However, in this specific instance, I think being strict on the scientist is okay.
The scientist made a mutation that was supposed to make the babies immune to HIV. But it turns out that the alleged CCR5 mutation that he induced a) didn't make the babies immune to all HIV strains only most of them (there are CXCR4 based HIV strains) [1], and b) made them more susceptible to other diseases like west nile virus infection [2].
He didn't just do that, he also circumented the approval infrastructure of science that is supposed to watch out for unethical experiments. If such behaviour is rewarded, people will start to do ethically questionable experiments. Biology is still in the shadow of past ethical violations like e.g. the HeLa cell samples that were being used throughout science without consent from the donor. Or think of the contergan scandal where the manufacturer company advertised that pregnant women should take contergan. Later it turned out that it had an absolutely harmful effect on fetuses, crippling them for life. The worst thing about this: the company knew about the effects but chose to hide the facts from the public.
I don't say that we shouldn't do germline edits at all. We should just be really careful at what we are doing, especially at the start. After all we are forcing (we don't ask the baby) a human being to live with that genetic change for an entire lifetime and their offspring.
Instead, we could maybe have started with one of those really ugly hereditary diseases that end in certain early death due to an unpopular mutation of a certain gene... Changing that gene to the popular variant would be objectively helpful and meaningful.
[1]: https://www.nature.com/articles/d41586-018-07545-0 [2]: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2118086/