-
Notifications
You must be signed in to change notification settings - Fork 35
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Add Bastin-1999-92 and Zanchi-2022-170 #1188
Conversation
@mathildavz, if you click on DETAILS in the test, that failed, you will see that the error code is about an "invalid column name". This should show up if you run the test on your system before. So question: did you run |
@mathildavz, you have still |
That is: Zanchi-2022-170.tsv-metadata.json has this the column |
@LinguList I did successfully. However, now that I think about it, I changed something about Zanchi-2022-170 and forgot to run it again. |
Thank you for finding the mistake! @LinguList |
So was this the problem?
|
Okay, one more thing that should be done now. I'll explain here. You have the LINK which is part of the URL, the end, right? So if you re-write the metadata now as:
This means, our computer code knows that LINK is a URL-Part and that the concatenation of the base URL |
@mathildavz, the tests show still an error, related to the JSON file this time. Can you check on your computer again, that the |
One more point: we do not allow spaces in our headers, so can you adjust the header in Zanchi for |
This is the error here: the
|
Okay. @tarotis and @AnnikaTjuka could you review this contribution? Annika is in joliday this week, but we are not in a hurry and can do this in the beginning of next week then. |
And @MuffinLinwist, please also check this contribution and discussion, so that you can later add the list by sidwell. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I can also help with the review if need be. One small thing: Please don't include the PDFs directly in the commit/PR (I've also noticed that we've got another PDF in sources/
that shouldn't be there). Rather, include them in conceptlists.tsv
(PDF
column) and make me aware of the PDFs that should be included (here, via mail, Mattermost). I'll then place them on our dedicated storage space for binary files.
Thanks, @chrzyki, we should at some point make sure that I could also upload pdfs if need be, so we do not need to ask you every time. And, @mathildavz, please make a "git rm sources/Bastin1999.pdf" and "git commit" and send the pdf to Christoph in mattermost. |
And if you have time to check, Christoph, this is also very welcome of course ;) |
I'll happily provide my first review. Thanks Mathilda for adding the two concept lists. Regarding the .bib-file, I have two comments:
In the Bastin concept list, there is one typo and one change I would suggest:
Further, I have the following suggestions for the Zanchi-2022 list:
There are two entries which feature with two different meanings:
In both cases, I would argue that the concept behind the meaning is the same, and that the difference is merely syntactic. Would you agree? Possible concept additions I was surprised not to find in the list yet:
|
Thanks, @Tarotis. Very useful comments. For sink and similar cases, transitivity is important. And sink in Concepticon is intransitive, so I would not map it. Similar care should be taken for the other concepts. |
Burn in Concepticon has a general reading, not specified with resp to transitivity. Burning is intransitive. So mapoing depends from the gloss and the usage. |
For cold feel cold I suggest to check which concepts are linked so far. It may just be linked to a concept set "freeze" if we have this and they do not have freeze in their conceptlist already. |
The concepticon description for |
Did you check examples? Or broader/narrower? This is more important than the definition, which may be problematic at times. And don't we have another sink in the same list? |
But this just needs to be checked and then we see. I am writing from mobile and cannot check right now |
@mathildavz and @Tarotis, I propose some checks on the suggestions that I consider critical and should be checked, also illustrating how checking in Concepticon is usually carried out.
If you check https://concepticon.clld.org/values, and search for
BURN is the least specified notion. But BURNING is explicitly intransitive, and the Role Frame by Zanchi et al. is s burns, so I would leave the mapping.
Probably not, the role frame in Zanchi is X crams Y into Z.
TIRED is an adjectival reading in Concepticon (meaning, it is a state), feel sleepy is a verb. We have one other "feel sleepy" (https://concepticon.clld.org/values/Nagano-2013-1256-27) which is also unlinked so far. So it should not be linked, rather we can later make a new Concepticon Gloss.
If you check again the values linked to this concept set, you can see that it is always intransitive, and even more important, we have linked the entry "sink" in Zanchi et al. already to this concept set and want to avoid multiple links. So I'd argue: we should update the definition, as it is misleading, and we should correct this entry, which is also wrong, as it is explicitly transitive. We can make a new concepticon gloss "SINK (SOMETHING)" and link both to this entry. We should make an issue to discuss this.
Intransitive "roll" in Zanchi et al. is already mapped to this concept. So rather leave it and check links actively (https://concepticon.clld.org/values/, important to check there whenever in doubt).
Adjective vs. verb is not trivial, again, checking https://concepticon.clld.org/values/ is essential, and checking list-internally is also important. |
I propose to change the following (for now) based on your helpful comments @Tarotis and your remarks @LinguList:
I would not map Same goes for |
Thank you for the useful comments, I see that I should pay more attention to the examples and not rely on the descriptions as I did previously. |
This is a useful enterprise for all of us here :) What I consider most important when doing the mapping is:
|
The discussion will go on when Annika joins next week. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Great work! I only left a few more comments.
Bastin-1999-92-12 12 breast sein 1402 BREAST | ||
Bastin-1999-92-13 13 burn brûler 2102 BURN | ||
Bastin-1999-92-14 14 cloud nuage 1489 CLOUD | ||
Bastin-1999-92-15 15 cold (weather) froid 1287 COLD |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
We have a more specific concept: 2483 COLD (OF WEATHER)
Bastin-1999-92-38 38 horn corne 1393 HORN (ANATOMY) | ||
Bastin-1999-92-39 39 kill tuer 1417 KILL | ||
Bastin-1999-92-40 40 knee genou 1371 KNEE | ||
Bastin-1999-92-41 41 know savoir 3626 KNOW |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Other instances of savoir are mapped to 410 KNOW (SOMETHING).
Bastin-1999-92-70 70 seed n semence 714 SEED | ||
Bastin-1999-92-71 71 sit être assis 1416 SIT | ||
Bastin-1999-92-72 72 skin (human) peau (humaine) 2613 SKIN (HUMAN) | ||
Bastin-1999-92-73 73 sleep n sommeil 1585 SLEEP |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
SLEEP is the verb meaning, I'd map to 2972 SLEEP (STATE).
Bastin-1999-92-84 84 tree arbre 906 TREE | ||
Bastin-1999-92-85 85 two deux 1498 TWO | ||
Bastin-1999-92-86 86 walk marcher 1443 WALK | ||
Bastin-1999-92-87 87 warm (weather) chaud 1232 WARM |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
More specific: 2110 WARM (OF WEATHER)
@@ -1,4 +1,26 @@ | |||
|
|||
@Inproceedings{Zanchi2022, | |||
Title = {PaVeDa - Pavia Verbs Database: challenges and perspectives}, | |||
Author = {Zanchi, Chiara and Luraghi, Silvia and Combei, Claudia R.}, |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I'd either use Claudia R. also in conceptlists.tsv or Claudia Roberta here.
} | ||
|
||
@Book{Bastin1999, | ||
Title = {Continuity and divergence in the {B}antu languages: perspectives from a lexicostatistic study}, |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Uppercase after colon
Zanchi-2022-170-49 49 APPEAR appear S appears appearer, appear beneficiary, appear causer, appear location 1076 EMERGE (APPEAR) | ||
Zanchi-2022-170-50 50 DIE die S dies dieer, killer/death causer, cause of death, dying location 1494 DIE | ||
Zanchi-2022-170-51 51 FALL fall S falls fallee, fall causer, fall instrument, fall goal, fall source 1280 FALL | ||
Zanchi-2022-170-52 52 FEEL COLD feel-cold S is cold freezing person, feeling cold locus, chill causer |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
We have 3819 COLD (FEELING COLD), but it is categorized as a property.
Zanchi-2022-170-80 80 HEAR hear E hears M hearer, heard sound, hear instrument, hear causer 1408 HEAR | ||
Zanchi-2022-170-81 81 WANT want A wants X wanter, wanted thing, want beneficiary, want causer 1784 WANT | ||
Zanchi-2022-170-82 82 SINK sink S sinks sunken entity, sink-(caus)er, sink location 1088 SINK (DESCEND) | ||
Zanchi-2022-170-83 83 BOIL boil S boils boiled thing, boiler 792 BOIL (OF LIQUID) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I'd map to the broader concept: 2493 BOIL
Zanchi-2022-170-106 106 ENTER enter X enters somewhere enterer, entered place 749 ENTER | ||
Zanchi-2022-170-107 107 REST rest X rests resting person 168 REST | ||
Zanchi-2022-170-108 108 DRINK drink X drinks Y drinking person, drunken thing 1401 DRINK | ||
Zanchi-2022-170-109 109 FEEL SLEEPY feel-sleepy E feels sleepy sleepy head |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
It could be mapped to 3620 SLEEPY since other glosses are also mapped to properties.
Zanchi-2022-170-135 135 MAKE APPEAR make-appear A makes P appear appearer (tr), appear causer (tr) | ||
Zanchi-2022-170-136 136 FELL fell A fells P feller, fellee 355 CUT DOWN | ||
Zanchi-2022-170-137 137 DROP drop A drops P dropper, dropped thing 2866 DROP (SOMETHING) | ||
Zanchi-2022-170-138 138 SEAT seat A seats P (somewhere L) seater, seatee, seating place 1476 CHAIR |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
CHAIR is a noun. I'd unmap.
Thanks @AnnikaTjuka, so I suggest @mathildavz wraps this up, incorporating all of your comments (with which I agree) tomorrow, and we can merge this PR. |
Our next PR will be done by @MuffinLinwist (Carlos) on a dataset by Sidwell that is easy to map. Here, I'd ask @mathildavz then to moderate (taking the role I took this time), once Carlos is ready. But this may be on Wednesday only. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thanks for integrating the changes. Approved!
@chrzyki Since you requested changes, @mathildavz will also need your approval to merge the PR. |
@chrzyki, have your requested changes been addressed? If so, I'd merge this now. |
I just saw that you indeed followed the requests, @mathildavz. So I merged right now. |
Pull request checklist
concepticon notlinked --gloss "NEW_GLOSS"
Additional information
...