<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
</head>
<body>
<p>Hi,</p>
<p>Having to suffer the FBX (in)famous case, I fully support
Sybren's point here. I would also strongly advocate against
working with any standard or extension of a standard which
specifications are not publicly available. This is road to Hell
and never-ending maintenance.</p>
<p>Cheers,<br>
Bastien<br>
</p>
<div class="moz-cite-prefix">On 4/25/22 11:47, Sybren A. Stüvel via
Bf-usd wrote:<br>
</div>
<blockquote type="cite"
cite="mid:CAFOOXSbbbGR30d16NHdzVPXSsym_DZp+G=xiB2uEaUx-atW7eg@mail.gmail.com">
<meta http-equiv="content-type" content="text/html; charset=UTF-8">
<div dir="ltr">
<div class="gmail_quote">
<div dir="ltr" class="gmail_attr">On Thu, 21 Apr 2022 at
18:45, John Harrison via Bf-usd <<a
href="mailto:bf-usd@blender.org" moz-do-not-send="true"
class="moz-txt-link-freetext">bf-usd@blender.org</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">
<div style="overflow-wrap: break-word;">
<div>Implementing an importer as part of the same effort
makes sense to me (more work of course) because it
balances the feature set and will help with testing
(import->export & compare). [...] Regarding font
point size, I would use Reality Composer as “ground
truth” and see how it maps font point sizes to designer
units (meters usually), then document my findings and
follow suite.</div>
</div>
</blockquote>
<div><br>
</div>
<div>Public specifications are always better. If there is no
other way, we can try and follow what some proprietary
application is doing, but that opens the door to all kinds
of nasty issues. The most important aspect is that triagers
and Blender developers won't be able to reproduce bug
reports. Being able to go through an export-import cycle and
see that the result is the same is a good start, but that's
not enough to ensure the implementation is correct.<br>
</div>
<div><br>
</div>
<div>We have been in this situation with Alembic, where
different applications had different interpretations of
certain data, and Blender was stuck in the middle. I don't
want to repeat this process with USD. Especially when
implementing support for a 3rd party extension, I feel that
we have to be careful and have a thorough understanding of
that extension before writing any code.<br>
</div>
<div> </div>
<div>Sybren<br>
</div>
</div>
</div>
<br>
<fieldset class="moz-mime-attachment-header"></fieldset>
<pre class="moz-quote-pre" wrap="">_______________________________________________
Bf-usd mailing list
<a class="moz-txt-link-abbreviated" href="mailto:Bf-usd@blender.org">Bf-usd@blender.org</a>
List details, subscription details or unsubscribe:
<a class="moz-txt-link-freetext" href="https://lists.blender.org/mailman/listinfo/bf-usd">https://lists.blender.org/mailman/listinfo/bf-usd</a>
</pre>
</blockquote>
<pre class="moz-signature" cols="72">--
----------------
Bastien Montagne
Blender Developer</pre>
</body>
</html>