Klaas Wynne
27 March 2009
Libraries: so 20th century
Tags: literature, RSS, searches, web of sciencePosted in PhD life, Technical (ms word, tex), Web 2.0
I used to go to the library. Every couple of weeks or so I would go and check the journals, browse through their tables of content (TOCs), and flip through the pages. You would find odd articles in areas there weren’t quite your own. Slowly over time journals got bigger and were published more frequently. Then they started emailing TOCs out, which seemed like a pretty good improvement as you could just read them on your computer. Gradually, I started to run out of time: the TOCs were just too long and there were too many of them. Virtual journals (such as the Virtual Journal of Ultrafast Science) offered some hope but they don’t cover all journals such as Elsevier journals, so you would still end up having to read a bunch of emailed TOCs.
For a couple of years, I tried an RSS reader. A free reader, such as NetNewsWire for the Mac or FeedDemon for Windows (http://www.newsgator.com/individuals/) can collect RSS feeds containing TOCs from you chosen journal. FeedDemon has to possibility of searching through your feeds. That worked pretty well until I switched to a Mac and found that NetNewsWire couldn’t search through feeds. By the time the TOCs were simply to huge to read them “raw”.
Thus, I switched to Web of Science (WoS) searches. I have searches for topics like:
“boson peak” or fragility or “fragile to strong” or “liquid fragility” or “dynamic crossover” or “glass transition” or “liquid liquid” or “mode coupling theory” or polyamorphism or “Stokes Einstein” or “strong liquid” or supercooled of “Vogel Fulcher” or “Vogel Tammann” or “Widom line”
But that often gives results from journals I don’t like, so I added a list of my favoured journals to the search too:
“physical review *” or “accounts of chemical research” or “applied physics letters” or “biophysical journal” or “chemical physics” or “chemical physics letters” or “chemical reviews” or “ieee journal of quantum electronics” or “journal of applied physics” or “journal of chemical physics” or “journal of molecular liquids” or “journal of physical and chemical reference data” or “journal of physical chemistry *” or “journal of physics condensed matter” or “journal of the american chemical society” or “journal of the optical society of america *” or “nano” or “nano letters” or “nature” or “nature materials” or “nature nanotechnology” or “nature photonics” or “nature physics” or “optics communications” or “optics express” or “optics letters” or “phys chem chem phys” or “physchemcomm” or “proceedings of the national academy of sciences *” or “review of scientific instruments” or science[SO]
These searches run once a week and are delivered to me through an RSS feed set up on WoS. I’ve been running this set-up since October and am reasonably happy. However, whenever I search “by hand”, that is click around semi-randomly on WoS following whatever seems interesting, I tend to find articles in my area that were not picked up by my search. That’s a bit worrying.
How do you search for papers? A colleague of mine (they shall remain anonymous) told me that they didn’t look for papers at all but waited for people to send them papers of interest. Surely some sort of macho thing to say… But seriously, I would be interested to know how other people keep current with the scientific literature.
Or subscribe to our RSS feed.
27 Mar 2009 15:11, Jacopo Bertolotti
I just use google reader as a rss aggregator and I periodically run through the whole list of titles. This is quite time consuming but I manage to keep the time needed for it on a reasonable level limiting the number of journals (at present I check 10 of them).
Google reader also allow to search through the feeds (using the very same syntax as in standard google) but I never checked it carefully so I don’t know how user frindly it is.
By the way: some journals (like the whole “Physical Review something” family) have rather crappy rss. Physical Review Letters, for instance, deliver via rss only the first 30 articles. Therefore it might happen that you miss something if you relay only on rss.
27 Mar 2009 18:04, Sanli
I use google reader as well. I have about 20 journals in my subscription list. I also follow citations to “important” papers of my own interest (about 50) through google reader. Web of science provides the rss-feeds for citations but it is much lagging behind the journal output. Often, I have already seen the titles few weeks ahead. It is also full of duplicated entries, but in general it is very efficient in sense of relevance.
Above that, I use the sharing option and suggest titles to my fellow group members (about 8 who use the service). They do the same as well and this reduces the chance of missing a really interesting article. However, the number of articles that I share is often an order of magnitude more than those that my colleagues share.
I am very willing to exchange my google reader shared items with somebody in my field who is an active (rss)reader.
If you are interested Jacopo, send me an Email.
1 Apr 2009 16:51, Alex
Hi! My blog is designed to help PhD students to get through their programs by giving them a very realistic picture of what academia is really like. I was wondering whether you can add a link to my blog. I would really appreciate it.
2 Apr 2009 17:55, Ad Lagendijk
@Alex,
Alex, your comment is out of place here. But certainly we are interested in your blog. But links go two ways: that is what we call exchange of links.
What would also help, is to allow on your site trackbacks.
14 Apr 2009 16:39, Ben Hoganson
I’m a Research Librarian for a laboratory and stumbled across your blog.
I was caught by the title of this blog post, and somewhat saddened by it, as I feel that it defines libraries as a thing of the past. We use RSS feeds, blogs, wikis, commercial databases, alerts, citation analysis tools, the open web, etc. on a regular basis. Many of us also have presences in Second Life and other virtual worlds. We also purchase more and more eBooks as space constraints require us to get rid of our paper collection. Libraries are definitely adapting to how information’s disseminated, but we still are very relevant, though less visible; my job as a Research Librarian is to save scientists and engineers time on the information side so they have more time to do their primary research.
However, I need to mention that the print collection is still very important and it would be wise to not disregard the collection. There was a paper published recently that showed how Google Scholar indexed quite a bit of the recent scientific literature compared with EI Compendex, but steadily declined the further back you go. Depending on your field of study, it’s very important to use this older literature.
If you have any questions about research, feel free to contact me at hogansonb (at) gmail.com, or befriend me on Facebook. If you work for a university or a company that has a library, I encourage you to seek them out. Their job is to help you.
-Ben Hoganson
P.S. Great idea for a blog. I commend you.
15 Apr 2009 10:02, Riccardo Sapienza
Hi,
I have the same problem. I first started with RSS feed readers, but then I was overwhelmed by the feeds… Recently I do manual search only on selected journals, I use Papers for mac, to organize them like iTunes, and I share a lot of novel articles via email of pdf, which also overcome the problem of limited journal access.
I am very interested in the google reader bookmark sharing, Sanli, Jacopo, shall we try it, at least as an experiment?
15 Apr 2009 10:19, Klaas Wynne
@Ben: I agree with you on older literature. I use Web of Science, which (for me) only goes back to 1970. It is sometimes important to be able to go back further. For example, I really enjoyed it when the American Physical Society put the Physical Review online back to 1893. Having said that, in practice I rarely use anything older than 10 years.
@Riccardo: Papers is stunningly good. I was going to write something about it later.
15 Apr 2009 17:17, Riccardo Sapienza
Hi Klaas,
some people use Sente for Mac, but I definitely prefer Papers.
R.
16 Apr 2009 10:11, jacopo
Searching a bit around the only non-Mac equivalent of Papers I found was JabRef. It has the nice properties of being multi-platform and opensource but I wasn’t completely satisfied (it’s still a bit “rough”)
16 Apr 2009 17:15, Klaas Wynne
Sente seems to be a bit like Papers but I have no personal experience. JabRef looks like a free version of EndNote with links to PDFs but not an integrated PDF reader/searcher.
16 Apr 2009 20:41, Ad Lagendijk
I like multi-platform (Windows + Mac + Linux-families) applications a lot. I think that anybody developing software that is targeted especially for scientists should develop it as a multi-platform application. Programs, however good they are for a certain scientific goal, that are developed for one platform only will (or should) never make it in the scientific world. Developing multi-platform is extremely difficult and usually the solution that developers choose, is using a web-interface for their application. In such a way the multi-platform complications are on the shoulders of the developers of browsers.