As someone who works in a higher education institution, I am “supposed to” participate in research and regularly communicate with the scientific community. This obligation is much easier to meet nowadays with the Internet. Earlier, personal correspondence, institution supported bulletins and regular review of scientific correspondence open letters were some of the ways that a researcher could employ to keep in touch and keep up with current affairs and developments in the scientific community. There are many on-line communities, forums, open and closed groups and “social platforms” created for the purpose of simplifying communication and cooperation between research and education workers.

Among these platforms is ResearchGate. As ResearchGate describes itself, it is “a network dedicated to science and research.” There, users can “Connect, collaborate and discover scientific publications, jobs and conferences.” Also, they say that it is “All for free”. Truth be told, all of these statements are true, broadly and technically speaking. But as someone who uses this platform and who has been using it for years, there are certain issues under the surface that might not always be obvious to someone who is just starting to use this service.

In this post will try to explain why I have started to dislike ResearchGate more and more and I will also try to explain why I find certain “changes” that ResearchGate has been making lately bad for their “business” as well as our own (speaking as a user).

First, let us start with their own statement that the network is dedicated to science and research. Outright, I will disagree with this and will, instead, argue that ResearchGate is dedicated to providing a platform for self-promotion of its users. Some might reason that this is not a bad thing and with that I will agree, but I will still point out the flawed design of the platform. I will illustrate this with the following scenario.

Say that I create a profile for one John Doe. Afterwards, I start listing my “research portfolio” or “references” as they are popularly called. After this is done my profile is pretty much finished. All I have to make sure is that I regularly update my profile by adding any new research paper that I might have (be it published in a journal, in conference proceedings or elsewhere). Now, when we look at the second part of the ResearchGate’s own statement, it can be used to “connect, collaborate and discover scientific publications, jobs and conferences”. By adding all my publications, I’ve helped them achieve the goal of providing someone what is needed to discover scientific publications. As for jobs and conferences, I will not dispute this, because there are obviously mechanism of ResearchGate through which conferences can be promoted as well as mechanisms for posting job offers etc. This part is fine, but we still have the “connect and collaborate” to deal with.

This is where things get a bit messier. Frankly, there is no systematic moderation done when it comes to “friend requests”. Of course, ResearchGate does not call this feature a “friend request”, as most of the people who will send you requests over time are total strangers to you.

Continuing with the scenario, John Doe receives requests from “researchers” from all around the world and accepts them, hoping that these people might be interested in cooperating in joint research or development or any other form of collaboration, which ResearchGate says it is supposed to provide a mechanism for. Instead, most of the times, after accepting to connect with someone, there will be no further communication (like there was any real communication in the first place) and in the odd case when there is, it would roughly follow this all-familiar pattern:

John Doe receives a message from a certain Johnny Dee, thanking him for accepting to connect on Research Gate (thanking you is a strategy to get on your good side with proper manners and behaviour) and also, asking you to endorse him for certain skills. Being a good and naive newbie, John Doe endorses Johnny Dee and in return is himself endorsed for some random skill. Since this is the first endorsement, opening a completely new way of using ResearchGate, John Doe might not react abruptly and protest for being wrongly endorsed. After all, who doesn’t like to be praised from time to time, even if, sometimes, it is just a bit misguided? In his defence, Johnny Dee doesn’t really know him, so it was an honest mistake. How could he really now? His “real” friends and colleagues will not make such mistakes and one mistaken endorsement will eventually be covered up with dozens of proper ones.

Now, let us allow for some time to pass and then let us look at John Doe’s endorsements. Before that, we should be a bit more familiar with John Doe. John Doe used to work as a web programmer and he teaches web programming. He is well versed in PHP, HTML5, CSS#, JavaScript, AngularJS, Node.js and related technologies. Now, look at his endorsements. It would seem that John Doe is mostly endorsed for having great marketing skills, for knowing to code in Assembly language, for being a great electrical engineer and for being really good at using Microsoft Excel. How could this be? Surely, John’s 100 of a total of 200 connections who have endorsed him at some point in time know him better than that, don’t they? Well, no. Of his 200 and so connections, he actually knows only about 30 people personally and of those 30, only 10 have endorsed him at some point. So only about 10% of his endorsements are from people who actually known him, his work, his professional achievements etc. The rest just endorse in hope to be endorsed back, because they need those endorsements alongside a lot of publications (real authorship questionable) for creating some fake brand of themselves.

This factious scenario, so far, illustrates how using ResearchGate to promote oneself is a misguided effort because no one will really take this part of your profile seriously. IN my personal example, at the moment, my topmost endorsement is that I am good at JSP. I totally despise using JSP or using Java for web programming in general. I never liked making web applications in Java and I will never suggest this to anyone, nor will I ever do this myself. Instead, I use PHP. So, I keep being endorsed for JSP while I don’t really like it. The second, third and all further endorsements (Programming languages, Linux, ICT, Databases etc.) are correct, as I really do have skills in those fields.

Regarding the “connect, collaborate and discover scientific publications, jobs and conferences” statement, one can certainly connect, but there is no systematic moderation and validation. You are personally responsible for connecting only with people that you personally know. As for discovering scientific papers, this is possible, but regarding it being legal or not is a different questions. Say that John Does had published a paper in a Journal. John was required to sign off copyright to the publisher so that the publisher can earn money by selling the journal which includes John’s paper. But, on ResearchGate, John did not just list his publication and abstract. John had uploaded his full-paper. After all, everyone does it, so why wouldn’t he do so too? Now, John is in direct violation of his paper’s publisher’s copyright. So, using ResearchGate in this way is probably, most of the time, illegal. So, ResearchGate promises discovery of papers, but researches are misled in thinking that they could and should upload the full paper, fearing that otherwise, their papers will be discriminated in ResearchGate’s search results. And they are. Then a user searches for something on ResearchGate, hoping to find articles related to his research, ResearchGate will list all found publications, but those that have been uploaded will have a nice preview icon, while those that were not will have a blank generic icon and will most probably be skipped by the user. This way, ResearchGate systematically discriminates certain publications and knowing this, users will out to upload their full paper even though they know that by doing this they are probably breaking laws. So, ResearchGate, by design requires you to break copyright laws.

Furthermore, regarding the same “connect, collaborate and discover scientific publications, jobs and conferences” statement, collaboration is completely different question. Say that John want’s to find collaborators for a certain research project. ResearchGate requires you to create a Project in order to pique an interest in researchers. Once John does this, he is publically telling the world what he is working on. Sometimes, a certain research project’s name is enough to give someone an idea about what he is working on. A decent researcher will offer help and collaboration in the proposed research and he might find a new college to work with. But a dishonourable person will try to steal the idea, do his own research with the help of his own corrupt partners who will publish their research in some fast publishing journal and practically steal the right to John’s idea by being the first to publish it.

Of all that ResearchGate offers I find only one feature useful and completely worthy of my time and that is the Question and Answer functionality. If one wants to help his fellow researchers, this is one of better ways to do it. Sharing experience and knowledge is good, but it is much better when one can help on a specific topic, instead of writing about generalised concepts etc. When a ResearchGate user needs help, he asks a question and those who can answer can do so. Answers that are wrong or misleading can be reported and useful ones can be recommended. Of course, ResearchGate can change this as they did once before by removing the option to up-vote and down-vote answers with the option to recommend them. However, as many other things regarding ResearchGate, even this has its cons.

Say that someone asks a question that John can answer because it is related to PHP and the way variables can be passed by reference to functions. John sees the question and sees that it was asked by a new user (profile still incomplete). He chooses to give a detailed and well written answer. His answer is read by the user who asked the question and, maybe, he is thanked for it in response to the question. The user does not recommend Johns question, even though there are other less favourable and not-as-complete answers given. Was John’s effort in vain? Well, partially. On one hand, John did a good thing by collaborating, but on the other he might have done something bad, such as give someone a free and easy solution to their school homework or research assignment. Being someone who works as an educator, I feel that just handing out solutions to homework is the worst kind of “service” or help you can give. The beneficiary will not really benefit from your help. So, John was not careful enough and might have actually did more harm than good. This is not really John’s fault. John just answered a question. But, ResearchGate lets anyone register. There is no real moderation of registered users. There is no verification of users regarding them really being faculty members of a university, an institute etc. So, her as well, ResearchGate is just being irresponsible as everywhere else.

In my own opinion, ResearchGate is a very poorly designed platform with many misleading features that does not really help to better the academic community. Instead, I believe that it is hindering its development by side-tracking researchers.

I will probably continue to use ResearchGate, mostly for its Question and Answer feature, because I believe that I should help if I can. Nonetheless, if I see researchers starting Projects named “Sampling brewed roasted coffee bean drinks at optimal venues for extended time intervals with familiar individuals” I will certainly rethink my continued use of this “social media” platform.