Month: September 2009

What happens when your research topic goes out of fashion?

sarikan

What? My research never goes out of fashion! How can it go out of fashion when the problem is not solved, and my PhD is not even remotely appearing in the horizon?

Well, you may be right, but when I say fashion, I mean the hype your research topic can create. The amount of people wondering what is that thing that you’re talking about, and the most important of all, the grants! Yes, grants by venture capitals, research institutions, EU Framework Programme calls, you name it.  Why the grants are labelled as most important of all? Well, I’m terribly realistic sometimes, and there are some topics out there, which are so big and complex that you can’t produce concrete outcomes without spending millions of hours of work on it, and an effort of this magnitude can’t exist only by people’s personal commitments.

I’ve been following the EHR domain for about 7 years now, and I’ve been also following the popular research topics via many grants and calls for projects, especially in the context of EU FP 6&7. Something I’ve noticed is, EHR domain is being mentioned less and less in the research grants, and people writing project proposals are now trying to avoid focusing on EHR aspects as the core of their projects. Reason?, well, according to many, EHR is now a topic that has reached saturation in research domain. We know that the state of EHR is nowhere near where it was meant to be, there is a huge amount of work waiting to be done, and the domain will need a lot of support. However, the focus of research around health IT seems to be shifting to some topics like genomics based medicine, bioinformatics, gene therapy etc. etc.  Don’t believe me? Check out this call for example. Take a look at the contents.

So the focus in research is shifting, but previous research in EHR did not bring a lot of change into this domain. Or did it? If it did, why am I not aware of it? There are bits and pieces of outcomes here and there, which I can use. These outcomes are the results of research projects, funded by various organizations, but did we achieve the goals we declared we had in our minds when we applied for all those research grants? My answer to this question is No. It appears though, that EHR is something that everybody is tired of hearing. This is a problem, we do not have the systems we want, but already people are beginning to skip to next hot topics in research, and EHR community will have to link its efforts, and its existing intellectual investment to these new frontiers, otherwise we’ll be disconnected from resources of funding. Sorry for being so much focused on funding, but I see great danger ahead if EHR domain can not sustain its research, and unless we put EHR work into larger contexts, we’ll have serious problems.

I believe openEHR and 13606 can survive this transition, and they can be relevant and useful for new research topics, and this is mostly due to their two level modelling approach. With right reference models these initiatives can help other research communities by providing them a better, computable platform for their work. HL7 has its industry roots, and it will stay alive due to its nature and probably due to a lot of money which will be spend in USA, but I can’t see it surviving the next wave of data, with its current design. HL7 still seems to be looking for a way of achieving the features of openEHR and 13606, only with different names, so they have some way to go.

Therefore, I think we need to consider other forms of data in relevant, but different medical domains as requirements that must be handled with our existing research topics. The work around EHR should be considered for alignment to handling of genomic data etc. Otherwise, from a purely academic research point of view, EHR research may become obsolete before reaching its target level in especially clinical practice domain.

Google Sidewiki

sarikan

Google has just released sidewiki, a toolbar that allows you to attach comments to web pages. This is a step that will allow  a (supposedly) large number of users to attach information to a resource, and Google will try to rank the comments using their ratings, which will be provided by other users. To me, it is an advanced version of  of del.icio.us, which provides a more granular access to other people’s interest in the things you are interested in. If it takes off due to large number of users Google has access to, it may become a very useful tool . It may even have the potential to change the way we do searches on Google, if it allows searching around the chosen page, using comments and commentors’ profiles. This is another case of Google providing a slightly better implementation of an existing idea, leveraging its size. Let’s see what’ll happen.

JRuby team gone, who is next?

sarikan

Oracle’s acquisition of Sun is having minor bumps, and in the process is making what would happen anyway happen faster. Cryptic? Well, how about this: JRuby team has left Sun. Their work was important, quite important actually. I’ve done some Ruby work before, of course starting with Ruby on Rails projects and after that some db utilities. I found out that scalability was the most important issue, and these guys were doing a great job with JRuby, so that ROR could run on Java. Java means scalability (among many other good and bad things).

Now they are gone to a rails company. I’m curious about the future of many other small but critical projects. Oracle is probably going to kill them all, and IBM won’t hesitate to increase the tension around Java related stuff since Oracle is a bigger enemy than Sun. We’ll end up losing valuable assets at the end.  Good luck to JRuby guys, and thanks for all of their great work.