There is a brief article in Nature Physics addressing the issue of poor writing skills in scientific journal papers and other communications. The article mentions that writing in technical fields is a skill that is assumed to be had by students at the beginning of their research careers. I think that it's fairly obvious that this is very often a false assumption. The article also mentions the common misunderstanding among scientists that verbose and complicated sentences communicate a deeper understanding possessed by the authors. In reality, however, this practice only serves to befuddle readers and acts as a barrier for the proliferation of one's research.
I find it interesting that several experts suggest a linear style of writing journal papers in which ideas are presented in series. This reminds me of how a manual is written. I suppose journal papers should ultimately communicate the steps that were taken to obtain a given set of data and their corresponding analyses.
The article can be found here.