maybe academic partners is not the correct term. But if you for instance need HPC but writing HPC infrastructure is not your expertise (hence you won't get funding for that) there might be other universities that do research on HPC and might be quite interesting getting in touch with a new programming language were they could bring in their HPC expertise.
Cheers, Tobias Am Freitag, 22. April 2016 05:55:37 UTC+2 schrieb Sheehan Olver: > > > Yes software development “because it is needed” for research works well, > and indeed there are many cases where people have built academic successful > careers while developing software. > > The catch is that there’s always a lot of time consuming maintenance tasks > involved in keeping open source software useable, and it feels unfair to to > ask people on temporary contracts to do this…so I’m looking for another > way: right now either "lab technician”-like positions or hiring > consultants. > > For your “academic partner” suggestion, how would that work? I can > imagine grants funding postdocs, lab technicians or possibly consultants, > but I’m not sure where academic partner fits in. > > > > > > > On 22 Apr 2016, at 1:42 PM, 'Tobias Knopp' via julia-users < > julia...@googlegroups.com <javascript:>> wrote: > > Well you mentioned "Thread support for Julia" and I could imagine that > there are academic partners that could work on this if they could get > funding. > > The problem with finding a postdoc that does software development is > indeed tricky. Usually this will not be full time but people will work on a > scientific topic and write software "because it is needed". There are > various people around Julia that work this way. Best example is Jeff who > made his Phd while developing Julia. The fftw library was also developed as > part of a research work. > > Cheers > > Tobias > > Am Freitag, 22. April 2016 01:32:51 UTC+2 schrieb Sheehan Olver: >> >> >> Maybe I don’t understand what you mean by “academic partner”, but I can’t >> imagine academics jumping at the chance of working on someone else’s code… >> >> There is also a problem with including a postdoc with the expectation of >> her/him carrying out basic software tasks: time spent on software >> development is not really a help getting an academic job. >> >> So the consulting model seems like a good alternative, I’ll look into it. >> >> >> >> >> On 22 Apr 2016, at 1:52 AM, 'Tobias Knopp' via julia-users < >> julia...@googlegroups.com> wrote: >> >> An alternative would be to instead make a collaboration with an academic >> partner that is capable of working on this. There are various large open >> source projects that use this model. >> >> Cheers >> >> Tobias >> >> >> Am Donnerstag, 21. April 2016 16:56:57 UTC+2 schrieb Steven G. Johnson: >>> >>> >>> >>> On Wednesday, April 20, 2016 at 11:58:02 PM UTC-4, Sheehan Olver wrote: >>>> >>>> Has anyone included "Julia Computing" charges in a grant proposal? Is >>>> this something that should be encouraged? >>>> >>> >>> That would count as hiring a consultant with your grant funding. Some >>> funding agencies allow this (e.g. NSF allows it in the USA), others may >>> not. See e.g. the NSF guidelines ( >>> http://www.nsf.gov/pubs/policydocs/pappguide/nsf14001/aag_5.jsp -- >>> scroll down to "Consultant Services"). >>> >>> I've seen this kinds of thing in grants a few times. You generally get >>> a "letter of intent" from the consultant indicating what they will do and >>> their fees. >>> >>> Seems like a great thing to do if you can do it, but I would only expect >>> it to be possible in a small minority of grants --- you need to make a >>> convincing case that the consulting is essential for the core goal of the >>> research project, and that hiring a consultant is more cost-effective than >>> doing it in-house. >>> >> >> >
