Basic usage question

[Simon Cozens]

I'm writing an application which accesses lexicons, and the first task is to
turn the various lexicons I have to hand (each with their own idiosyncratic
XML schemata) into a standard format so that retrieval and display of entries
can be done nicely. Searching for a standard XML format for dictionary data
brought me to LIFT.

So I have happily written a bunch of XSLT files and got some valid LIFT output
from them. One dictionary is giving me a bit of pause, though. It has some
etymological information but it's fairly free-format and so not sufficiently
well formed for me to put it into an <etymology> tag. So I am throwing it into
a general <note>. Is this too naughty?

The other problem I have is that it also refers to other entries in the same
lexicon, and I don't know the best way to encode a cross-reference.

Here is one example of an etymology part of an entry:

<strongs_derivation>from <strongsref language="GREEK" strongs="1"/> (as a
negative particle) and a presumed derivative of <strongsref language="GREEK"
strongs="2767"/>;</strongs_derivation>

(Yes, it's *that* lexicon.)

So, what should that look like in LIFT?

So far I've got as far as:

<note type="general">
<form lang="en"><text>from a compound of </text></form><form
lang="gr"><text>Α ἄν</text></form><form lang="en"><text>and a presumed
derivative of </text></form><form lang="gr"><text>κεράννυμι</text></form>
</note>

But it would be very useful to give these Greek forms cross-reference links
back to <entry id="G00001"> and <entry id="G02767">

How should I do it?

Simon