[sword-devel] Marking transChange within a <w> ?

[Chris Little]

On 11/5/2010 12:09 PM, DM Smith wrote:
> On 11/05/2010 02:48 PM, Chris Little wrote:
>>
>>
>> On 11/5/10 11:35 AM, David Haslam wrote:
>>
>>> The example from Mt 21:31 illustrates one of the 17 instances.
>>> <w lemma="strong:G3588 strong:G3962" morph="robinson:T-GSM
>>> robinson:N-GSM"
>>> src="8 9">of<seg subType="x-added" type="x-transChange">his</seg>
>>> father</w>?
>>
>> (Disclaimer: I don't know Greek, and I'm partly just guessing about
>> Greek grammar here.) I think this is just encoded incorrectly.
>> Arguably, we're hampered by some limitation of the engine, but I would
>> encode this as:
>>
>> <w lemma="strong:G3962" morph="robinson:N-GSM" src="8">of</w>
>> <transChange type="added"><w lemma="strong:G3588"
>> morph="robinson:T-GSM" src="9">his</w></transChange>
>> <w lemma="strong:G3962" morph="robinson:N-GSM" src="8">father</w>
> Chris,
>
> The problem with this is that in the KJV italics represents words that
> the translators felt that they had to add to the translation to reflect
> the proper meaning and for which there was no underlying Greek word.
>
> That implies for the KJV <transChange type="added">xxx</transChange>
> should never wrap <w>...</w> with a lemma referring to the word that it
> translated ;)

No... the problem is I don't know Greek, so I don't know what I'm 
talking about. :) But I take your point.

I think your suggestion that we could use x-split is probably the 
easiest/quickest path to getting everything to work in the short term.

--Chris