You've got a good point about negative results, but I don't think I agree with the rest.
There's nothing wrong with peer review as such, but the current research climate doesn't help it at all. In many countries, research grants are tied to "measurable, objective results", e.g. articles published, preferably in highly-ranked journals. And so researchers want to publish as much as possible, in as highly-ranked journals as they can get into. (Leading to an explosion in research, so no one really has the time to follow all the research in their own field, or even doing thorough peer review.) Journals are ranked among other things from how often they are cited. Negative results aren't often cited. Replicated tests are only cited in systematic reviews. Setting aside money for replication would be a good idea, but journals shouldn't need to fear for their ranking for publishing less glamorous articles either (or rather: ranking shouldn't be taken seriously). Most importantly, publishing shouldn't be so strongly encouraged. Far too much is published already, and much of it just isn't very good.
As for online publishing: that has been the norm the last decade, and is absolutely dominant now. Comment areas? Like Slashdot? God forbid.