The reason you're going bankrupt is because you've convinced yourself the $5 coffee is the solution.
If you want a small change that makes a real difference take the bus one day a week.
2,662 karma · joined November 26, 2008
[ my public key: https://keybase.io/tylermenezes; my proof: https://keybase.io/tylermenezes/sigs/Zowanx-mSryas0BzGDIAbFfk4icTRxWreK_NVWdoTyo ]
The reason you're going bankrupt is because you've convinced yourself the $5 coffee is the solution.
If you want a small change that makes a real difference take the bus one day a week.
1: https://www.zdnet.com/article/former-mozilla-exec-google-has...
[1]: https://mentorsintech.com/ [2]: https://www.nytimes.com/2012/10/01/technology/microsoft-send...
I'm not, especially as the post then moves straight into dissolving the DEI committee and moving its responsibilities to HR.
That lawsuit was dismissed, in fact the article you linked says as much.
- Reliable 1-5 hours a month of downtime. There are months where it's literally 1 nine of uptime. Their status page was actually down for a while (I think they forgot to renew it?)
- Simple and obvious bugs in the official API clients have not been fixed for years (for example, Promises in the official JS library throw the string `[object Object]` for all exceptions.)
- The API has no way to query the structure of a table, you have to infer it from the first few records. (The auto-generated API documentation is actually wrong if you don't have a fully-filled-in row somewhere in the first few rows.)
- You can't modify a table configuration by API... except you can _sometimes_ by adding a parameter which is documented to try to convert dates, but actually modifies field definitions for a number of fields. Documentation for which fields is entirely in scattered forum posts from users.
- Booleans in the API are `true` or `undefined`. Seriously.
This is all /barely/ worth it because Google Docs doesn't work when you need a large structured database with many collaborators. But man is it bad software.
Things are always fine when you're starting, but if you don't understand the system it's hard to troubleshoot when it breaks. Understanding Kubernetes is hard. Obviously there are ways to outsource cluster management, but we don't have the budget of a VC funded startup.
We ended up choosing Hashicorp's products -- Nomad for orchestration, Consul for service discovery, and Vault for secret management. Each one is just a binary with a <20 line config file, and it takes ~two days to read and understand the docs well enough that we have been able to troubleshoot issues quickly.
Unfortunately it's not possible to productively resolve issues with the Signal team, something you can find documented again and again.
(My own experience: I had to justify the the user impact of 30+sec freezes on every sent message, confirmed by multiple people. Bug was closed wontfix.)
This is a known thing with Moxie and the culture he's created at Signal and it's unfortunate that he's still starting drama with everyone instead of doing any self-reflection.
But you can't verify that the open-source matches what's on the Play Store, you can't link to the official server with your own build, and you can't run your own because the server is no longer open source. (Last released a year ago)
If a different private company wants to host the neonazis, they can go for it. It turns out it's /difficult/ for them because most people find them repulsive.
I don't hear anyone making this argument for child abuse images for what it's worth.
That's been fixed now :)