Nathan Heffley @nathan@heffley.social

Interesting to see Twitter regularly have incidents during peak load hours (https://twitter.com/danluu/status/1518677214035726341), e.g., here's downdetector for the past two days.

When Twitter had 2 DCs, this used to be quite rare outside of failover tests, which effectively doubled load. Most failover tests would pass without incident, implying Twitter could handle > 2x normal load, so of course the site would rarely have non-failover load issues.

But now it's normal to have an incident on a random weekday.

🔗 Always use [closed, open) intervals
https://fhur.me/posts/always-use-closed-open-intervals
#best-practices

One of the things that I think is sad about the decimation of Twitter eng is that Twitter was doing a lot of interesting (and high ROI) eng work that, at younger companies, is mostly outsourced at great cost.

A few examples off the top of my head:

The now gutted HWENG group was so good at server design that, in a meeting with Intel, the Intel folks couldn't believe the power envelope Twitter achieved and Google thought we were lying about our costs during cloud price negotiations.

Hello web devs, @SaraSoueidan has started taking pre-orders for her #PraticalA11y course at https://practical-accessibility.today/. This course is going to be amazing!! Highly recommend. I have already pre-ordered myself.

Finding #pest much more intuitive than #phpunit. Almost enjoyable! JS on the other hand gave me grief today. Tomorrow is another day. 🤪

as sad as i am about the rapid destruction of twitter, the (re)surge in blogs and newsletters has been lovely

back to the #indieweb!