[Foundation-l] what do we do in the event the Foundation fails? - Re: Policy governance ends

[David Gerard]

On 18/04/07, Platonides <Platonides at gmail.com> wrote:

> > See also http://davidgerard.co.uk/notes/2007/04/10/disaster-recovery-planning/
> > - what do we do in the event the Foundation fails?
> > (answer: anyone who wants to wipe Wikipedia from the face of the earth
> > only has until the next good en:wp history dump.)

> en.wikipedia would survive. I'm more afraid of the little wikis. They
> have full
> dumps, right, but how many copies of them are out there?
> And not so many people will download the full dump and share it if
> something breaks here. Jeff Merkey, and some statistics studies would,
> but ihaveanemptydomainandwanttoleechcontent.com won't.


If the need for archived dumps is popularised - and made easy - it
might be more plausible.

Outboard 400GB USB disk: £70. Plug that in, BitTorrent dumps to it,
have something which updates the dumps when there's a new one ...
there should be enough people who can afford that to do it.

At the moment I'm quite worried about Commons. The image dump is HUGE
and I don't think there's been a good dump of it for a while. And
300-400GB of images is quite a lot for people to keep around casually.


> I guess that if a hurricane broke the tampa cluster now we could
> get the content back from the feed subscriptions :/
>
> -Who are you?
> As you know the user table is secret because it's what is used to
> identify users. A centralized system is the easier.
> With OpenID you would need to store the user url. And on that url there
> would be a user table stored (maybe even by WMF, too). Not an
> improvement for a general breakage.


Eh?


> I was going to suggest we could publish some user db data with yet
> another hashing layer, but we can't. We sure have too many users with a
> one-letter password.
> But don't worry. Users table are so small that Brion can save them under
> his pillow ;-)


Let's assume Brion is 100% trustworthy with personal information (I
think we can actually assume this is true). The WMF is eaten by
invading Martian badgers.

* How much work to verify any person is who they say they are?
* Who does he give the results to?

In the second case, a public list of "this old-userID is this OpenID"
would be something that wouldn't violate a confidence - if each person
made the match by logging in and submitting the OpenID they wanted to
correlate and publicise for credit.


> About the forks
> Wikimedia projects have something more than the database. They have a
> great userbase.
> As long as WMF accomplishes what users want, WMF will have the supremacy
> over db copiers. Many people copy our articles, but so long, only
> citizendium has been a threat, and that because it takes a different
> approach (and had a lot of publicity).


Citizendium is not a threat - it's validation of our model and of open
content in general.

Really. Can you imagine anyone starting a new encyclopedia on the
Britannica model?


> If WMF defrauds its userbase, there will be forks anyway, as there were
> on the past, like the [[Enciclopedia Libre]], which is still being
> developed in paralell.
> If data is destroyed by a natural disaster, WMF will need to hold firm
> to avoid that disperssion.


I think multiple forks would actually be the best way to preserve
things. One will accumulate more weight.

Wiki encyclopedias aren't going away. They're the only reasonable
model for the 21st century, in fact ... and we did that. We just have
to make sure our work is preserved in case of disaster.


- d.