[Latingp] Latin-Cyrillic cross-script analysis
Michael.Bauland at knipp.de
Fri Jul 27 15:29:14 UTC 2018
On 26.07.2018 21:12, Bill Jouris wrote:
> I have reviewed these. In several cases, I believe it is possible to
> resolve the issue using the same principle:
> When both inspectors have the same rating for the highest ranked font,
> and the only difference is between fonts, there seems no reason to look
> further. Just take the agreed highest rating (for example 2) and move on.
I agree with Bill. We decided to look at several fonts in order to find
more cases for possible variants (not to find less cases). Therefore it
makes sense to always take the rating from the highest ranked font. In
case both inspectors agree, there is nothing more to do.
> I also noticed a couple of cases where the first inspector had a rating
> of 5 (i.e. no candidate found) while the second inspector had something
> else. Perhaps the first inspector could go back and consider the
> candidate which, apparently, arose after his initial inspection. Even
> if that results in a rating of 4 (different), it would at least give us
> someplace to start discussion.
Done. Those were my cases and I changed my original 5 to the rating
referring to the newly found candidate.
Furthermore I added my vote to the occasions were first and second
inspector disagreed (and Bill's suggestion from above does not work). If
the others could do the same, we could probably apply some majority vote.
Have a nice weekend
| knipp | Knipp Medien und Kommunikation GmbH
Dipl.-Informatiker Fon: +49 231 9703-0
Fax: +49 231 9703-200
Dr. Michael Bauland SIP: Michael.Bauland at knipp.de
Software Development E-mail: Michael.Bauland at knipp.de
Amtsgericht Dortmund, HRB 13728
Chief Executive Officers:
Dietmar Knipp, Elmar Knipp
More information about the Latingp