Closed zuphilip closed 6 years ago
Currently, we only use the more generic nameString property, since the OCR component also does not detect which is the first and which is the lastname.
In #58 we also figured out, that seperating between first and last names is difficult in principle.
Sorry, I should be more specific. I am aware of the problems with automatically splitting a name string into first and last name. That is not what I suggest here.
But sometimes we already have first and last names e.g. from Crossref. OpenCitations seem to save them then also into two different fields. Moreover, when we have to input them manually, then we want to do that consistent, e.g.
lastName, firstName
firstName lastName
However, the currently used lastName firstName
is very uncommon, or do I miss the point here?
Picture shows our captured data vs. the original data from Crossref for http://api.crossref.org/works/10.2307/2757850
Alright, I can check when the data is present and display it accordingly.
Would you prefer <firstname> <lastname>
over <lastname>, <firstname>
?
In any case I would need to default to <namestring>
, whenever there is no structure.
For inputing lastName, firstName
is more accurate and allows to split the name string into parts. However, we should according to the OpenCitation model save this either as two parts or in the form firstName lastName
.
Just to remember later:
For editing:
,
:
,
as lastname, the remainder as firstnamefirstname
and lastname
) and unstructured variant (namestring
) for safety<lastname>, <firstname>
format, if possible else fallback to nameString
What if some organization name contains a ,
Guess we wont get around putting separate fields into the form.
190d0fc
Currently we see sometimes
familyName givenName
which seems unusal. The OC model suggest as one possibilitygivenName familyName
or to split these parts up. When we begin to input these names manually we should also have a clear guideline.