Hi all;
I think that Jamie has started an important topic. I don't think that WMF is going to usurp Wikipedia and the sister projects now or in the future, but it is statistically possible. If we want to protect us, the human knowledge and our work of this hypothetical scenario, we need complete full dumps frequently. But this scenario is a malicious one, and I think that there are many more dangerous posibilities, and unfortunately, they are common.
For example, small or massive lost of data due to natural disasters, crackers attacks, stolen passwords, hardware and software bugs, sudden crazy sysops, and _human errors_. Is WMF ready for that?
Long time ago I searched info about that, but I only found these links[1][2]. Recently, I have been concerned about this again. Most of the Wiki[mp]edia projects are small, and their full backups are updated every week[3] and they can be stored everywhere, but the largest ones like English Wikipedia gets outdated soon[4] (now, it is +200 days old).
I don't know so much about the infrastructure and how WMF servers are allocated around the world, so, I want to ask a simple question:
In the case of a complete disaster in the "main" servers, will WMF be able to restore all the Wiki[mp]edia contain using backups?
We got a terrible fright when 3000 images were deleted accidentally in 2008[5] and I think that not all were recovered.
When people ask about images dump the most common reply is: "Are you going to store 7 TB (Commons)?" I can't store that at home of course, but, I'm sure that a few universities or entities around the world can, not only for backup purposes, for researching too (in full resolution or thumbs).
Also, I think that we need to start mirroring Wiki[mp]edia dumps to other servers around the globe, as the common GNU/Linux ISOs mirrors do. Also, Library of Congress said some time ago that they are going to save a copy of all the tweets sent to Twitter.[6] When are they going to save a copy of Wiki[mp]edia? I hope we have learnt a bit since Library of Alexandria was destroyed.
I don't want that an error moves us back to January 15, 2001.
Regards, emijrp
[1] http://wikitech.wikimedia.org/view/Disaster_Recovery [2] http://wikitech.wikimedia.org/view/Offsite_Backups [3] http://download.wikimedia.org/ [4] http://en.wikipedia.org/wiki/User:Emijrp/Wikipedia_Archive [5] http://lists.wikimedia.org/pipermail/wikitech-l/2008-September/039265.html [6] http://www.wired.com/epicenter/2010/04/loc-google-twitter/
2010/9/8 Jamie Morken jmorken@shaw.ca
Hi,
I was involved in an open source project that was usurped by one of the main developers for the sole reason of making money, and that project continues now to take advantage of the community to increase the profit of that developer. I never would have thought such a thing was possible until I saw that happen. If that developer wasn't acting greedy, there would now be open source hardware for radio transceivers of all types, but instead there is only open source software for radio of all types. I find it a shame, and when I was working on that project I could *feel* it being usurped! I unfortunately may be paranoid as I feel the same thing here with the wikimedia foundation usurping wikipedia. If you don't believe me, just consider that it is a very gradual process, like getting people used to not being able to download image dumps anymore, and ignoring ALL requests to restore this functionality. Also failing to provide full history backups of the flagship wiki. These two facts allow the wikimedia foundation to maintain the control of intellectual property that wasn't created by the people. If you want the wikimedia foundation to respect you as volunteers, you will have to DEMAND respect by making sure that they never usurp the project. I think the best way to do this is to make sure we can all download up to date full history with images wikipedia's so a fork at any time is possible. Sure it may be paranoid, but trust me it is worth it to be paranoid regarding a project as important as wikipedia. I have been in situations like this before, I wish I had acted before even if I was wrong! I wouldn't even be speaking now except for reading the heart-felt words of volunteers in this thread that are unhappy with how the wikimedia foundation is running. We need to organize to get wikimedia foundation to release images tarballs, they are only ignoring multiple requests to do so, so far.
cheers, Jamie
Wikitech-l mailing list Wikitech-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikitech-l
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
I couldn't have said it as well. I agree with the concerns of Jamie and their importance.
On 13/09/2010 22:14, emijrp wrote:
Hi all;
I think that Jamie has started an important topic. I don't think that WMF is going to usurp Wikipedia and the sister projects now or in the future, but it is statistically possible. If we want to protect us, the human knowledge and our work of this hypothetical scenario, we need complete full dumps frequently. But this scenario is a malicious one, and I think that there are many more dangerous posibilities, and unfortunately, they are common.
For example, small or massive lost of data due to natural disasters, crackers attacks, stolen passwords, hardware and software bugs, sudden crazy sysops, and _human errors_. Is WMF ready for that?
Long time ago I searched info about that, but I only found these links[1][2]. Recently, I have been concerned about this again. Most of the Wiki[mp]edia projects are small, and their full backups are updated every week[3] and they can be stored everywhere, but the largest ones like English Wikipedia gets outdated soon[4] (now, it is +200 days old).
I don't know so much about the infrastructure and how WMF servers are allocated around the world, so, I want to ask a simple question:
In the case of a complete disaster in the "main" servers, will WMF be able to restore all the Wiki[mp]edia contain using backups?
We got a terrible fright when 3000 images were deleted accidentally in 2008[5] and I think that not all were recovered.
When people ask about images dump the most common reply is: "Are you going to store 7 TB (Commons)?" I can't store that at home of course, but, I'm sure that a few universities or entities around the world can, not only for backup purposes, for researching too (in full resolution or thumbs).
Also, I think that we need to start mirroring Wiki[mp]edia dumps to other servers around the globe, as the common GNU/Linux ISOs mirrors do. Also, Library of Congress said some time ago that they are going to save a copy of all the tweets sent to Twitter.[6] When are they going to save a copy of Wiki[mp]edia? I hope we have learnt a bit since Library of Alexandria was destroyed.
I don't want that an error moves us back to January 15, 2001.
Regards, emijrp
[1] http://wikitech.wikimedia.org/view/Disaster_Recovery [2] http://wikitech.wikimedia.org/view/Offsite_Backups [3] http://download.wikimedia.org/ [4] http://en.wikipedia.org/wiki/User:Emijrp/Wikipedia_Archive [5] http://lists.wikimedia.org/pipermail/wikitech-l/2008-September/039265.html [6] http://www.wired.com/epicenter/2010/04/loc-google-twitter/
2010/9/8 Jamie Morken jmorken@shaw.ca
Hi,
I was involved in an open source project that was usurped by one of the main developers for the sole reason of making money, and that project continues now to take advantage of the community to increase the profit of that developer. I never would have thought such a thing was possible until I saw that happen. If that developer wasn't acting greedy, there would now be open source hardware for radio transceivers of all types, but instead there is only open source software for radio of all types. I find it a shame, and when I was working on that project I could *feel* it being usurped! I unfortunately may be paranoid as I feel the same thing here with the wikimedia foundation usurping wikipedia. If you don't believe me, just consider that it is a very gradual process, like getting people used to not being able to download image dumps anymore, and ignoring ALL requests to restore this functionality. Also failing to provide full history backups of the flagship wiki. These two facts allow the wikimedia foundation to maintain the control of intellectual property that wasn't created by the people. If you want the wikimedia foundation to respect you as volunteers, you will have to DEMAND respect by making sure that they never usurp the project. I think the best way to do this is to make sure we can all download up to date full history with images wikipedia's so a fork at any time is possible. Sure it may be paranoid, but trust me it is worth it to be paranoid regarding a project as important as wikipedia. I have been in situations like this before, I wish I had acted before even if I was wrong! I wouldn't even be speaking now except for reading the heart-felt words of volunteers in this thread that are unhappy with how the wikimedia foundation is running. We need to organize to get wikimedia foundation to release images tarballs, they are only ignoring multiple requests to do so, so far.
cheers, Jamie
Wikitech-l mailing list Wikitech-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikitech-l
foundation-l mailing list foundation-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
Also, I think that we need to start mirroring Wiki[mp]edia dumps to other servers around the globe, as the common GNU/Linux ISOs mirrors do. Also, Library of Congress said some time ago that they are going to save a copy of all the tweets sent to Twitter.[6] When are they going to save a copy of Wiki[mp]edia? I hope we have learnt a bit since Library of Alexandria was destroyed.
They've actually just reached out to us to discuss archiving all of the Wikimedia projects :)
Discussions are in their early stages but I'll happily update as I know more.
--tomasz
wikimedia-l@lists.wikimedia.org