Nascent

I use CiteULike:
http://www.citeulike.org
it can parse many reference styles, and you can tag and share your references.

I'm a colleague of Attila's, and we work in the same lab. I'm sure he told you that there's really not much interest in the web in our lab outside of the two of us, so I'd like to address your questions, with examples from our group.

You asked who NPG should be marketing to. I think you need to be addressing the group of people who don't know what a blog is. Attila and I understand the value provided by Scintilla and Connotea already, but many of our colleagues only use for the internet is to do pubmed searches and send email. They need to hear about the useful tools available from a very authoritative source using traditional channels such as print and conference presentations. I haven't read the printed version of Nature in years, so I don't know if this exists, but is there a notation in the heading of an article mentioning Connotea? I would think a small box saying something like, "See what your colleagues are saying about this article at http://Connotea.org/ArticleURI" would be somewhat effective. Holding brown-bag sessions at conferences on Connotea and Scintilla would be useful, too. Calling the sessions, "The overwhelming burden of keeping current with the literature: why you're missing critical developments in your field" would be a fairly provocative title, or you could go with, "How not to get scooped."

This approach would tie in with the grasstroots (netroots?) effort than the lab innovators are making, and it would make it easier for us, as well. I would think that in the older, more established labs you'll need this kind of approach.

Let me tell you how we share our protocols currently. For years we had a printed binder with the lab's protocols, and a couple years ago, someone typed it up and made a Word document, which was put in zip disks and distributed. I talked about putting it in a wiki, and the concerns I heard from my colleagues were that no one would be able to find it or use it, that it would take too much time to set up, and that it would provide competitors with an unfair advantage and possibly even leave us open to malicious mischief from anonymous editors. Of course, you know as well as I that edits are logged and reverts are a simple matter, but when you start explaining the technical solution to a technical problem, you've already lost a non-techical audience. Our protocols are now in a shared folder on a lab computer, in the same Word document. I think that a Nature-branded protocol database instead of a homebrew MediaWiki installation would have been much more easily accepted.

I was our lab's organizer for Journal Club for a while, and I used Tulane's Google docs-like application to distribute PDFs because only a few people have a google account and the university enforces a 50 Mb quota on our email accounts. However, once I passed the organization responsibility along (so that I could finish my thesis and get ready to graduate), they went back to emailing PDFs again. I would have liked to have just tagged the upcoming paper on Connotea, but no one signed up for an account when I invited them, because they had never heard of it before. Perhaps if they had previously heard about it in a presentation at a conference or seen a mention of it in the header of a Nature article, then my mention of it could have been the extra personal connection they needed to get started with it.

So that's the challenge you face getting more people interested in new web technologies. Nature is in a great position, because as a source of information, you have everyone's ear and almost everyone's trust. Why doesn't NPG leverage that position?

As a technical point, one thing that would really help usability would be to implement single sign-on for all the nature services that are open to the public. Blogs.nature.com, Connotea, Scintilla, Precedings, and Nature Network would all be more accessible to people who aren't used to remembering usernames and passwords if they didn't have to sign on to the various sites individually.

My guess is that the best way to convert more people into using web tools would be to develop them with young group leaders in mind. Setting up integrated lab managing tools that are more efficient than what we have today (mostly paper based). Get a group of them to give advice with the development. It would also help later for the promotion.
The group were I am working has a CMS (Plone) for sharing datasets, help coordinate projects, calendar of upcoming conferences and shared protocols.

Thanks for the comments.

We're with you on the single sign on, Mr G - it's on the way.

"Setting up integrated lab managing tools that are more efficient than what we have today (mostly paper based)"

This is interesting - I wonder if there are any CMS distributions designed specifically for use by reesarch labs?

Mr. Gunn wrote: "many of our colleagues only use for the internet is to do pubmed searches and send email"

I feel so relieved... I thought this was just something happening in this small corner of Spain and other non-first-rank-in-science places. But I see that internet illiteracy may be widespread among scientists all over the world. Are we just too busy? Lazy for anything that is not a priority? Why internet literacy is not considered a priority among scienctists??? Just ignorance of the benefits?

It seems that internet illiteracy is not only condoned, but backed by the "powers that be."
For instance, the sys admins of our system has this large, blanket WARNING to every user on the system about downloading unnecessary computer viruses, etc. that most of my more senior colleagues take as "don't look at anything other than email, journal articles, and ads from 'trusted vendor sites'." It would take too long to train each and every potential user how to do things the "right way." It makes me think about herding cats and training fleas.
Also, there's an overwhelming inertia against learning new software amidst the higher-ups especially when the present system is working "just fine." (I had to fight to get a copy of EndNote and no one else dares to use it unless they get supervisory direction.)
I think that marketing to a younger, more net-savvy demographic would save us a lot of time and frustration, simply because the people in charge don't really have time to "fuss with new stuff" that they are not familiar with. They will always resist and fear what they do not understand or cannot learn in under two hours. Example: My management had to have a meeting to decide to refuse my offer to fix some broken links on the website with WYSIWYG Dreamweaver. So now, they're outsourcing to an expensive expert to do what is a trivial, basic html task--no special software even needed. By simply targeting younger scientists in the next technological generation to try something honestly time-saving and useful should usher in net literacy, albeit on a geological scale. Otherwise, it will take a massive force-feeding effort from senior, senior levels to enforce such widespread learning.
And that requires a sound, economic argument as to how long-term cost-savings (i.e. in man hours, outside contractor fees, etc. etc. etc.) will appear quickly enough before something more critical pops up and diverts their attention. I'm not so sure that even a persistent and obvious compaign by NPG will imprint the importance of net literacy on scientists, a social group--I might add as a personal observation--that reveres the traditions of what-has-worked and consistently shuns maverick innovations without someone spending his/her entire career proving the approach to the overall community.
Subverting from a "netroots" angle (BWAHAha!!) sounds more prudent, mentally healthy, and fun.

P.S. I didn't know one of my favorite radio stations KCRW is Plone-based. Their GUI is slick!