At 7:50 Justin Searls talks about how he uses integration test to only test the happy path and will test variations in tests that are closer to the variations.
At nine minutes Noel wrap and talks about the integration test in the middle that is not an acceptance test, but does stitch together the smaller units.he says that when your earlier on, it can feel duplicative to do that. Also you don't have a sense for how to mark the collaborating objects effectively.
At 1402 a relatively slow test for the amount of value it provides
At 1725 he mentions that you can't use rails for domain discovery
At 28:00 "discovery tests", which is Justin's term.
podcast-tech-done-right#speed-of-repl-vs-rails1 234:30 "Justin: If it's faster to use a REPL to answer the question of 'If I do X will I get Y?', people will use the REPL and then as soon as the REPL session is over, it goes away forever. But if it's equally fast to just codify that in a quick test or check like it is in Smalltalk or, even environments like Java where your IDE you just hit ctrl-r and it will run the tests under the line in a half second. It definitely reduces that cost." podcast-tech-done-right#speed-of-repl-vs-rails1 2
podcast-tech-done-right#small-pieces-but-maybe-not-tdd1At 47 Noel Rappin says that writing code in small pieces is more important than doing TDD, and points to cut back and Erich Gamma's essay where they are doctrinaire about writing code in small pieces, but not necessarily about writing the test before writing the code. podcast-tech-done-right#small-pieces-but-maybe-not-tdd1
At 48 Sam talks about how he doesn't follow prescriptive practices and changes his style throughout the day, but he understands why the prescriptive practices are prescriptive