boxed, 9 months ago

@adamchainz 45 + 122 percent for the two options? Mastodons client seems rather confused about math.

adamchainz, 9 months ago

@boxed I'm seeing 27% / 73% now

boxed, 9 months ago

@adamchainz hmm, yea now it shows correct for me too, even after a hard restart. Weird bug.

markwalker, 9 months ago

@adamchainz Would certainly start a new branch & get ready for some copy/paste action.

adamchainz, 9 months ago

@markwalker What about a new worktree then…

markwalker, 9 months ago

@adamchainz well this is why I need your book, because I've never done that 😂

timb07, 9 months ago

@adamchainz Take off and nuke the entire site from orbit – it’s the only way to be sure.

adamchainz, 9 months ago

@timb07 Which Git command is that?

lilyf, 9 months ago

@adamchainz Usually fix the merge conflicts, but sometimes redo it.

adamchainz, 9 months ago

@lilyf “It depends”, the senior dev answer 😀

chris, 9 months ago

@adamchainz
Classic programmers answer: it depends :)

If there are 99 conflicts merging a commit with 100 total lines of code, you probably want to scrap it and start again. If there are 99 merge conflicts on 5,000 total lines, it is might be more appropriate to resolve the conflicts.

If the conflicts are all on your package lockfile, that's a lot of conflicts, but fixing them should be pretty easy.

Are the merge conflicts horse-sized, or duck-sized?

adamchainz, 9 months ago

@chris Very senior response, I like it.

And ha, I should do another horse/duck based poll

benjaoming, 9 months ago

@adamchainz The poll has 113 votes on ex-Birdsite vs 81 votes here.. I like how things are heading, and I admire you for the hard work of keeping up on 2 platforms 😍

(also: ex-Birdsite only got 1 comment, here there are 3)

adamchainz, 9 months ago

@benjaoming In the end, the vote numbers reversed. 212 cotes on Twitter, 145 here.

The lack of an “algorithm” means good posts drop out much faster on Mastodon. Much harder to "go viral" , which is a shame.

I still prefer to read Twitter, it tends to show higher value, more random content thanks to the “algorithm”.

benjaoming, 9 months ago

@adamchainz The stats didn't get much worse over time... 212 vs 145 is a 1:1.46 ratio and the previous ratio was 113 to 81 so 1:1.40.... that's not so much worse.

You have 953 followers in the Fediverse, so that's 0.15 votes per follower. On Birdsite, you have 9818 followers - so that's 0.021 votes per follower. So you have almost 7x as many votes per follower here.

I think the algorithm here is nicer judging from stats? ❤️

benjaoming, 9 months ago

@adamchainz in my experience, the merge conflicts always contain forgotten aspects, so I seek them out and indulge in the pain 😎

frenck, 9 months ago

@adamchainz I choose to fix all the merge conflicts. Is that masochistic? 😅

adamchainz, 9 months ago

@frenck Not necessarily! I think it may be best if the conflicts are all fairly predictable e.g. in import statements.

petervanderdoes, 9 months ago

@adamchainz
redo, especially if most of conflicts comes from code I haven't worked on

adamchainz, 9 months ago

@petervanderdoes Let’s say they’re all your doing 😈

petervanderdoes, 9 months ago

@adamchainz if it was all me there wouldn’t be merge conflicts. It’s a matter keeping your branches up to date. I’ve had a major rewrite branch for months, when it came to merging it into the regular dev branch, no merge conflicts. Keep merging while you develop in a regular basis. 😇

adamchainz, 9 months ago

@petervanderdoes Yes that’s definitely the way, though it could lead to fixing 99 conflicts incrementally.