Red tape. So much red tape. It can take literally years to get permission to release code as open source within NASA. It's not the scientists - they want to release their code. It's the lawyers.
Concur that it takes time (it took 2.5-3 years for me to open source github/nasa/coda), but in my experience at JSC it wasn’t red tape, but a lack of staffing in the export office. It seems reasonable to me that some amount of review be performed before something can be open sourced, and the effort wasn’t too much on my end. It just took a long time.
I release my work at JPL routinely. The process has been streamlined a LOT in the last few years, and now it usually takes on the order of a week or so.
I'm sure there's a joke here about the men in black wanting to slowly release all the reverse-engineered UFO operating system code.