My last post was about headers, and in the post before that I wrote optimistically about sIFR—at which I’ve now had a closer look.
The two psosts turn out to be related, since, in the words of its developers, sIFR is not really usable for ordinary text. Its main use is for headers, pull-quotes, etc. So all it effectively does is provide a slightly better alternative to textual images.
Proper universal font embedding is still a dream.
