In my previous posts on twitter I have been describing (in general terms, I will do some posts on some of the more technical aspects soon) how I have automated some of my tweets. The system, in brief, tweets scientific articles on a random basis – doing about twenty tweets a day. Each 24 hours, I change the topic so that if people are interested they can just follow me when I am tweeting about a topic that interests them.
INFORMS, in particular, have been very kind to retweet some of my tweets. I have not checked but, I suspect that the articles they retweet are those that are published in INFORMS journals!
This made me think that I should be a little more proactive in telling the publishers what a good job I am doing for them. As I update my database with articles that I tweet, I have started to add a field which indicates which publisher they come from. At the moment, the publishers I have on file are IEEE, INFORMS, Taylor & Francis, Science Direct and Wiley. All of these also have a twitter account, with the exception of Wiley. Well, maybe they do, but I cannot find it.
Over the past couple of days I have been implementing a system that chooses one of these publishers at random and also chooses a time interval to query. For example, it might be the last 3 months, the last 7 days, the last 11 months etc. It is now a simple matter to run an SQL query from PHP to extract how many tweets I have done for that publisher over the relevant time period. I can then format a tweet so that it says something like
@TandFRef You may have noticed us tweeting your papers? We have tweeted 147 of your papers in the last 29 days
I can then use my automatic tweeting system to post the tweet to twitter. To try and keep things looking fresh, the format of the message I post is randomly selected from a number of templates I have defined. Along, with the variation of the intervals I use, I hope that the publishers will find it useful and not too repetitive.
Of course, one of the ideas behind this is to raise my profile with the publishers, and also on twitter generally but, in doing so, I hope that people find the information useful. I am conscious that it could become intrusive though, so I only do a couple of these tweets a day. Hopefully the publishers will not mind. If they do I can, of course, remove them from the service.
In the future, I am thinking of extending the system so that I can tell people how many tweets I have done on (say) vehicle routing in the past n days/months. I could even combine it with the publisher information so that I can tell the publishers how many of their articles I have tweeted over the past few days/months on a given topic. But I’ll let this new system bed in first.