise. The DITA-OT includes a Maven plug-in and some Ant so there is expertise in this area. The biggest contribution might be at the specification stage and if you can get some DITA users interested in the other XMLGraphics actvities, you might get some help in documentation since most of the users are Technical Writers by profession.

So IIUC the goal of this would be to attract sponsors who use the
DITA-OT to fund specific developments?

I think that there is pent-up demand for customization and improvements to FOP.

Or are DITA developers willing to
learn what is necessary to contribute new features to FOP?
I can not speak to that since I do not know the developers that well to comment on the range of their abilities and interests nor do I know what is required to make any specific upgrade to FOP. I suspect that it will depend on the difficulty and the technology involved in the specific feature. They certainly can provide input on functional details and provide test suites to test the new features as incorporated into DITA-OT.

It may also attract some of the companies that develop portals, CMSs and editors for DITA (proprietary and open source) that see the XMLGraphics group as a place to acquire big chunks of functionality for their projects. They may have an interest in participating in the XMLGraphics development efforts in order to get features that they need or to contribute functionality that they have developed to existing projects. They may also want to draw on the expertise here to find better ways to process XML in their products.

b)I think that the current Apache members could make positive suggestions
about the processing and structure of the DITA toolkit.

Again, because of the different skill sets you might have put your hopes
a bit high I’m afraid.

I have a lot of respect for the Apache processes and products and have used FOP for quite a few years in a very minor way. I suspect that there are people here that are very good at XSLT and software architecture in general.

c) The DITA language specification is constantly undergoing updates and there are likely things that current FOP users could suggest as additions to allow
them to use a standard language to author documents.

FWIW you might want to become involved in the W3C Print ang Page Layout
Community Group:

d) The users of the DITA tool kit need some customization of FOP
configuration to produce nice looking documents and there is likely a lot of
experts here that could be engaged to do these projects.

We can certainly increase the collaboration between the two projects,
although I don’t think becoming part of XML Graphics is a requirement
for that. Just out of curiosity, what are your reasons for wanting to
become part of Apache?

This is a bit of a personal exploration and anything that I find needs to go back to the DITA-OT group as a suggestion for further investigation by the key people. I am a minor user of DITA (2 small projects - one internal software manual and one to create a study for a client on modernizing their Learning and Development processes). Apache software is an integral part of everything that we develop. I see a lot of frustration in the DITA group about the limitations of DITA production caused by features missing from FOP (perhaps some are just misunderstandings about how to output XSL-FO in a way that will get the job done).

DITA has a huge community of writers and documentation managers who are committed to producing reusable documentation using XML. It seems to me that putting these two communities together on the technical side would have a lot of mutual advantages. The document writers and front-end people who are using FOP directly might find that the DITA community has a lot of the answers for their document management questions and the DITA community would benefit from a relationship with people who understand the technical possibilities for conversion of XML to XML and non-XML products.

I haven't even considered the implications for SVG if the DITA community can get exposed to tools in this area(CAD to SVG).




On 27/03/2014 12:18 PM, Chris Bowditch wrote:
Hi Ron,

Thanks for your e-mail. I can certainly see why joining the 2 communities
seems like a good idea to help improve funding for FOP, and deliver the
features DITA requires. I'd certainly be interested in seeing what the list
of missing XSL-FO features are from a DITA perspective.

The challenge with joining multiple communities together is that it makes the merged project too complex and it scares users/developers away. This is a problem faced by FOP already. The codebase is so vast and complex, many
people struggle to get to grips with it.

I would be against DITA and FOP merging for this reason, plus the separation
of concerns already stated by others. That said, if DITA wanted to join
Apache, adding DITA as a sub project of XMLGraphics would seem like a
logical place for the DITA community. That way there would be some links between the 2 projects and an oversight of both from a common committee. This should help to improve interoperability a little although perhaps not
to the extent you had hoped for.



On 26/03/2014 14:42, Ron Wheeler wrote:
The DITA-OT (DITA Open Toolkit) (http://www.ditaopentoolkit.org/) is an
open source project that depends on FOP.
It takes DITA
(http://en.wikipedia.org/wiki/Darwin_Information_Typing_Architecture) XML
input and produces a number of different document types.
One of the main output types is PDF and it uses FOP to do this. It takes
DITA input (xml) and produces an intermediate set of files that are
processed by FOP to produce a PDF. It can also produce HTML
There is a Maven plug-in to control the production in an IDE environment. It calls all the bits and pieces required to take the raw DITA XML files
and output a document in PDF in the target folder.

There is some interest in adding DITA-OT to the Apache family and in my
opinion, the XMLGraphics group seems like a natural home and the FOP
sub-project might be a good place for DITA-OT to reside.

DITA-OT is a large community of users but a small community of developers. There are also a few enhancement ideas that would require FOP enhancements
to complete.
In addition, my own belief is that the FOP community could add some
technical advice to the DITA-OT community that would be helpful.

http://sourceforge.net/projects/dita-ot/ is the download page. DITA-OT is
currently distributed under the Apache License V2.0.

I am not one of the main players in the group but have taken on the task of seeing if there is a possibility of opening the discussion with the FOP group.

Would there be any interest in considering adding front-end XML processing
to the FOP production project?


Ron Wheeler
Artifact Software Inc
email: rwhee...@artifact-software.com
skype: ronaldmwheeler
phone: 866-970-2435, ext 102

Reply via email to