> --- Taras <[EMAIL PROTECTED]> wrote: > > Adam McDaniel wrote: > > >Taras wrote: > > > >I second that opinion. It is very annoying if you > > > >have pages that are basicly long tables of > > > >contents and they get cut into 3-4 pages. Makes > > > >scrolling to find the right word very painful. > > > > > > >Let's say that you have an HTML document that's > > >200k long. When you parse it it splits it up into > > >6 pages in the viewer (~32k each, etc etc). To view > > >just one of those pages there is a load time of 10 > > >seconds until you can actually begin to read and > > >scrolldown on that one page. > > > > > >Would you be willing to instead have a loadtime of > > >60 seconds just so the viewer can properly re-build > > >everything into one complete page, just like the > > >original? > > > > >However can't you do incremental loading? So > >somewhere the length of the document would be stored > >and scrollbar would be sized accordingly. then as you > >get close enough to the unloaded segment it would > >load it. I suspect that's how isilo gets around the > >problem > > I'm not a Palm OS programmer, just a user, but I > wonder if anyone has looked at the source code used by > Weasel Reader (GNU GPL). I can open The Three > Musketeers (DOC format, SD card) in less than 30 > seconds and at top scroll speed (about 22,000 cpm) I > notice no blips in the first 100k or so (I got bored > after that). > > I realize that Plucker might not be able to achieve > quite that speed due depending on the decompression > algorithm but, like the others who have commented, I > think the front end time spent is worth it in back end > usability.
FYI, using a doctored version of the Plucker parser which uses the "reserved" field next to the record type id as a continuation character to produce a "continuous" version (ie no click here for next part etc) I can open The Three Musketeers in modified plucker format in less than a second and scroll without blips for as much as I like (which is not much as I don't like scrolling) using OpieReader. Anyone want to upgrade to a Zaurus? (But see below, first). The main difference between the Zaurus reader and the Palm reader is, I think, that the Zaurus reader renders line by line rather than page by page and uses "number of expanded characters" (ie text only, no pictures) as a progress measurement (so doesn't need to look at the content of a record to find out how big it is). I'm not sure how well this would work on the Palm architecture. OTOH, the Zaurus reader was written as an eBook reader and there are innumerable advantages in using the Palm viewer for web-like plucks (probably for eBooks, too, for that matter - it is a far more mature product). The main reason I mention this (apart from illustrating how cool the Zaurus is ;^) ) is that maybe some of this might be useful for the Palm reader development - feel free to ask questions or for the source code (C++ so not directly useful). Cheers Tim _______________________________________________ plucker-list mailing list [EMAIL PROTECTED] http://lists.rubberchicken.org/mailman/listinfo/plucker-list
