Re: Announcing new font compression project

On Fri, Mar 30, 2012 at 2:05 PM, John Daggett <[email protected]> wrote:
> Vladimir Levantovsky wrote:
>
>> 3) I=92d like to commend Raph for his efforts in dissecting the MTX
>> and creating a new, much more efficient compression engine and its
>> reference implementation that allows to precisely quantify the
>> individual contribution of each processing/optimization step to
>> overall compression efficiency. By doing this he enables us to have
>> an informed and intelligent discussion where each piece of the
>> compression puzzle can be evaluated based on its complexity and
>> resulting data reduction percentage, and where the final decision on
>> the new compression algorithm and stream format can be made based on
>> real-life experimental results and not just paper specs. Great job,
>> Raph!
>
> I'd like to chime in here and also thank Raph and his group for doing all
> the legwork involved in putting this proposal together. =A0Just as Vlad d=
escribes,
> the work that's been done so far gives us enough tools to begin to analyz=
e
> the strengths and weaknesses of the proposal by comparing the results for
> real-world situations.

Now that I've had time to fully digest this and read the docs
carefully, I wanted to say: Wow. This is awesome. Very well
considered. I have passed it on to the rest of my team, but it looks
very well thought out.

It will be interesting to see the trade-off of compression gains
versus performance costs analyzed in detail for different elements of
the scheme. Will the ideal tradeoff be different for mobile devices?
They have both (often) lower bandwidth and usually less powerful CPUs,
upping the stakes for finding the sweet spot in that trade-off....

T

--=20
=93Why shouldn't truth be stranger than fiction? Fiction, after all, has
to make sense.=94
=A0=97Mark Twain