Re: New Morph Module telco time

Dear all,

as discussed yesterday, I added sample data for three languages under
https://drive.google.com/drive/folders/1hNOj5qm34J7uGJrsudI4eDuQAMCFFfzv?usp=sharing.
The modelling challenges are described in the document. Here, they have to
do with understanding how paradigms are being used, how to do recursive
morphology, how to model sequences and tree structures over morphological
segments and how to encode embiguity/alternative analyses efficiently. A
meta question is how to map a sequence of morphemes to a surface string
(allophony). As for the latter, I did not make it explicit because it may
be beyond the scope of the model. I think yesterday we decided to collect
requirements in an independent document, but not sure there is a link for
it yet, anywhere. In any case, this should not require novel vocabulary
(so, it doesn't really fit the wiki page:
https://www.w3.org/community/ontolex/wiki/Morphology), but to understand
how to use/define the existing terms.

Best,
Christian

Am Mo., 18. Jan. 2021 um 00:08 Uhr schrieb Bettina Klimek <klimek@infai.org
>:

> Hello Christian,
>
>
>
> thank you for your feedback. It is certainly a good idea to navigate the
> workflow along new use cases. I will incorporate this into the agenda
> before Wednesday.
>
>
>
> Also, I can confirm to have the morph module telco bi-weekly.
>
>
>
> Regards,
>
> Bettina
>
>
>
> *Von:* Christian Chiarcos <christian.chiarcos@googlemail.com> *Im Auftrag
> von *Christian Chiarcos
> *Gesendet:* Samstag, 16. Januar 2021 10:31
> *An:* public-ontolex@w3.org; Bettina Klimek <klimek@infai.org>
> *Cc:* chiarcos@informatik.uni-frankfurt.de
> *Betreff:* Re: New Morph Module telco time
>
>
>
> Hi Bettina,
>
>
>
> good to see you back. Next Wednesday should work, but I would ask you to
> have that bi-weekly, not weekly.
>
>
>
> As for the mid-term agenda: As soon as we have a first full version of the
> model at hand, we should apply it to a number of resources and use cases
> and to revise it as needed. I suggest to collect such additional sample
> data along with reviving the discussion. Of course, we had such data in our
> mind (and in our hands) when developing all of this in the first place, but
> a long time went by, more people got/get involved (I presume, in
> particular, out of the context of Nexus Linguarum), all of us have new
> projects with new data, and adding these new perspectives is a chance to
> make the model robust.
>
>
>
> Thanks,
>
> Christian
>
>
>
> Am .01.2021, 23:06 Uhr, schrieb Bettina Klimek <klimek@infai.org>:
>
>
>
> Hello all,
>
>
>
> thank you for filling in the Doodle poll.
>
>
>
> The new telco time will be every Wednesday between 1-2pm CET. The first
> telco will be on January 20th.
>
> You can join the call with this link:
> https://meet.google.com/rrs-garw-oxu?hs=122&authuser=0
>
>
>
> Looking forward to talking to you all very soon!
>
> Have a great weekend,
>
> Bettina
>
>
>
>