Re: Somebody must package djvulibre
On Fri, Jan 25, 2002 at 11:19:03AM -0500, Christian Hudon wrote:
> le mer 23-01-2002 ? 19:15, elf@florence.buici.com a ?crit :
> > I looked into packaging this at one time and decided that there wasn't
> > a compelling reason to do so. IIRC, the algorithms that they
> > distribute aren't complete. It doesn't include a compressor that is
> > on par with the patented code. Thought I didn't test it, it seemed to
> > be uninteresting except for viewers. This may be worthwhile if we are
> > (someone is) interesting in supporting images created on MS platforms.
> >
>
> Whether the free code's compressor is on par with the proprietary code's
> doesn't seem very relevant to me. The interesting question is whether
> the free code's compressor compresses scanned documents better than
> other free solutions out there. I haven't tried it out (I'd need a
> source for scanned documents first), but I wouldn't be surprised if it
> were the case.
As for sources of scanned documents, I remember that there was a set
of documents used to validate the CCITT and JBIG compressors. These
ought to be available on the net. Using them should give a valid
point for comparison.
Reply to: