rfcs
rfcs copied to clipboard
bytecodealliance
Number each RFC file by the PR number it was introduced in
In principle this seems reasonable, but I'll note that this will break a number of links across the Internet -- scattered throughout GitHub comments at least, and e.g. some of my own links in recent blog posts. We should probably have a defined standard for "stable link to an RFC": do we link to the PR itself (and then keep an up-to-date link in the PR's first comment to the final markdown path)? Or something else?
My thought was that post-migration, every PR would include an update that puts the PR number in the file name. From then on the RFC name/path shouldn't need to ever change and a link to the PR or the file can be used to identify it permanently.
As for handling existing links to RFC files that would no longer be correct, the only way to avoid breaking them that I can think of is to make the accepted folder essentially an archive keeping all of the files therein but replacing the contents of each with a link to the actual RFC in some new folder where all of the files are numbered. The issue being that this would make the structure of the RFCs repo a awkward/confusing.
Is there any way to estimate the amount of these links that exist in the wild?
Is there any way to estimate the amount of these links that exist in the wild?
An alternative to break old links would be to rename the files and then to add new files with the old name which only include a link to the new place.
