3 ms·

I wonder how often it is the case that code isn't considered an academic product per se, and so free to use. May have to make it very explicit.

webmaven 5y ago
> I wonder how often it is the case that code isn't considered an academic product per se, and so free to use. As an outsider with occasional glimpses into academia, I've observed a bimodal response: On the one hand, code seems to be seen as more available for reuse in terms of copy-and-paste, particularly in the exploratory phase. On the other hand, code reuse is somewhat less likely to trigger an acknowledgement and citation. This unacknowledged copy-and-paste borrowing of code seems in turn to be one influential factor inhibiting the derived code from being shared in turn. > May have to make it very explicit. That certainly can't hurt. I've certainly seen that reusable tools aren't cited as often as they are used, which in turn inhibits efforts from being made by academics in the production and refinement of such tools (the citation being the primary currency and reward mechanism for publication in academia). Explicitly asking for citations (and getting them) helps with that, though the more general issue of "releasing open source software isn't seen as 'publication' for academic status purposes such as being considered for tenure" remains a problem, though it varies considerably by field, institution, and department. But if given some thought and extra effort, a researcher might actually get more papers (possibly with different sets of collaborators) out of roughly the same body of work, especially if they look more broadly for appropriate conferences and journals. BTW, again as an outsider with occasional glimpses into academia, it seems to me that there are many vacant niches for cross-disciplinary journals and conferences focused on reusable assets such as datasets and tools necessary for research.

webmaven 5y ago

> I wonder how often it is the case that code isn't considered an academic product per se, and so free to use. As an outsider with occasional glimpses into academia, I've observed a bimodal response: On the one hand, code seems to be seen as more available for reuse in terms of copy-and-paste, particularly in the exploratory phase. On the other hand, code reuse is somewhat less likely to trigger an acknowledgement and citation. This unacknowledged copy-and-paste borrowing of code seems in turn to be one influential factor inhibiting the derived code from being shared in turn. > May have to make it very explicit. That certainly can't hurt. I've certainly seen that reusable tools aren't cited as often as they are used, which in turn inhibits efforts from being made by academics in the production and refinement of such tools (the citation being the primary currency and reward mechanism for publication in academia). Explicitly asking for citations (and getting them) helps with that, though the more general issue of "releasing open source software isn't seen as 'publication' for academic status purposes such as being considered for tenure" remains a problem, though it varies considerably by field, institution, and department. But if given some thought and extra effort, a researcher might actually get more papers (possibly with different sets of collaborators) out of roughly the same body of work, especially if they look more broadly for appropriate conferences and journals. BTW, again as an outsider with occasional glimpses into academia, it seems to me that there are many vacant niches for cross-disciplinary journals and conferences focused on reusable assets such as datasets and tools necessary for research.