RE: [dita] Positions on the Open Toolkit

["Troy Klukewich" <troy.klukewich@oracle.com>]

Hi Paul:

<quote>
The Open Toolkit is one implementation among several.  It is not necessarily a better "reference" in terms of defining DITA than any other.  
</quote>

Agreed. That's how I've always thought of it. In fact, it is probably more overtly technical and difficult than most implementations by design, given its open, under-the-hood approach.

<quote>
Just imagine replacing "Open Toolkit" with "XMetaL" or "Arbortext" in your email message below or in JoAnn's message and see how that sounds.
</quote>

Interesting, I think of the Open Toolkit as purely open source and a distinct solution from the commercial implementations with a distinct audience. I think of the OT as a "roll-your-own" solution with rough edges by design. If I don't want a commercial implementation for processing DITA doc types (and some of us don't), I either design my own or use the DITA OT as a basis.

But I can understand where you are coming from as a vendor with current or future implementations in mind. As I am not a vendor, I think of these things differently, especially as an early adopter. I don't rely on commercial tools to show me how to use an open source standard like DITA. If I did, I'd have to wait too long and I wouldn't be early. :-)

<quote>
It is not appropriate to use the OT to "know what the thing is supposed to look like when making evaluations", because that implies that the OT is right and any deviation from what it does is wrong, and I would object to that.
</quote>

In actual practice, the OT is usually the first thing that a technical implementer like me turns to when commercial tools don't even exist yet. If I want to run Learning and Training, I'm not going to wait until commercial tools like XMetal or Arbortext formally support it (perhaps) in a future release. I'm going to use the OT for an initial test.

It's not so much a matter of whether the OT is better or not, or right or not. As an early adopter, it's the only initial tool that I keep on hand for testing specializations. 

<quote>
Practical guidance in using the OT is fine, but it is not something an OASIS TC should do.  It is something that should be done by consultants or the DITA OT group itself outside of OASIS.
</quote>

Thus my concern about a chilling effect within OASIS itself for early adoption efforts now in the works, both for OASIS specialization groups and for the DITA Adoption TC. If these groups cannot provide anything of practical, hands-on value, with tutorial-type information to support early adoption, I would argue that they have no practical value to provide.

I did not understand the restrictions of OASIS as a standards body in this light, but you may well be right. I will seek more guidance from others as your concerns impact an OASIS initiative I am heading up.

I was planning to initiate an approach that would be of use to users and vendors alike for Learning and Training early adopters or evaluators. But if I cannot provide tutorial type information to generate the sample outputs with the OT or other tools as you indicate, then there is nothing to do. 

If I understand you correctly, we in OASIS can only provide abstract standards and hope someone else understands them, implements them, and promotes them on a first-come, first-serve basis, at some point in the future. Adoption is outside the purview of OASIS.

Troy

 

-----Original Message-----
From: Grosso, Paul [mailto:pgrosso@ptc.com]
Sent: Monday, April 13, 2009 1:13 PM
To: troy.klukewich@oracle.com; dita@lists.oasis-open.org
Cc: dita-adoption@lists.oasis-open.org
Subject: RE: [dita] Positions on the Open Toolkit


Just by way of clarification, I don't believe anyone at PTC
has a problem with showing output generated from a particular
implementation as sample output.

What's problematic is for an OASIS TC to provide tutorial
material on how to use a given implementation.  OASIS is
a standards organization, not a consulting organization,
so we shouldn't be providing training on how to use one
particular implementation.  We should be talking about how
to use DITA as a standard and as a methodology.

I guess it depends on what you mean by "reference implementation".
The Open Toolkit is one implementation among several.  It is not
necessarily a better "reference" in terms of defining DITA than
any other.  It is not appropriate to use the OT to "know what 
the thing is supposed to look like when making evaluations",
because that implies that the OT is right and any deviation
from what it does is wrong, and I would object to that.  Just
imagine replacing "Open Toolkit" with "XMetaL" or "Arbortext"
in your email message below or in JoAnn's message and see how 
that sounds.

Practical guidance in using the OT is fine, but it is not
something an OASIS TC should do.  It is something that should
be done by consultants or the DITA OT group itself outside
of OASIS.

paul

> -----Original Message-----
> From: Troy Klukewich [mailto:troy.klukewich@oracle.com] 
> Sent: Monday, 2009 April 13 14:11
> To: Grosso, Paul; dita@lists.oasis-open.org
> Cc: dita-adoption@lists.oasis-open.org
> Subject: RE: [dita] Positions on the Open Toolkit
> 
> I think Paul's e-mail brings up especially valid concerns for 
> vendors. 
> 
> Nevertheless, in terms of response, I am admittedly concerned 
> about a chilling effect that vendors could potentially have 
> on early adoption efforts for upcoming specializations such 
> as Learning and Training. 
> 
> If we cannot provide sample outputs from any implementation, 
> especially from the Open Toolkit, without concerns of bias, 
> we cannot refer to any outputs whatsoever. If this is the 
> case, we cannot demonstrate any practical outputs from DITA, 
> which is essential in early adoption within the marketplace.
> 
> In terms of workload and maintenance, it is not possible for 
> a small individual workgroup or committee to reasonably cover 
> examples of every possible implementation, now and in the 
> future, including the implementation at work in someone's 
> basement today. :-)
> 
> In terms of providing practical guidance to evaluators, 
> vendors, and users interested in a nascent specialization, 
> especially in a specialization that would have little formal 
> vendor support in early adoption, I would question excluding 
> everything because we cannot possibly include everything.
> 
> I have always thought that the purpose of the Open Toolkit 
> was to provide a reference implementation, both for users and 
> vendors alike at the early stages of adoption. We need to 
> know what the thing is supposed to look like when making 
> evaluations. This would hold true for vendors or content 
> development organizations. I think of the Open Toolkit as a 
> tool for both.
> 
> Perhaps the way to proceed for specialization groups is to:
> 
> 1. Establish an actual or ad hoc representative of the Open 
> Toolkit to provide practical guidance in using the Open 
> Toolkit to support a particular specialization, such as 
> Learning and Training, Help, or whatever.
> 
> 2. At the same time, invite participating vendors to provide 
> the same kind of parallel guidance for workgroups within OASIS.
> 
> 3. Provide a general, public announcement outside of OASIS. 
> If other vendors want to provide guidance, they can, either 
> inside or outside of OASIS, now or in the future (obviously, 
> they should be encouraged to join OASIS if they want to 
> contribute now). 
> 
> 4. Mount all of the examples of practical guidance somewhere common.
> 
> No one could be accused of excluding anyone, but adoption 
> efforts could proceed and be of practical, demonstrative use 
> to early adopters and vendors alike.
> 
> Just some ideas.
> 
> Troy Klukewich
> Information Architect
> Oracle
> 
> 
> -----Original Message-----
> From: Grosso, Paul [mailto:pgrosso@ptc.com]
> Sent: Monday, April 13, 2009 10:49 AM
> To: dita@lists.oasis-open.org
> Subject: RE: [dita] Positions on the Open Toolkit
> 
> 
> [I cannot post directly to the DITA Adoption TC, but I have
> no objection to someone forwarding this as deemed appropriate.]
> 
> While PTC feels that the approach taken by the DITA TC's Help 
> Subcommittee is a reasonable way to provide useful information 
> about specific implementations without including or excluding 
> any particular implementations, we would be uncomfortable if 
> OASIS or any of its TCs were to single out a particular 
> implementation--whether the Open Toolkit or another--for 
> special treatment.  This includes "providing guidance for the 
> OT" that goes beyond guidance that is relevant to any DITA 
> implementation.  Guidance for a particular implementation is 
> not something a DITA TC should undertake to the exclusion of 
> other DITA implementations.
> 
> paul
> 
> > -----Original Message-----
> > From: JoAnn Hackos [mailto:joann.hackos@comtech-serv.com] 
> > Sent: Wednesday, 2009 March 25 16:22
> > To: dita-adoption@lists.oasis-open.org; dita@lists.oasis-open.org
> > Subject: [dita] Positions on the Open Toolkit
> > 
> > Dear Friends,
> > We are all following this interesting thread. In this memo, I'm
> > suggested we give it a recognizable name. 
> > 
> > Part of the issue in front of us is the necessary distinction (I
> > believe)in the DITA TC between the DITA specification, which is an
> > official OASIS standard, and the DITA Open Toolkit, which is not. I
> > wonder if such a distinction might be important for the 
> publication of
> > guidelines that would have a substantial influence on DITA 
> adoption? 
> > 
> > Can we develop information that indicates how tools developers are
> > supporting the OT without inviting charges of bias? Mary 
> McRae's note
> > indicates that OASIS takes no official position on the nature of the
> > documents that are official statements from the technical 
> > committees (or
> > at least it hasn't as yet). I think we might be able to 
> produce a Fact
> > Sheet to accompany the Beginner's Guide for Getting Output 
> > from DITA(our
> > long, awkward working title from today's meeting of the Adoption TC)
> > that provides data about tools support for the OT and processing. I
> > think that was what the Help SC was trying so valiantly to do. If we
> > publish the Fact Sheet in the wiki and link to it from the 
> > Guide (which
> > we would publish with the OT, then we can invite anyone to add their
> > testimonial about their implementation (without advertising 
> -- so must
> > be reviewed before added). Would this not be similar to the 
> > testimonials
> > we now offer about using the DITA 1.2 specification? These are
> > statements of fact without judgments attached. 
> > 
> > Let me know what you think. The Adoption TC is committed to 
> providing
> > guidance for the OT, obviously with the involvement of 
> > Matthew Varghese
> > who has assumed responsibility for updating the OT Guide. No one in
> > today's discussion felt that the existing OT Guide was taking 
> > the right
> > direction or had the right audience's in mind.
> > 
> > Our present task is to create a possible TOC for such a 
> Guide based on
> > the goals we identify for the different levels of the user 
> > community. So
> > far, we've identified those levels as 1) techie folks who 
> > prefer working
> > at the command line 2)ordinary folks who wouldn't dream of 
> > command lines
> > to save their lives, and 3)intermediate folks who are 
> willing to learn
> > the techie stuff if they don't have any other choice.
> > 
> > We envision explaining how to use ANT in very simple terms with code
> > that can be cut and pasted. We envision explaining that there 
> > are tools
> > that hide all of that. We envision providing some very basic
> > instructions for modifying the FO to change the look of the 
> > output based
> > on examples and lines of code.
> > 
> > That was the tenor of this morning's (25 March 
> 2009)discussion. Let me
> > know what you think.
> > 
> > Best regards,
> > JoAnn
> > 
> > JoAnn Hackos PhD
> > President
> > Comtech Services, Inc.
> > joann.hackos@comtech-serv.com
> > Skype joannhackos
> > 
> > 
> > 
> > 
> ---------------------------------------------------------------------
> > To unsubscribe from this mail list, you must leave the OASIS TC that
> > generates this mail.  Follow this link to all your TCs in OASIS at:
> > https://www.oasis-open.org/apps/org/workgroup/portal/my_workgr
> > oups.php 
> > 
> > 
> 
> ---------------------------------------------------------------------
> To unsubscribe from this mail list, you must leave the OASIS TC that
> generates this mail.  Follow this link to all your TCs in OASIS at:
> https://www.oasis-open.org/apps/org/workgroup/portal/my_workgroups.php
> 
> 
> 

---------------------------------------------------------------------
To unsubscribe from this mail list, you must leave the OASIS TC that
generates this mail.  Follow this link to all your TCs in OASIS at:
https://www.oasis-open.org/apps/org/workgroup/portal/my_workgroups.php