Hello,
Proberly some of you will flame me, but I will take that for granted.
My idea is to combine all the European language wikipedia's (except english and maybe spanish and portuguese) on one server and using one database, located somewhere in Europe.
Here is why:
* We will not be depending on the US anymore.
- As this can be considered a political issue, it is mostly a technical issue. Because when there are problems, we (Europeans) can solve this on hours, people in the US are still at sleep. This also includes fixing things on location.
- Also the performance will not be influenced by the English-language wikipedia.
Problems:
* Finding a reliable server that is free and only for us. * Finding European sysadmins for this server.
Which wikipedias:
# Basque (Euskara) # Bosnian (Bosanski) # Catalan (Català) # Czech (?eský) # Danish (Dansk) # Dutch (Nederlands) # Esperanto # Estonian (Eesti) # Finnish (Suomi) # French (Français) # Frisian (Frysk) # German (Deutsch) # Hungarian (Magyar) # Interlingua # Irish (Gaeilge) # Italian (Italiano) # Latin (Latina) # Latvian (Latviešu) # Lithuanian (Lietuviškai) # Low Saxon (Nedersächsisch) # Norwegian (Norsk) # Polish (Polska) (# Portuguese (Português) <<< Maybe not, because most speakers are living outside Europe) # Romanian (Român?) # Romanica # Russian (???????) # Serbocroatian (Srpskohrvatska) # Slovenian (Slovensko) (# Spanish (Castellano) <<< Maybe not, because most speakers are living outside Europe) # Swedish (Svenska) # Welsh (Cymraeg)
Cheers,
Jeroenvrp nl.wikipedia.org
Jeroen wrote:
Hello,
Proberly some of you will flame me, but I will take that for granted.
Disagreement is not a flame if it is done respectfully.
My idea is to combine all the European language wikipedia's (except english and maybe spanish and portuguese) on one server and using one database, located somewhere in Europe.
Here is why:
- We will not be depending on the US anymore.
I'm as critical as any foreigner (relative to the US) about their policies, but that alone does not make the case for the move. Ideally, it should not make any difference where the server(s) is located if we are operating in a global environment As a Canadian I would be just as concerned about having the French Wikipedia dominated by European interests.
Of course, I still think that it would be worthwhile to have every Wikipedia automatically backed up in a second country. This sort of reflects a street smart self-preservation instinct. Given the centralizing tendencies in the European Community and it's love for bureaucracy, anything based in the EC would have to be backed up outside that community since any two countries in the EC are now different in name only.
- As this can be considered a political issue, it is mostly a technical
issue. Because when there are problems, we (Europeans) can solve this on hours, people in the US are still at sleep. This also includes fixing things on location.
This is a round the clock operation. Hopefully, problems requiring on-location attention will remain very few, but we can expect long waitsin those cases no matter when the problem arises.
- Also the performance will not be influenced by the English-language
wikipedia.
This is not an argument for moving out of the US; only for moving them to another server.
Eclecticology
If it's just a technical issue, and not a political issue, then I have no objections to doing this *under the Wikimedia Foundation*.
However, when you start your comments with this:
- We will not be depending on the US anymore.
I immediately suspect that it's political.
- As this can be considered a political issue, it is mostly a technical
issue. Because when there are problems, we (Europeans) can solve this on hours, people in the US are still at sleep. This also includes fixing things on location.
But, things tend to break at random times around the clock. The US-based servers sometimes crash at night, when people are sleeping here.
One solution to this is to give access to the power port to people living in other parts of the world.
- Finding a reliable server that is free and only for us.
Well, everything is free as long as I pay for it, and if it can be shown that it is technically superior to host European wikis in Europe, I'm willing to do that.
- Finding European sysadmins for this server.
That's not hard -- Magnus and Erik spring to mind immediately, but of course there are many eligible people in Europe.
Again, let me say: I strongly support taking technical measures to ensure responsiveness and uptime for everyone on the planet. I strongly oppose taking measures to divide the wikipedia community politically.
--Jimbo
Jeroen-
Hello,
Proberly some of you will flame me, but I will take that for granted.
My idea is to combine all the European language wikipedia's (except english and maybe spanish and portuguese) on one server and using one database, located somewhere in Europe.
Crashes that are so hard that they require someone to do a fix on location should be expected to take about 24 hours to fix -- that's normal and OK. The present crash took about 7 hours to fix (we still don't know what caused it), and it would have taken just a few minutes if I or Tim could have done the reboot instead of waiting for Jimbo to wake up. Hopefully Brion will be back soon as well.
Splitting up the wikis can cause the following problems: * makes software updates harder -- we are trying to move towards a common codebase * makes it impossible to combine the different tables into a single database, which would have various advantages * makes it more difficult to combine at least user accounts and interlanguage links into a single DB * could lead to cultural split -- different power structures, different policies etc.
Furthermore, Jimbo is in the process of finalizing the Wikimedia Foundation, which will allow us to take donations. Given enough money, we can improve our current setup. A reasonably redundant database setup with replication should be more resistant to this kind of failure.
Realistically, with a decent setup, we don't need to split up our databases -- it's the single point of failure (pliny) that is the problem, not that the server is based in the US.
Regards,
Erik
Erik Moeller wrote:
Jeroen-
Hello,
Proberly some of you will flame me, but I will take that for granted.
My idea is to combine all the European language wikipedia's (except english and maybe spanish and portuguese) on one server and using one database, located somewhere in Europe.
Crashes that are so hard that they require someone to do a fix on location should be expected to take about 24 hours to fix -- that's normal and OK. The present crash took about 7 hours to fix (we still don't know what caused it), and it would have taken just a few minutes if I or Tim could have done the reboot instead of waiting for Jimbo to wake up. Hopefully Brion will be back soon as well.
Splitting up the wikis can cause the following problems:
- makes software updates harder -- we are trying to move towards a common
codebase
- makes it impossible to combine the different tables into a single
database, which would have various advantages
- makes it more difficult to combine at least user accounts and
interlanguage links into a single DB
- could lead to cultural split -- different power structures, different
policies etc.
Furthermore, Jimbo is in the process of finalizing the Wikimedia Foundation, which will allow us to take donations. Given enough money, we can improve our current setup. A reasonably redundant database setup with replication should be more resistant to this kind of failure.
Realistically, with a decent setup, we don't need to split up our databases -- it's the single point of failure (pliny) that is the problem, not that the server is based in the US.
Regards,
which reminds m... Any status update on the wikimedia foundation?? I know as of know Its officially a non-profit, but still hasn't achieved IRS tax exempt status. So, where are we with that? Have the papers been filed, or not? And how many more steps do we have afterwards. It'd be nice to have an estimated time by when this could be done. Is it going to take a couple of weeks or a couple of months? Just curious, and eager to donate....
Lightning
which reminds m... Any status update on the wikimedia foundation?? I know as of know Its officially a non-profit, but still hasn't achieved IRS tax exempt status. So, where are we with that? Have the papers been filed, or not? And how many more steps do we have afterwards. It'd be nice to have an estimated time by when this could be done. Is it going to take a couple of weeks or a couple of months? Just curious, and eager to donate....
Alex is an attorney who does this kind of thing all the time, and he is now advising me. We chatted last week, and are scheduled to chat again this week. At that time, I'll be submitting to him the completed application, by-laws, etc., for him to review and make suggestions. Also, I have to register with Florida and California as a charity, etc. If I've done a good job, it'll be a matter of sending the apps to the relevant agencies.
However, we can actually start accepting donations before that, he tells me. So I'm setting up a bank account and a merchant account and a paypal account for all of that.
I would say within 2 weeks to 1 month, we'll be ready to do a press release (to Slashdot, for example) and to have a pledge week or something like that.
--Jimbo
Jimmy Wales wrote:
which reminds m... Any status update on the wikimedia foundation?? I know as of know Its officially a non-profit, but still hasn't achieved IRS tax exempt status. So, where are we with that? Have the papers been filed, or not? And how many more steps do we have afterwards. It'd be nice to have an estimated time by when this could be done. Is it going to take a couple of weeks or a couple of months? Just curious, and eager to donate....
Alex is an attorney who does this kind of thing all the time, and he is now advising me. We chatted last week, and are scheduled to chat again this week. At that time, I'll be submitting to him the completed application, by-laws, etc., for him to review and make suggestions. Also, I have to register with Florida and California as a charity, etc. If I've done a good job, it'll be a matter of sending the apps to the relevant agencies.
However, we can actually start accepting donations before that, he tells me. So I'm setting up a bank account and a merchant account and a paypal account for all of that.
I would say within 2 weeks to 1 month, we'll be ready to do a press release (to Slashdot, for example) and to have a pledge week or something like that.
--Jimbo
Slick. Let us know when that's up and which one charges you a smaller fee. Also, Are you going to put up a musicbrainz-style finances page? I really like that about Music Brainz, I think it encourages people to donate because they see exactly where the money they donate is going. It's just a list people who donated either with ther name or no name or a pseudonym if they want to and how much they donated as well as a list of expenses and It's updated every day or so I would guess. I think that kind of thing really helps the credibility of projects and makes people more comfortable with donating their money.
Lightning
wikitech-l@lists.wikimedia.org