Making Wikipedia stable

PROBLEMS:

When a person loads an article, they get whatever good and/or bad is in it at that second
Wikipedia is perceived to be not credible due to vandalism/spam/poor edits
Vandalism and spam is encouraged since it shows up immediately
- instant gratification, no approvals process to be thwarted by
Edit histories can be helpful, but can get loaded with edits that are reverted (edit+revert=2 useless edits)
Good articles can degrade over time due to poor edits
- note Wikipedia:Featured article review process, though it is partly due to standards being raised and articles not meeting the increased standards (I think I read somewhere that someone did a study)
All details of articles should be vetted

SOLUTIONS:

Have protected "Final" or "Stable" versions of articles
- new tab at the top; current "article" tab is renamed "Draft" or "Current"
- these versions are loaded by default
- could possibly use/convert the Featured Article process to come up with them
  - need to deal with edits made during the process to come up with really-final version
    (what if graders don't approve edits made after they've checked it out?)

Have a system where edits need to be accepted
- all edits submitted could be retained so that future editors can see if any possibly-worthwhile edits have been turned down
- should be able to go through only edits that actually make it into article (no vandalism/spam/etc and reverts)
- edits could still be allowed into articles if they haven't been approved after a certain time
- who approves edits? (don't want people policing themselves via sock puppets/friends)
  - use digg(?) system where edits are rated?

UPDATE: After speaking with Ward Cunningham, I think the first one would be the better option. It does not requiring creating the bureaucracy to do the other option, which Ward was worried about.

2016: A partial step to implementing this that should be politically and technically feasible to do would be allowing others to mark individual edits as questionable or bad and why (vandalism/spam/false info/etc). Make reverting be an option in this. It would separate edit histories into wheat and chaff and allow editors/reviewers to focus on whether other edits are good and correct.

Also, having stable versions would improve the editing process in that just the edits since the last stable version can be focused on and the current version of the article can be compared to a version that was agreed upon to be good. It provides structure in that editing can "jump" from stable version to stable version with each being a break point/separator instead of having one giant, continuous less-manageable and less-review-able heap.

A goal for "good" (i.e. reviewed, stable) articles should be to have every edit checked, and the stable version process would provide an opportunity and encouragement to do that, as well as split up article histories into smaller chunks.

An argument was made at Wikipedia talk:Stable versions proposal/Archive 1#Conditional Support that "if all the apples of the world were to spontaneously turn purple, we can be the first to say so". In my view, it is best to err on the side of caution, and to be stable and verified rather than cutting-edge up to date. It is fine to not say apples have turned purple if they hadn't at the time the article was last approved (written). It is better to let something not be included than it is to allow a vandal to add it when it isn't actually true.

Also add structure to talk pages by making them editable only by section (remove full page editing) and adding formal archiving (section collapsing) feature.