Geeks should learn to write.

In 1923 the Nobel prize in Medicine was awarded to Sir Frederick Banting and John Macleod of the University of Toronto for the discovery of insulin. A flurry of controversy ensued as Banting's lab assistant, Charles Best, was overlooked for the award in spite of his large contribution to the discovery. Another co-discoverer James Collip  that shared the patent for insulin with Best and Banting was also snubbed. This has been viewed most commonly as pandering to Macleod's already established reputation but it could be seen in another light.
Macleod's contribution was legitimizing the initial findings from the insulin discovery that were met with wide skepticism. His expert skills as an orator and writer allowed insulin to reach the world and his influence caused Banting to suspect Macleod of trying to take all the credit. Putting aside the personal issues, the Nobel committee  reasoned that the entire discovery might not have taken off if not for Macleod spreading it effectively.
Writing is the principal method of communication worldwide and most students in technical fields still unfortunately do not get explicit education for it. There are untold amounts of useful discoveries sitting in abandoned shelves or servers due to the poor writing abilities of their authors. It is not without reason a portion of most research grants include publishing funds as scientific discoveries essentially do not count if not communicated. I like that the University of Toronto is trying to approach this issue with the Writing Instruction for TAs (WIT) program in many STEM departments.  This is extra funding for TAs to mark and give feedback on writing. I have personally been through numerous chemistry courses with this programs and have felt the benefit on my lab reports. In the end, the ideal is to be able to read any paper or book and not tear one's hair out because you can't understand an author even though you both speak the same language.