Начало основного контента
Test Run Latency
Testing tools seem to focus on throughput: how many tests can we run per unit time? Kent suggest an alternative: mean time to first failure. How long does a programmer have to wait until seeing a test fail (in the case that one fails)?
Kent will demonstrate what is possible with low-latency test runs with TCR, Test && Commit || Revert, an alternative programming workflow to TDD. He will discuss the wider role feedback latency plays in software development. The talk will close with conversation and audience questions.