Two thumbs up for CiteULike

After getting tired of the slowness and lack of features of Connotea I decided to try out CiteULike. My only regret is that I wish I made the switch about 6 months ago, since CiteULike seems superior to Connotea in almost every way.

In particular, I really like the built in priority list for indicating the level of importance for me to read a certain article. This is much better than my old Connotea method of using a "to_read" tag. Also, the use of tag clouds (instead of simple lists) not only for my personal tags but also for authors is a really nice feature. I was really surprised to see that you can upload a pdf of the paper being cited for storage on CiteULike. I'm not sure how much I will use this feature, but it is a nice option for those papers that you might not have easy access to all the time. Lastly, the search function is simple yet effective and the site runs smoothly without any waiting or timeouts.

The transfer of my bookmarks from Connotea to CiteULike was easy. I simply exported my bookmarks from Connotea in RIS format (which of course took about a minute of waiting) and then imported them into CiteULike. The only small hiccup is that CiteULike doesn't handle multiple words as a single tag (ie. no spaces allowed), but CiteULike did convert all of these cases using dashes instead of spaces.

Overall, I am a happy social bookmarker again and you can follow what I am reading here:
http://www.citeulike.org/user/mlangill

Sorry Connotea, it's not you, it's me.

I used to really like Connotea and the idea of being able to easily bookmark and share journal articles was what drove me to encourage many people in my lab to start using it. I truly believed it would evolve into a tool that would be essential for researchers and would be a major improvement to existing methods of tracking new papers.

Well, boy was I wrong. I haven't seen any improvements in Connotea for at least a year including a search engine that would be better replaced by Google. Even worse for the past few months Connotea website has been getting slower so that instead of looking up my papers there I just go back to searching Pubmed. Finally, for the past few days Connotea seems completelly useless so that papers I was hoping to cite for my near future thesis writing days are inaccessible.

Like any bad relationship I was hoping it would get better over time, but I have waited long enough and I'm breaking it off before I get any more committed. If the site does ever become accessible again I plan on exporting my bookmarks and taking them elsewhere. Any recommendations?

How do you find science papers to read?

I use a variety of methods to keep me updated on new papers that may be of interest to me, but I am always on the look out for a new tool or method that makes this job easier. Currently, I use the following methods (in order of relevance):

1. TOC email alert of certain journals (Nature, PLOS, EMBO, etc)
2. Pubcrawler
3. myNCBI updates
   
4. Papers mentioned in blogs
5. RSS feed of Connotea groups or similar users

Of course I still have to filter through tons of papers to find ones that I want to read and I often end up missing papers that are not caught by any of these methods. Ideally, it would be great if I had a tool that recorded what I read, and figured out what new articles I would probably be interested in (giving more weight to recently read papers). I think the information from the papers I have bookmarked and tagged in Connotea would be a great starting place for such a tool.

I wonder if anyone else has some tips or tricks that help them filter through the journal haystack?