<html>
<head>
<meta content="text/html; charset=UTF-8" http-equiv="Content-Type">
</head>
<body text="#000000" bgcolor="#FFFFFF">
<div class="moz-cite-prefix">On 10/12/2012 03:16 PM, Chris Little
wrote:<br>
</div>
<blockquote cite="mid:507896BE.6080508@crosswire.org" type="cite">On
10/12/2012 1:40 PM, Daniel Owens wrote:
<br>
<blockquote type="cite">Gary Holmlund and I are working on a
problem related to the Westminster
<br>
Hebrew Morphology (WHM) module. We need a consensus on markup
practices
<br>
for OSIS lemma.
<br>
<br>
I was having a problem getting natural Hebrew lemma to look up
an entry
<br>
and display it in the mag window. Gary discovered that if "H" is
<br>
prefixed to lemma in WHM, the BibleTime mag window works with
Hebrew
<br>
lemma (as opposed to Strong's numbers).
<br>
<br>
My understanding is that this is not typical OSIS best practice
but a
<br>
SWORD convention. I resisted at first, but now I think there is
some
<br>
wisdom to using this method. We need some way to distinguish
between
<br>
Hebrew and Aramaic words, which can be identical in form but not
in
<br>
meaning. WHM uses @ for Hebrew and % for Aramaic. I suggested to
Gary
<br>
that we compromise and simply change @ to H and % to A,
modifying
<br>
BibleTime to strip A and H and use that to look for the entry in
the
<br>
correct lexicon.
<br>
<br>
The markup would look like this:
<br>
<br>
Hebrew (from Deuteronomy): <w lemma="whmlemma:Hאבד"
<br>
morph="whmmorph:some_value">תֹּאבֵדוּן֮</w>
<br>
<br>
Aramaic (from Jeremiah): <w lemma="whmlemma:Aאבד"
<br>
morph="whmmorph:some_value">יֵאבַ֧דוּ</w>
<br>
<br>
The main problem I see is that other front-ends may not follow
the
<br>
process of looking for G or H and then stripping the character
before
<br>
looking up the entry.
<br>
<br>
Could we come to a consensus on this?
<br>
</blockquote>
<br>
Could you confirm that this is the behavior in some front end
other than BibleTime? From my perspective it just sounds like a
BibleTime bug.
<br>
<br>
This is certainly bad OSIS encoding. It is also not a Sword
convention. If anything is implemented that requires a language
prefix like this, it represents a bug, whether in Sword or in
BibleTime.
<br>
<br>
--Chris<br>
</blockquote>
Here is a quote of a comment from Xiphos source code:<br>
<br>
<meta http-equiv="content-type" content="text/html; charset=UTF-8">
Strong's words are specified as a prefix letter H or G (Hebrew
or <br>
Greek) and the numeric word identifier, e.g. G2316 to find
\"θεός\" (\"God\").<br>
<br>
So it appears to use the H or G method. Is there is documentation
about a better way to do this?<br>
<br>
Gary <br>
</body>
</html>