PubMed and me: Why my coughs are more productive than my searches

My story begins with yet another PubMed search. I like PubMed. It's generally a friendly sort of search engine and seems to require less commitment than my previous engine, MEDLINE. With PubMed I don't have to choose which years I want searched. It seems to know me and understands that I want it to do a thorough job without asking me too many questions. It also has that handy “related articles” option, which I play with a great deal. Best of all, I can get PubMed to search just for articles with abstracts and not waste my time with the others.

At any rate, I am working with PubMed and, for perhaps the hundredth time, I am amazed at the number of journal article titles that contain a colon as punctuation. This has always struck me as a cheat: a way of having two titles for the price of one. I know this observation can't be unique to me and wonder if anyone has ever bothered to publish an article on the subject. With PubMed up and running, I decide I might as well do a search. (I later realize that this diversion is evidence of a high degree of distractibility and that my frontal lobes need a bit more executive function, but I am not thinking about this at the time.)

My first search phrase is “use of the colon.” It doesn't work: I get 76 368 items on 3819 pages. The first article is “Role of yoghurt in the prevention of colon cancer.” Maybe one of the 76 367 others is more relevant, but as I have already implied, my limited attention span isn't going to let me check them all out. I try a different request and plug in “punctuation.” This is somewhat better in that only 105 items are found. Some are pretty bizarre, for example, “Adaptation, punctuation, and information: a rate-distortion approach to non-cognitive learning.” I can't tell what the article might be about, even with the use of the cheating colon. Another one sounds potentially more helpful. It is called “Grammar and punctuation in scientific writing,” but it has no abstract, so I move on. Most of the other articles seem to be about genetic code punctuation. There is a lot of chatter about “prosody” as well.

I decide to be more direct, and I punch in “:” but PubMed doesn't like that much and sends back “syntax error in query: improper beginning of expression.” I can't believe it — PubMed itself is using a colon.

I try another tack and enter “article titles:” I am thinking that if I slip the colon in with some words, PubMed might take it. I am wrong. PubMed comes right back at me with “syntax error in query: text phrase expected after range operator.” No yield from the search, and I have to endure another colon. At this point I drop the punctuation and just put in “article titles.” This approach is more profitable, generating 78 responses. One item that catches my eye is “Misleading and unserious article titles,” by M.J. Tos, published in something called Ugeskr Laeger in 1998. It was written in Danish. I know that, not because I try to find this article, but because PubMed tells me it is. I wonder about this article. First, I wonder if “unserious” is actually a word. I also wonder why M.J. could not bring himself/herself to use the word “funny.” I decide not to read the article, as I have a feeling that M.J. and I might not see eye to eye on the issue of humorous writing. I do give M.J. credit for not using a colon in the title, however.

I try clicking “related articles” off this title. It gives me 120 items. The subsearch is worth it just to read some of the titles generated. My favorite is “Are we using the terms exactitude, validity, precision, reproducibility and accuracy correctly? A proposal for discussion.” Alas, the article is written in Spanish, and PubMed doesn't offer me an abstract.

Another item that turns up on this subsearch is “Unlocking the article inside you.” I like the sound of that. I wonder if the locked-up article will be written by my inner child. I wonder how it's going to get out. I also see “Why should a surgeon publish?” I like that one, too. The title is direct and doesn't need a colon.

I then strike gold. I discover an article called “Editorial pet peeves and the colonized title.” It was published in the Western Journal of Nursing Research in December 2000. The author is P.J. Brink. (If you want to find it, put in “colonized title.” The article will come up, along with one by J. Basuray called “Nurse Miss Sahib: colonial culture-bound education in India and transcultural nursing.” I'm guessing they are not about the same thing.)

The Brink article sounds exactly right, but it doesn't have an abstract. For a fleeting moment I consider a trip to the library, but that seems a bit extreme, so I click on “related articles.” That's when I hit my PubMed low: the message reads “link data retrieval error (IQ).” Now that is just getting too personal. I don't know how PubMed could even know my IQ, and I don't think it is polite for it to throw the whole issue in my face.

I am getting testy at this point, so I back into my earlier search and find another possible hit in an article called “Titles,” written by somebody named Brenner and published in Current Biology in 1998. Because it sounds interesting, I decide to follow up on the tempting little offer that appears for me to see the whole article by logging onto BioMedNet. I have had some bad experiences on the Internet and tend to be a bit cautious about signing in to anything new, so I log in using one of my many pseudonyms, “Teddy.” I am told that the user name has already been taken, so I add “bear.” Somebody else has already logged onto BioMedNet using “teddybear,” so in the end I have to go with “teddybearfur.” I click to tell them I am from Antarctica, because that is one of the choices they give me, and advance.

BioMedNet asks me about my interests. I'm okay with that, as it seems like a sincere overture. I picture the gang at BioMedNet saying, “Hey, I wonder what Teddybearfur from Antarctica really cares about?” They don't give me a chance to say I like long walks or chocolate, but I do check off “neuroscience.” It's then that I notice the little box at the bottom that you have to check that says “I do not wish to receive information from carefully selected, reputable companies which may be of interest to me.” Those statements always bother me. It's the use of the negative. I think what it is supposed to mean is that, if you check the box, they won't let any companies contact you. My concern is that they don't say anything about unselected or disreputable companies. If I check the box, am I actually saying that I don't want them to make sure that the companies are reputable or maybe that they should let people send me stuff that will bore me? I worry for a while, but check it anyway.

I'm almost there, ready to read Brenner's work, when I get held up by the need to agree to the “terms and conditions.” They sound appropriate in general, but I get hung up on the following: “BioMedNet may modify any part of the Terms and Conditions, and may discontinue or revise any or all aspects of BioMedNet at its sole discretion and without prior notice.” In other words, I have to agree to terms and conditions that may change at any time without so much as a call to see how I feel about it. I can't see how that's fair. I decide I don't want the article that much anyway, don't sign, and back my way out.

By this time I have wasted an hour. I am no closer to finishing the article I am supposed to be working on. I decide to turn PubMed off and go have a coffee. Maybe if my inner child has a little caffeine it will get to work and write something for me.