From: Kent Karlsson (kent.karlsson14@comhem.se)
Date: Mon Jul 16 2007 - 14:58:28 CDT
Asmus Freytag wrote:
> > While I would appreciate if more fonts supported the fj ligature,
> > I would expect no rendering system or font to insert a dotted
> > circle between an f followed by a j just because they don't
> > support that ligature.
> Neither would I. It would be a cause for extreme customer
> dissatisfaction ;-)
> > Instead they just output an f followed
> > by a j, though the result sometimes is not perfect (but much
> > better than getting a dotted circle in-between).
> >
> This is because this particular fallback is so eminently
> preferable in
> this example - not (only) because Unicode says so, but from
> the logic of how the script works.
And this is not script dependent. Any insertion of spurious
(graphic) characters leads to customer dissatisfaction...
> Clearly a matter of opinion. The standard suggests that it's
> better to have some base character, and, in the absence of
> a higher level protocol, that this should be the NBSP.
I think that should be the *normal* case also when there is a
higher-level protocol. Things like "show *ALL* combining characters
with a dotted circle", "show invisibles", etc. are out of scope
for this discussion.
> > That would be better than inserting spurious dotted circles.
> >
> I see that you just can't abide by dotted circles ;-)
I cannot abide insertion of spurious characters in normal
display mode. And I don't see why anyone else should.
/kent k
This archive was generated by hypermail 2.1.5 : Mon Jul 16 2007 - 14:59:35 CDT