[throwaway0x7E6]

>Ads UI is very buggy and login with SSO and 2FA broken

if all of those have somehow deteriorated in 1-2 weeks, this is further justification for axing the people who made them.

[mjmsmith]

If Twitter goes down, it just shows that all those people were incompetent, and he should have fired them.

If Twitter stays up, it just shows that all those people were unnecessary, and he should have fired them.

[adgjlsfhk1]

2fa broke because Elon ordered all the micro services to be shut down

[Robotbeat]

I've heard this claim repeated, but I've been unable to find a link to causal evidence. Do you happen to have a citation?

[reverius42]

I've had a support ticket open since Saturday for 2FA. (Didn't try sooner than that, but the timing of when it stopped working suggests it's possibly related to Elon's tweet.) No response on my support ticket and it still doesn't work.

[Robotbeat]

Thanks. I’ve seen folks report problems with 2FA (which since resolved). Correlative evidence is indeed evidence, but yet it is still not as strong as causal evidence.

[shagie]

If I've got something that needs an hour of care every week by a sales person, https://xkcd.com/1205/ puts it at about 10 days (two weeks) of work to make it more efficient.

However, if shaving that 1h a week off takes more than 10 days of work, it is probably better to leave it as it is.

https://www.fincher.org/tips/General/SoftwareDevelopment/Bug...

This bug tracking works on Pain / Effort / Frequency and Risk / Effort / Verifiability

Realizing that I'm making up numbers here (and you can make them up too for the story we're telling each other):

    Pain   : 2 (low pain)
    Effort : 2 (low effort)
    Freq.  : 4 (fairly often)

We've got a PEF of 8 there. Note that this is the pain for the hypothetical sales person - not the pain for the end advertiser who the sales person works with.

So, what's the risk of making the change to make it so that the sales person doesn't have to do that 1h/week task? How much effort does it take? How easy is it to verify that its actually working?

And so now we've got bugs where we can put them on a grid and say "these fixes are the low hanging fruit" while "these other ones are risky and don't get us much."

Its quite possible that that this issue would never get into the "this is painful and so we need to do the analysis for it" or "this is low hanging fruit that can be fixed and have a big impact on 'stuff'".

Nonetheless, you've got something that takes 1h/week from someone who knows what they're doing with their accounts... and they're not working there anymore. Even in a perfect org, this could fall apart rather quickly.