Needless editorializing

This commentary is hyperbolic and isn't supported by the actual timeline when you open up the thread and look at the time stamps.

It begins here, <https://lwn.net/ml/linux-kernel/20220629150102.1582425-2-...>, Wed, 29 Jun 2022 17:01:02 +0200. I immediately asked, "will this break my stuff? If not, can you mention why in the commit message?" Christoph didn't like this, and we went back in forth to the tune of "please don't intentionally break my stuff, or if you're not, say why not; I haven't researched it yet myself", "no, your situation is BS, I'll break what I want", "that's not helpful", "why should I have to explain myself", and so on. My last message to Christoph was at Wed, 29 Jun 2022 19:42:28 +0200, at which point he ducked out. That span comprised basically the entire thing related to Corbet's statement, which you can read again:

> The way this solution came about could have been better, though. The kernel community works best when developers work together toward a common goal rather than argue over who is doing things incorrectly. That did eventually happen here, but it took some time to get to that point.

Meanwhile, my entreaties not to intentionally break my stuff and some mentions of various Android source files garnered the attention of Android developers, and Kalesh jumped in at Wed, 29 Jun 2022 12:05:23 -0700. We had a discussion on whether this should be a runtime knob or a compile time switch. I wrote inline sample patches for each method, just so there'd be something concrete to work with and discuss, and mentioned whichever one they want is fine. There was some discussion on describing a device versus describing behavior. And then by the time I woke up in the morning, we had a viable patch (based on one of my earlier inline samples) for the kernel, and then for Android, and then Greg took the patches, and the whole thing was done.

So... I don't think Corbet's editorializing is even minutely correct:

> The way this solution came about could have been better, though. The kernel community works best when developers work together toward a common goal rather than argue over who is doing things incorrectly. That did eventually happen here, but it took some time to get to that point.

Alex raised the initial issue. Christoph wrote a patch for it. That patch broke my stuff, and Christoph didn't want to work on that. So he ducked out, and I took the baton and worked with the Android developers. And this all happened in the span of a day or two. This sounds a lot more like kernel developers working together toward a common goal and accomplishing something in a short span of time. Yes, there was some minor argument in there. But it didn't seem to affect anybody actually getting anything done in a swift and timely manner.

So while _maybe_ there's some minorly newsworthy content (one config knob is gone, now there's another, interesting forward secrecy considerations in the process), the dramatization and fingerwagging in this piece seems unnecessary.