Post #1794775
2026-04-29 11:03 UTC
Replies (7)
-
@tastapod@mas.to 2026-04-30 08:33
oh, adorbs, he blocked me because 'you obviously don't understand how web services work'. Ok then.
-
@carlosenog@hachyderm.io 2026-04-29 11:09
@tastapod It's still a disaster... my guy additionally, you can check 2026, If you want, of course
-
@civitz@mastodon.uno 2026-04-29 11:13
@tastapod 99.5 is ~40 Hours downtime per year, or around 10-15 minutes down time per working day. I’m not saying you’re wrong, but the graph would not be legible without the zoom. I agree on the misdirection though, we need to double check 🙂
-
@fasterthanlime@hachyderm.io 2026-04-29 11:14
@tastapod Yeah, this is a weird call out. As someone else pointed out, uptime has genuinely dipped into the 90s. I don't think this is a great example of lying with graphs. I also don't think GitHub needs anyone defending them right now. And I also think that your post makes it look like the author of Ghostty is misleading people (When actually, it's just some rando who shared their article). Anyway, I agree with the sentiment, I disagree with the choice of target.
-
@kerryb@mastodon.social 2026-04-29 11:17
@tastapod And immediately makes me wonder what happened in March 2016.
-
@adredish@neuromatch.social 2026-04-29 11:45
@tastapod Scale depends on what the graph supposed to show. If you want to ask "has something changed after MS acquired github?", then this is a *correct* scale. It shows a clear and significant change in uptime (such as MS's inability to maintain github to the industry term "5 9's of reliability"). On the other hand, if the question is "is github up more often than it is down?", then you should show the graph with a scale going to 0. Yes, one should always check the y axes on the graph. And one should always make the decision on the y axes based on the question at hand.
-
@campuscodi@mastodon.social 2026-04-29 22:37
@tastapod I'm just gonna block you cause you're obviously clueless about how web services work