We lost 54k GitHub stars • Hacker News Board

Re: We lost 54k GitHub stars

#201

Post by jonahx » Thu, Apr 14, 2022, 11:31 PM UTC

wpietri wrote: post #123

cmeacham98 wrote: post #111

Earlier quoted context omitted.

I specifically dislike the "Lessons" section, as it throws all the blame on github and doesn't mention the seemingly obvious advice: "make sure you're not on autopilot when taking potentially dangerous actions, on github or any website". Yes, GitHub probably should show the stars in the warning UI, and hopefully that will prevent some of these mistakes. But GitHub makes it pretty hard to make this mistake already - t…

I suggest you read "The Field Guide to Understanding 'Human Error'". You'd learn a lot. https://www.amazon.com/Field-Guide-Understanding-Human-Error... My view is that expecting humans to stop making mistakes is much less effective than fixing the systems that amplify those mistakes into large, irreversible impacts.

There is also an audiobook version, for some reason not linked from there:

https://www.amazon.com/Field-Guide-Understanding-Human-Error...

Re: We lost 54k GitHub stars

#202

Post by lexx » Thu, Apr 14, 2022, 11:32 PM UTC

ncmncm wrote: post #159

lexx wrote: post #129

On the plus side. 54k know and use this project not because they have it starred. So it will regain the stars rapidly, and it will again hit all the trending metrics, so new people will discover it also. No biggie

You wholly miss the point.

I just add to the conversation. There are great points discussed. No reason to repeat them. I just add

Re: We lost 54k GitHub stars

#203

Post by layer8 » Thu, Apr 14, 2022, 11:32 PM UTC

wpietri wrote: post #136

forty wrote: post #60

What can you purchase with GitHub stars? More seriously, why do they matter? Is it a prestige thing only or are there practical consequences to losing the stars?

Starring a repo is a way of tracking it. So now there are 55k people who were trying to keep an eye on the project who have silently lost that. It's also an important indicator of community size and authenticity, important factors in the decision to use the code.

> silently

I’d argue that GitHub should send them a notification that the repo they starred has been made private.

Re: We lost 54k GitHub stars

#204

Post by r3trohack3r » Thu, Apr 14, 2022, 11:33 PM UTC

capableweb wrote: post #53

ghoomketu wrote: post #17

That comparison spot the difference pic is really scary. I had to check it 2 times myself before I could spot that the last line is different. Now that you're on the HN's frontpage, hopefully somebody from Github upper management will see this and all will be good soon thanks to gold old HN (especially when they did the same thing themselves and restored it). Wish you the best.

> That comparison spot the difference pic is really scary. I had to check it 2 times myself before I could spot that the last line is different. That's not the only difference. The other difference is that you also have to type the full repository name, including the organization name. If they wanted to delete the correct repository, they would have to type httpie/.github, but instead they typed httpie/httpie. It's u…

> That's not the only difference. The other difference is that you also have to type the full repository name, including the organization name.

This is not true.

We had an internal tool at a previous job that required you to type the name of a service's cluster into a field before deleting it. Automated deployments sometimes got left behind - and users would go in periodically and clean them up.

If you watched folks, they would type the name of the cluster being deleted into that box UNTIL they had to do this many times in a row. As soon as it was a "batch" job, folks would go through the motion of:

1. click delete

2. double click service name

3. ctrl+c

4. click text input

5. ctrl+v

6. press "Confirm"

It was remarkable, after someone did this once, I don't think I ever observed them going back to typing in the cluster name. That algorithm of "copy and paste" carried over into every prompt they saw of that form moving forward.

I'm guilty of this too - I no longer type the prompted name for this form of confirmation dialogue, it's a copy-and-paste.

Re: We lost 54k GitHub stars

#205

Post by pm90 » Thu, Apr 14, 2022, 11:33 PM UTC

This kind of shit (accidentally doing something that in retrospect feels dumb and stupid but made sense if you were in a certain brain space) happens all the time and it’s nice the author wrote about it. I could easily see myself doing this.

It should be possible for GitHub to restore. If not now, then in the future.

Re: We lost 54k GitHub stars

#206

Post by HellsMaddy » Thu, Apr 14, 2022, 11:34 PM UTC

In America, we blame individuals when systems fail. We don't have to, we can actually design systems to be better. Design isn't just about making things look pretty. Good design reduces entropy.

https://www.youtube.com/watch?v=Ra_0DgnJ1uQ

https://en.wikipedia.org/wiki/Human_factors_and_ergonomics

Re: We lost 54k GitHub stars

#207

Post by mekster » Thu, Apr 14, 2022, 11:34 PM UTC

nemothekid wrote: post #143

sefrost wrote: post #3

It does strike me as unfair that GitHub themselves made the exact same mistake but were able to fix it with database backups.

When you think about it, it seems reasonable, I think about what would happen if it were me. If I was working at GitHub, and I nuked something, then the rest of my day would be un-nuking that thing. On the other hand if I was working on something totally unrelated, and then you my PM came to me and said "can you stop what you're doing and unfuck this thing" it would obviously depend on what I was working on.

It's your customer (repo owners) asking for support and throwing it away being busy is just lazy attitude. You had prior incidents and the bugs were left as is and apparently it will happen again.

Re: We lost 54k GitHub stars

#208

Post by ncmncm » Thu, Apr 14, 2022, 11:36 PM UTC

layer8 wrote: post #203

wpietri wrote: post #136

Earlier quoted context omitted.

Starring a repo is a way of tracking it. So now there are 55k people who were trying to keep an eye on the project who have silently lost that. It's also an important indicator of community size and authenticity, important factors in the decision to use the code.

> silently I’d argue that GitHub should send them a notification that the repo they starred has been made private.

And another that it was made public again. Or neither, and just not delete shit that there was no need to delete.

Re: We lost 54k GitHub stars

#209

Post by pm90 » Thu, Apr 14, 2022, 11:36 PM UTC

r3trohack3r wrote: post #204

capableweb wrote: post #53

Earlier quoted context omitted.

> That comparison spot the difference pic is really scary. I had to check it 2 times myself before I could spot that the last line is different. That's not the only difference. The other difference is that you also have to type the full repository name, including the organization name. If they wanted to delete the correct repository, they would have to type httpie/.github, but instead they typed httpie/httpie. It's u…

> That's not the only difference. The other difference is that you also have to type the full repository name, including the organization name. This is not true. We had an internal tool at a previous job that required you to type the name of a service's cluster into a field before deleting it. Automated deployments sometimes got left behind - and users would go in periodically and clean them up. If you watched folks,…

Lol, I do this whenever they ask me to re-enter my email/password for verification even though I know exactly the purpose this mechanism is trying to enforce.

Humans will always take the easiest route. We just have to accept that and come up with better solutions that aid humans or fit to our very weird nature.

Re: We lost 54k GitHub stars

#210

Post by actually_a_dog » Thu, Apr 14, 2022, 11:36 PM UTC

Meta: why is the word "how" omitted from the title? I've seen this a couple of times recently and I don't understand why people are omitting words like "how" and "why." The only thing I see in the guidelines is about omitting "gratuitous" numbers and adjectives, neither of which "how" and "why" are.