Personally I think way too many people believe they *must* have a CMS to do any
kind of technical documentation.  I think you can get by with source control if
you don't need a lot of sophisticated search or workflow functionality.

What I mean is if you need to

    * search for, "a paragraph that details the safety information for part
TWR-345, version 2.3, in configuration R"
    * trigger a workflow when a particular event happens (i.e., a
deliverable/file is updated)

then you probably need a CMS.  These things can be done without a CMS, but they
are harder to do.

A SCMS will give you the benefits of archiving, disaster recovery, traceability,
locking, etc., that a CMS can give you, and, most likely, your organization
already has a SCMS in place that you can use without having to go through the
trouble and cost of implementing it yourself.  If you work in a software
development shop, you'll also gain a lot of benefit from using the same
repository that your software developers use.

In the end, I think Eliot's advice is the best anyone can give you - rather than
trying to predict your needs, do a small case, then, based on that, see what you
actually need, and implement exactly that.

BTW, this isn't a bash on CMS.  Using a CMS is great, but in the real world you
need to make sure the cost is worth the benefit.

     Steve

Stilo to present at DITA Europe 2009
MAKING DITA WORK
Monday 16 November 2009, 1.30pm - 2.30pm

If you're already attending or planning to attend this year's DITA Europe Conference, being held next week - 16-17 November 2009 - at the Holiday Inn, Munich City Centre, be sure to attend the following presentation by Stilo's Joe Gollner:

MAKING DITA WORK
Monday 16 November, 1.30pm - 2.30pm

Joe's presentation will take a closer look at some of the challenges organizations encounter as they attempt to move from initial exploratory investments in DITA to implementing a new environment for creating, managing and publishing content assets.  These new environments must be able to shoulder the weight of production demands, and for these reasons, as well as others, the transition from exploration to operation can be a very challenging one.  Based on numerous past and present DITA projects that have specifically focused on addressing these challenges, this presentation will provide attendees with real-world insights into what implementation challenges require attention and what practical steps can be taken to address them.  Read the full presentation outline at: http://www.infomanagementcenter.com/DITAeurope/2009/abstracts.htm#gollner

Based in our offices in Ottawa, Canada, Joe is the Vice President Enterprise Publishing Solutions at Stilo.  He has over 20 years of experience in designing and implementing enterprise content management and publishing solutions, and in the last five years he has led some of the largest and most ambitious DITA implementations ever undertaken.

Joe would love to meet with you at DITA Europe to discuss your particular DITA implementation challenges.  If you would like to arrange a meeting, please email us at: howens@... to arrange a mutually convenient time.

And don't forget to stop by our stand, where we will be featuring our range of specialist tools and online services for XML content processing.  Whatever your XML content conversion and transformation requirements, we can provide in-house or outsourced solutions tailored to your specific business needs and all supported by our content processing experts.  Visit stilo.com at: http://www.stilo.com/XMLContentProcessing/tabid/159/Default.aspx  for more information.

If you would like further information on DITA Europe 2009 and to register, please visit the conference website at: http://www.infomanagementcenter.com/DITAeurope/2009/index.htm

Kind regards
Helen Owens
Marketing Communications Manager
Stilo International

Hi,

I recently encountered this problem, and after searching on the internet, I
couldn't find a solution.

Say, I have a container topic (title.xml), a topic only has a title in it and is
used to provide a TOC. This same topic is used multiple times within a ditamap
to show different sets of TOCs.

A sample ditamap:
<map title="Map Title">
<topicref id="0001" href="title.xml">
<topicref href="a.xml">
</topicref>
</topicref>
<topicref id="0002" href="title.xml">
<topicref href="b.xml">
</topicref>
<topicref href="c.xml">
</topicref>
<topicref href="d.xml">
</topicref>
</topicref>
</map>

I used the latest dita-OT, and it renders index.html fine:

Title
--a

Title
--b
--c
--d

However, the link on the Titles brings to the same page showing only the link to
a. What I thought would happen is it creates a new Title file for the second
Title section, and puts the links to b,c,d in the second title file.

I tried "copy-to" attribute on the second title topicref, but that only makes a
copy of the first Title, the links inside of them are still the same.


Any comments are appreciated!


Jason

There was a good bit of discussion about this a week or so ago.

In your case, with a relatively small library and modest reuse, you can
definitely get by with a code control system and some general organization
and naming rules. I don't think a CMS will help you that much simply because
you don't have that much content to manage. Clear naming and organizational
conventions and a bit of author discipline will get you a long way.

For example, for our product RSuite CMS, we have a library of about 12
manuals with some reuse. The source is managed in CVS and various members of
the product engineering team contribute to the docs. We've established some
simple organizational conventions. I use Ant tasks to automate publishing
the docs with the Open Toolkit to an internal site and packaging them for
customer delivery.

You might ask why we don't use RSuite CMS and the short answer is the
Shoemaker's Children problem, but also because RSuite does not (yet) support
the use of DITA for technical documentation use cases (it is focused on
Publishing use cases where the DITA XML is not normally directly edited but
generated from Word or InCopy). While RSuite can import and export and
process (with the Toolkit) DITA as well as or better than any system out
there, we're still working on our XML editor integrations, which are
required to really support technical documentation.

In general, I advise groups starting out with DITA to focus on authoring and
production and leave the CMS for later--just getting up to speed with XML
authoring and production from DITA is sufficiently challenging that adding a
CMS to mix can seriously delay a project. By the same token, because you can
get started with free or inexpensive editing tools and with the free DITA
Open Toolkit, you can put together a demonstration or prototype quickly and
with minimal cash outlay.

Once you've the team up to speed with authoring and you can produce your
deliverables, you will also understand your business processes and the
technology much better and be able to make a much more informed decision
about what CMS functionality you need. You'll also have a existing body of
content that reflects your business that will make a more realistic test of
a CMS system.

Another thing to keep in mind, that I don't think I've mentioned in this
forum before, but it generally takes at least two revision cycles to
determine that your CMS doesn't actually work. By having more data in DITA
and more experience with working with it, you can short circuit that process
somewhat.

Cheers,

Eliot
----
Eliot Kimber
Senior Solutions Architect
"Bringing Strategy, Content, and Technology Together"
Main: 610.631.6770
www.reallysi.com
www.rsuitecms.com


On 11/10/09 1:53 AM, "nanr93" <nancyr@...> wrote:

> Hi,
>
> My company - a smallish startup - is heading fast toward a DITA solution.
> Here's what we've got:
>
> - A product documentation package - Comprised of about 15 documents (currently
> written in Word) with maybe 30-40% content reuse. The package is distributed
> in both PDF form and as OLH.
>
> - An endless number of Tech Notes - one-page docs - that the system engineers
> produce and distribute ad hoc.
>
> - An online knowledge base - Currently pretty small, but we're hoping it will
> grow with the company. The content is posted primarily by the SEs. Eventually
> it will include all the Tech Notes and the product documentation package, too.
>
> Here's my question: is it realistic to put together a DITA solution without a
> CMS? We're considering DITA Exchange (any impressions you might want to share
> are most welcome), but the cost of the POC is non-refundable. We're a little
> leery of shelling out that kind of money for something we might not like.
>
> Thanks for your help,
>
> Nancy
>

--

Hi Nancy,

To answer the base question, you can do a DITA solution without a CMS. You just
spend lots of time manually managing components and you have to decide whether
you want to pay that price.

One thing I suggest is to look at all the products available and that the CMS
supports your workflow and not just a collection of features. It's best to make
sure that what you purchase fits what you do and that you not force-fit what you
do into a suggested tool.

HTH,

Julio J. Vazquez
SDI Global Solutions

--- In dita-users@yahoogroups.com, Dan Tran <dantran@...> wrote:
>
> On item 1, DITA + CMS is perfect solution together with  build server to
> fully generate DITA outputs ( online help, PDF, etc )
>
> On item 2, and 3.  Use Wiki.  DITA without CMS is bad, but make SE to use
> CMS is also bad idea.
>
> -Dan
>
> On Mon, Nov 9, 2009 at 11:53 PM, nanr93 <nancyr@...> wrote:
>
> >
> >
> > Hi,
> >
> > My company - a smallish startup - is heading fast toward a DITA solution.
> > Here's what we've got:
> >
> > - A product documentation package - Comprised of about 15 documents
> > (currently written in Word) with maybe 30-40% content reuse. The package is
> > distributed in both PDF form and as OLH.
> >
> > - An endless number of Tech Notes - one-page docs - that the system
> > engineers produce and distribute ad hoc.
> >
> > - An online knowledge base - Currently pretty small, but we're hoping it
> > will grow with the company. The content is posted primarily by the SEs.
> > Eventually it will include all the Tech Notes and the product documentation
> > package, too.
> >
> > Here's my question: is it realistic to put together a DITA solution without
> > a CMS? We're considering DITA Exchange (any impressions you might want to
> > share are most welcome), but the cost of the POC is non-refundable. We're a
> > little leery of shelling out that kind of money for something we might not
> > like.
> >
> > Thanks for your help,
> >
> > Nancy
> >
> >
> >
>

Hello,
I would like to add funcionality (xsl) to htmlhelp transtype at
dita.map.htmlhelp.hhp pipeline. How to?

I tried a plugin like:

<plugin id="htmlhelp.extenstion">

   <feature extension="dita.conductor.target.relative" value="build.xml"
type="file"/>

<feature extension="depend.dita.map.htmlhelp.hhp.pre" value="mytarget"
type="text"/>

</plugin>

Where mytarget is defined in the build.xml as a ant xsl call.

Dear Chad,

I mean that either nothing or almost nothing appears.

Sometimes there are two little vertical bars on either side of the page
where the image is expected to appear and there is empty space about the
same height as a line of text.

Other times, there is not even that.

The images are all gif files.


Until I included the "custom_pdf.xsl", all these images appeared as I expect
them to, the only problem being that sometimes the page would break between
the image and the previous text within the <cmd> or <note>.


Best regards,

-Brian.






Messages in this topic (5)
________________________________________________________________________
________________________________________________________________________
2a. Re:
     Posted by: "Chad" chad.dybdahl@... chad.dybdahl
     Date: Mon Nov 9, 2009 6:09 am ((PST))

When you say the images don't appear, do you mean that nothing at all
appears in the space where you are expecting your images?

Or is there a generic placeholder image inserted instead?
What format are your images in (png, jpg, etc)?

We have a source control management tool, of course, and I understand a lot of
places use their SCM tools in place of a CMS. But no, we don't have a CMS yet.

--- In dita-users@yahoogroups.com, Dan Tran <dantran@...> wrote:
>
> Hi Nancy,
>
> I am not sure how a tech company can survive without a CMS? dont you already
> have one?
>
> -Dan
>
>
>
> On Tue, Nov 10, 2009 at 12:28 AM, nanr93 <nancyr@...> wrote:
>
> >
> >
> > Thanks, Matt. The problem is cost, of course.
> >
> > Nancy
> >
> >
> > --- In dita-users@yahoogroups.com <dita-users%40yahoogroups.com>,
> > "mathew_varghese2" <mathew.techwriter@> wrote:
> > >
> > > Nancy,
> > >
> > > Given the high degree of reuse, I think you should use a CMS. They add to
> > some of DITA's native reuse features, esp conref.
> > >
> > > Best regards,
> > >
> > > Matt
> > >
> > > --- In dita-users@yahoogroups.com <dita-users%40yahoogroups.com>,
> > "nanr93" <nancyr@> wrote:
> > > >
> > > > Hi,
> > > >
> > > > My company - a smallish startup - is heading fast toward a DITA
> > solution. Here's what we've got:
> > > >
> > > > - A product documentation package - Comprised of about 15 documents
> > (currently written in Word) with maybe 30-40% content reuse. The package is
> > distributed in both PDF form and as OLH.
> > > >
> > > > - An endless number of Tech Notes - one-page docs - that the system
> > engineers produce and distribute ad hoc.
> > > >
> > > > - An online knowledge base - Currently pretty small, but we're hoping
> > it will grow with the company. The content is posted primarily by the SEs.
> > Eventually it will include all the Tech Notes and the product documentation
> > package, too.
> > > >
> > > > Here's my question: is it realistic to put together a DITA solution
> > without a CMS? We're considering DITA Exchange (any impressions you might
> > want to share are most welcome), but the cost of the POC is non-refundable.
> > We're a little leery of shelling out that kind of money for something we
> > might not like.
> > > >
> > > > Thanks for your help,
> > > >
> > > > Nancy
> > > >
> > >
> >
> >
> >
>

Thanks, Dan. We have a KB solution in place. It's not quite as open to input as
a regular Wiki, but it's not bad.

--- In dita-users@yahoogroups.com, Dan Tran <dantran@...> wrote:
>
> On item 1, DITA + CMS is perfect solution together with  build server to
> fully generate DITA outputs ( online help, PDF, etc )
>
> On item 2, and 3.  Use Wiki.  DITA without CMS is bad, but make SE to use
> CMS is also bad idea.
>
> -Dan
>
> On Mon, Nov 9, 2009 at 11:53 PM, nanr93 <nancyr@...> wrote:
>
> >
> >
> > Hi,
> >
> > My company - a smallish startup - is heading fast toward a DITA solution.
> > Here's what we've got:
> >
> > - A product documentation package - Comprised of about 15 documents
> > (currently written in Word) with maybe 30-40% content reuse. The package is
> > distributed in both PDF form and as OLH.
> >
> > - An endless number of Tech Notes - one-page docs - that the system
> > engineers produce and distribute ad hoc.
> >
> > - An online knowledge base - Currently pretty small, but we're hoping it
> > will grow with the company. The content is posted primarily by the SEs.
> > Eventually it will include all the Tech Notes and the product documentation
> > package, too.
> >
> > Here's my question: is it realistic to put together a DITA solution without
> > a CMS? We're considering DITA Exchange (any impressions you might want to
> > share are most welcome), but the cost of the POC is non-refundable. We're a
> > little leery of shelling out that kind of money for something we might not
> > like.
> >
> > Thanks for your help,
> >
> > Nancy
> >
> >
> >
>

Thanks, Matt. The problem is cost, of course.

Nancy

--- In dita-users@yahoogroups.com, "mathew_varghese2" <mathew.techwriter@...>
wrote:
>
> Nancy,
>
> Given the high degree of reuse, I think you should use a CMS. They add to some
of DITA's native reuse features, esp conref.
>
> Best regards,
>
> Matt
>
> --- In dita-users@yahoogroups.com, "nanr93" <nancyr@> wrote:
> >
> > Hi,
> >
> > My company - a smallish startup - is heading fast toward a DITA solution.
Here's what we've got:
> >
> > - A product documentation package - Comprised of about 15 documents
(currently written in Word) with maybe 30-40% content reuse. The package is
distributed in both PDF form and as OLH.
> >
> > - An endless number of Tech Notes - one-page docs - that the system
engineers produce and distribute ad hoc.
> >
> > - An online knowledge base - Currently pretty small, but we're hoping it
will grow with the company. The content is posted primarily by the SEs.
Eventually it will include all the Tech Notes and the product documentation
package, too.
> >
> > Here's my question: is it realistic to put together a DITA solution without
a CMS? We're considering DITA Exchange (any impressions you might want to share
are most welcome), but the cost of the POC is non-refundable. We're a little
leery of shelling out that kind of money for something we might not like.
> >
> > Thanks for your help,
> >
> > Nancy
> >
>

Nancy,

Given the high degree of reuse, I think you should use a CMS. They add to some
of DITA's native reuse features, esp conref.

Best regards,

Matt

--- In dita-users@yahoogroups.com, "nanr93" <nancyr@...> wrote:
>
> Hi,
>
> My company - a smallish startup - is heading fast toward a DITA solution.
Here's what we've got:
>
> - A product documentation package - Comprised of about 15 documents (currently
written in Word) with maybe 30-40% content reuse. The package is distributed in
both PDF form and as OLH.
>
> - An endless number of Tech Notes - one-page docs - that the system engineers
produce and distribute ad hoc.
>
> - An online knowledge base - Currently pretty small, but we're hoping it will
grow with the company. The content is posted primarily by the SEs. Eventually it
will include all the Tech Notes and the product documentation package, too.
>
> Here's my question: is it realistic to put together a DITA solution without a
CMS? We're considering DITA Exchange (any impressions you might want to share
are most welcome), but the cost of the POC is non-refundable. We're a little
leery of shelling out that kind of money for something we might not like.
>
> Thanks for your help,
>
> Nancy
>

Hi,

My company - a smallish startup - is heading fast toward a DITA solution. Here's
what we've got:

- A product documentation package - Comprised of about 15 documents (currently
written in Word) with maybe 30-40% content reuse. The package is distributed in
both PDF form and as OLH.

- An endless number of Tech Notes - one-page docs - that the system engineers
produce and distribute ad hoc.

- An online knowledge base - Currently pretty small, but we're hoping it will
grow with the company. The content is posted primarily by the SEs. Eventually it
will include all the Tech Notes and the product documentation package, too.

Here's my question: is it realistic to put together a DITA solution without a
CMS? We're considering DITA Exchange (any impressions you might want to share
are most welcome), but the cost of the POC is non-refundable. We're a little
leery of shelling out that kind of money for something we might not like.

Thanks for your help,

Nancy

Hi all,

I have my ${customization.dir} configured to point to another directory where I
have my

   common/var/en_US.xml
   fo/xsl/custom.xsl
   fo/font-mappings.xml
   catalog.xml

Define

The content of catalog.xml is

     <uri name="cfg:fo/xsl/custom.xsl" uri="fo/xsl/custom.xsl"/>
     <uri name="cfg:fo/font-mappings.xml" uri="fo/font-mappings.xml"/>


I think DITA OT does not pickup my custom font-mappings.xml file since
if I remove that file, DITA-OT Ant build dont seem to care and runs find.

However if my custom.xsl is missing, DITA-OT Ant build bails out.

Any suggestion to troubleshoot this problem??

Big thanks

-Dan

Hello.

See the following note for the solution.

http://tech.groups.yahoo.com/group/dita-users/message/15025

Eric

ekimber wrote:
>
> On 11/9/09 2:47 PM, "wyvzeyeview" <lwwhite5@...
> <mailto:lwwhite5%40hotmail.com>> wrote:
>
> > When I run a build (OT 1.5 M22) using a ditamap generated through
> Serna 4.2
> > (DITA 1.1), I get the following fatal error:
> >
> > C:\DITA-OT1.5\build_preprocess.xml:25: [DOTJ012F][FATAL] Failed to
> parse the
> > input file 'test.ditamap' due to below exception. Please correct the
> input
> > based on the exception message.:test.ditamap Line 4:Attribute
> "xmlns:xsi" must
> > be declared for element type "map".
> >
> > In the map, this is the offending line:
> >
> > <map xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance
> <http://www.w3.org/2001/XMLSchema-instance>"
> > xsi:noNamespaceSchemaLocation="urn:oasis:names:tc:dita:xsd:map.xsd:1.1">
> >
> > I notice that 1.1 ditamaps I create using other editors have this as
> the map
> > declaration:
> >
> > <map xmlns:ditaarch="http://dita.oasis-open.org/architecture/2005/
> <http://dita.oasis-open.org/architecture/2005/>">
> >
> > which works.
> >
> > What's going on? I can always manually edit the ditamap, but would
> like to
> > understand the underlying issue.
>
> It looks like something is trying to validate a schema-based map with
> a DTD.
> The DITA DTDs do not declare the XSD namespace, so you can't just slap
> a DTD
> declaration on an XSD-based document and have it work--you have to remove
> the XSD-specific attributes.
>
> You may need to check the files created by Serna in a text editor since
> Serna (and most other graphical XML editors) won't show you the DOCTYPE
> declaration, if there is one.
>
> Cheers,
>
> E.
>
> ----
> Eliot Kimber | Senior Solutions Architect | Really Strategies, Inc.
> email: ekimber@... <mailto:ekimber%40reallysi.com>
> <mailto:ekimber@... <mailto:ekimber%40reallysi.com>>
> office: 610.631.6770 | cell: 512.554.9368
> 2570 Boulevard of the Generals | Suite 213 | Audubon, PA 19403
> www.reallysi.com <http://www.reallysi.com <http://www.reallysi.com>> |
> http://blog.reallysi.com <http://blog.reallysi.com>
> <http://blog.reallysi.com <http://blog.reallysi.com>> |
> www.rsuitecms.com <http://www.rsuitecms.com <http://www.rsuitecms.com>>
>
>

On 11/9/09 2:47 PM, "wyvzeyeview" <lwwhite5@...> wrote:

> When I run a build (OT 1.5 M22) using a ditamap generated through Serna 4.2
> (DITA 1.1), I get the following fatal error:
>
> C:\DITA-OT1.5\build_preprocess.xml:25: [DOTJ012F][FATAL] Failed to parse the
> input file 'test.ditamap' due to below exception. Please correct the input
> based on the exception message.:test.ditamap Line 4:Attribute "xmlns:xsi" must
> be declared for element type "map".
>
> In the map, this is the offending line:
>
> <map xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
> xsi:noNamespaceSchemaLocation="urn:oasis:names:tc:dita:xsd:map.xsd:1.1">
>
> I notice that 1.1 ditamaps I create using other editors have this as the map
> declaration:
>
> <map xmlns:ditaarch="http://dita.oasis-open.org/architecture/2005/">
>
> which works.
>
> What's going on? I can always manually edit the ditamap, but would like to
> understand the underlying issue.

It looks like something is trying to validate a schema-based map with a DTD.
The DITA DTDs do not declare the XSD namespace, so you can't just slap a DTD
declaration on an XSD-based document and have it work--you have to remove
the XSD-specific attributes.

You may need to check the files created by Serna in a text editor since
Serna (and most other graphical XML editors) won't show you the DOCTYPE
declaration, if there is one.

Cheers,

E.


----
Eliot Kimber | Senior Solutions Architect | Really Strategies, Inc.
email:  ekimber@... <mailto:ekimber@...>
office: 610.631.6770 | cell: 512.554.9368
2570 Boulevard of the Generals | Suite 213 | Audubon, PA 19403
www.reallysi.com <http://www.reallysi.com>  | http://blog.reallysi.com
<http://blog.reallysi.com> | www.rsuitecms.com <http://www.rsuitecms.com>

When I run a build (OT 1.5 M22) using a ditamap generated through Serna 4.2
(DITA 1.1), I get the following fatal error:

C:\DITA-OT1.5\build_preprocess.xml:25: [DOTJ012F][FATAL] Failed to parse the
input file 'test.ditamap' due to below exception. Please correct the input based
on the exception message.:test.ditamap Line 4:Attribute "xmlns:xsi" must be
declared for element type "map".

In the map, this is the offending line:

<map xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
xsi:noNamespaceSchemaLocation="urn:oasis:names:tc:dita:xsd:map.xsd:1.1">

I notice that 1.1 ditamaps I create using other editors have this as the map
declaration:

<map xmlns:ditaarch="http://dita.oasis-open.org/architecture/2005/">

which works.

What's going on? I can always manually edit the ditamap, but would like to
understand the underlying issue.

Thanks,
Leigh

The Center for Information-Development Management asks you to consider bringing your best practices to the upcoming Content Management Strategies/DITA North America conference.

We invite you to share your experience building a content management solution in your organization, collaborating with content producers throughout your organization (training, support, engineering, documentation), delivering content in new media, using wikis and blogs to communicate with internal and external customers, and managing change. Management and technical topics are both welcome in the program.

For the DITA NA conference, we welcome discussions of selecting a DITA solution, implementing DITA in your organizations, as well as technical topics about DITA processing, specialization, and so on.

The 12th Annual Content Management Strategies/DITA North America Conference will be held at the Hyatt Regency in Santa Clara, CA April 19-21, 2010.

Please submit your presentation proposal by December 8, 2009. at http://www.cm-strategies.com/

If you have questions or would like to learn more, please contact our office at  +1-303-232-7586 or email us at info@....

We look forward to seeing you at CMS/DITA NA 2010!

 

Hello Julio,   Thanks for this valuable suggestion. Using the ‘My link text’ the text is wrapped and moved to the next line. However, the ‘href link’ of the xref attribute is itself too long that when I click the link in the PDF output, the link breaks and the following message is displayed: The specified file <https://text........> does not exist. The last 20-25 letters are not accepted in this <https://text.........>   Please let me know the solution for this.   Regards, Kalpana

I have the same problem. In the TOC that I'm generating with the DITA Open
Toolkit and FOP, the numbers at the end of the dot leaders only sort of kind of
line up. I wouldn't call them left aligned or centered or right aligned, and I
want them right-aligned. I've put a demo at
http://www.snee.com/temp/tocdemo.pdf.

Is this one of those things that works better with a non-free renderer like
RenderX or Antenna House?

thanks,

Bob

>The section/title parts of the TOC still look the same as the topic/title
parts,

As a follow-up for those who are interested, I figured out how to fix that. In a
modified version of the __toc__topic__content xsl:attribute-set from
demo/fo/cfg/fo/attrs/toc-attr.xsl, I added

   or contains(@class, ' topic/section '

in the two places where it computes a $level value so that it would treat
section titles as lower levels in the TOC.

Bob

On 11/9/09 8:59 AM, "THOMAS" <ttowle1@...> wrote:

> Did you get anywhere with this? I am just now entering the phase of storing my
> XML Dita files into a database and am always eager to use OpenSource
> technologies (which is why I use DITA OT:)
> It seems that the real backbreaking work after creating content is storing it
> so it can be used as a single source of content tagged appropriately in
> reusable chunks which leads right to the database as the backend to store
> these chunks. And there are so many chunks created by Dita so the CMS or a
> database either xml or relational or object oriented flavor.
> But alas how to get them into a database. I suppose that is where the CMS
> world really ties into the Dita movement.
> So what did you think of EXist xml database? Good or bad experiences are both
> interesting to me.
> Thx and good luck.

I am in the process of putting into open source, through the DITA for
Publishers SourceForge project, a general DITA processing Java library that
includes facilities for determining the set of dependencies used from a
given root map (the map's "bounded object set" of DITA and non-DITA
resources). It will also include key management facilities. This library is
driven directly by the requirements for importing and exporting DITA content
to and from a CMS system.

Once I release it, which should be in the next couple of weeks (I'm trying
to get a general key management implementation available so we can use it in
part to test the 1.2 spec) it should be relatively easy to use it as the
base for eXist-specific importers and exporters.

I will announce more here as I have news to report.

Cheers,

Eliot

----
Eliot Kimber | Senior Solutions Architect | Really Strategies, Inc.
email:  ekimber@... <mailto:ekimber@...>
office: 610.631.6770 | cell: 512.554.9368
2570 Boulevard of the Generals | Suite 213 | Audubon, PA 19403
www.reallysi.com <http://www.reallysi.com>  | http://blog.reallysi.com
<http://blog.reallysi.com> | www.rsuitecms.com <http://www.rsuitecms.com>

Did you get anywhere with this? I am just now entering the phase of storing my
XML Dita files into a database and am always eager to use OpenSource
technologies (which is why I use DITA OT:)
It seems that the real backbreaking work after creating content is storing it so
it can be used as a single source of content tagged appropriately in reusable
chunks which leads right to the database as the backend to store these chunks.
And there are so many chunks created by Dita so the CMS or a database either xml
or relational or object oriented flavor.
But alas how to get them into a database. I suppose that is where the CMS world
really ties into the Dita movement.
So what did you think of EXist xml database? Good or bad experiences are both
interesting to me.
Thx and good luck.

--- In dita-users@yahoogroups.com, "escocesito" <escocesito@...> wrote:
>
> I'm evaluating using the eXist XML database with Dita files. To store the
files is easy but I'm now looking at how to integrate with the Dita Open Toolkit
to generate documents.
> The Dita toolkit works fine if all the files are on the hard disk but is it
possible for the toolkit to retrieve the files direct from a database?
>
> Maybe it possible to specify in a href or something that a file is actually in
the database rather than on the hard disk?
>
> Or do all the files have to be downloaded to the harddisk first? In this case
is it possible to customize the genlist part of the process to get all the
required files?
>
> Has anyone any experience of using XML databases with DITA? There doesn't
appear to be a lot of documentation on the subject.
>
> Many thanks in advance.
>
>    Colin
>

When you say the images don't appear, do you mean that nothing at all appears in
the space where you are expecting your images?

Or is there a generic placeholder image inserted instead?
What format are your images in (png, jpg, etc)?

Hi Jeff,

If you want the faqquest and faqans to be divs, you need a different structure
than what the faq specialization provides. What you're looking for is to
specialize section for both of these. The problem you'll run into is that DITA
does not allow nested sections so that you'll need to add an outputclass
attribute to each section so you can write an XSLT to style the output.

Even with what you've done, in your sample, you don't show an XSLT to take the
faqquest and faqans you've defined to change the styling so that all you'd see
is the default styling from the OT for lines. You could possibly send the output
from these two elements to an HTML span with different attributes or with a
class that your CSS styles the way you would like.

You have a number of options, but rather than toss what you have, work on an
XSLT that tests for the new classes you've defined and produces HTML that you
can style either statically in the XSLT or using CSS based on an assigned class.

HTH,

Julio J. Vazquez
SDI Global Solutions

--- In dita-users@yahoogroups.com, "jvb_uwyo" <jvb@...> wrote:
>
> Hi again Julio. Let me try again with a more specific question. As I said, I'm
now trying to modify the html output produced for my faq. So I copied the
faq.mod file, renamed it hjfaq.mod, and modified it. All I changed was the
element specialization section, like so:
> <!-- ============ Element specialization declarations ============ -->
> <!ATTLIST faq         class  CDATA "- topic/topic       faq/faq ">
> <!ATTLIST faqbody     class  CDATA "- topic/body        faq/faqbody ">
> <!ATTLIST faqgroup    class  CDATA "- topic/section     faq/faqgroup ">
> <!ATTLIST faqlist     class  CDATA "- topic/ol  faq/faqlist ">
>
> <!ATTLIST faqitem     class  CDATA "- topic/li  faq/faqitem ">
> <!ATTLIST faqquest    class  CDATA "- topic/lines     faq/faqquest ">
> <!ATTLIST faqans      class  CDATA "- topic/lines     faq/faqans ">
>
> I used topic/lines here, because I didn't know what else to use. I really just
want the faqquest and faqans to be divs with a class attribute that I specify.
>
> Then I copied the faq_shell.dtd file to hjfaq_shell.dtd and modified it like
so:
> <!-- (c) Copyright IBM Corp. 2004, 2005 All Rights Reserved. -->
>
> <!--Specify the topics that can be included within a FAQ topic-->
> <!ENTITY % faq-info-types "faq">
>
> <!--Specify the domain-specific vocabulary modules (none here,
>     but real version would probably provide some vocabulary)-->
> <!ENTITY included-domains "">
>
> <!--Embed the base topic module to define the base elements-->
> <!ENTITY % topic-type PUBLIC "-//OASIS//ELEMENTS DITA Topic//EN"
"../../dtd/topic.mod">
>   %topic-type;
>
> <!--Embed the faq module to get the faq elements-->
> <!ENTITY % faq-typemod PUBLIC "-//IBM//ELEMENTS DITA FAQ//EN" "hjfaq.mod">
>   %faq-typemod;
>
> Finally, I wrote a test faq file whose contents is:
> <?xml version="1.0" encoding="utf-8"?>
> <!-- (c) Copyright IBM Corp. 2004, 2005 All Rights Reserved. -->
> <!DOCTYPE faq PUBLIC "-//IBM//DTD DITA FAQ//EN" "hjfaq_shell.dtd">
> <?xml-stylesheet type="text/css"
href="../../Applications/oxygen/frameworks/dita/DITA-OT/demo/faq/faq.css"?>
> <faq id="Testfaq" xml:lang="en-us"><title>Test FAQ</title>
>     <faqbody>
>         <faqgroup><title>Example</title>
>             <faqlist>
>                 <faqitem>
>                     <faqquest>An example question?</faqquest>
>                     <faqans>Create a DTD module in which the DTD elements
derive from elements in an
>                         existing DTD module.</faqans>
>                 </faqitem>
>             </faqlist>
>         </faqgroup>
>     </faqbody>
> </faq>
>
> When I ran the DITA to OT transformation for XHTML, it produces the same
output html as before. Namely, faqitem is transformed to <tr>, faqquest to <td>,
and faqans to <td>.
>
> So I guess I'm not sure how to get new mod file to be used.
>
> What should I do?
>
> Thanks in advance,
>
> Jeff
> ----
> --- In dita-users@yahoogroups.com, "juliov27612" <julio_v27612@> wrote:
> >
> > Hi Jeff,
> >
> > Can you post some sample source so we have a better idea of the conref and
the topic in which the source content resides?
> >
> > Julio J. Vazquez
> > SDI Global Solutions
> >
> > --- In dita-users@yahoogroups.com, "jvb_uwyo" <jvb@> wrote:
> > >
> > > I'm using DITA-OT 1.5 from oxygen 11. I'm trying to create a faq that
pulls from a user manual I have already written. I'm using the demo faq stuff
that comes with DITA-OT. I put conrefs in my faq elements, but they seem to be
unrecognized and don't show up in the xhtml or pdf target. Can anyone give me a
pointer on how to make this happen?
> > >
> > > Thanks,
> > >
> > > Jeff
> > > ----
> > >
> >
>