I'm not so happy with the result. The parchment is too three-dimensional, which clashes with the font. I'd recommend regular (or calligraphy) paper (maybe printed with a non-white background, if your printer is good enough), and then "aging" it. Tea or coffee is usually a good method, although simple crumpling might suffice.
Also, either for a medieval source or a 1920s document, block lettering seems a bit off.
But this really shows what's possible if you have programmatic control over your typesetting. I really should do some (La)TeX again…
For a 1920s document, I would actually expect it to be (badly) typewritten, but perhaps I'm injecting too much of modern myth of the times. Certainly cursive handwriting was considered to be vital at the time.
As for medieval, you can actually do pretty well with the medieval Italics or my personal favorite, the Carolingian miniscule. Blackletter is good for late period but it's very difficult for modern folks to read.
Not as hard as some period cursive variants (e.g. German Sütterlin). But once you go that far back, you'd also have to emulate the language, and I don't think that most GMs are a) up to that and b) willing to inflict this upon their players. Some fake Shakespearean might do, of course…
One could scour the net for a handwritten font that looks slightly arabic, considering where the Mythos' most famous madman came from…
Back in the days, I had some Postscript document that actually included some font, based on Edsger Dijkstra's handwriting. Given that PS is a full-fledge programming language, too, that could be the basis for some crazy per-letter randomization. (Although I bet TeX/MetaFont wizards could do something similar, and there might even be some OpenType fonts with enough glyph variants out there already)
Also, either for a medieval source or a 1920s document, block lettering seems a bit off.
But this really shows what's possible if you have programmatic control over your typesetting. I really should do some (La)TeX again…