Duncan G M Stuart

@JohnHMcWhorter @lexiconvalley …also Leveret is a great word

The latest @lexiconvalley is all about rabbits, and well, you just have to be there. Hares and bunnies, too. Really you'll get some linguistics lessons without knowing it going in. open.substack.com/pub/lexiconval…

@JohnHMcWhorter @lexiconvalley Really surprised at this idea that hares is some kind of academic term. As a Brit the word feels very normal. I live in Gothenburg, Sweden where there are a Lot of hares anywhere there’s green space - they feel like a quite different beast: instantly recognisable and distinct.

@honeybadgerapp Where's the status page for Honeybadger.io - the service seems to be down right now?

🚨Awesome new feature alert!🚨 We just added incident management for our status pages! Check it out. honeybadger.io/blog/incident-…

@centristjim1 @OwenJones84 No, my friends in Kyiv (20s/30s professionals) are saying it, so it’s not purely a western media invention.

@OwenJones84 Sounds like western media spin to me. Very hard to tell if anything is true right now

Is there credible information that the Russian invasion is failing amidst lack of supplies and tough Ukrainian resistance?

@CuriousScutter Ah that was one of my favourite songs as a romantic teenager ☺️

@bonsaistudio @nvictorme @housecor I feel like people are somehow scared of the concept of rebasing, maybe because it's an unfamiliar word, maybe because it's not something they're initially exposed to. Which is a shame, because to me it feels conceptually MUCH simpler than a merge.

@nvictorme @housecor This, definitely this. Nobody is really listening to me and even if I show them how to safely pull —rebase the branch, are super scary to screw up things…!! And I’d also add interactive rebase when adding fixes from PR reviews instead of tens of meaningless commits. #git #rebase

What important software development truth do very few people agree with you on?

@TkDodo @housecor To be clear I'm not trying to say "this is the only way to do it and all other ways are plain wrong", more like "I tangibly experience the value of doing it this way, on a pretty regular basis" and that's why I think yes, it is worth it.

@DGMStuart @housecor Yeah they are often not meaningful, I agree. But does it matter so much that they are there? And is it worth the effort to squash them? And rarely there is a time when having a record of things that didn't work is actually helpful to not repeat mistakes, like the missed case in x

@TkDodo @housecor iii. I completely disagree that the record of the mistake is useful: in my experience leaving it unsquashed more often leads to the reader _repeating_ the mistake, because we're usually finding commits via blaming lines, not reading the history sequentially.

@TkDodo @housecor ii. I feel like it's really not very much effort: much less effort than writing a good commit message

@TkDodo @housecor i. For me it renders the commit messages kind of useless, since they're misleading: the message for 1 says it's the change x, but actually change x is the combination of 1, part of 2 and 3. It's work for a future reader to reassemble that context.

@joelanman @CuriousScutter Spreadsheets are after all Turing Complete

@TkDodo @housecor I think we're talking about two different things: I'm definitely not a fan of squashing *features* down to a single commit. I'm talking about when the commit history looks like: 1. change to x 2. case missed in x, plus a change to y 3. fix typo in x None of these are meaningful

@DGMStuart @housecor I ❤️ using commit messages as documentation on the _why_, but I think that squashing removes those comments. Like, I have 5 file changes and 10 commits, each on 2-3 LOC with lots of context + description in the commit message, then we squash the PR into: "implemented feature X".

@MoAlhaddar @TkDodo @housecor Interesting: what useful information do you get from the fact that a merge happened?

@TkDodo @housecor --no-ff all the way too! I'm so angry that this is not the default merge strategy on local merges

@TkDodo @housecor For me commit messages can be incredibly useful documentation of why code is the way it is, and squashing commits into meaningful changes is necessary to make that happen. When a change is split over multiple commits, we’re asking the future reader to do work to understand it.

@housecor That you shouldn't squash / force-push away commits. "clean history" is pretty overrated.

@avdi Do you have an opinion on The Prime Directive of Retrospectives? funretrospectives.com/prime-directiv… It feels like it calls for "assume positive intent" in a specific bounded context, trying to avoid the opposite (assigning blame, which usually doesn't end well).

How TF is "assume positive intent" in a corporate context not an obviously toxic value to everyone. Some of y'all have never been on the receiving end of abuse enablement or gaslightingy religious leadership or codependent family AND IT SHOWS

Finally removed Octopress dependencies from my Jekyll blog (Octopress has been unmaintained for many years now) and wrote some thoughts about the arc and vision of that project: dgmstuart.github.io/blog/2022/02/1…

Still not getting used to hearing the @_bikeshed podcast talking about I gem I'm a maintainer on (Pundit). I mean I didn't develop any of the features they're complimenting, but still.

I made a bingo app in @reactjs (for an actual game I run), and added #Wordle-style shareable emoji grids (just for the lols) dgmstuart.github.io/blog/2022/02/1…

@dxw dxw.com seems to sensibly show the site description instead. Also nice to see that the ?author=1 query still refers to almighty zule

In need of expert WP advice, I turn to my old accomplices at @dxw: a site for the nonprofit dance association I’m part of wcj.se, shows the main site author in the share block when shared to discord, instead of the site description. Discord is calling ?author=1 ?