I think of writing Internet Drafts as very akin to writing code. And that's
partially because it is: we edit a source format that is different from
what the finished product is.The fact that we often use a code review tool
like GitHub really cements that mindset too. But, since we're there, why
not take a page from software engineering best practices? A very common one
is to use small pull requests. (If you're not familiar with this, type
"small pull requests" in your favorite search engine and read one of the
many blog posts on the topic.) When I write code, I will often fix other
nits I find in the source file I'm modifying while working on a new feature
I'm building. I'll probably also refactor the code to make some function
accessible. Conceptually, this phase is similar to a single pass. But then,
once my new feature works, I don't just throw this over the fence to my
reviewer. I take some extra time to split this PR into smaller ones: first
an editorial one, then a refactor, then finally the new feature. That makes
it orders of magnitude easier for my reviewer, because there is an inherent
asymmetry here: I have the full context for everything I just did in my
brain, and they don't. I just saved them an order of magnitude more time
than what it cost me. It's trivial to review a small change that says
"moved function foo to file bar, no other changes" or "reordered sections 2
and 7", but if that change also contains editorial changes or a new
feature, now the reviewer needs to painstakingly read the whole thing.

To loop back to the topic at hand, all we're asking is to split the AUTH48
changes into a handful of small sub-changes. This doesn't fundamentally
change how editing is done, and would not drastically increase editing
time. Please consider running an experiment to measure this. I'm sure we'll
find it's not a huge lift.

Thanks,
David

On Wed, Apr 16, 2025 at 1:41 PM Ted Lemon <[email protected]> wrote:

> Er, hopefully that message I intended just for EKR that went to everyone
> instead didn't come across as rude. I'm butthurt because it was hard, not
> because anybody did anything wrong. :)
>
> _______________________________________________
> rfc-interest mailing list -- [email protected]
> To unsubscribe send an email to [email protected]
>
_______________________________________________
rfc-interest mailing list -- [email protected]
To unsubscribe send an email to [email protected]

Reply via email to