Feature requests. 2nd surname & author id

[eber]

Hi,
I'd like to thank all developers contributing to distribution of knowledge via PKP software, I think you are doing a great job.

I have a couple of features that could be interesting to include in future updates of OJS:

1.- Authors from Spain and most Latinamerican countries use two surnames. Have a look at: http://en.wikipedia.org/wiki/Spanish_naming_customs and http://yuba.stanford.edu/~molinero/html/surname.html
It would be very helpful if OJS could include an extra field to store the second surname of authors.

2.- Although it's a difficult task to have a global unique author identifier, It would be helpful if OJS could have a unique code for every author that collaborates at any time in the journal. AFAIK every time you introduce a new article and its correspondent author, a new authority record is created in the database. Wouldn't it be better if you could have a unique record for every author?

¿What do you think?

Cheers.

[jmacgreg]

Hi eber,

Many thanks for your kind words, and your suggestions! Feedback such as yours is tremendously helpful in the development of our software. Please see below for some comments:

1.- Authors from Spain and most Latinamerican countries use two surnames. Have a look at: http://en.wikipedia.org/wiki/Spanish_naming_customs and http://yuba.stanford.edu/~molinero/html/surname.html
It would be very helpful if OJS could include an extra field to store the second surname of authors.

The "Last name" field in the User Profile form can handle spaces, so it is currently possible to list two surnames in that field. For example, If in my journal profile I were to write "James" for my first name, and "MacGregor Phinney" for my last name, my name would be displayed as either James MacGregor Phinney or MacGregor Phinney, James, depending on where it was displayed (eg. in my biography or in an author listing, respectively). The same goes for the author information during submission. Is there a case where having the two surnames separated might be useful? A quick fix might be to change the form language to read "Last name(s)" rather than "Last name", at least in Spanish, where it would make sense that there might be more than one name expected.

2.- Although it's a difficult task to have a global unique author identifier, It would be helpful if OJS could have a unique code for every author that collaborates at any time in the journal. AFAIK every time you introduce a new article and its correspondent author, a new authority record is created in the database. Wouldn't it be better if you could have a unique record for every author?

We are in the process of tackling this. I'm not entirely sure about having global author identifiers, per se -- but we will be providing the ability for submitting authors to look up authors that already exist within the system as they create the article's metadata. This would hopefully cut down on typos, misattribution, etc., and would also make it easier for the submitting author to fill out the submission metadata.

If you have any other comments, questions or suggestions, please let us know!

Cheers,
James

[eber]

Thanks for your quick answer,

The "Last name" field in the User Profile form can handle spaces, so it is currently possible to list two surnames in that field. For example, If in my journal profile I were

Your proposal will solve an appearance problem, but It's more a matter of interoperation and normalization. As well as you use isolated fields for Name and Surname, in spanish people databases we use different fields for different surnames. You cannot be sure that two words in the same field mean two different surnames, in spanish there are surnames composed by two words, or even three.
If you want to establish relationships between author records from different sources, it's much easier if you have first and second surname isolated.
Would it be possible to make it a configuration setting? That way you won't see the second surname field unless you enable this setting.

Related to this problem, all interoperation formats, and metadata schemas that mainly come from english world, don't have this solved, one unique field for surnames

We are in the process of tackling this. I'm not entirely sure about having global author identifiers, per se -- but we will be providing the ability for submitting authors to look up authors that already exist within the system as they create the article's metadata. This would hopefully cut down on typos, misattribution, etc., and would also make it easier for the submitting author to fill out the submission metadata.

This sounds great because using one unique code for authors in the system will help assigning all the articles to the same author record, when exporting and consolidating information in other systems.

Thanks for your time.
Cheers.

[jmacgreg]

Hi Eber,

I sent this thread along to the rest of the group, and we've had a good time discussing it.

Regarding point one: this is, to a degree, a display issue, but the larger issue here is that there are many, many ways for names to be created and displayed across the world. One other such example is Indonesian, which reverse the order of given name and surname. Name modelling can be tremendously diverse, and hence problematic. We can help the matter by centralizing how we deal with names within the codebase (I have submitted this as a feature request: see http://pkp.sfu.ca/bugzilla/show_bug.cgi?id=5553), but any further work (eg. developing support for two second names, or reversed names, etc.) would probably have to come from the community, at least from the beginning. We are happy to accept patches towards this, and to help maintain the patch or plugin. Of course, Bug 5553, above, would have to be resolved first.

Regarding point two: this brought up discussion of the ORCID project (http://www.orcid.org/) which is trying to solve just these kinds of problems. If an acceptable standard or protocol exists for any sort of given function, we like to use it rather than reinvent the wheel ourselves; ORCID is still just a work in progress, but it does show promise, and we'll continue to keep an eye on it. In the meantime, we can likely do some sort of workaround that would introduce a "soft link" between author information and user accounts: see http://pkp.sfu.ca/bugzilla/show_bug.cgi?id=5554 and also http://pkp.sfu.ca/bugzilla/show_bug.cgi?id=4653.

Any other comments welcome!

Cheers,
James