<div dir="ltr"><div class="gmail_default" style="font-family:verdana,sans-serif"><span style="font-size:small">Back at the time, Adobe (represented in the discussions by me and David Lemon) was in favor of doing ss01-99, and Microsoft objected—for much the reasons John suggests.</span></div><div class="gmail_default" style="font-family:verdana,sans-serif"><span style="font-size:small"><br></span></div><div class="gmail_default" style="font-family:verdana,sans-serif"><span style="font-size:small">Of course, that was 15 years ago, and thinking may have evolved. I remain in favor of allowing up to 99. If not all apps can expose them all, that would be too bad, but I would rather have them in the spec, and apps encouraged to evolve their interfaces, rather than not.</span></div><div class="gmail_default" style="font-family:verdana,sans-serif"><span style="font-size:small"><br></span></div><div class="gmail_default" style="font-family:verdana,sans-serif"><span style="font-size:small">Trivia: I believe you can tell that this feature was at least partly created by non-programmers, because the numbering starts at 01 instead of 00. ;)</span></div><div class="gmail_default" style="font-family:verdana,sans-serif"><br></div></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Wed, Mar 20, 2019 at 3:07 PM John Hudson <a href="mailto:john@tiro.ca">john@tiro.ca</a> [mpeg-OTspec] <<a href="mailto:mpeg-OTspec-noreply@yahoogroups.com">mpeg-OTspec-noreply@yahoogroups.com</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
<u></u>
<div style="background-color:rgb(255,255,255)">
<span style="display:none"> </span>
<div id="gmail-m_2204528273174580476ygrp-mlmsg">
<div id="gmail-m_2204528273174580476ygrp-msg">
<div id="gmail-m_2204528273174580476ygrp-text">
<p>On 20032019 12:40 pm, Ken Lunde <a href="mailto:lunde@adobe.com" target="_blank">lunde@adobe.com</a> [mpeg-OTspec] wrote:<br>
> At this point, based on the discussions thus far, I doubt that anyone <br>
> can provide a convincing argument against registering 'ss21' through <br>
> 'ss99' as additional Stylistic Set features. <br>
<br>
Well...<br>
<br>
The original intention of the Stylistic Set features was to provide <br>
access for coordinated sets of design variants of complete or <br>
significant portions of a character subset, e.g. all lowercase letters, <br>
grouped by shared style. The initial use case was the OpenType-ification <br>
of the Poetica and Zapfino families, in which the stylistic sets had <br>
been shipped as separate fonts in their pre-OT incarnations. The <br>
decision to limit the number of Stylistic Set features to twenty was <br>
influenced by a couple of factors: one was that the number of stylistic <br>
sets in Poetica and Zapfino was four, so twenty seemed like quite a lot, <br>
and the other was that a smaller number was more likely to get buy-in <br>
from applications needing to give some kind of UI real-estate to the <br>
features, possibly à la InDesign with a menu listing (I'll leave aside <br>
the whole other topic of poor UI design for OpenType Layout).<br>
<br>
What began to happen fairly soon after the Stylistic Set features were <br>
registered and began to show up in applications is that font makers <br>
began using them to provide access to variants of individual characters <br>
instead of sets of characters, e.g. multiple variants of an ampersand, <br>
each mapped to a different Stylistic Set feature. And used in this way <br>
the features very quickly get used up and people start asking why there <br>
aren't more.<br>
<br>
Meanwhile, SIL registered the 0–99 Character Variant features, which not <br>
only, by design, provide access to variants of individual glyphs, but <br>
also recommend doing so using GSUB one-to-one-of-many lookups, rather <br>
than one variant per feature. [It is technically possible, of course, to <br>
build Stylistic Set feature using such lookups, but application UI for <br>
these features tends only to expose the first enumerated variant.]<br>
<br>
The Character Variant features — including enumerated variants — are <br>
supported in CSS, but not in common desktop applications, and so font <br>
makers have continued to use Stylistic Set features to access individual <br>
character variants, and continued to complain that twenty isn't enough <br>
to accommodate this use.<br>
<br>
I don't know if this is a 'convincing argument', but it seems to me that <br>
if one is going to have to ask application makers to add support for 80 <br>
new Stylistic Set features, why not ask them to support the existing 100 <br>
Character Variant features instead?<br>
<br>
JH<br>
<br>
-- <br>
<br>
John Hudson<br>
Tiro Typeworks Ltd <a href="http://www.tiro.com" target="_blank">www.tiro.com</a><br>
Salish Sea, BC <a href="mailto:tiro@tiro.com" target="_blank">tiro@tiro.com</a><br>
<br>
NOTE: In the interests of productivity, I am currently<br>
dealing with email on only two days per week, usually<br>
Monday and Thursday unless this schedule is disrupted<br>
by travel. If you need to contact me urgently, please<br>
use some other method of communication. Thank you.<br>
<br>
</p>
</div>
<div style="color:rgb(255,255,255);height:0px"></div>
</div>
</blockquote></div><br clear="all"><div><br></div>-- <br><div dir="ltr" class="gmail_signature"><div dir="ltr"><div><div dir="ltr"><div dir="ltr" style="font-family:arial;font-size:small"><font face="verdana, sans-serif"><span style="line-height:18px">“</span><span style="color:rgb(41,47,51);line-height:22px;white-space:pre-wrap">If I don’t use fancy words, you won’t know I’m an expert.</span><span style="line-height:18px">”</span><span style="line-height:18px"><br></span><span style="border:0px;margin:0px;padding:0px">—Marcel Matley, document examiner</span></font></div></div></div></div></div>