Online classes really emphasize how irrelevant closed book/notes/neighbor tests have always been. In cute terms, this environment is the strongest possible cou

Online classes really emphasize how irrelevant closed book/notes/neighbor tests have always been. In cute terms, this environment is the strongest possible counterexample to "you won't always have a calculator with you!". All students are literally doing the test on a computer with internet access in the present tense. Their connectivity is considered so robust that a student may lose points or time if their internet drops during the test, and yet open internet access (or even just textbook and notes!) is somehow not an obvious resource to allow students during tests. Professors frequently copy tests from other sources or previous semesters, and this obviously saves an enormous amount of effort, but ironically their copied work is what makes cheating on their tests so much more accessible. Professors make loud noises about cheating, and some are so hyperfixated on preventing cheating that they demand ridiculous requirements like proctoring software. They clearly aren't really prioritizing cheating prevention in their efforts, otherwise they'd write new tests more frequently (or at all!). Cheating prevention measures that don't require greater instructor effort are far more commonplace. When a professor literally copies a test, they shouldn't be surprised to see their students doing the same. Tests that are suddenly ineffective with open internet access are already complete garbage for 3 reasons: 1. Some students will always cheat, and as long as teaching is remote, some students will always have internet access. Open access means that honesty is no longer discouraged. 2. Verifying granular details/vocab on the internet is completely normal in real work, even among experts. 3. Any test without resources like notes or internet relies far too heavily on memorization. This might be justifiable if there were evidence that standard test studying strategies (cramming) were effective at guaranteeing long term memorization. Anecdotally, I would fail the majority of my finals from previous semesters without time to (re)prepare. Students should be encouraged to understand the inner workings of the topic, in terms of analysis and potential states of that system, not to memorize dozens of details and vocabulary terms. Within my Computer Science curriculum, the most effective and important content (between tests, projects, and lectures) has been programming projects by far. I'm dubious that tests are effective for any curriculum with relevant real world projects, though CS may be better than other subjects at making projects the primary measurement of understanding, and replacing the rigor that testing creates.