Connotea has been around for a few years now, and to be honest with you, not a huge amount has changed with it in that time. Our usage increases and more citations are being bookmarked in the site every day, but we have been putting on our thinking caps and trying to brain storm where we could take the site to make it more useful to scientists.
We have are developing a shortlist of suggestions for features that we hope to develop, but I’d like to hear any thoughts that any of you might have.
- Ian
Integrate it into more external sites, eg Friend Feed?
I know this isn’t really an exciting request, but could more time be spent on speeding up the website? I find it frustrating that it can take anywhere from 3-15 seconds to just load a list of my bookmarks.
Also, a more advanced search would be nice so that you could search using numerous fields including journal name and author. I created this rough search mostly for myself, but it is fairly slow (and not very pretty) due to most of the searching being done remotely.
Quite often I end up just going to Pubmed or Google to find a publication because Connotea is quite slow.
Hopefully, I am not being too negative. I still think Connotea is a great tool for researchers!
Have you seen this blog post by Eva Amsen? She describes her ideal online citation software.
Hi Anna,
Yes, I saw a link to that post. There has also been a lot of discussion over at FrendFeed
- Ian
I started exploring connotea only recently, to see if it can be a tool for the association of medical doctors i work for. Untill now we have been mailing or favorites to each other and integrate it in bookmarks, or commenting on articles on blogsites.
So I think connotea would be a huge step in the future for us. Advantages of connotea above other social bookmarking-websites are
Problems I encountered
I am really an average user, a part-time researcher and not entirely new but same time not to advanced in web2.0
I do not think this is the time to invest in many new tools. Better assure first that the ones you have are more thriving. If I compare with what the IT-communities are doing on open source, we nature-scientists (and especially the medical professionals) have a long way to go.
I agree with Morgan, in that I’d like the site to have quite a bit more speed and polish, rather than a many new features. The RSS feeds for group libraries, for example, only seem to erratically sync with the actual list of publications. Meanwhile, if I post a bookmark privately to one group, the bookmark still seems to show up in another group’s library; it’s only visible there to me, I think (I’m not sure if it’s visible to other members in that other group), but the fact that the entry is there makes the settings for the bookmark very confusing.
The only real new feature I’d like would be simpler group management (e.g. can request to join or leave a group yourself). I like the group interface for Facebook quite a bit.
What I would much like to do is to have Connotea interfacing with a word processor, say MS or Open Word, for example, and help me formatting my bibliography in a manuscript, according to a given journal style. Martin Fenner has been writing quite a bit about this on his blog.
To keep a list of my preferred publications online is only doing half the job, for me.
I am using EndNote on my PC, and I know there is a thing called EndNote Web, but I believe that is a proprietary package.
What (more) can Connotea do for me?
Oh, gosh, yes. Integrate Connotea into OpenOffice the way Endnote is into Word, and I guarantee overnight success. Well, if you could put it into Word it would be better (same-day?), but I understand there may be some barriers there. It’s the one major reason I haven’t adopted it (limited time, fair number of offers).
I would like to see integration of Connotea with desktop reference managers, e.g. Endnote, Papers or Mendeley. Then there would be no need for integration with Microsoft Word or OpenOffice. I also wish we had Connotea integration in Nature Network.
Okay – I’m with Martin. Make Connotea compatible with Mendeley and I will be very happy, too. But it would still be nice to have the references easily insertable into text documents.
Martin and Heather, I am not sure about Papers and Mendeley, but I don’t see the strict need of integrating EndNote with Connotea. Can’t Connotea plug straight into a Word Processor? I guess the CEO at EndNote (or whichever name the company has) is not going to like me very much.
Please explain, I know you know better.
Massimo, for me, reference managers have to do (at least) four things:
Most applications can do only 1-2 things well. So we can either wait for the application that does everything well, or have applications that talk to each other. Connotea is good at sharing references, but it is not good for the other three points. The Macintosh application Papers is good at the first two. Endnote is only really good at #4.
My ideal solution would store all my references online and allows me to share them with others. There would be different ways to add references to that solution and it could be a different application that creates the bibliography.
Endnote is owned by Thomson Reuters (they also do Web of Science and Impact Factors, among many other things), and they are probably interested in a complete solution. Connotea (Nature Publishing Group), CiteULike (Springer) and 2collab (Elsevier) are owned/supported by scientific publishers and might have different interests.
Hi Martin,
thank you for taking some time to explain this in detail. I believe that EndNote, and possibly RefManager too, can do both 1&4 pretty well, the quality of their remote searches hinging on the quality of the database they link you to.
I think I can see now the advantage of linking EndNote to Connotea. Perhaps, one could instruct EndNote, one day, to query shared databases, such as our own Connotea libraries. It should not really take much in fact, as they already have developed a large number of filters that connect to dozens of archives. Our own connotea libraries already have rss feeds, anyways.
Martin, Massimo and Heather,
The integration of Connotea and Word has been performed. Imagine clicking from Connotea website to cite in word, in a style that you can choose, modify, create.
I won’t say more as I believe I would scoop on Ian’s stuff… Hopefully, more to come end of September. Stay tuned.
To answer this from Massimo,
“Martin and Heather, I am not sure about Papers and Mendeley, but I don’t see the strict need of integrating EndNote with Connotea. Can’t Connotea plug straight into a Word Processor? Please explain, I know you know better.”
The answer is yes, but you always will need to download a local client to take control of your word processor. Connotea can’t take control of word from withing a webbrowser.
There are adavantages in having a local client installed:
1- Synchronisation
2- Full customization of bibliographic styles for instance
3- Advanced search options
4- Automated pdf handling etc…
I can’t speak for Endnote to Connotea, but Endnote (a commercial system in its own right) does not work for Nature journal publication. Apparently, Endnote tells me, they provide a button that allows the user to strip out the Endnote code after exporting/formatting the references. This is important for Nature, because in our production workflows we use our own reference (and a great deal of other text structuring and styling) macros which are not compatible with EndNote’s codes.
Maxine
word has an option do that. You need to remove ‘field codes’.
When you cite with EndNote or other tool, a field is created with hidden data. You can see the data by pressing ‘alt + F9’. See how to remove field codes in word
Hope this help.
Maxine,
a question though. Then what would you want as citations + bibliography, just plain text?
Thanks, Mounir. We can strip out macros, et al., from Word of course, but it takes a heck of a lot of time! Nicer for us if the author does it. We need to do this so that we can start processing the ms ourselves (all this is in our detailed guide to authors). We (Nature) do lots of text and other structuring, including references (including querying them online to check they are correct) using automatic editing tools which we have developed. These tools are also used for a host of other structuring, eg figure captions, greek and special characters, complex equations and so on. These tools provide customised xml output to the journal’s DTD which is picked up by our typesetter and further refined for web publication. It is a bulk processing system because of the number and speed of manuscripts going through the system.
Endnote is but one of several independent commercial reference management systems and it isn’t compatible with any of that. From our point of view (journal display) EndNote is only doing what we do automatically anyway (in parallel with a lot of other processes). I appreciate that EndNote is useful for scientists in other ways than displaying reference lists in correct style and order, but as a journal, that particular concern is the only one we have.
And of course, not all authors use EndNote…..Nature’s authors are from every discipline of science and from every country. So our preference is for the final ms to be in simple form and we do the rest.
as a journal, that particular concern is the only one we have.
I should clarify that – the only one we have as far as displaying the paper in typeset format is concerned. Of course we do lots of web-based things as well, including providing dois, linking references into various databases such as CrossRef, PubMed and ThomsonScientific so the reader can click straight through, etc. What I’m trying to say here in longwinded fashion is that we have tools in place to do all of this, so EndNote and other similar commercial Word macros are not useful to the journal production process.
Maxine
alright well thank you for all this info. I have little idea of what you guys are going through, in the publication process. I am amazed at the level of sophistication. Now then I have a question: if you have so much to do, why not contract with a software company (like mine [but I won’t put the name here] ;-) to create a software program to automate all these tasks (or most of them) and distribute them freely to users. Thus this would take off a lot of the burden you guys have to deal with by shifting a some tasks to the user. I mean you can ask users to download a program that would automate most of the task you have to perform (that would be virtually free) and you would pass the software development cost to a submission fee or publication charges. Nature can ask anything I guess…
I mean that in the long run, this would probably save a lot of time, effort and money. Besides authors would be thrilled to have their papers published even faster. Win/win…
I would like to second the person above me who asked for it to search authors and to search better and faster. I thought my comments were for my own record keeping, so I use them to help myself search. didn’t realize other poeple could see them.
I am a not very unsophisticated computer user, but I have not cracked the meaning of groups here (or why I should make a home page just yet), I thought it would be groups of users? and then you would be able to share certain references with certain other users? that would be great, but it is not obvious how to do this.
Another idea would be to integrate this type of function with Biomedexperts, where your network (perhaps more broadly defined than only co-authors) could see what you were reading… thanks, good luck, I like Connotea, but wish I could use it more, stephr
Robin – If you drop by we can take a look at it and see if we can figure out what you want to do and whether Connotea or another tool can do it.
The acknowledgment of Connotea’s stagnant state of affairs in the face of its many glaring performance issues is galling. How can you even dream of adding ancillary features to a system that is so severely lacking in the performance of its basic, core, objectives? As Morgan, Eric, and Robin have commented upon previously, and to which I see you have ignored, actually using Connotea is so maddeningly slow as to become prohibitive in the basic function of bookmarking and reference management. Actually being able to search one’s database is a pipe dream I’ve given up upon months ago. Indeed, that these performance issues have persisted, and in fact worsened, on the scale of months is simply another indictment upon the indifferent and/or incompetent current state of affairs.
Hi Rob,
I am well aware of our current speed issues. This stems from two underlying issues. The first is growth in the usage of the site and the second is the underlying structure of our MySQL queries, which are not currently scaling well.
To address this we are currently in the process of installing new hardware. This will be the first significant hardware update since the service was launched. Our main plan to deal with slow query issues is to serve the DB from a machine with 32GB of memory. The idea is that the DB server should be able to keep the entire DB in memory, providing a significant speedup.
This may also help with the search problems, but our plan for dealing with search is to move away from our current system and implement a new method for structuring the search index. Currently we slave the MySQL DB to a search DB and search is done from within MySQL. Creating a Lucene index of the content seems like it has a better chance to scale, and we have been looking into this.
These are definitely significant problems, and we are spending most of our available time addressing them. That does not preclude thinking about where we want to be in a year or two years, which was the main purpose for this post.
I’d like Connotea to automatically flag up duplicates within my own bookmarks. Two of us compile the research library for PSI-Nature Structural Genomics Knowledgebase on Connotea. We now have over 250 bookmarks and so it’s easy to add the same one twice.
We export it as XML (and then we do a tiny bit of processing), and so it’s perfect for our website. When the speeded up version arrives, it’ll be a dream.
And another thing – I’d like Connotea to “see” Science Direct properly. I try to avoid looking at articles on SD but sometimes there is no way round it…