Bullshit

Posted Oct 31, 2025 9:36 UTC (Fri) by nim-nim (subscriber, #34454)
Parent article: Security quote of the week

Even if there was no alternative to hyperscalers there is no good excuse for depending on a single hyperscaler.

That’s the usual tragedy of the commons, doing things right is harder, it’s not my fault if I chose the easy lock-in option, someone else should have made commons more attractive.

Bullshit

Posted Oct 31, 2025 12:49 UTC (Fri) by Wol (subscriber, #4433) [Link]

> That’s the usual tragedy of the commons, doing things right is harder,

The real (as in original, historic) tragedy of the commons was outsiders (or insiders) seeing the opportunity of a quick buck. Actually, not that different from here - "Take a setup that's working fine, persuade the big stakeholders that they will be better off doing it differently, leave all the real commoners a lot poorer".

Cheers,
Wol

Bullshit

Posted Oct 31, 2025 13:25 UTC (Fri) by jzb (editor, #7867) [Link]

If you read the thread linked, Whittaker said that it is "surprising to me that so many people were surprised to learn that Signal runs partly on AWS". (Emphasis mine.) The quote here did not imply that they are solely on AWS: she was reacting to people who were shocked that they were using AWS at all.

I'm not sure what the percentage is or how Signal is architected, but AIUI they use Azure and/or Google Cloud as well. For what it's worth, I use Signal and never noticed any issues during the AWS outage. I did see other folks reporting problems, though. But, AFAICT, they are not solely set up on AWS infra.