It's somewhat better in FOSS than in science-land because there are generally at least a few people who find a new FOSS project genuinely interesting and worth supporting, whereas scientific conformity is much less forgiving.