Musings on Conferences

So I have a couple of thoughts on conferences I would like to air out.  I think there is a lot of problems in the current structure as I have seen in my research career.  Some organizations put on better conferences than others, this is not in doubt.  But the general quality of a quality conference is slightly unnerving to me.

1. The publishable delta is larger than you know.  There is a lot of thinking that the smallest publishable delta should be published (and sought after).  I have problems with this model as it encourages people to do trivial uninterested research in general.  Additionally the smallest publishable delta is larger than most people think.  It is obvious that some institutions are publication machines without any thought on the quality of the research.
2. Most people give terrible presentations.  This one probably wouldn't surprise most people as a lot of people in general give poor presentations.  Most colleges have a public speaking class that is required, I know I did.  The course was a joke honestly.  This is however a very important skill and it is one that I actually did take away from graduate school partly from my advisor whom I left.  It is important to be able to explain your research to anyone and it is important to describe it clearly.  Most people I have seen recently fail at both of those.
3. Most people can't write coherent papers.  I am amazed at the poor quality of writing that is accepted in a lot of journals.  I have been an editor a couple of times now and there are occasionally good co-editors, however in general I feel like the reviewers just rubber stamp most work.  Additionally I feel like reviewers should not know who wrote the paper because it biases them.  I mostly feel like my time is wasted when I review papers because my work is larger ignored or unnoticed.  There are good papers and better journals but most are just full of crap (including the 'premiere' journals).
4. All because you can make a graph of your work doesn't make it interesting, research, or publishable.  I can't tell you how many presentations I have been to where the presentation was just talking about an experiment and not results.  Or some stuff which results with one graph.  Amazing.  There is no peer review of the academic merit there.  Not everything should be published.  While I agree that no result is a result you have to have some reason for someone to care beyond 'I need to publish for my career'.
5. Fifteen minutes is not long enough for anything.  The common thread I have seen in a lot of conferences is to allow for 15 minute presentations.  You know that isn't enough time.  I really feel like 30 minutes is what is needed (however 15 minutes is more than enough to bore the hell out of me).  No one ever gets done in 15 minutes and no real information can be transferred to the audience in that time.
6. This follows from the previous one but posters and WAY better than presentations.  I know everyone thinks that presentations are better and more respectable than posters but I love posters both as a presenter and as an observer.  I like talking about the material, being able to move forwards and backwards through the material, not having a time limit, the amount of questions you can address, etc.  I just feel like it is a more effective way to get information across.  However I will say, there are a lot of REALLY bad posters.  If your poster has over 200 words on it you are probably doing something wrong.  I once saw a poster with, I kid you not, 5000 words on it.  Could not believe it, needless to say I didn't read it.

This all comes on the heels of an interesting discussion on slashdot.org about Chinese papers and employees.  I don't actually have any direct experience with the material talked about in the discussion but my experiences with outsourcing are less than glowing.  What is happening to research?  Perhaps I was spoiled because I went to a big name university where I met and interacted with really, really, smart people.  I know people who did interesting and compelling research (though not all).  Perhaps the majority of research is this search for the smallest publishable delta and they fall short.  Seems like a bad thing if it is.