Maybe not secret then, but to be released at a later date. Or released in one of those journals that most of cannot access.
About the later date: I think it's reasonable that those that put together the proposal and thought up the project have a chance at getting their results out first. Otherwise, why would you go through the effort of planning and executing observations if you could just poach someone else's data as it comes down the pipe? Think of it as a 1-year patent on the research project.
Some argue this isn't good enough (and I have some sympathy for those arguments). But it's very different to a deliberate action to hide or obscure information.
That said, it's very true that this is worlds different from maliciously withholding data.
Go down to your local university library. Go over to one of the computers (or hop on the wifi). Read the article. Email it to yourself. Not that hard now was it?
This is not ideal, but it's not that you _cannot_ access it, it's just not convenient.