Publish your code only after you have made the journal publications / conference papers. I have witnessed a researcher getting robbed of his work when another r

Publish your code only after you have made the journal publications / conference papers. I have witnessed a researcher getting robbed of his work when another researcher took his almost complete code from github and submitted faster to a journal for publication. Now both of the researchers have to be cited, but only one of them did the discovery work.