[Lingtyp] IPA for linguists' names

[Jocelyn Aznar]

Dear Yuri,

Le 06/09/2026 à 07:25, Yuri Koryakov via Lingtyp a écrit :
> On the one hand, all information not added by people themselves is taken 
> from public online sources; people add further information voluntarily. 
> On the other hand, this table does not contain any sensitive personal 
> information – no emails, place of residence, place of work, position, etc.

Note that you suggest to people to provide email address in the comment 
section.

In some countries (like the US), the pronoun to use is actually a 
sensitive personal information, as it reveal gender info. Native 
language can be a proxy for ethnic identity which also can be 
problematic in many countries. Google Doc, especially with a free 
account, is not safe for personal info, not to mention that someone/AI 
could vandalize the spreadsheet and insert/modify the data, and when 
collecting personal info, the database creator is responsible of 
providing a database which is secured enough according to the 
sensitivity of the info collected.

More practical issue, I can imagine in the US that inviting a researcher 
that asked for the pronoun they could be problematic, either for the 
researcher or the lab inviting them. As it is information available on 
Google Doc, this will be automatically scanned by the IA that the border 
police uses.

Birthday can also be problematic actually, as the other disambiguators, 
as they allow to identify the person, which is normal for the purpose of 
this database, two people having the same name might want different ways 
of being addressed. Better in this case ORCID or professional URL which 
are publicly available data.

Somehow, GDPR might seem annoying, but, in this day and age of IA, any 
info available on internet is being scrapped and compiled, and 
automatically analyzed by more and more fascist governments and 
organizations for whatever reasons they have, or will have. Somehow, I 
wonder more and more about the relevance of open access data in this new 
context.

>> - There isn't only one valid pronunciation of my name.
> 
> The IPA column allows you to specify multiple pronunciation variants, 
> both equivalent and language-specific.

Right.

>>
>> - I'm ok with people having difficulties pronouncing my name (at the 
>> end of the day, no one can master the pronunciation of the many 
>> phonologies we work with in our profession).
>> - Pushing people to pronounce your name following your expectations 
>> can cause situations in which they feel really ill at ease because 
>> they cannot pronounce your name.
> Well, we don't force people to pronounce the names exactly this way, but 
> we do give them a reference point. For example, when I just see the name 
> Jocelyn, I don't even know what the first sound might be – [ʤ / ʒ / x / 
> j] or something else.

Of course I understand this issue.

>> - I'm also ok with people adapting my name if it is too hard to 
>> pronounce, that's usually what happens in practice.
>> - From a more technical angle,it would make more sense if the IPA 
>> transcription follows the same order of the first name + other name.
> Okay, maybe. But this way one can show the order in which a person 
> prefers to pronounce their name. (By the way, the most common order in 
> this column is: first name + other name.)

This is not so clear from the instruction neither, and this is not 
what's systematically done from the data that was already populated.

For the purpose of this database, I would have a dedicated column for 
the preferred name for a seminar/event, not just "Other names" which can 
lead to confusion as it might contain many info in different order. 1 
info : 1 column.

I think this project would be better handled through the server of an 
institution (well, more of our institutions rely on Microsoft/Google and 
co., but if they pay for the services, the data might be actually 
properly handled), with a better interface, like a form in which people 
can enter these info and a page that displays those info.

Best,
Jocelyn