Cloudseal's team is moving on to new adventures. In this final blog post from our current website, we look back on what we've done: https://t.co/UZqDhYmD21
My recipe for effective software debugging is to follow the scientific method:
Step 1: Ask a specific question
In other words, don't tweak random settings or collect random data/logs. Be purposeful in every experiment. Justify every action [1/5]
This 1 min video shows how a dev will use Cloudseal to reproduce errors and go straight to debugging. You can run in Cloudseal anywhere you run Linux programs.
https://t.co/4uICYkv8Cf
This new blog is written from a functional programming perspective -- connecting FP (in the small) and determinism (in the large): from pure functions, to pure programs.
https://t.co/uUdNWv3bHT
This article about flaky tests and the comments are good. But people need to understand: if we invested (a lot) more in hardware/platform support for recording execution, we could make *every* flaky test easy to debug, changing the game completely. https://t.co/nRipWFdlEj
@samth@rocallahan IMHO, the ideas in this thread regarding controlling and randomizing different environmental factors still make sense when trying to expose flakes (with or without record/replay). We just want to do that in one fell swoop, rather than piecemeal.
https://t.co/c2yIAROOt1...
That is, traditional efforts like Google Blaze/Bazel, or the fabulous https://t.co/z3OHWZGj3Y, require fixing thousands of packages individually. But a reproducible container can enforce the property in one fell swoop.
This blog post explains how build reproducibility doesn't have to be whack-a-mole, and doesn't need to require patching tool chains.
https://t.co/C1iogIRKGT
After a successful first year in operation, gBETA Indy is continuing its accelerator program with five new companies in its Spring 2019 cohort: @cafebabytogo@CloudsealTeam@karyosoft, Plan Forward & @qualifi_hr --> https://t.co/o3njbUELgE
> We are looking for a Haskell software engineer (...). Our core technology is a new containerization mechanism for Linux/x86 programs that enables reproducible builds and functional tests that never flake. - @rrnewton
https://t.co/JnM1rhGqOX