2 ms·
Thanks! I do think that some data would be good, but at the same time, the kind of data that's relevant here is really hard to get in any concrete terms, so mi
by steveklabnik 9y ago
Thanks!
I do think that some data would be good, but at the same time, the kind of data that's relevant here is really hard to get in any concrete terms, so might actually be harmful.
> a "paper" of any sort
Remember that this isn't a paper of just any sort: it's a whitepaper. So it has a specific audience: CTOs, VPs, and other technical management types. Audience is important for what you have to include in any work.
- faitswulff 9y agoThat makes sense, thanks!
- steveklabnik 9y agoNo problem. I bet CTOs would love “we had a 30% reduction in defect rate” or data like that, for sure, it’s just so hard to collect...
- faitswulff 9y agoOh! Actually, I'm paging through "Rust in production at Figma" (https://blog.figma.com/rust-in-production-at-figma-e10a0ec31929 https://blog.figma.com/rust-in-production-at-figma-e10a0ec31...) and I think this is closer to what I would expect in a "paper" - while acknowledging that am predisposed toward thinking "graphs and data" when I think of a paper, I do think these kinds of blog posts have more of an impact (and seem more true to the idea of a "paper") on me than the current Rust whitepapers. That said, I will happily concede that I have no idea who the target audience is for the whitepapers and probably don't know the accepted industry definition of a whitepaper. However, I'm very happy to see Rust flourish either way.