EDIT: downvoting doesn't bother me, but absent comment I have no idea what part of it you disagreed with...

It seems to me that Greenwald is assuming his own conclusion here, that the chat logs provide evidence of Manning's innocence. They could, of course; and I imagine that Manning's Lawyer will subpoena them for just that reason: if he can not get them from Wired, then he can get them from Lamo or the FBI. (Lamo apparently says the FBI took his computer with the logs on them, but until he says so under oath that doesn't mean a thing.)

There are alternative possibilities, which Greenwald does not address, that could well justify Wired's withholding other parts of the chat logs and still maintaining a high standard of journalistic ethics.

First: the possibility that the logs contain more evidence of Manning's or Assange's guilt. Lamo makes statements to multiple journalistic sources such as the NYT and Washington Post as well as to Wired; so whether Poulsen reports his remarks first or not, the fact that such allegations are being made is a matter of public record. Poulsen neither confirms nor denies Lamo's allegations with reference to the chat logs. Greenwald argues that he should, as Lamo's accusations amount to prosecuting Manning and Assange in public. However, accusations without evidence are just talk. If Poulsen releases the transcripts, many will see the contents as conclusive evidence, whether they support or undermine Lamo's assertions, or even if they are ambiguous. That could make a fair trial for Assange impossible, if indeed he is charged. It could also amount to conducting Manning's (military) trial in public in a way that undermines his rights.

As yet, there's no proof that Bradley Manning was 'Bradass87' or that he was the one conducting those conversations (vs someone else using his login, say), or that the logs are a true and unedited copy of actual chats. But the logs that are public already are being treated as the indisputable truth. That they are public makes it very difficult for Manning's lawyer to attack their admissibility as evidence: if he does so, many people will dismiss his arguments as lies. Just the fact that the public logs have the 'feel' of a real conversation is enough for many people to decide they're authentic, for the same reason that people often remember movies better than they do real history. Even if it can be shown that the chat logs are authentic records of conversations between Manning and Lamo, they are only evidence that Manning believed and said certain things to Lamo. As far as the logs talk about Assange, their legal value is that of hearsay evidence. But opponents of Assange will treat such hearsay remarks as proven facts (they're not even close), while his supporters will say none of it is admissible at all because it's hearsay (not true either). Regardless of which side Poulsen or anyone else is on, making that material public risks undermining justice.

Personally, I think it would have been better not to release any transcripts at all in order to avoid biasing a trial (either of Manning or of Assange, if he is charged later). Mind, that's very much a personal opinion. I'm also against police releasing mugshots or making people do perp walks because so many assume that criminal defendants must be guilty because they look bad. But suppose, for argument's sake, that there's so much evidence of Manning being the leaker that the facts are not really in question, and that's why Poulsen felt ethically justified in releasing partial transcripts.

This leads me towards the other possibility which might have led Poulsen to release only partial transcripts - the question of Manning's motivations. In the published logs he sounds like someone offended by the government's low ethical standards, unhappy in his job situation, and alternatively bragging and remorseful about the magnitude of the revelations. So his defense might go ahead on the basis of 'yes, he broke his oath and leaked all these documents, but he was young and idealistic and believed himself to be acting on a higher purpose, thus he's really only guilty of exercising very poor judgment.' But (in theory) if the unpublished logs include him having a bad day and going on an extended rant to the effect that 'bin Laden was right' or 'the American empire deserves to collapse' or something? Such sentiments are not unknown, and considering that members of Congress have already been calling for the kid's execution - although none of the charges against him carry the death penalty - if evidence of a 'bad' motivation comes before the public there's likely to be lynch mobs forming outside the camp where he's being held. More likely, there's an entirely different set of possible motivations.

[Quoting OP article] When I first wrote back in June about Wired's concealment of these chat logs, the excuses Poulsen gave were quickly proved to be false.

That's very much a matter of opinion. The parts that were not in Wired's original release which later appeared in the WaPo (orange and red text at the link above) all seem to me to have a bearing on these subjects. 'Sensitive information' could refer to the removal of Assange's name/identifying detail in some places, but mostly Wired's redactions seem to concern such personal matters as Manning's emotions, mental stability, and life situation, which might interpreted as damning or mitigating by different people. Greenwald says several parts are neither, but doesn't say which parts or (more importantly) why he thinks so.

The personal information here appears to show Manning as a troubled kid who turned to the army in search of the stability that was missing in his life, but became deeply disillusioned with the institution and country he served. It appears that the army had become disillusioned with Manning too, and that he was being discharged for an 'adjustment disorder.' What that means is uncertain, but there's been a lot of speculation to the effect that Manning was gay or had gender identity issues. That's a pretty contentious subject as far as it relates to the military (see the depth of feeling in the recent debate about repealing DADT) and the extent to which it could have affected Manning personally, and the extent to which any lack of stability on Manning's part would be his fault or that of the army's, are questions which could have a significant bearing on his sentence. And because the issue is very contentious, it tends to split opinion further among those already holding polarized opinions: if you think Manning's a bad person, then his personal hangups may compound his moral failings or be an unacceptable sympathy play; if you think he's a good person, then his personal hangups may be an additional justification for leaking the secrets of an unfairly discriminatory, or an unacceptable attempt to smear him as a sexual deviant.

It seems to me that Poulsen decided to publish those parts of the logs which showed Manning's access to, awareness of, and methods for leaking classified information. In other words, the published excerpts offer answers to the question of where multiple different leaks of military and diplomatic information came from (one single intelligence analyst tasked with cross-referencing army intel with that from other branches of government); whether the leaking was a deliberate act, or an accidental result of bureaucratic incompetence (deliberate, with understanding of content, likely publicity, and negative impact); and whether the allegations were credible (obviously the government intranet does not include a 'leak to public?' option, and the weak point was the CD-RW supposedly filled with Lady Gaga tunes (am I the only person who finds this detail bizarrely ironic?)). All this information is concerned with factual issues. Courts-martial have to establish proof beyond a reasonable doubt as do civilian courts, but they differ in terms of things like jury requirements, which may make it easier to deal with the publication of factual evidence than would be the case for a civilian trial. But while it may be easy to prove the acts occurred, mental state, criminal responsibility, and would be an appropriate sentence are much trickier questions. Because such matters are inherently subjective, the standards for evaluating them are different from those for establishing material fact, and there are different risks of potential bias.

My guess is that Wired's legal department (rather than Poulsen as an individual) would have gone through the transcripts marking different sections as factual or subjective, with lesser or greater possibilities of bias and thus legal liability for publication.