Here's a nice quick review of soft vs hard lock of
Post# of 72440
http://blog.clinfosource.com/guidelines-hard-...-database/
It should be noted, that a soft lock is simply limiting the number of people who are able to edit information, and does not mean that data can't be revised if needed. Even a hard lock is not necessarily set in stone.
As far a journal publication delays, I came across this paper discussing that exact issue:
http://openaccesspublishing.org/oa11/article.pdf
As far as time lines, I've seen data showing publication as quickly as a couple of weeks, up to several months, with several steps in between submission and final pub. It seems to depend a lot on what the data covers, and how much there is. In our case, the K trial lasted for freaking ever, and I imagine the sheer amount of collected data is daunting, meaning many revisions.