Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
taeric
on Oct 15, 2014
|
parent
|
context
|
favorite
| on:
Curry-Howard, the Ontological Ultimate
The cognition overhead of 100% test coverage is probably absurd, as well.
gcanti
on Oct 15, 2014
|
next
[–]
Probably you are right, but for a library I think it should be an obligation to provide 100% coverage. For a generic program, maybe I sould provide 100% coverage at least of the "critical paths".
taeric
on Oct 15, 2014
|
parent
|
next
[–]
Might also be worth it to emphasize the "intended paths."
Of course, that is probably just saying that there should be documentation.
cantankerous
on Oct 15, 2014
|
prev
[–]
Especially given that you can't reasonably enforce the rigor of your test suite by way of your language.
taeric
on Oct 15, 2014
|
parent
[–]
Oddly, you could, in a way. Just treat your language and compiler as a tool in the toolchain. Add in a testing segment and you can set rules on it.
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search: