You're absolutely right on journalists working with intelligence agencies on appropriate redactions on the like, I don't want to ignore that. Happened with Reality as well. Despite that, there is always risk. I'm not trying to downplay the legitimacy of all leaks ever, just stating that as a fact. These dumps are huge, and considering that the agencies in question already didn't want to redact them (and chose to just classify them instead) it's hard to believe that journalists would do that same job better than them. Journalists when they don't get meetings precisely on their terms go into hero mode and try to redact themselves only. Snowden is a great example of this, so I'm not sure why you bring it up.
Mistakes happen all the time, and that doesn't even get into poor redaction methods that can be reversed, ESPECIALLY since the groups being targeted are most likely able to put pieces together that journalists can't. The mysterious nine character name might be a total mystery to the interns at NYT, but if you're an insurgent with the context of the rest of the report and (most likely) a few good names to guess with it's far from real mystery.
To reiterate, the lives in question are likely still classified, and revealing even their deaths can come at significant cost. Telling terrorists that they have the right guy presents about zero benefit to anyone but them, so there's no reason it'd be public information.
Furthermore, lives are a pretty low bar. I don't think it should be controversial to say that the US government, for instance, should be a few steps ahead of violent extremists. Debating the specifics of how and what measures are appropriate is another very important and necessary conversation, but saying that redacted leaks are totally fine is like giving just the suits of your cards to your opponent. Nothing like this is harmless, that's just myopic.
The risk is low but indeed not zero. We can construct a scenario where an operative is capture but which identity is unknown, and five years later the person is still in captivity, and suddenly a leak is announced to be published in which agency get knowledge that the captured person name will be included, and the intelligence agency are for some reason still unable to rescue the person before the article later get published.
There is however a lot of conditions for that to happen, which is why the general claim that a leak could endanger lives should be seen as rare, unlikely, while possible event.
To make a guesstimate, journalist and government official risked more life by the additional traveling by plane and car in order to discuss and publically address the leaked documents of Chelsea Manning than the risk exposed by the leaked documents themselves. The agencies involved was likely competent enough to eliminate all higher risks well before the publication date.
> Furthermore, lives are a pretty low bar
I don't think anyone object to that. Leaks should be seen as having a high risk of disrupting operations and increasing resource costs. I would expect that pulling out operatives, protecting collaborators, replacing operatives, and operations that fails are all very costly. The trade between an informed citizens and costs is something which should be more often discussed in politics. Journalists can sometimes reduce the costs with careful work, but it not a clear cut and sometimes they will make a mistake and sometimes its the government that goes to far in hiding too much information from its citizens.
Mistakes happen all the time, and that doesn't even get into poor redaction methods that can be reversed, ESPECIALLY since the groups being targeted are most likely able to put pieces together that journalists can't. The mysterious nine character name might be a total mystery to the interns at NYT, but if you're an insurgent with the context of the rest of the report and (most likely) a few good names to guess with it's far from real mystery.
To reiterate, the lives in question are likely still classified, and revealing even their deaths can come at significant cost. Telling terrorists that they have the right guy presents about zero benefit to anyone but them, so there's no reason it'd be public information.
Furthermore, lives are a pretty low bar. I don't think it should be controversial to say that the US government, for instance, should be a few steps ahead of violent extremists. Debating the specifics of how and what measures are appropriate is another very important and necessary conversation, but saying that redacted leaks are totally fine is like giving just the suits of your cards to your opponent. Nothing like this is harmless, that's just myopic.