Evaluation Report Comments

Read through the report and just a couple of comments:

   - In the description of the web corpus there's no mention of what fonts
   were used. I suggest mentioning that the font corpus consisted entirely of
   fonts from the open source Google Fonts collection (found here:
   https://github.com/google/fonts).
   - You link to the raw results in the appendix of the simulation results
   document, may also be worth pointing out the raw results can be found in
   CSV form here:
   https://github.com/w3c/PFE-analysis/tree/master/results/07-08-2020
   - I mentioned this at the meeting today, but I'll mention it again here:
   patch subset (and possibly range request) can be made to automatically
   behave as 'whole font' dependent on the network conditions. This could be
   used to ensure that in the worst performing cases (such as 2G) it has
   identical performance to whole font. This is important because it means
   that site operators don't need special case 2G and 3G font loading if they
   decide to use a PFE technique. I think it would be good to point this out
   in the eval report.