[Wikipedia-l] Stable versions policy

Wed Dec 21 11:34:27 UTC 2005

On 12/21/05, Wikipedia Romania (Ronline) <rowikipedia at gmail.com> wrote:
> There is something that you don't seem to understand. By making the stable
> version default to the majority of readers, then Wikipedia's instant
> editability becomes compromised. I'll give you an example - say a stable
> version has been created for an article about the Romanian economy. Monthly,
> stats change, and I'd also like update some prose about the role of ICT in
> the Romanian economy. So, as an editor, I go to the editable version and
> change it, and - bang - as you said, it's instantly there.

There are some independent issues there have been mixed up, confusing
the discussion:

1) Using the review process to somehow decide that a certain revision
of an article is the good one. That's the stable revision.

2) What to present readers. The latest revision? The last stable
revision? Some combination of the two?

3) What to do if some flaw is found in the stable revision, and there
are 827 other revisions after that, so that it's not possible to just
fix a spelling error. A special editing of the stable revision would
effectively "fork" the article.

I trust that point 3) is not a big problem - if an article goes
downhill after a stable revision, it's more important to fix the new
revisions than fix minor errors in the stable one.

Discussions about point 2) should not obfuscate the point that a way
for selecting good revision is needed. What we'll precisely do with
them can be argued separately.

Alfio