First Experiences with Google Scholar Citations
Aside from different efforts to tackle the author identity problem on a global scale, such as ORCID, Google launched its own Citations product today as an addition to Google Scholar.
What is it?
Basically, Google Scholar Citations offers public researcher profiles, together with citations, h-index and i10 index calculations. By default, you are asked to manually enter new publications, as Google decides on the sources it automatically gets your publications from.
There is very little you need to do in order to setup your account. Only your Name is required in the first step, to see a list of Articles that Google has selected for you.
To avoid fake registrations, you are required to enter an email address at your institution, before you can list the institution name in a public profile.
In the next step, matched articles are presented as "Article Groups". Although this is not explicitly mentioned, it seems that each of the groups seems to correspond with a source in which Google has retrieved the results. Although you can click links to find the actual articles at a later stage, it remains unclear which sources Google selected.
You can either add a whole group of articles at once, or dig into the group to manually deselect the ones that don't make sense.
As a last step, you are asked whether Google can automatically update your list. It is clear that asking as little involvement from the researchers is definitely a plus, it remains to be seen how correct these automated lists will be in the long run.
Inclusion of Repository Metadata
When setting up my own profile, it was immediately clear that Google retrieved information from our own public DSpace repository, @mire Labs. This makes you think about the sources Google considers for inclusion there. Certainly, our domain is not an academic one. Secondly, is it possible that Google detected the highwire press metadata (in the header), and uses these? These points remain unclear for now.
There is support to export a selected list of citations to BibTeX, Endnote and RefMan.
Although it's a little bit hidden, the main dropdown menu allows you to manually add new references to your list.
Examples of "Work in progress"
Here are just a few examples in which it's clear that the service still needs some fine tuning. First of all there are records that are scarcely populated. Luckily enough, you are able to manually edit and improve the record. There seems to be no metadata from which exact source the record was harvested and when. However, the citation links to an external system, in case of most of my items, @mire labs.
The second example shows that some metadata parsing still needs some work as well.
The first day of the public launch of a new product is definitely too early for hard conclusions, although there are a few clear observations.
It's clear that Google adds value to its successful Google Scholar search engine with this product. If you search for an author name, and the search string matches with an existing public profile, the profile will be accessible from the results.
Asking as little involvement and work from the user as possible is the way to go. If it turns out that the data isn't good enough, you can always push for more involvement. To me, this seems like a better and more challenging approach than starting with an empty box and asking someone to fill out thousands of fields.
Where are the ads? Although I haven't seen any ads, there are reports that Google Scholar sometimes serves ads. If it turns out there are no ads at all, it remains to be seen what the business model behind the service really is.