On Sat, May 30, 2009 at 9:33 PM, Brian Brian.Mingus@colorado.edu wrote:
How does Google Wave help the WMF achieve its goals?
Not sure, it doesn't really exist yet. I'm sure there will be numerous ways in which it can do it, though.
Wikipedia has already become a dominant information source for the 1.5
billion people with Internet access thanks to Google.
How does being a dominant information source for people help the WMF achieve its goals?
We need to focus on getting Wikipedia to the 5.2 billion people who can't
access it.
Indeed. What languages do these 5.2 billion people speak? Are they connected to the Internet? If not, what's stopping them? Do they have a telephone, a computer, electricity, television, running water? If not, what is stopping them from being able to get these things? Are the problems things that are well geared toward the expertise of the WMF, or are we better off letting other non-profits with more specialized expertise fix them?
Personally, I didn't even know the number was 5.2 billion. Should I do this further research myself, or can someone answer these questions for me?
2009/5/31 Anthony wikimail@inbox.org:
On Sat, May 30, 2009 at 9:33 PM, Brian Brian.Mingus@colorado.edu wrote:
How does Google Wave help the WMF achieve its goals?
Not sure, it doesn't really exist yet. I'm sure there will be numerous ways in which it can do it, though.
While we could move this mailing list over to Waves and get access to fancy games of su-doku while we flame each other, I can't see any great ways it can help, at least directly. I'm open to being surprised, though.
Wikipedia has already become a dominant information source for the 1.5
billion people with Internet access thanks to Google.
How does being a dominant information source for people help the WMF achieve its goals?
I don't think being dominant helps in any way, but being an information source for people basically *is* the WMF's goal.
We need to focus on getting Wikipedia to the 5.2 billion people who can't
access it.
Indeed. What languages do these 5.2 billion people speak? Are they connected to the Internet? If not, what's stopping them? Do they have a telephone, a computer, electricity, television, running water? If not, what is stopping them from being able to get these things? Are the problems things that are well geared toward the expertise of the WMF, or are we better off letting other non-profits with more specialized expertise fix them?
Personally, I didn't even know the number was 5.2 billion. Should I do this further research myself, or can someone answer these questions for me?
Wikipedia [1] tells me there are 1.58 billion internet users world wide. It also tells me [2] there are 6.78 billion people in the world. That leaves 5.20 billion non-internet users.
If we look at [1] we see the countries with the lowest percentage of people with internet access are North Korea (well, actually that's unknown, but I think we can safely assume it is as close to zero as makes no odds) and Myanmar - there probably isn't a great deal we can do to help them. Countries a little higher up the list we might be able to help with OLPC-style schemes. Printed versions of Wikibooks might be useful. While we can't directly help with things like getting people access to clean water, education is a very important part of any long term scheme to get people out of poverty, and we can certainly help there.
1. http://en.wikipedia.org/wiki/List_of_countries_by_number_of_Internet_users 2. http://en.wikipedia.org/wiki/World_population
On Sun, May 31, 2009 at 8:38 AM, Thomas Dalton thomas.dalton@gmail.comwrote:
2009/5/31 Anthony wikimail@inbox.org:
On Sat, May 30, 2009 at 9:33 PM, Brian Brian.Mingus@colorado.edu
wrote:
How does Google Wave help the WMF achieve its goals?
Not sure, it doesn't really exist yet. I'm sure there will be numerous
ways
in which it can do it, though.
While we could move this mailing list over to Waves and get access to fancy games of su-doku while we flame each other, I can't see any great ways it can help, at least directly. I'm open to being surprised, though.
If Waves works anything like email, then it will be possible to use it when not directly connected to the Internet. How's that for helping get Wikipedia to people without Internet access?
Wikipedia has already become a dominant information source for the 1.5
billion people with Internet access thanks to Google.
How does being a dominant information source for people help the WMF
achieve
its goals?
I don't think being dominant helps in any way, but being an information source for people basically *is* the WMF's goal.
The goal is for the information to get to the people. Being the source is only the means to that goal.
Personally, I didn't even know the number was 5.2 billion. Should I do
this
further research myself, or can someone answer these questions for me?
Wikipedia [1] tells me there are 1.58 billion internet users world wide. It also tells me [2] there are 6.78 billion people in the world. That leaves 5.20 billion non-internet users.
Ah. :( That's not very helpful... What does it mean to be an "internet user"?
While we can't directly help with things like getting
people access to clean water, education is a very important part of any long term scheme to get people out of poverty, and we can certainly help there.
I'd say education is lower on the list than most people think, unless you mean education as a means to get the hell out of the place you're living in.
2009/5/31 Anthony wikimail@inbox.org:
If Waves works anything like email, then it will be possible to use it when not directly connected to the Internet. How's that for helping get Wikipedia to people without Internet access?
Not very. Waves, like email, should work for people with intermittent internet access, but it's pretty useless for people with no internet access.
Wikipedia has already become a dominant information source for the 1.5
billion people with Internet access thanks to Google.
How does being a dominant information source for people help the WMF
achieve
its goals?
I don't think being dominant helps in any way, but being an information source for people basically *is* the WMF's goal.
The goal is for the information to get to the people. Being the source is only the means to that goal.
Fair enough, but I think that's mostly a technicality. Information travels along a chain, wherever the WMF is in that chain you can probably call it a/the source.
Personally, I didn't even know the number was 5.2 billion. Should I do
this
further research myself, or can someone answer these questions for me?
Wikipedia [1] tells me there are 1.58 billion internet users world wide. It also tells me [2] there are 6.78 billion people in the world. That leaves 5.20 billion non-internet users.
Ah. :( That's not very helpful... What does it mean to be an "internet user"?
Don't know, I didn't look at the methodology of the statistics (rookie mistake, I know). Hang on... Ok, it looks like the data comes fro a variety of sources, so I doubt there is a common methodology. Take the statistics with a large pinch of salt!
While we can't directly help with things like getting
people access to clean water, education is a very important part of any long term scheme to get people out of poverty, and we can certainly help there.
I'd say education is lower on the list than most people think, unless you mean education as a means to get the hell out of the place you're living in.
It depends if you are thinking long term or short term. Short term, education is pretty useless, but long term it is extremely important. I don't think getting people out of poverty by moving them is a sustainable strategy, you need to make the poor villages wealthier. That means enabling them to move beyond subsistence living, which education is a key part of.
Hoi, Wave in its reference implementation relies on HTML 5. This means that it requires a modern browser. With a browser it is possible to access data that is on a LAN or on the local computer. This would allow us to have "Wikipedia" type content stored locally or on a LAN. One question is how will resources will react when newer data becomes available, will it synchronise? When a resource available to Wave *can *be updated, it makes no difference if it on a system on a LAN or on the WAN / Internet when it is the availability of data that is essential.
When you think of Wave as a replacement for e-mail it is easy to forget about the other aspects of Wave. At its most basic it provides a platform for data that can be approached with a browser.. This line of thought begs the question, to what extend the Wave reference implementation will be able to use gears for local data storage.. Thanks, GerardM
2009/5/31 Thomas Dalton thomas.dalton@gmail.com
2009/5/31 Anthony wikimail@inbox.org:
If Waves works anything like email, then it will be possible to use it
when
not directly connected to the Internet. How's that for helping get Wikipedia to people without Internet access?
Not very. Waves, like email, should work for people with intermittent internet access, but it's pretty useless for people with no internet access.
Wikipedia has already become a dominant information source for the 1.5
billion people with Internet access thanks to Google.
How does being a dominant information source for people help the WMF
achieve
its goals?
I don't think being dominant helps in any way, but being an information source for people basically *is* the WMF's goal.
The goal is for the information to get to the people. Being the source
is
only the means to that goal.
Fair enough, but I think that's mostly a technicality. Information travels along a chain, wherever the WMF is in that chain you can probably call it a/the source.
Personally, I didn't even know the number was 5.2 billion. Should I
do
this
further research myself, or can someone answer these questions for me?
Wikipedia [1] tells me there are 1.58 billion internet users world wide. It also tells me [2] there are 6.78 billion people in the world. That leaves 5.20 billion non-internet users.
Ah. :( That's not very helpful... What does it mean to be an "internet user"?
Don't know, I didn't look at the methodology of the statistics (rookie mistake, I know). Hang on... Ok, it looks like the data comes fro a variety of sources, so I doubt there is a common methodology. Take the statistics with a large pinch of salt!
While we can't directly help with things like getting
people access to clean water, education is a very important part of any long term scheme to get people out of poverty, and we can certainly help there.
I'd say education is lower on the list than most people think, unless you mean education as a means to get the hell out of the place you're living
in.
It depends if you are thinking long term or short term. Short term, education is pretty useless, but long term it is extremely important. I don't think getting people out of poverty by moving them is a sustainable strategy, you need to make the poor villages wealthier. That means enabling them to move beyond subsistence living, which education is a key part of.
foundation-l mailing list foundation-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
On Sun, May 31, 2009 at 9:42 AM, Gerard Meijssen gerard.meijssen@gmail.comwrote:
One question is how will resources will react when newer data becomes available, will it synchronise?
That seems to be part of the protocol. You'd set up a bot which makes the updates, and add it. Someone on the LAN would have to run a server, but that server would only need intermittent access. In fact, a bot might even be set up to allow *write* access through this intermittent connection.
Now my understanding is that the protocol for interserver communication isn't completed, and who knows it may be vaporware. But it's an intriguing possibility. (As I said in a previous message, finally the platform I need for P2Pedia is here.)
When you think of Wave as a replacement for e-mail it is easy to forget
about the other aspects of Wave.
Yeah, Wave became interesting to me as soon as I stopped thinking of it as a replacement for e-mail, and more like a distributed private (access controlled) wiki.
2009/5/31 Anthony wikimail@inbox.org:
Now my understanding is that the protocol for interserver communication isn't completed, and who knows it may be vaporware. But it's an intriguing possibility. (As I said in a previous message, finally the platform I need for P2Pedia is here.)
Wave sounds more like a MovieOS version of Usenet. How would you do p2pedia via NNTP?
- d.
On Sun, May 31, 2009 at 10:14 AM, David Gerard dgerard@gmail.com wrote:
2009/5/31 Anthony wikimail@inbox.org:
Now my understanding is that the protocol for interserver communication isn't completed, and who knows it may be vaporware. But it's an
intriguing
possibility. (As I said in a previous message, finally the platform I
need
for P2Pedia is here.)
Wave sounds more like a MovieOS version of Usenet. How would you do p2pedia via NNTP?
I wouldn't.
Did you watch the Wave preview video? Wave allows you to edit other people's messages and to update your own. Usenet doesn't allow that.
2009/5/31 Gerard Meijssen gerard.meijssen@gmail.com:
Hoi, Wave in its reference implementation relies on HTML 5. This means that it requires a modern browser. With a browser it is possible to access data that is on a LAN or on the local computer. This would allow us to have "Wikipedia" type content stored locally or on a LAN. One question is how will resources will react when newer data becomes available, will it synchronise? When a resource available to Wave *can *be updated, it makes no difference if it on a system on a LAN or on the WAN / Internet when it is the availability of data that is essential.
What would be the point of that? Wave is good for collaborating, not reading. If all they are going to be doing is reading Wikipedia then just download a dump and put in on a local apache server for your LAN.
Hoi, Much of the Wave functionality demonstrated is superior to what is available in MediaWiki. Consider a LAN with OPLC systems, consider a Wave server on the school server.. It would be pretty damn good to be able to have all kinds of activities that makes use of the functionality that is part of the reference implementation. Consider what a talk page would look like when with the Wave "back" functionality.
Anthony said it before, think here of what Wave can do when you concentrate on its Wiki aspects. Now consider a talk page.. when it changes you may get an e-mail, you can go there and edit it... All this functionality is there and more, Wave allows you to have a real time conversation.. And all of this can happen on your LAN, your WAN or your Internet.
Given its license I am excited, given the demonstrated functionality I am excited. The first thing is to get MediaWiki content into Wave. A developer friend of mine has read the developer documentation and thinks he can do this. What is needed is for him to have access to a Wave environment where he can experiment. Thanks, GerardM
2009/5/31 Thomas Dalton thomas.dalton@gmail.com
2009/5/31 Gerard Meijssen gerard.meijssen@gmail.com:
Hoi, Wave in its reference implementation relies on HTML 5. This means that it requires a modern browser. With a browser it is possible to access data
that
is on a LAN or on the local computer. This would allow us to have "Wikipedia" type content stored locally or on a LAN. One question is how will resources will react when newer data becomes available, will it synchronise? When a resource available to Wave *can *be updated, it makes
no
difference if it on a system on a LAN or on the WAN / Internet when it is the availability of data that is essential.
What would be the point of that? Wave is good for collaborating, not reading. If all they are going to be doing is reading Wikipedia then just download a dump and put in on a local apache server for your LAN.
foundation-l mailing list foundation-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
It's interesting thinking about it this way, because Wave could potentially even replace Wikipedia. Transfer the contents of a Wikipedia article to Wave, and make a widget to display the current article (or, even better, the latest approved version). Now anyone can start a mirror with virtually no costs by just putting the widget up on a website. I wonder how access control works. Can any Wave recipient give read/write access to anyone else? That'd be problematic. Can a bot be added to control the access? If not it seems like an easy extension to add.
It'd be a major money drain on Google, though. They'd bear virtually all of the hosting costs, and it's hard to see how they'd get anything in return. Something would have to give eventually. But once the protocol is out and the reference implementation is out, maybe it wouldn't matter. People could start up their own servers.
Adapt or perish, WMF.
On Sun, May 31, 2009 at 10:33 AM, Gerard Meijssen <gerard.meijssen@gmail.com
wrote:
Hoi, Much of the Wave functionality demonstrated is superior to what is available in MediaWiki. Consider a LAN with OPLC systems, consider a Wave server on the school server.. It would be pretty damn good to be able to have all kinds of activities that makes use of the functionality that is part of the reference implementation. Consider what a talk page would look like when with the Wave "back" functionality.
Anthony said it before, think here of what Wave can do when you concentrate on its Wiki aspects. Now consider a talk page.. when it changes you may get an e-mail, you can go there and edit it... All this functionality is there and more, Wave allows you to have a real time conversation.. And all of this can happen on your LAN, your WAN or your Internet.
Given its license I am excited, given the demonstrated functionality I am excited. The first thing is to get MediaWiki content into Wave. A developer friend of mine has read the developer documentation and thinks he can do this. What is needed is for him to have access to a Wave environment where he can experiment. Thanks, GerardM
2009/5/31 Thomas Dalton thomas.dalton@gmail.com
2009/5/31 Gerard Meijssen gerard.meijssen@gmail.com:
Hoi, Wave in its reference implementation relies on HTML 5. This means that
it
requires a modern browser. With a browser it is possible to access data
that
is on a LAN or on the local computer. This would allow us to have "Wikipedia" type content stored locally or on a LAN. One question is
how
will resources will react when newer data becomes available, will it synchronise? When a resource available to Wave *can *be updated, it
makes
no
difference if it on a system on a LAN or on the WAN / Internet when it
is
the availability of data that is essential.
What would be the point of that? Wave is good for collaborating, not reading. If all they are going to be doing is reading Wikipedia then just download a dump and put in on a local apache server for your LAN.
foundation-l mailing list foundation-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
foundation-l mailing list foundation-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
Hoi, Wave might replace parts of MediaWiki but it would not replace Wikipedia... To appreciate this, you have to realise what it is the WMF stands for.. It is content first and foremost. MediaWiki is our current software. It is great software and it has great functionality. When the Wave software is able to replace aspects of MediaWiki, you will find that it is possible to integrate it into MediaWiki.
If there is a problem, it is with MediaWiki. Given its license it cannot contribute back to Wave.. Wave functionality can be incorporated in MediaWiki because it does not have a viral license. Thanks, GerardM
2009/5/31 Anthony wikimail@inbox.org
It's interesting thinking about it this way, because Wave could potentially even replace Wikipedia. Transfer the contents of a Wikipedia article to Wave, and make a widget to display the current article (or, even better, the latest approved version). Now anyone can start a mirror with virtually no costs by just putting the widget up on a website. I wonder how access control works. Can any Wave recipient give read/write access to anyone else? That'd be problematic. Can a bot be added to control the access? If not it seems like an easy extension to add.
It'd be a major money drain on Google, though. They'd bear virtually all of the hosting costs, and it's hard to see how they'd get anything in return. Something would have to give eventually. But once the protocol is out and the reference implementation is out, maybe it wouldn't matter. People could start up their own servers.
Adapt or perish, WMF.
On Sun, May 31, 2009 at 10:33 AM, Gerard Meijssen < gerard.meijssen@gmail.com
wrote:
Hoi, Much of the Wave functionality demonstrated is superior to what is available in MediaWiki. Consider a LAN with OPLC systems, consider a Wave server on the school server.. It would be pretty damn good to be able to have all kinds of activities that makes use of the functionality that is part of the reference implementation. Consider what a talk page would
look
like when with the Wave "back" functionality.
Anthony said it before, think here of what Wave can do when you
concentrate
on its Wiki aspects. Now consider a talk page.. when it changes you may
get
an e-mail, you can go there and edit it... All this functionality is
there
and more, Wave allows you to have a real time conversation.. And all of this can happen on your LAN, your WAN or your Internet.
Given its license I am excited, given the demonstrated functionality I am excited. The first thing is to get MediaWiki content into Wave. A
developer
friend of mine has read the developer documentation and thinks he can do this. What is needed is for him to have access to a Wave environment
where
he can experiment. Thanks, GerardM
2009/5/31 Thomas Dalton thomas.dalton@gmail.com
2009/5/31 Gerard Meijssen gerard.meijssen@gmail.com:
Hoi, Wave in its reference implementation relies on HTML 5. This means
that
it
requires a modern browser. With a browser it is possible to access
data
that
is on a LAN or on the local computer. This would allow us to have "Wikipedia" type content stored locally or on a LAN. One question is
how
will resources will react when newer data becomes available, will it synchronise? When a resource available to Wave *can *be updated, it
makes
no
difference if it on a system on a LAN or on the WAN / Internet when
it
is
the availability of data that is essential.
What would be the point of that? Wave is good for collaborating, not reading. If all they are going to be doing is reading Wikipedia then just download a dump and put in on a local apache server for your LAN.
foundation-l mailing list foundation-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
foundation-l mailing list foundation-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
foundation-l mailing list foundation-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
On Sun, May 31, 2009 at 11:53 AM, Gerard Meijssen <gerard.meijssen@gmail.com
wrote:
Wave might replace parts of MediaWiki but it would not replace Wikipedia... To appreciate this, you have to realise what it is the WMF stands for..
It stands for the Wikimedia Foundation.
It is content first and foremost.
No, it's a board of directors, and a staff, and a bunch of servers - none of which are really required and all of which can and should be replaced.
If you want to say that whatever it is that replaces Wikipedia *is* Wikipedia, fine, though you might have some trademark issues until the WMF actually dissolves.
MediaWiki is our current software. It is great software and it has great functionality. When the Wave software is able to replace aspects of MediaWiki, you will find that it is possible to integrate it into MediaWiki.
But MediaWiki is not a distributed platform. That's the problem. It's too centralized, both technically and politically.
If there is a problem, it is with MediaWiki. Given its license it cannot
contribute back to Wave.. Wave functionality can be incorporated in MediaWiki because it does not have a viral license.
Huh? Wave does not have a license. Parts of it will be proprietary, parts will be GPL, parts will be BSD, parts will be public domain.
Hoi, Wave DOES have a license and it is neither BSD nor GPL. It is however very much open source. You can create a GPL licensed implementation of Wave but that would not be acceptable as production code that is to be used in association with proposed changes to the protocol. Thanks, GerardM
NB you may check out the WMF wiki because imho you do not get it.
2009/5/31 Anthony wikimail@inbox.org
On Sun, May 31, 2009 at 11:53 AM, Gerard Meijssen < gerard.meijssen@gmail.com
wrote:
Wave might replace parts of MediaWiki but it would not replace
Wikipedia...
To appreciate this, you have to realise what it is the WMF stands for..
It stands for the Wikimedia Foundation.
It is content first and foremost.
No, it's a board of directors, and a staff, and a bunch of servers - none of which are really required and all of which can and should be replaced.
If you want to say that whatever it is that replaces Wikipedia *is* Wikipedia, fine, though you might have some trademark issues until the WMF actually dissolves.
MediaWiki is our current software. It is great software and it has great functionality. When the Wave software is able to replace aspects of MediaWiki, you will find that it is possible
to
integrate it into MediaWiki.
But MediaWiki is not a distributed platform. That's the problem. It's too centralized, both technically and politically.
If there is a problem, it is with MediaWiki. Given its license it cannot
contribute back to Wave.. Wave functionality can be incorporated in MediaWiki because it does not have a viral license.
Huh? Wave does not have a license. Parts of it will be proprietary, parts will be GPL, parts will be BSD, parts will be public domain. _______________________________________________ foundation-l mailing list foundation-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
2009/5/31 Gerard Meijssen gerard.meijssen@gmail.com:
Hoi, Much of the Wave functionality demonstrated is superior to what is available in MediaWiki. Consider a LAN with OPLC systems, consider a Wave server on the school server.. It would be pretty damn good to be able to have all kinds of activities that makes use of the functionality that is part of the reference implementation. Consider what a talk page would look like when with the Wave "back" functionality.
Wave has some great features for those of us that edit Wikipedia, but we're talking about people reading it. The only way to edit it is via a live internet connection, I can't see anything else working unless we get some vastly improved edit conflict handling. All people reading Wikipedia need is a plain HTML file per article, nothing more.
On Sun, May 31, 2009 at 11:09 AM, Thomas Dalton thomas.dalton@gmail.comwrote:
2009/5/31 Gerard Meijssen gerard.meijssen@gmail.com:
Hoi, Much of the Wave functionality demonstrated is superior to what is available in MediaWiki. Consider a LAN with OPLC systems, consider a Wave server on the school server.. It would be pretty damn good to be able to have all kinds of activities that makes use of the functionality that is part of the reference implementation. Consider what a talk page would
look
like when with the Wave "back" functionality.
Wave has some great features for those of us that edit Wikipedia, but we're talking about people reading it. The only way to edit it is via a live internet connection, I can't see anything else working unless we get some vastly improved edit conflict handling.
If you watched the Wave presentation you'll see that there is quite a bit of edit conflict handling already built in (they showed three people editing the same page simultaneously).
All people reading Wikipedia need is a plain HTML file per article, nothing more.
Easier said than done, though. The static HTML Wikipedia dumps haven't been updated since June 2008. With Wave, updates are instantaneous.
2009/5/31 Anthony wikimail@inbox.org:
If you watched the Wave presentation you'll see that there is quite a bit of edit conflict handling already built in (they showed three people editing the same page simultaneously).
I did watch it. That was live, they could see each other editing and avoid each other. There was no conflict. You are talking about people without live internet connections.
All people reading Wikipedia need is a plain HTML file per article, nothing more.
Easier said than done, though. The static HTML Wikipedia dumps haven't been updated since June 2008. With Wave, updates are instantaneous.
MediaWiki updates are instantaneous. Where is the improvement in people reading Wikipedia via Wave?
On Sun, May 31, 2009 at 12:37 PM, Thomas Dalton thomas.dalton@gmail.comwrote:
2009/5/31 Anthony wikimail@inbox.org:
If you watched the Wave presentation you'll see that there is quite a bit
of
edit conflict handling already built in (they showed three people editing the same page simultaneously).
I did watch it. That was live, they could see each other editing and avoid each other. There was no conflict. You are talking about people without live internet connections.
Watch it again then. There was at least one conflict, and they even pointed it out and mentioned conflict resolution.
Clearly if two people edit exactly the same text one of the edits is going to fail. But that's not the common situation.
All people reading Wikipedia need is a plain HTML file per article,
nothing
more.
Easier said than done, though. The static HTML Wikipedia dumps haven't
been
updated since June 2008. With Wave, updates are instantaneous.
MediaWiki updates are instantaneous. Where is the improvement in people reading Wikipedia via Wave?
One advantage, the one we were talking about, is that you don't need to be connected to the Internet while you're reading it.
Yes, you could in theory implement such a thing without Wave, but unless the WMF starts offering free live feeds (even for intermittent connections), it's not going to be updated like Wave is. Maybe the WMF won't support access to Wikipedia articles through Wave, but then, a fork will, and the WMF's goal ("to empower and engage people around the world to collect and develop educational content under a free licensehttp://en.wikipedia.org/wiki/en:free_contentor in the public domain, and to disseminate it effectively and globally"), is still met, even if their particular implementation of it isn't.
2009/5/31 Anthony wikimail@inbox.org:
On Sun, May 31, 2009 at 12:37 PM, Thomas Dalton thomas.dalton@gmail.comwrote:
2009/5/31 Anthony wikimail@inbox.org:
If you watched the Wave presentation you'll see that there is quite a bit
of
edit conflict handling already built in (they showed three people editing the same page simultaneously).
I did watch it. That was live, they could see each other editing and avoid each other. There was no conflict. You are talking about people without live internet connections.
Watch it again then. There was at least one conflict, and they even pointed it out and mentioned conflict resolution.
Clearly if two people edit exactly the same text one of the edits is going to fail. But that's not the common situation.
Edit conflicts with live editing aren't an issue, manual resolution is trivial. Edit conflicts with significant delays are a much bigger problem and require automated merging, which isn't always possible, and is often very difficult.
All people reading Wikipedia need is a plain HTML file per article,
nothing
more.
Easier said than done, though. The static HTML Wikipedia dumps haven't
been
updated since June 2008. With Wave, updates are instantaneous.
MediaWiki updates are instantaneous. Where is the improvement in people reading Wikipedia via Wave?
One advantage, the one we were talking about, is that you don't need to be connected to the Internet while you're reading it.
Yes, you could in theory implement such a thing without Wave, but unless the WMF starts offering free live feeds (even for intermittent connections), it's not going to be updated like Wave is. Maybe the WMF won't support access to Wikipedia articles through Wave, but then, a fork will, and the WMF's goal ("to empower and engage people around the world to collect and develop educational content under a free licensehttp://en.wikipedia.org/wiki/en:free_contentor in the public domain, and to disseminate it effectively and globally"), is still met, even if their particular implementation of it isn't.
We already have dumps (the latest dump of all enwiki primary content finished a couple of hours ago and is 4.8 gig), all we would need to do is make incremental dumps available so people don't have to download the whole thing repeatedly. That would be pretty easy to program compared to rewriting the whole of MediaWiki to function via Waves.
On Sun, May 31, 2009 at 2:49 PM, Thomas Dalton thomas.dalton@gmail.comwrote:
Edit conflicts with live editing aren't an issue, manual resolution is trivial. Edit conflicts with significant delays are a much bigger problem and require automated merging, which isn't always possible, and is often very difficult.
Why do edit conflicts with significant delays require automated merging? What's wrong with sending back a message that "your edit failed due to an edit conflict", or even better "there was an conflict with your edit - it has been sent to a queue for manual processing"? Sure, third worlders won't be able to get into an edit war on the English Wikipedia version of [[George W. Bush]], but that doesn't mean they can't contribute to one of the millions of lower traffic articles.
We already have dumps (the latest dump of all enwiki primary content
finished a couple of hours ago and is 4.8 gig), all we would need to do is make incremental dumps available so people don't have to download the whole thing repeatedly.
Great. Do it.
That would be pretty easy to program compared to rewriting the whole of MediaWiki to function via Waves.
Google has already done that, except it's not MediaWiki, it's something much much more powerful and easy to use.
2009/5/31 Anthony wikimail@inbox.org:
On Sun, May 31, 2009 at 2:49 PM, Thomas Dalton thomas.dalton@gmail.comwrote:
Edit conflicts with live editing aren't an issue, manual resolution is trivial. Edit conflicts with significant delays are a much bigger problem and require automated merging, which isn't always possible, and is often very difficult.
Why do edit conflicts with significant delays require automated merging? What's wrong with sending back a message that "your edit failed due to an edit conflict", or even better "there was an conflict with your edit - it has been sent to a queue for manual processing"? Sure, third worlders won't be able to get into an edit war on the English Wikipedia version of [[George W. Bush]], but that doesn't mean they can't contribute to one of the millions of lower traffic articles.
There would be an extremely large number of conflicts that never got resolved, since the message that got sent back would be sent some time after the person had lost interest in the edit. I don't see why we're discussing these people editing, the important thing is that they can read it. If they don't have access to Wikipedia chances are they don't have access to reliable sources to edit using anyway.
We already have dumps (the latest dump of all enwiki primary content
finished a couple of hours ago and is 4.8 gig), all we would need to do is make incremental dumps available so people don't have to download the whole thing repeatedly.
Great. Do it.
A little out of my skillset, I think (I haven't looked at the new dump system, but it is probably beyond me to understand well enough to do more than trivial bug fixes). If you can show a demand for it, though, I expect someone will do it sooner or later.
That would be pretty easy to program compared to rewriting the whole of MediaWiki to function via Waves.
Google has already done that, except it's not MediaWiki, it's something much much more powerful and easy to use.
And if we were starting Wikipedia in a few months time, perhaps we would base it around Waves, but we aren't, so that's irrelevant. It would need a massive amount of work to convert everything over. Even if we did that work, it wouldn't make any difference to the 5.2 billion people this thread is about. Until you start making sense, I'm out - have a good day.
On Sun, May 31, 2009 at 3:10 PM, Thomas Dalton thomas.dalton@gmail.comwrote:
2009/5/31 Anthony wikimail@inbox.org:
On Sun, May 31, 2009 at 2:49 PM, Thomas Dalton <thomas.dalton@gmail.com wrote:
Edit conflicts with live editing aren't an issue, manual resolution is trivial. Edit conflicts with significant delays are a much bigger problem and require automated merging, which isn't always possible, and is often very difficult.
Why do edit conflicts with significant delays require automated merging? What's wrong with sending back a message that "your edit failed due to an edit conflict", or even better "there was an conflict with your edit - it has been sent to a queue for manual processing"? Sure, third worlders
won't
be able to get into an edit war on the English Wikipedia version of
[[George
W. Bush]], but that doesn't mean they can't contribute to one of the millions of lower traffic articles.
There would be an extremely large number of conflicts that never got resolved, since the message that got sent back would be sent some time after the person had lost interest in the edit.
1) there would be more that got through than that caused a conflict. 2) it wouldn't have to be the original person who resolved the conflict.
I don't see why we're discussing these people editing, the important thing is that they can read it.
Fair enough. I pointed it out as a possibility. You're the one who turned it into a long discussion.
If they don't have access to Wikipedia chances are they don't have access to reliable sources to edit using anyway.
Who is the "they" we're talking about here. Google Wave won't help everyone in every situation, but it does have potential to help the WMF meet its goals.
That would be pretty easy to program compared to rewriting the whole of MediaWiki to function via Waves.
Google has already done that, except it's not MediaWiki, it's something
much
much more powerful and easy to use.
And if we were starting Wikipedia in a few months time, perhaps we would base it around Waves, but we aren't, so that's irrelevant. It would need a massive amount of work to convert everything over.
I'm sure you can import html into a Wave.
The WMF might not do it, but someone will.
Even if we did that work, it wouldn't make any difference to the 5.2 billion people this thread is about. Until you start making sense, I'm out - have a good day.
Adios, don't let the door hit, et. al.
On Sun, May 31, 2009 at 9:17 AM, Thomas Dalton thomas.dalton@gmail.comwrote:
2009/5/31 Anthony wikimail@inbox.org:
If Waves works anything like email, then it will be possible to use it
when
not directly connected to the Internet. How's that for helping get Wikipedia to people without Internet access?
Not very. Waves, like email, should work for people with intermittent internet access, but it's pretty useless for people with no internet access.
How many people in the world have intermittent Internet access? I guess you're right if you were just saying that Wave doesn't directly help someone with absolutely no Internet access. If so, I misunderstood, because my comment was that Google Wave can "help the WMF achieve its goals".
Personally, I didn't even know the number was 5.2 billion. Should I
do
this
further research myself, or can someone answer these questions for me?
Wikipedia [1] tells me there are 1.58 billion internet users world wide. It also tells me [2] there are 6.78 billion people in the world. That leaves 5.20 billion non-internet users.
Ah. :( That's not very helpful... What does it mean to be an "internet user"?
Don't know, I didn't look at the methodology of the statistics (rookie mistake, I know). Hang on... Ok, it looks like the data comes fro a variety of sources, so I doubt there is a common methodology. Take the statistics with a large pinch of salt!
That's a pretty important question. If not being an internet user just means that you have internet at the library and not in your home, the method to reach such people is much different. Considering that the United States is listed at 74.7%, I'm sure "internet user" is defined too strictly for the purpose of saying who "can't access [Wikipedia]".
While we can't directly help with things like getting
people access to clean water, education is a very important part of any long term scheme to get people out of poverty, and we can certainly help there.
I'd say education is lower on the list than most people think, unless you mean education as a means to get the hell out of the place you're living
in.
It depends if you are thinking long term or short term. Short term, education is pretty useless, but long term it is extremely important.
I still think it's less important than most people make it out to be. I'd say politics is much more important than education in terms of getting people out of poverty long term, and a good education does not at all guarantee a good set of political beliefs.
I don't think getting people out of poverty by moving them is a
sustainable strategy, you need to make the poor villages wealthier. That means enabling them to move beyond subsistence living, which education is a key part of.
I think you're ignoring the political reality of the situation. Poor villages are poor due to 1) lack of natural resources and/or 2) a bad political environment. In either case, the long term solution is to leave until the conditions have improved. Yes, lack of education is a problem, but poor education is a result of 1 and/or 2, not the other way around.
That's a pretty important question. If not being an internet user just means that you have internet at the library and not in your home, the method to reach such people is much different. Considering that the United States is listed at 74.7%, I'm sure "internet user" is defined too strictly for the purpose of saying who "can't access [Wikipedia]".
I'm pretty sure "internet user" doesn't refer to the availability of a connection, but rather whether or not a person actually uses the Internet. My grandparents were married for nearly 75 years before they died, living in the same house until the end. My grandfather was an avid internet user and would've been happy to let my grandmother use the Internet had she shown any interest, the computer with the connection was right in her home, yet she never connected, even when he was not home or if he was ill. She simply felt she had better things to occupy her time.
Your idea that the fact that the statistic counts over 25% of US Americans as not being internet users must mean a very strict definition is used is not necessarily correct. There are plenty of people in the USA, particularly the elderly and rural poor, who do not use the Internet. 20% of US Americans live in rural areas, 12% are age 65 and over. I'm not saying that nobody from those groups uses the Internet, I'm just saying that 75% seems like an accurate number to me of people who use the Internet.
Mark
On Sun, May 31, 2009 at 11:15 AM, Mark Williamson node.ue@gmail.com wrote:
Your idea that the fact that the statistic counts over 25% of US Americans as not being internet users must mean a very strict definition is used is not necessarily correct.
Touche. I was equating "internet user" with "someone with internet access". That, broadly defined, is the statistic that this thread is about: "getting Wikipedia to the ??? people who can't access it". Apparently that ??? is significantly less than 5.2 billion, but more information on the demographics of those people would be helpful. In fact, if there are a few people seriously interested in this, maybe we could start a focus group to better define the problem and come up with a half dozen proposed solutions.
wikimedia-l@lists.wikimedia.org