On Wed, 2018-01-10 at 04:26 +0000, Máirín Duffy wrote:
> Hi William,
>
> I'm one of the Fedora UX designers and was pointed to this thread by
> charcol on another list.
>
> I would like to express enthusiastic support for having by default a
> display name field and a legal name field. This is of particular
> benefit and interest to women as you clearly recognize. My legal name
> is not what people know me by yet it's important for various gov't
> docs that my legal name be used. In most contexts the name I go by is
> more appropriate and recognizable for everyone. Red Hat allows me as
> an employee to choose a displayname for my first name but not my last
> - my legal last name is not what I go by. This has definitely caused
> me some serious real world challenges.
Thanks! I'm glad to see that this idea is supported. I have always
wanted to improve this, and this seems like a great time to achieve it.
>
> One question I have about having multiple names in one field, from a
> UX designer POV - often when retrieving lists of names for display in
> interfaces, in a Western context they are often listed lastname
> firstname in alphaorder. If the field is freeform and the person
> inputs first middle last or even more names, how can the lastname be
> identified so that a given user can be located in an alphabetically
> ordered list?
You can sort on displayname, but that's about it - The challenge get's
worse in this case.
firstname lastname
singlename
firstname familyname1 familyname2
firstname middlename lastname
Which do we sort on? I know in spain they have a non-hypenated pair of
family names. And then you have individuals with single names.
So sadly, there is no *truly* consitent answer here.
In the "surname" case you would have:
[ surname field ]
firstname lastname
singlename ????
firstname familyname1 familyname2
firstname middlename lastname
So already, we have a broken design in trying to sort on "lastname,
firstname" in the western context because of the individual with the
singlename.
Additionally, some people may not even enter the surnames correctly - I
know one spanish lady who consistently had issues with the Australian
government insisting that familyname2 was her surname. So she would be
put into systems as:
[ surname field ]
firstname familyname1 familyname2
Again, the surname doesn't "sort" correctly, because her true
familyname1 is not in the surname field.
With displayname you can only sort by "order of the displaynames". So
we can at least consistently sort this given the scenario above!
To *search* for surnames however, now you can do a substring search in
the displayName field. I'm not sure of your LDAP profficency, but the
search would be:
(displayName=*lastname)
So for me, I would choose my display name to be:
"William Brown"
To find me would be:
(displayName=*Brown)
For a legal version, you now need to search the legal name field, and
again, the same search syntax is possible
(legalName=*Brown)
> This seems like a core front end use case and I wonder if condensing
> down to one field is going to cause problems for systems connecting
> to the directory.
*thankfully* most LDAP connections default to the current system of uid
OR displayName, so this is already taken care of!
> Another consideration - names may be listed in different orders
> depending on locale. Eg typical Western format is given middle
> surname but other locales (Japan comes to mind) is surname given. Can
> applications connecting to the directory be able to display names
> appropriately for a given locale if there isnt a way to parse them
> out correctly? This locale ordering isnt an issue for single names,
> but 3+ names make it I am
> imagining nearly impossible to programmatically parse the user input
> in any reliably correct way.
We already have complete UTF8 handling in the server, with sorting and
ordering available.
In terms of "displaying this in correct order" you are right that
Japanese prefer family name: This is still already a challenge in the
current design of the server regardless of the field used because of:
uid
cn
givenName
surname
None of these properly encapsulate the "international" nature of names.
They are very "western". If we display:
givenName surname
We work for western cultures, but not japanese
If we do:
surname, givenName
We may work better for japanese, but this isn't consistent to western
standards.
And finally, to make it fun - singlenames. How can we handle this?
Thus we arrive at the conclusion - we have a cultural and social
concept, that technology can not represent simply. To have a single
"displayName" and "legalName" field, allow expression of all cultures
names and styles, and we can choose how they are filled in in a way
that is meaningful to a human. We can use ldap's fast querying to help
humans search these fields and still get meaningful data, and we have
to as humans, apply context to the name to understand it (and respect
it).
>
> Hope this feedback is helpful.
It is thank you!
>
> Cheers,
> ~m
> _______________________________________________
> 389-devel mailing list -- 389-devel@lists.fedoraproject.org
> To unsubscribe send an email to 389-devel-leave@lists.fedoraproject.o
> rg
--
Sincerely,
William Brown
Software Engineer
Red Hat, Australia/Brisbane
_______________________________________________
389-devel mailing list -- 389-devel@lists.fedoraproject.org
To unsubscribe send an email to 389-devel-le...@lists.fedoraproject.org
