This Gristmill anti-offset post by Joseph Romm about geoengineering is more justified than their carpet-bombing of temperate forestry projects, but still a bit of an overstatement. It's completely appropriate to attack the stupid, fools-rushing-in approach of Planktos as it attempts to sell offsets for the unproven and dubious ocean-fertilizing project. But still, while no geoengineering ideas are currently ready for implementation, they shouldn't be completely ruled out.
Romm pulls back slightly, potentially allowing geoengineering in the future "(1) only after we have exhausted every plausible mitigation strategy, and (2) only after we have done rigorous, small-scale experiments to prove its safety and effectiveness." He's completely right about #2, but if it's safe and effective, why not use it? The ocean fertilizing approach has a lot of problems and may be untenable, and the same may hold true for the other geoengineering dreams, but part of showing seriousness about climate change means not stacking the deck against a particular approach.