On Mon, Feb 10, 2014 at 3:41 AM, Rainer M Krug <rai...@krugs.de> wrote:
> > > On 02/09/14, 19:44 , Georg Baum wrote: > > Rainer M Krug wrote: > > > >> The idea would be that a round-trip framework is envisaged, which > >> provides the facilities to easily expand it from one export backend > >> (docx) to another (possibly odt? markdown?). > >> > >> IMPORTANT: this would NOT change ANYTHING in the existing export / > >> import features, as these are geared to export / import the documents as > >> good as possible, with maintaining as many features as possible in the > >> document. > >> > >> The round-trip would guarantee that: > >> > >> A document authored in LyX would result in a e.g. docx with a LIMITED > >> set of features, but that a re-import would result in the SAME .lyx > >> file. features and formats not supported by the backend should be stored > >> in a metadata file. > >> > >> The important point here is *limited set of features*! > >> > >> In addition, the framework should be easily, possibly only by using > >> config files, able to be extended to other formats. > > > > I don't understand the difference between round trip and the existing > > export/import here. Why is it important? If the additional metadata is > > stored in a different file, it could simply be generated for the standard > > export, and be used by the standard import (if it exists). > > > > The goal of the export/import is to support as many features as possible. > > This is needed for round trip as well. The only difference I see is the > > additional metadata file, so the roundtrip framework vs. export/import > > difference reduces to a switch whether the metadata file should be > generated > > (for export) or used (for import). Or did I understand anything wrong? > > The difference is that for round-trip, i.e. working together with > co-authors and getting comments back, a different set of features are > relevant. These are mainly concerned about content and not that much > formating. The import - export is concerned with both. In addition, a > round trip has to be symmetric, i.e. that exported features have to be > available in the re-importd as well - this is not the case in the export > and import. Lastly, round-trip is for editing, and export - import is > for editing and final consumption (reading). > > I actually disagree on this point: the most useful doc-export facility for LyX would be equally focused on semantic content and not on formatting. In other words, it would be just half (or slightly less than half) of the round-trip project. The rationale is simple: exporting to doc(x) makes sense and is actually required when working with a third party (typically, for Lyx's main audience, with a publisher) who will then either provide final formatting directly with Word (the worst case) or will use the doc(x) file as import into a real typesetting program (InDesign, etc). In neither case formatting instructions are relevant. I think it is a losing proposition to aim for the preservation of format when exporting to Word---and in fact it is the reason why, in my experience, *all* latex-to- word- (or to-odt) or lyx-to-word exporters actually fail in practice. It is impossibly hard to provide the same pdf look that (la)tex produces with Word. And the use cases in which this conversion is required are exceedingly rare. Far more pressing for our user base is the need to guarantee a hassle-free 100% valid export to a "sanitized word format" which is narrowly restricted, on both sides, to the semantic information contained in LyX. To put it more bluntly (and to repeat what I and others have stated many times in the past): LyX is barely usable right now for any academic work in the Humanities, due to the necessity to deliver doc documents to virtually any publisher. If you are a student, you are similarly asked by professor to submit drafts in Word format. </rant>I had to convert a ~50,000 words book from LyX to Word last month and it took me 2 full days. I think I tried all exporters known to men (and women). They all failed to various degrees. In the end, I had better luck converting the file from the pdf (!!!!!) output to word and then reinserting manually all footnotes (all 450 of them). I am facing the prospect of converting a 200,000 words manuscript in a few months and I am already sweating at night at the very idea. <\rant> Anyway: I am willing to "mentor" a student through the process of producing a LyX-to-Word semantic-only exporter. Scare quotes are necessary, because I would have to learn as much as the student. If Rob can provide some guidance and expert advice (both as a previous mentor and obviously as an expert in the area) I think we may have something working by the end of the summer. What I know is that *I* will absolutely a solid word-exporter by that time. Cheers, Stefano -- __________________________________________________ Stefano Franchi Associate Research Professor Department of Hispanic Studies Ph: +1 (979) 845-2125 Texas A&M University Fax: +1 (979) 845-6421 College Station, Texas, USA stef...@tamu.edu http://stefano.cleinias.org
