Hi, I’d like to chime in. Cookbook tries to address few problems:
- crate discoverability - admittedly if readers are already aware about particular examples in crate doc, then our work is done
. But it’s likely that new developers will:
- not know about your excellent crate a all
- will be confused by a plethora of competing crates and would like to be informed on a community consensus for the “blessed” crate for the job
- provide a learning curve - starting from trivial ramping up to more complex cases in a uniform coding and writing style (not restricting crate authors to conform to any particular one)
- provide a problem solution approach to common programming problems. One that is not focused on showing all of the crate bells and whistles but rather the most idiomatic solution for given problem. Each example would be heavily linked to crate and std docs allowing readers to find a more sophisticated solutions.
Currently the cookbook examples are built and tested with skeptic on per commit basis with locked dependencies and nightly via travis cron job with wildcard dependencies to get buildbreak if crate api moves/ evolves. Granted it adds some maintenance burden for the cookbook devs.
I imagine that most other crate maintainers would prefer this as well, for the same reasons.
Well this is a matter of taste really. Some maintainers expressed that they want the added discoverability even if some examples are duplicated but I understand that this may not be always the case and your concerns are perfectly valid.
If I were to judge the ring examples, some of these might be a little beyond the cookbook scope. I would venture that typical cookbook reader might come searching for way to “hash and verify file contents” or “sign and verify a message” and someone searching specifically for “Ed25519 signing” would already get a ring docs as google result. But I might be wrong here, that is why we always solicit crate ideas prior to implementation.
I’m not saying that cookbook solves any of the mentioned problems particularly well (or at all) ATM. It currently suffers from lack of a good story behind it. But this is only a first iteration on its form and design (which will likely change heavily). We would love to hear your thoughts on the cookbook.
Sorry for making this so long 
