On 5/21/2014 8:16 AM, Philipp Gesang wrote:

http://www.informit.com/articles/article.aspx?p=2213858&WT.mc_id=Author_Knuth_20Questions

Quote:

“The people at MSP wrote special software that converts my TeX source text into suitable input to other software that creates pdf files.”

I wonder why using a TeX engine capable of producing PDF output directly wasn’t an option.

Don't overestimate the large publishers. They often have no clue what interactive pdf is or what is needed to make it, and that it can be done easily. Also, as they have no r&d, they seldom know that producing both an advanced screen pdf and printable one from the same source without adaption is an option. This apart from the fact that there is really nothing new happening: it's just that we now have devices that run on batteries longer (we made out first interactive docs 15-20 years ago and there were simply no (cheap) devices then. But the techniques for producing those docs (as well as the features) are mostly the same. Things that are normal for us texies are alien for most publishers, also because they outsource these things. They are not aware of tex (maybe only know the name) and of what it can(not) do. A publisher who knows how to apply tex and friends has definitely an advantage but that demands a horizon different from stockholders. In fact, awareness of pdf as option is (imo) mostly surfacing because html viewers opened their eyes. We can produce highly interactive documents but I still have to meet the first publishers who really wants them (given that they see it as option). It is not uncommon for publishers to waste many man years (hiring extra people) on getting stuff on the web using the same sources that are used to get the paper: paper becomes a side effect. (This is also why I would not be surprised if within a decade those publishers will disappear and be replaced by smaller ones and/or self publishing, but that's another matter.) Concerning DEK's books: given that the layout is kind of fixed (as Don puts a lot of admirable effort in the whole look and feel) I assume we're talking of interactive variants of the same layout. In that case it's a matter of adapting macros (not too hard, would even be fun to do) to get the interactivity in there. I'm pretty sure that Don knows well what happens in tex community (including recent developments like luatex) but that doesn't change the fact that in the end the publisher rules. Maybe these e-books have some banding (like printing the user's name someplace) but that is also ancient trickery (which in my opinion is also a bad thing: imagine your paper books or music you buy being branded). Hans (I'm still waiting for 500+ dpi e-ink kind of devices as reading from a too bright lcd screen is no real option for me yet, apart from a bit of browsing.) ----------------------------------------------------------------- Hans Hagen | PRAGMA ADE Ridderstraat 27 | 8061 GH Hasselt | The Netherlands tel: 038 477 53 69 | voip: 087 875 68 74 | www.pragma-ade.com | www.pragma-pod.nl -----------------------------------------------------------------

On 5/21/2014 8:16 AM, Philipp Gesang wrote:

http://www.informit.com/articles/article.aspx?p=2213858&WT.mc_id=Author_Knuth_20Questions

Quote:

“The people at MSP wrote special software that converts my TeX source text into suitable input to other software that creates pdf files.”

I wonder why using a TeX engine capable of producing PDF output directly wasn’t an option.

Ok, I think this is what sums it: quote: "My principal goal was to make the books easily searchable—and that's a much more challenging problem than it seems, if you want to do it right. Secondarily, I wanted to let readers easily click on the number of any exercise or equation or illustration or table or algorithm, etc., and to jump to that exercise; also to jump readily between an exercise and its answer." (1) The interactive bit is rather trivial: cross references, tocs and indices, linking answers and questions, are already generated so there one only needs to add some hyperlinking code (maybe some additional housekeeping if backlinks or in the case of indices linked backlinks are supported but it is probably kept simple in order not to spoil the look and feel.) (2) The searching might be more complex as traditional 8 bit fonts are used in plain tex, so there is some tounicode magic needed (not too hard either, mostly housekeeping and careful checking) ... math can be somewhat complex there as not all is copyable, given that it's needed, but on the other hand, searching is also non trivial: does a reader search for "unicode symbol math italic i" or for just an "i"? Copying math demands a proper tounicode, but I bet no viewer can search for both kind of tagging at the same time (math alphabet stuff). Searching regular text of course is no issue. (3) The question is: has more been added than just basic access as in books? Crosslinking books and so? So, no rocket science, mostly careful keeping the original (not interfering with breaks and so) so that is probably where most of the work went into. If pdftex was used, one thing to keep in mind is that adding whatsits can interfere sometimes. This can be avoided by luatex but then one has to accept the fact that the output can differ (no lig recomposition), although that can be checked by diffing pdf's and making local exceptions. A nice challenge: kind of making an electronic facsimile i.e. don't touch (spoil) the original. Hans ps. An interesting audience, those asking the questions. ----------------------------------------------------------------- Hans Hagen | PRAGMA ADE Ridderstraat 27 | 8061 GH Hasselt | The Netherlands tel: 038 477 53 69 | voip: 087 875 68 74 | www.pragma-ade.com | www.pragma-pod.nl -----------------------------------------------------------------