Request for comments on Prow Documentation design doc

62 views
Skip to first unread message

Linus Arver

unread,
Jan 10, 2022, 8:14:18 PM1/10/22
to d...@kubernetes.io, kubernetes-...@googlegroups.com
Hello folks,

Please have a look at this PR and its associated design doc. TL;DR is that I want to revamp how Prow user-facing docs are done, but would like community feedback before I proceed. Thank you!

Linus (@listx on GitHub)


Linus Arver

unread,
Jan 27, 2022, 5:17:00 PM1/27/22
to d...@kubernetes.io, kubernetes-...@googlegroups.com
Friendly ping for folks to give any additional feedback for this effort.

(NOTE: In the last message, the "PR" link actually pointed to the Issue, not PR. Oops. The PR can be found here).

I am not seeing much resistance to the main idea of adding statically-generated, user-facing docs. (The design doc also addresses possible automation improvements around docs, but these are not as essential as actually *having* a central user-facing docs site in the first place.)

Please feel free to comment in-line in the design doc or in this thread (or on the PR) if you have any more feedback. Thanks!

Priyanka Saggu

unread,
Jan 28, 2022, 1:42:54 AM1/28/22
to lin...@google.com, d...@kubernetes.io, kubernetes-...@googlegroups.com
Good day, Linus.

Thank you so much for compiling the design document. I noticed some discussion about sig-docs' previous effort in the "freshness checker" section of the document.

I'm not sure if there has been any recent progress on the "freshness checker" bit in the sig-docs. However, I was able to locate the following historical document and issue if they add any additional information/context for assistance:
Thank you once more!

Priyanka Saggu

Linus Arver

unread,
Jan 28, 2022, 11:00:57 AM1/28/22
to Priyanka Saggu, d...@kubernetes.io, kubernetes-...@googlegroups.com
Hello Priyanka! Thank you so much for the excellent pointers! I will add this as prior art in my design doc. I will also try contacting the people involved in this previous effort to try to build some momentum around staleness-checking automation.
Reply all
Reply to author
Forward
0 new messages