[encoding] Last Call Comment: Choice of encoding details

This is a Last Call comment on 
http://www.w3.org/TR/2014/WD-encoding-20140603/.

Extensive tests conducted by a student of mine have shown that for 
single-byte encodings, Opera <=12 (i.e. Presto rendering engine) is the 
most conformant, often perfect, while other rendering engines have more 
errors.

As an example, for "windows-874", Presto is perfectly conforming, while 
Firefox shows 23 errors, and IE, Chrome, Safari, and Opera >=16 show 
exactly the same 8 errors.

If the goal of this spec, as I understand, is to help browsers 
transition to uniform treatment of encodings, and if this allegedly can 
be best achieved by considering browser market share, then some of the 
encoding definitions should be carefully revisited.

When it comes to passing CR, does the WG intend to count the old Opera 
browser? I don't think this would be appropriate, because this browser 
is no longer being developed or distributed.

Regards,   Martin.

Received on Wednesday, 2 July 2014 11:16:03 UTC