Comment by fictioncircle | Hacker News Reader

Comment by fictioncircle | Hacker News Reader

Parent Full thread

Frankly, the problem with this Op Ed is there is a very simple solution available to anyone who is concerned about a "doomsday" scenario for Wikipedia:

https://en.wikipedia.org/wiki/Wikipedia:Database_download#Wh...

Mirror it and make it available read-only until that day comes as a separate "disaster recovery" organization and raise the minimal funding required for that function (1 employee + 1 dedicated server) would be sufficient to the task. More if you wanted to make it usable but at that point, you aren't really a dumb mirror.

> pages-articles.xml.bz2 – Current revisions only, no talk or user pages; this is probably what you want, and is approximately 13 GB compressed (expands to over 58 GB when uncompressed).

You just need to mirror this regularly and maintain a reasonable depth of the backup. (Say, once a month for the past 12 months)

Then, whenever this terrible event destroys Wikipedia you are on the clear due to you operating under the past licenses for the data and people will need a new place to go if Wikipedia is genuinely destroyed.

I understand how this is valuable in offsetting load immediately but I think the graver threat here is the disbanding of the contributor community and resultant loss of network-effect value.

I imagine the cost of the disorganization or loss of volunteer contribution effort in the splintering of WP spinoffs/mirrors is orders of magnitude larger than load offset from mirroring.

> I imagine the cost of the disorganization or loss of volunteer contribution effort in the splintering of WP spinoffs/mirrors is orders of magnitude larger than load offset from mirroring.

I'm sure it would be but at the end of the day I'm not sure that would be, ultimately, bad if a monolithic entity failed because of mismanagement.

Nothing stops them from sharing information/pages or linking to each other.

I understand what you mean. I also understand that we're reasoning in the abstract here but I don't think it's fair to apply traditional economic theory to a social mission (accessible information?) and I'm not convinced that failure here is a favorable outcome for it.

You are right, though, that I might be overestimating the cost of coordination post-dissolution. It's interesting to think about how that might be done effectively.

I've seen that kind of diaspora before in internet forums and ran some for awhile based on that principle.

It does work more or less in that case. I'm not 1000% sure it'll work with Wikipedia but in the case I was personally involved in it did.

No one outside of WMF is downloading and storing the dumps they provide to the public?

In the past, I have used these for parsing and database setup practice, making CSV and local databases that require no network access. I am mainly interested in the reference urls found in wikipedia pages.

> No one outside of WMF is downloading and storing the dumps they provide to the public?

No one who is actively marketing and encouraging contributions on a scale people are aware of them.

If they existed, you'd have told me who they were. ;)

Simply storing the data isn't enough, you'd need to maintain it and make it publicly available as a mirror of the site and encouraging people to interact so you have a community to work with in the event of Wikipedia's failure.

I guess my point is the OP should be doing that instead of bitching.

Reply on news.ycombinator.com