Re: Why integrating DITA into XMLGraphics makes sense.
Hi Ron, I agree that some education of both parties is needed. Its no longer true that FOP is a buggy limited piece of software. That might have been true 3-4 years ago, but it is becoming a very mature product now. That is a clear sign that the DITA experts are out of date. We need to work with the DITA experts to keep them informed of progress. Also input from the DITA experts on the top 10 missing features or bugs would be helpful for us in prioritizing work too. Of course, DITA folks might see a solution in XEP or Antenna house, but both are expensive, and the last time I checked AH only supports PDF output, not PS or AFP. Thanks, Chris On 25/05/2014 15:40, Ron Wheeler wrote: A good starting point: http://thecontentwrangler.com/2008/04/11/choosing_an_xml_schema_docbook_or_dita/ A good discussion about how DITA-OT uses XSL and XSL-FO to create PDF from DITA XML. http://www.scriptorium.com/whitepapers/ditaotpdf/DITA-PDF-tweaks.pdf I am trying to get the FOP side to be aware of the importance of DITA as a standard for documentation so the FOP developers will pay some attention to the needs for improved FOP features and perhaps give advice to the DITA-OT developers to use FOP in the best possible way. I am trying to get the DITA side to stop considering FOP to be a static thing that can not be changed and to start to contribute ideas and funding to make FOP do the things that it needs to do. I also want to encourage the DITA-OT team to enter into discussions with the FOP experts to make sure that DITA-OT uses FOP in the best possible way. This problem with the leading dots is a good example of the problem. When the problem was raised by a documentation author, one of the leading DITA experts proposed the solution to the problem was to stop trying to make FOP work since it is buggy and inconsistent rather than suggesting that the user ask the question in the FOP user forum. When I brought the problem here, an answer was provided that looks like a simple change to DITA-OT's FOP configuration that seemed to be a solution to this problem that is well understood here. Ron On 24/05/2014 9:22 PM, Glenn Adams wrote: I'm not familiar with DITA, but if a DITA product depends on FOP, then DITA as a group or its sponsors should consider funding the work they would like to see done in FOP. Simply asking the few developers in the FOP project to support DITA priorities won't guarantee any results. On Sun, May 25, 2014 at 5:38 AM, Ron Wheeler rwhee...@artifact-software.com mailto:rwhee...@artifact-software.com wrote: You are right , of course. However, this very large community ( one DITA LinkedIn group has 4,600 members, the Technical Doc group has over 14,000 members) does not see themselves as users of FOP but only as users of DITA-OT which in turn has a dependency on FOP. As far as I know DITA is the most popular XML language for constructing documents and it seems odd that there is almost no connection between the Apache efforts in XML and the biggest potential set of users and drivers of demand for the things that XMLGraphics is producing. I will pass on the information to the forum where the question was raised. Thanks On 23/05/2014 6:05 PM, Luis Bernardo wrote: I think this only shows that the person is not going to the source (i.e., the FOP user mailing list) to request help. The example shown can be greatly improved by using fo:leader width=100% leader-pattern=use-content./fo:leader instead of fo:leader leader-pattern=dots / The FOP implementation repeats 3 dots (...) when using the leader-pattern=dots which is not very intelligent since it can lead to misalignment many times. But by specifying the leader content as just one dot the result can be greatly improved, although I agree that there is room for improvement in this feature. On 5/23/14, 4:14 PM, Ron Wheeler wrote: The conversation below shows one of the reasons why I would like to see DITA become part of the XMLGraphics family. One of the most experienced and influential DITA practitioners is giving advice about the suitability of FOP for producing correct PDFs. Ron - This is an issue with the FOP XSL-FO engine (one of many). For top-qualify PDF output you really need to license Antenna House XSL Formatter or RenderX XEP, both of which produce excellent results. FOP simply has too many bugs and limitations to be generally useful for production PDF generation using XSL-FO, unfortunately. Cheers, XXX On 5/23/14, 9:54 AM, yyy [dita-users] dita-us...@yahoogroups.com mailto:dita-us...@yahoogroups.com wrote: [Attachment(s) #TopText from Mark Peters included below] Hi, Using DITA-OT 1.7 and the default org.dita.pdf2 plugin, I
Re: Why integrating DITA into XMLGraphics makes sense.
A good starting point: http://thecontentwrangler.com/2008/04/11/choosing_an_xml_schema_docbook_or_dita/ A good discussion about how DITA-OT uses XSL and XSL-FO to create PDF from DITA XML. http://www.scriptorium.com/whitepapers/ditaotpdf/DITA-PDF-tweaks.pdf I am trying to get the FOP side to be aware of the importance of DITA as a standard for documentation so the FOP developers will pay some attention to the needs for improved FOP features and perhaps give advice to the DITA-OT developers to use FOP in the best possible way. I am trying to get the DITA side to stop considering FOP to be a static thing that can not be changed and to start to contribute ideas and funding to make FOP do the things that it needs to do. I also want to encourage the DITA-OT team to enter into discussions with the FOP experts to make sure that DITA-OT uses FOP in the best possible way. This problem with the leading dots is a good example of the problem. When the problem was raised by a documentation author, one of the leading DITA experts proposed the solution to the problem was to stop trying to make FOP work since it is buggy and inconsistent rather than suggesting that the user ask the question in the FOP user forum. When I brought the problem here, an answer was provided that looks like a simple change to DITA-OT's FOP configuration that seemed to be a solution to this problem that is well understood here. Ron On 24/05/2014 9:22 PM, Glenn Adams wrote: I'm not familiar with DITA, but if a DITA product depends on FOP, then DITA as a group or its sponsors should consider funding the work they would like to see done in FOP. Simply asking the few developers in the FOP project to support DITA priorities won't guarantee any results. On Sun, May 25, 2014 at 5:38 AM, Ron Wheeler rwhee...@artifact-software.com mailto:rwhee...@artifact-software.com wrote: You are right , of course. However, this very large community ( one DITA LinkedIn group has 4,600 members, the Technical Doc group has over 14,000 members) does not see themselves as users of FOP but only as users of DITA-OT which in turn has a dependency on FOP. As far as I know DITA is the most popular XML language for constructing documents and it seems odd that there is almost no connection between the Apache efforts in XML and the biggest potential set of users and drivers of demand for the things that XMLGraphics is producing. I will pass on the information to the forum where the question was raised. Thanks On 23/05/2014 6:05 PM, Luis Bernardo wrote: I think this only shows that the person is not going to the source (i.e., the FOP user mailing list) to request help. The example shown can be greatly improved by using fo:leader width=100% leader-pattern=use-content./fo:leader instead of fo:leader leader-pattern=dots / The FOP implementation repeats 3 dots (...) when using the leader-pattern=dots which is not very intelligent since it can lead to misalignment many times. But by specifying the leader content as just one dot the result can be greatly improved, although I agree that there is room for improvement in this feature. On 5/23/14, 4:14 PM, Ron Wheeler wrote: The conversation below shows one of the reasons why I would like to see DITA become part of the XMLGraphics family. One of the most experienced and influential DITA practitioners is giving advice about the suitability of FOP for producing correct PDFs. Ron - This is an issue with the FOP XSL-FO engine (one of many). For top-qualify PDF output you really need to license Antenna House XSL Formatter or RenderX XEP, both of which produce excellent results. FOP simply has too many bugs and limitations to be generally useful for production PDF generation using XSL-FO, unfortunately. Cheers, XXX On 5/23/14, 9:54 AM, yyy [dita-users] dita-us...@yahoogroups.com mailto:dita-us...@yahoogroups.com wrote: [Attachment(s) #TopText from Mark Peters included below] Hi, Using DITA-OT 1.7 and the default org.dita.pdf2 plugin, I notice that some TOC entries are randomly misaligned slightly. The leader extends one or two extra dots. For example (not sure if the formatting will come through correctly. An image is also attached): Chapter 1: RTI4T System Overview...11 RTI System Diagram.12 System Components...12 RTI Network Diagram...15 Summary of RTI Setup Tasks...15 Like I said, the misalignment is random. It's not all H1s or H2s, for example, which would probably be relatively easy to fix. Even more puzzling, the XSL-FO attributes in the temp
Re: Why integrating DITA into XMLGraphics makes sense.
You are right , of course. However, this very large community ( one DITA LinkedIn group has 4,600 members, the Technical Doc group has over 14,000 members) does not see themselves as users of FOP but only as users of DITA-OT which in turn has a dependency on FOP. As far as I know DITA is the most popular XML language for constructing documents and it seems odd that there is almost no connection between the Apache efforts in XML and the biggest potential set of users and drivers of demand for the things that XMLGraphics is producing. I will pass on the information to the forum where the question was raised. Thanks On 23/05/2014 6:05 PM, Luis Bernardo wrote: I think this only shows that the person is not going to the source (i.e., the FOP user mailing list) to request help. The example shown can be greatly improved by using fo:leader width=100% leader-pattern=use-content./fo:leader instead of fo:leader leader-pattern=dots / The FOP implementation repeats 3 dots (...) when using the leader-pattern=dots which is not very intelligent since it can lead to misalignment many times. But by specifying the leader content as just one dot the result can be greatly improved, although I agree that there is room for improvement in this feature. On 5/23/14, 4:14 PM, Ron Wheeler wrote: The conversation below shows one of the reasons why I would like to see DITA become part of the XMLGraphics family. One of the most experienced and influential DITA practitioners is giving advice about the suitability of FOP for producing correct PDFs. Ron - This is an issue with the FOP XSL-FO engine (one of many). For top-qualify PDF output you really need to license Antenna House XSL Formatter or RenderX XEP, both of which produce excellent results. FOP simply has too many bugs and limitations to be generally useful for production PDF generation using XSL-FO, unfortunately. Cheers, XXX On 5/23/14, 9:54 AM, yyy [dita-users] dita-us...@yahoogroups.com wrote: [Attachment(s) #TopText from Mark Peters included below] Hi, Using DITA-OT 1.7 and the default org.dita.pdf2 plugin, I notice that some TOC entries are randomly misaligned slightly. The leader extends one or two extra dots. For example (not sure if the formatting will come through correctly. An image is also attached): Chapter 1: RTI4T System Overview...11 RTI System Diagram.12 System Components...12 RTI Network Diagram...15 Summary of RTI Setup Tasks...15 Like I said, the misalignment is random. It's not all H1s or H2s, for example, which would probably be relatively easy to fix. Even more puzzling, the XSL-FO attributes in the temp files are identical for nodes at the same level that have different alignments. For example, here are two H1-level nodes. The first node is slightly misaligned. But the indent values are identical for both nodes. fo:block start-indent=25pt + (1 * 30pt) + 14pt fo:block end-indent=22pt font-size=10pt font-weight=normal last-line-end-indent=-22pt text-align=justify text-align-last=justify text-indent=-14pt fo:basic-link internal-destination=_OPENTOPIC_TOC_PROCESSING_d70e1310 line-height=150% fo:inline end-indent=14ptRTI System Diagram/fo:inline fo:inline keep-together.within-line=always start-indent=-14pt fo:leader leader-pattern=dots/ fo:page-number-citation ref-id=_OPENTOPIC_TOC_PROCESSING_d70e1310/ /fo:inline /fo:basic-link /fo:block /fo:block fo:block start-indent=25pt + (1 * 30pt) + 14pt fo:block end-indent=22pt font-size=10pt font-weight=normal last-line-end-indent=-22pt text-align=justify text-align-last=justify text-indent=-14pt fo:basic-link internal-destination=_OPENTOPIC_TOC_PROCESSING_d70e1334 line-height=150% fo:inline end-indent=14pt System Components /fo:inline fo:inline keep-together.within-line=always start-indent=-14pt fo:leader leader-pattern=dots/ fo:page-number-citation ref-id=_OPENTOPIC_TOC_PROCESSING_d70e1334/ /fo:inline /fo:basic-link /fo:block /fo:block I'm viewing the PDFs in Abobe Reader, but that hasn't made a difference in the past. Any idea what's going on? Thanks for any insights. yyy -- Ron Wheeler President Artifact Software Inc email:rwhee...@artifact-software.com skype: ronaldmwheeler phone: 866-970-2435, ext 102 -- Ron Wheeler President Artifact
Re: Why integrating DITA into XMLGraphics makes sense.
I'm not familiar with DITA, but if a DITA product depends on FOP, then DITA as a group or its sponsors should consider funding the work they would like to see done in FOP. Simply asking the few developers in the FOP project to support DITA priorities won't guarantee any results. On Sun, May 25, 2014 at 5:38 AM, Ron Wheeler rwhee...@artifact-software.com wrote: You are right , of course. However, this very large community ( one DITA LinkedIn group has 4,600 members, the Technical Doc group has over 14,000 members) does not see themselves as users of FOP but only as users of DITA-OT which in turn has a dependency on FOP. As far as I know DITA is the most popular XML language for constructing documents and it seems odd that there is almost no connection between the Apache efforts in XML and the biggest potential set of users and drivers of demand for the things that XMLGraphics is producing. I will pass on the information to the forum where the question was raised. Thanks On 23/05/2014 6:05 PM, Luis Bernardo wrote: I think this only shows that the person is not going to the source (i.e., the FOP user mailing list) to request help. The example shown can be greatly improved by using fo:leader width=100% leader-pattern=use-content./fo:leader instead of fo:leader leader-pattern=dots / The FOP implementation repeats 3 dots (...) when using the leader-pattern=dots which is not very intelligent since it can lead to misalignment many times. But by specifying the leader content as just one dot the result can be greatly improved, although I agree that there is room for improvement in this feature. On 5/23/14, 4:14 PM, Ron Wheeler wrote: The conversation below shows one of the reasons why I would like to see DITA become part of the XMLGraphics family. One of the most experienced and influential DITA practitioners is giving advice about the suitability of FOP for producing correct PDFs. Ron - This is an issue with the FOP XSL-FO engine (one of many). For top-qualify PDF output you really need to license Antenna House XSL Formatter or RenderX XEP, both of which produce excellent results. FOP simply has too many bugs and limitations to be generally useful for production PDF generation using XSL-FO, unfortunately. Cheers, XXX On 5/23/14, 9:54 AM, yyy [dita-users]dita-us...@yahoogroups.com dita-us...@yahoogroups.com wrote: [Attachment(s) #TopText from Mark Peters included below] Hi, Using DITA-OT 1.7 and the default org.dita.pdf2 plugin, I notice that some TOC entries are randomly misaligned slightly. The leader extends one or two extra dots. For example (not sure if the formatting will come through correctly. An image is also attached): Chapter 1: RTI4T System Overview...11 RTI System Diagram.12 System Components...12 RTI Network Diagram...15 Summary of RTI Setup Tasks...15 Like I said, the misalignment is random. It's not all H1s or H2s, for example, which would probably be relatively easy to fix. Even more puzzling, the XSL-FO attributes in the temp files are identical for nodes at the same level that have different alignments. For example, here are two H1-level nodes. The first node is slightly misaligned. But the indent values are identical for both nodes. fo:block start-indent=25pt + (1 * 30pt) + 14pt fo:block end-indent=22pt font-size=10pt font-weight=normal last-line-end-indent=-22pt text-align=justify text-align-last=justify text-indent=-14pt fo:basic-link internal-destination=_OPENTOPIC_TOC_PROCESSING_d70e1310 line-height=150% fo:inline end-indent=14ptRTI System Diagram/fo:inline fo:inline keep-together.within-line=always start-indent=-14pt fo:leader leader-pattern=dots/ fo:page-number-citation ref-id=_OPENTOPIC_TOC_PROCESSING_d70e1310/ /fo:inline /fo:basic-link /fo:block /fo:block fo:block start-indent=25pt + (1 * 30pt) + 14pt fo:block end-indent=22pt font-size=10pt font-weight=normal last-line-end-indent=-22pt text-align=justify text-align-last=justify text-indent=-14pt fo:basic-link internal-destination=_OPENTOPIC_TOC_PROCESSING_d70e1334 line-height=150% fo:inline end-indent=14pt System Components /fo:inline fo:inline keep-together.within-line=always start-indent=-14pt fo:leader leader-pattern=dots/ fo:page-number-citation ref-id=_OPENTOPIC_TOC_PROCESSING_d70e1334/
Re: Why integrating DITA into XMLGraphics makes sense.
I think this only shows that the person is not going to the source (i.e., the FOP user mailing list) to request help. The example shown can be greatly improved by using fo:leader width=100% leader-pattern=use-content./fo:leader instead of fo:leader leader-pattern=dots / The FOP implementation repeats 3 dots (...) when using the leader-pattern=dots which is not very intelligent since it can lead to misalignment many times. But by specifying the leader content as just one dot the result can be greatly improved, although I agree that there is room for improvement in this feature. On 5/23/14, 4:14 PM, Ron Wheeler wrote: The conversation below shows one of the reasons why I would like to see DITA become part of the XMLGraphics family. One of the most experienced and influential DITA practitioners is giving advice about the suitability of FOP for producing correct PDFs. Ron - This is an issue with the FOP XSL-FO engine (one of many). For top-qualify PDF output you really need to license Antenna House XSL Formatter or RenderX XEP, both of which produce excellent results. FOP simply has too many bugs and limitations to be generally useful for production PDF generation using XSL-FO, unfortunately. Cheers, XXX On 5/23/14, 9:54 AM, yyy [dita-users] dita-us...@yahoogroups.com wrote: [Attachment(s) #TopText from Mark Peters included below] Hi, Using DITA-OT 1.7 and the default org.dita.pdf2 plugin, I notice that some TOC entries are randomly misaligned slightly. The leader extends one or two extra dots. For example (not sure if the formatting will come through correctly. An image is also attached): Chapter 1: RTI4T System Overview...11 RTI System Diagram.12 System Components...12 RTI Network Diagram...15 Summary of RTI Setup Tasks...15 Like I said, the misalignment is random. It's not all H1s or H2s, for example, which would probably be relatively easy to fix. Even more puzzling, the XSL-FO attributes in the temp files are identical for nodes at the same level that have different alignments. For example, here are two H1-level nodes. The first node is slightly misaligned. But the indent values are identical for both nodes. fo:block start-indent=25pt + (1 * 30pt) + 14pt fo:block end-indent=22pt font-size=10pt font-weight=normal last-line-end-indent=-22pt text-align=justify text-align-last=justify text-indent=-14pt fo:basic-link internal-destination=_OPENTOPIC_TOC_PROCESSING_d70e1310 line-height=150% fo:inline end-indent=14ptRTI System Diagram/fo:inline fo:inline keep-together.within-line=always start-indent=-14pt fo:leader leader-pattern=dots/ fo:page-number-citation ref-id=_OPENTOPIC_TOC_PROCESSING_d70e1310/ /fo:inline /fo:basic-link /fo:block /fo:block fo:block start-indent=25pt + (1 * 30pt) + 14pt fo:block end-indent=22pt font-size=10pt font-weight=normal last-line-end-indent=-22pt text-align=justify text-align-last=justify text-indent=-14pt fo:basic-link internal-destination=_OPENTOPIC_TOC_PROCESSING_d70e1334 line-height=150% fo:inline end-indent=14pt System Components /fo:inline fo:inline keep-together.within-line=always start-indent=-14pt fo:leader leader-pattern=dots/ fo:page-number-citation ref-id=_OPENTOPIC_TOC_PROCESSING_d70e1334/ /fo:inline /fo:basic-link /fo:block /fo:block I'm viewing the PDFs in Abobe Reader, but that hasn't made a difference in the past. Any idea what's going on? Thanks for any insights. yyy -- Ron Wheeler President Artifact Software Inc email:rwhee...@artifact-software.com skype: ronaldmwheeler phone: 866-970-2435, ext 102
Re: Why integrating DITA into XMLGraphics makes sense.
You get what you pay for. If you want to invest resources to improve FOP, you are free to do so. On Sat, May 24, 2014 at 12:14 AM, Ron Wheeler rwhee...@artifact-software.com wrote: The conversation below shows one of the reasons why I would like to see DITA become part of the XMLGraphics family. One of the most experienced and influential DITA practitioners is giving advice about the suitability of FOP for producing correct PDFs. Ron - This is an issue with the FOP XSL-FO engine (one of many). For top-qualify PDF output you really need to license Antenna House XSL Formatter or RenderX XEP, both of which produce excellent results. FOP simply has too many bugs and limitations to be generally useful for production PDF generation using XSL-FO, unfortunately. Cheers, XXX On 5/23/14, 9:54 AM, yyy markpeters.w...@gmail.com [dita-users]dita-us...@yahoogroups.com dita-us...@yahoogroups.com wrote: [Attachment(s) #TopText from Mark Peters included below] Hi, Using DITA-OT 1.7 and the default org.dita.pdf2 plugin, I notice that some TOC entries are randomly misaligned slightly. The leader extends one or two extra dots. For example (not sure if the formatting will come through correctly. An image is also attached): Chapter 1: RTI4T System Overview...11 RTI System Diagram.12 System Components...12 RTI Network Diagram...15 Summary of RTI Setup Tasks...15 Like I said, the misalignment is random. It's not all H1s or H2s, for example, which would probably be relatively easy to fix. Even more puzzling, the XSL-FO attributes in the temp files are identical for nodes at the same level that have different alignments. For example, here are two H1-level nodes. The first node is slightly misaligned. But the indent values are identical for both nodes. fo:block start-indent=25pt + (1 * 30pt) + 14pt fo:block end-indent=22pt font-size=10pt font-weight=normal last-line-end-indent=-22pt text-align=justify text-align-last=justify text-indent=-14pt fo:basic-link internal-destination=_OPENTOPIC_TOC_PROCESSING_d70e1310 line-height=150% fo:inline end-indent=14ptRTI System Diagram/fo:inline fo:inline keep-together.within-line=always start-indent=-14pt fo:leader leader-pattern=dots/ fo:page-number-citation ref-id=_OPENTOPIC_TOC_PROCESSING_d70e1310/ /fo:inline /fo:basic-link /fo:block /fo:block fo:block start-indent=25pt + (1 * 30pt) + 14pt fo:block end-indent=22pt font-size=10pt font-weight=normal last-line-end-indent=-22pt text-align=justify text-align-last=justify text-indent=-14pt fo:basic-link internal-destination=_OPENTOPIC_TOC_PROCESSING_d70e1334 line-height=150% fo:inline end-indent=14pt System Components /fo:inline fo:inline keep-together.within-line=always start-indent=-14pt fo:leader leader-pattern=dots/ fo:page-number-citation ref-id=_OPENTOPIC_TOC_PROCESSING_d70e1334/ /fo:inline /fo:basic-link /fo:block /fo:block I'm viewing the PDFs in Abobe Reader, but that hasn't made a difference in the past. Any idea what's going on? Thanks for any insights. yyy -- Ron Wheeler President Artifact Software Inc email: rwhee...@artifact-software.com skype: ronaldmwheeler phone: 866-970-2435, ext 102