Jim Gettys jg at
Thu Apr 9 06:31:01 PDT 2009

If you look at our paper here:

You'll discover that the font metadata turned out to be as large as
the glyphs actually used.

And client side fonts with server caching therefore turns out to be a
wash as far as bits transferred in practice, while avoiding a huge
number of round trips to get font metadata (which was always
inadequate), and *horrible* for application performance at starup.

So the X11 core font design is fundamentally a mistake, which we fixed.
                       - Jim

On Thu, 2009-04-09 at 15:18 +0200, Olivier Galibert wrote:
> On Thu, Apr 09, 2009 at 02:05:51PM +0100, Alan Cox wrote:
> > Each to their own, I know which I find easier to read and I know what
> > extensive studies say people prefer as well.
> To each their own indeed.
> > Also be careful with the images to view them full size - if your browser
> > is scaling them because of their size you'll get very funny and
> > misleading results.
> Oh yeah, I'm careful with that.
> > Not of course that anti-aliased fonts are really an X issue anyway - it a
> > client side matter...
> I've always wondered why.  It makes no sense.  The network-oriented
> nature of X means you should do your best to send as little data as
> possible, and prerendered pixmaps are nowhere near minimal.  Why isn't
> fontconfig/xft and even pango in the server where it seems to belong?
>   OG.
> _______________________________________________
> xorg mailing list
> xorg at
Jim Gettys <jg at>

More information about the xorg mailing list