Re: [Foundation-l] One Wikipedia Per Person (regarding the distribution of and the ability to read Wikipedia)
On Mon, Jun 1, 2009 at 8:14 PM, Thomas Dalton thomas.dal...@gmail.comwrote: While I can't imagine how I managed it now, I don't remember struggling with browsing Wikipedia on a 56K modem. In fact, I think I browsed it on a 36.6K modem... If it is what you are used to, it really doesn't seem that bad. As long as you can download an article (with images) faster than you can read it, it at least serves the basic purpose of providing access to knowledge. But in my opinion Wikipedia (like any encyclopedia) is an absolutely terrible source of knowledge standing alone. An encyclopedia can provide a broad outline of a topic to evaluate which topics you are interested in learning more about, point you to some resources for further reading, and remind you of the answer to some specific questions as they come up, but an encyclopedia cannot stand alone. Education requires access to the rest of the library as well. And it also requires things that probably won't be found in any library (or Wikibook). How to bribe the local police comes to mind. And then there's the whole world which is excluded from Wikimedia projects for being allegedly POV. Perhaps if the definition of POV had been better designed this wouldn't be such a problem, but considering that WP:POV says such things as Hard facts are really rare, I think it's quite obvious NPOV knowledge is not sufficient. So unless you're going to create a very targeted library for each individual, I think that means full internet access (even that is quite incomplete though, especially if you ignore non-free resources like e-books and audiobooks). Going through all the trouble of providing a netbook and wireless connection and then crippling it to only be capable of accessing Wikipedia (and presumably the rest of the Wikimedia sites) would be incredibly wasteful. If full Internet access is too expensive for one individual, have it shared among many. If even that is too expensive, probably because sufficient sharing is infeasible due to low population density, then the solution should be explicitly temporary. Enough generalities, though. ___ foundation-l mailing list foundation-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
Re: [Foundation-l] One Wikipedia Per Person (regarding the distribution of and the ability to read Wikipedia)
On Mon, Jun 1, 2009 at 8:14 PM, Thomas Dalton thomas.dal...@gmail.comwrote: While I can't imagine how I managed it now, I don't remember struggling with browsing Wikipedia on a 56K modem. In fact, I think I browsed it on a 36.6K modem... If it is what you are used to, it really doesn't seem that bad. As long as you can download an article (with images) faster than you can read it, it at least serves the basic purpose of providing access to knowledge. But in my opinion Wikipedia (like any encyclopedia) is an absolutely terrible source of knowledge standing alone. An encyclopedia can provide a broad outline of a topic to evaluate which topics you are interested in learning more about, point you to some resources for further reading, and remind you of the answer to some specific questions as they come up, but an encyclopedia cannot stand alone. Education requires access to the rest of the library as well. And it also requires things that probably won't be found in any library (or Wikibook). How to bribe the local police comes to mind. And then there's the whole world which is excluded from Wikimedia projects for being allegedly POV. Perhaps if the definition of POV had been better designed this wouldn't be such a problem, but considering that WP:POV says such things as Hard facts are really rare, I think it's quite obvious NPOV knowledge is not sufficient. So unless you're going to create a very targeted library for each individual, I think that means full internet access (even that is quite incomplete though, especially if you ignore non-free resources like e-books and audiobooks). Going through all the trouble of providing a netbook and wireless connection and then crippling it to only be capable of accessing Wikipedia (and presumably the rest of the Wikimedia sites) would be incredibly wasteful. If full Internet access is too expensive for one individual, have it shared among many. If even that is too expensive, probably because sufficient sharing is infeasible due to low population density, then the solution should be explicitly temporary. Enough generalities, though. ___ foundation-l mailing list foundation-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
Re: [Foundation-l] One Wikipedia Per Person (regarding the distribution of and the ability to read Wikipedia)
On Tue, Jun 2, 2009 at 4:59 PM, David Gerard dger...@gmail.com wrote: 2009/6/2 George Herbert george.herb...@gmail.com: OLPC is focused on kids. That's important. Perhaps a sister program to provide one OLPC or like device per village, with a more adult development / educational / practical hands on skills data set load would be appropriate. http://www.pixelqi.com/ are displaying their first prototypes this week at Computex in Taipei. Put one of those in a cheap cheap cheap netbook with lotsa flash. Expensive this year, half the price every second year following. Yeah, OLPC without the focus on kids is just a netbook, and there are already netbooks cheaper than OLPCs. ___ foundation-l mailing list foundation-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
Re: [Foundation-l] One Wikipedia Per Person (regarding the distribution of and the ability to read Wikipedia)
On Tue, Jun 2, 2009 at 2:13 PM, Anthony wikim...@inbox.org wrote: On Tue, Jun 2, 2009 at 4:59 PM, David Gerard dger...@gmail.com wrote: 2009/6/2 George Herbert george.herb...@gmail.com: OLPC is focused on kids. That's important. Perhaps a sister program to provide one OLPC or like device per village, with a more adult development / educational / practical hands on skills data set load would be appropriate. http://www.pixelqi.com/ are displaying their first prototypes this week at Computex in Taipei. Put one of those in a cheap cheap cheap netbook with lotsa flash. Expensive this year, half the price every second year following. Yeah, OLPC without the focus on kids is just a netbook, and there are already netbooks cheaper than OLPCs. OLPC will last far longer out in the field than alternate platforms, and uses less power. Durability matters... -- -george william herbert george.herb...@gmail.com ___ foundation-l mailing list foundation-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
Re: [Foundation-l] One Wikipedia Per Person (regarding the distribution of and the ability to read Wikipedia)
On 2009-06-01 00:18, Anthony wrote: On Sun, May 31, 2009 at 6:05 PM, Thomas Daltonthomas.dal...@gmail.comwrote: 2009/5/31 Ray Saintongesainto...@telus.net: Assuming that I were somewhere in rural Africa, and perfectly functioning hardware with Wikipedia software loaded in dropped in front of me from the sky like a magic Coke bottle from the Gods, how much would I then be able to use that gift to get a better yield from my little patch of poor farm-land? Wikipedia could be *part* of a solution, it's never going to be a solution on its own. Wikipedia could be useful as part of an education system, but it can't be the whole thing. I just found another statistic. Mobile networks cover roughly 80-90% of the worlds population. For them, using that mobile network is probably the most cost effective solution. For the rest, giving them enough of an education to have the means to come live with the rest of us, is probably the most cost effective solution. You also found any statistics on what prices for internet access through mobile networks are? What proportion of the world's people can afford a internet connection in the first place, and how many can afford a connection which is useful to browse wikipedia? I'm just curious as I know someone - a westerner - working in Africa and finding internet access hideously expensive. (chat and email ok, but she tells that she avoids browsing the net as the cost is per downloaded MB) ___ foundation-l mailing list foundation-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
Re: [Foundation-l] One Wikipedia Per Person (regarding the distribution of and the ability to read Wikipedia)
2009/6/1 mike.wikipe...@gmail.com mike.wikipe...@gmail.com: You also found any statistics on what prices for internet access through mobile networks are? What proportion of the world's people can afford a internet connection in the first place, and how many can afford a connection which is useful to browse wikipedia? I'm just curious as I know someone - a westerner - working in Africa and finding internet access hideously expensive. (chat and email ok, but she tells that she avoids browsing the net as the cost is per downloaded MB) Indeed. That's why I was suggesting not using a regular ISP but rather having them phone a WMF number to get direct access to the appropriate data, sort of like the old BBS system. That could be subsidised/free except for the cost of the phone call (which could hopefully be subsidised/free, but would require negotiation with the network provider). ___ foundation-l mailing list foundation-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
Re: [Foundation-l] One Wikipedia Per Person (regarding the distribution of and the ability to read Wikipedia)
On Sun, May 31, 2009 at 7:50 AM, Fajro fai...@gmail.com wrote: And why partner with Google? There are Free alternatives in development: http://www.apertium.org/ http://wiki.apertium.org/wiki/Main_Page I tried this with a first paragraph from en.wikipedia, translating to Spanish and back. Worked surprisingly well, even though it renamed New Jersey to New Sweater... Magnus ___ foundation-l mailing list foundation-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
Re: [Foundation-l] One Wikipedia Per Person (regarding the distribution of and the ability to read Wikipedia)
Berber isn't a unitary or standardised language. As far as I'm aware, we have a WP in one of the Berber languages only right now, Kabyle: http://kab.wikipedia.org/ Mark skype: node.ue On Sun, May 31, 2009 at 3:50 AM, geni geni...@gmail.com wrote: 2009/5/31 Brian brian.min...@colorado.edu: Given currently existing technology, and technology that we can reasonably assume to be available within the next decade, how can the WMF best achieve its goal of giving every person free access to our current best summary of all human knowledge? Dead tree technology. Wikipedia based encyclopedias in the most widely used languages. Select the 40K most important articles (that will be fun). 40K was 2002 encarta and most people I knew who used it felt that that was a fairly complete encyclopedia. There are a number of languages with less than 40K articles. The problem ones are: Bengali (19K) Hindi (32K) Punjabi (1.4K) Javanese (19K) Tamil (18K) Marathi (23K) Sindhi (.3K) very low I'm not sure there is a Berber language wikipedia. Can't find it nor a Tamazight one. Anyone know what's going on here? Oriya (.5K) again very low Kannada (6K) Azeri (20K) Sundanese (14K) Hausa (.1K) very low Pashto (1.3K) although you might have a hard time finding volunteers to distribute anything in those areas. Uzbek (7K) Yoruba (6K) Amharic (3K) Strangely Telugu and Malayalam do break the 40K barrier. I've not included the various Chinese languages in this list because I don't understand how spoken languages map to written languages in china. Now a lot of those languages are Indian which since they tend to be fairly closely related and bilingualism is fairly common Bengali, Hindi, Punjabi and English should cover most cases. So how to fill the gaps? Auto translation is one option but not one I like.. Seeing if we can obtain funding to pay people to write articles is another. -- geni ___ 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
Re: [Foundation-l] One Wikipedia Per Person (regarding the distribution of and the ability to read Wikipedia)
geni wrote: 2009/5/31 Brian brian.min...@colorado.edu: Given currently existing technology, and technology that we can reasonably assume to be available within the next decade, how can the WMF best achieve its goal of giving every person free access to our current best summary of all human knowledge? Dead tree technology. Wikipedia based encyclopedias in the most widely used languages. Select the 40K most important articles (that will be fun). 40K was 2002 encarta and most people I knew who used it felt that that was a fairly complete encyclopedia. There are a number of languages with less than 40K articles. The problem ones are: Bengali (19K) Hindi (32K) Punjabi (1.4K) Javanese (19K) Tamil (18K) Marathi (23K) Sindhi (.3K) very low I'm not sure there is a Berber language wikipedia. Can't find it nor a Tamazight one. Anyone know what's going on here? Oriya (.5K) again very low Kannada (6K) Azeri (20K) Sundanese (14K) Hausa (.1K) very low Pashto (1.3K) although you might have a hard time finding volunteers to distribute anything in those areas. Uzbek (7K) Yoruba (6K) Amharic (3K) I think Gujarati (6K) must be in this list. Strangely Telugu and Malayalam do break the 40K barrier. Not surprising: Malayalam is one of the Indian state with the best literacy rate. Telugu is the language of Andhra Pradesh, the 5th Indian state by population, and the South Indian language with largest speaking population. Regards, Yann -- http://www.non-violence.org/ | Site collaboratif sur la non-violence http://www.forget-me.net/ | Alternatives sur le Net http://fr.wikisource.org/ | Bibliothèque libre http://wikilivres.info | Documents libres ___ foundation-l mailing list foundation-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
Re: [Foundation-l] One Wikipedia Per Person (regarding the distribution of and the ability to read Wikipedia)
mike.wikipe...@gmail.com wrote: On 2009-06-01 00:18, Anthony wrote: On Sun, May 31, 2009 at 6:05 PM, Thomas Daltonthomas.dal...@gmail.comwrote: 2009/5/31 Ray Saintongesainto...@telus.net: Assuming that I were somewhere in rural Africa, and perfectly functioning hardware with Wikipedia software loaded in dropped in front of me from the sky like a magic Coke bottle from the Gods, how much would I then be able to use that gift to get a better yield from my little patch of poor farm-land? Wikipedia could be *part* of a solution, it's never going to be a solution on its own. Wikipedia could be useful as part of an education system, but it can't be the whole thing. I just found another statistic. Mobile networks cover roughly 80-90% of the worlds population. For them, using that mobile network is probably the most cost effective solution. For the rest, giving them enough of an education to have the means to come live with the rest of us, is probably the most cost effective solution. You also found any statistics on what prices for internet access through mobile networks are? What proportion of the world's people can afford a internet connection in the first place, and how many can afford a connection which is useful to browse wikipedia? I'm just curious as I know someone - a westerner - working in Africa and finding internet access hideously expensive. (chat and email ok, but she tells that she avoids browsing the net as the cost is per downloaded MB) Last I asked, broadband Internet access in India was about INR 1500 (32 US$), which is at least a week day salary for an Indian worker. True, in theory, there are Internet cafes, but last I tried (in 2007) they can be really used for looking at Wikipedia (too slow). Anyway the priorities are very far from being able to access any online resources. Even when there is a phone, often it doesn't work because people can't pay the bill. Yann -- http://www.non-violence.org/ | Site collaboratif sur la non-violence http://www.forget-me.net/ | Alternatives sur le Net http://fr.wikisource.org/ | Bibliothèque libre http://wikilivres.info | Documents libres ___ foundation-l mailing list foundation-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
Re: [Foundation-l] One Wikipedia Per Person (regarding the distribution of and the ability to read Wikipedia)
2009/6/1 Yann Forget y...@forget-me.net: Last I asked, broadband Internet access in India was about INR 1500 (32 US$), which is at least a week day salary for an Indian worker. True, in theory, there are Internet cafes, but last I tried (in 2007) they can be really used for looking at Wikipedia (too slow). 1500 rupees for how long? And do you mean week's salary or day's salary? It can't be both! What is the point of these internet cafes if the connection is too slow to browse a predominantly text website? ___ foundation-l mailing list foundation-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
Re: [Foundation-l] One Wikipedia Per Person (regarding the distribution of and the ability to read Wikipedia)
This is a good thought-experiment to rerun regularly : working through what 'all human knowledge to each person in his/her own language' means (practical approximations of all, each, and own, c). I think at a minimum, without trying to directly solve high-upkeep projects such as hardware manufacture or physical distribution, this should include making available reasonably good/complete/comprehensible * USB-key distributions * half-offline cell-phone/portable distributions * compact offline distributions (for laptops and computers) * 'full' distributions for PCbangs, net cafes, and school computer labs * the above in one- and two-language editions, All of this should be done regularly, with the best that can be done at a given time; regularly snapshotted with a process that improves over time. Someone has already suggested this isn't the best place to have this discussion. I think it's a pretty good one for at least another while, but wouldn't mind seeing a dedicated group and on-wiki discussion grow out of this and tap into the WP-1.0 and (old-school) WikiReader energy whose contributors rarely chat here. SJ (to the comment that there's not enough space on handhelds to store a 'full' WP snapshot, that's no longer true... offline readers that can keep articles compressed and Flash prices that drop faster than WP grows make it easy enough for most single languages. that said, many people for whatever reason still have a hard time downloading GB of programs or files; which is the real reason to maintain svelte subsets.) SJ On Mon, Jun 1, 2009 at 1:53 PM, Brian brian.min...@colorado.edu wrote: http://meta.wikimedia.org/wiki/OWPP http://meta.wikimedia.org/wiki/One_Wikipedia_Per_Person - Offline Handheld Wikipedia Reader - Dead Tree Technology - Wikipedia TV Also Wikipedia Radio. And don't forget distributed ideas like WP-over-DNS... https://dgl.cx/wikipedia-dns ___ foundation-l mailing list foundation-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
Re: [Foundation-l] One Wikipedia Per Person (regarding the distribution of and the ability to read Wikipedia)
Thomas Dalton wrote: 2009/6/1 Yann Forget y...@forget-me.net: Last I asked, broadband Internet access in India was about INR 1500 (32 US$), which is at least a week day salary for an Indian worker. True, in theory, there are Internet cafes, but last I tried (in 2007) they can be really used for looking at Wikipedia (too slow). 1500 rupees for how long? And do you mean week's salary or day's salary? It can't be both! What is the point of these internet cafes if the connection is too slow to browse a predominantly text website? Sorry, INR 1500 for a month. Well, you can still send a few mail (think it is like a 56 K connection). But most people have TV, so broadcasting some content could reach a lot of people. Yann -- http://www.non-violence.org/ | Site collaboratif sur la non-violence http://www.forget-me.net/ | Alternatives sur le Net http://fr.wikisource.org/ | Bibliothèque libre http://wikilivres.info | Documents libres ___ foundation-l mailing list foundation-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
Re: [Foundation-l] One Wikipedia Per Person (regarding the distribution of and the ability to read Wikipedia)
2009/6/2 Yann Forget y...@forget-me.net: Thomas Dalton wrote: 2009/6/1 Yann Forget y...@forget-me.net: Last I asked, broadband Internet access in India was about INR 1500 (32 US$), which is at least a week day salary for an Indian worker. True, in theory, there are Internet cafes, but last I tried (in 2007) they can be really used for looking at Wikipedia (too slow). 1500 rupees for how long? And do you mean week's salary or day's salary? It can't be both! What is the point of these internet cafes if the connection is too slow to browse a predominantly text website? Sorry, INR 1500 for a month. Well, you can still send a few mail (think it is like a 56 K connection). But most people have TV, so broadcasting some content could reach a lot of people. While I can't imagine how I managed it now, I don't remember struggling with browsing Wikipedia on a 56K modem. In fact, I think I browsed it on a 36.6K modem... If it is what you are used to, it really doesn't seem that bad. ___ foundation-l mailing list foundation-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
Re: [Foundation-l] One Wikipedia Per Person (regarding the distribution of and the ability to read Wikipedia)
On Sat, May 30, 2009 at 11:38 PM, Brian brian.min...@colorado.edu wrote: Given currently existing technology, and technology that we can reasonably assume to be available within the next decade, how can the WMF best achieve its goal of giving every person free access to our current best summary of all human knowledge? Consider that Google Translate has the best machine translation corpus, I think it is too early for this. Don't forget that there aren't a Wikipedia, but Wikipedias. Each language version of Wikipedia has slightly different viewpoints/bias. Which will you chose to be the source for the translations then? And why partner with Google? There are Free alternatives in development: http://www.apertium.org/ http://wiki.apertium.org/wiki/Main_Page -- △ ℱajro △ ___ foundation-l mailing list foundation-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
Re: [Foundation-l] One Wikipedia Per Person (regarding the distribution of and the ability to read Wikipedia)
I would guess a partership with Google would be a good idea because: 1) They are the best (according to Brian) and 2) If we were to go through with this proposal we'd want the translation technology now, not in X years when the technology catches up with google, if at all. And with many OSS/free projects, the X could be insanely high. On Sunday, 31 May 2009 2:50 pm, Fajro wrote: And why partner with Google? There are Free alternatives in development: http://www.apertium.org/ http://wiki.apertium.org/wiki/Main_Page -- △ ℱajro △ -- fl http://en.wikipedia.org/wiki/user_talk:fl ___ foundation-l mailing list foundation-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
Re: [Foundation-l] One Wikipedia Per Person (regarding the distribution of and the ability to read Wikipedia)
Hoi, Currently the translation engine by Goole works for some twenty languages. We have Wikipedias in over 250 languages and we localise in over 300. If we are to collaborate with Google on this, we should partner in the building of translation engines for our other languages. We could and we should consider this when the software was to be open source. Thanks, GerardM 2009/5/31 Foxy Loxy foxyloxy.wikime...@gmail.com I would guess a partership with Google would be a good idea because: 1) They are the best (according to Brian) and 2) If we were to go through with this proposal we'd want the translation technology now, not in X years when the technology catches up with google, if at all. And with many OSS/free projects, the X could be insanely high. On Sunday, 31 May 2009 2:50 pm, Fajro wrote: And why partner with Google? There are Free alternatives in development: http://www.apertium.org/ http://wiki.apertium.org/wiki/Main_Page -- △ ℱajro △ -- fl http://en.wikipedia.org/wiki/user_talk:fl ___ 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
Re: [Foundation-l] One Wikipedia Per Person (regarding the distribution of and the ability to read Wikipedia)
Proprietary algorithms aren't what make their system better - it's that they have a larger corpus. Google has published a trillion token dataset for machine translation researchers but it's presumably just a subset of what they now have. The data that makes their system so good is already available public but it is not (yet) within the scope of the WMF to harvest all copyrighted information in order to increase the performance of already published machine translation algorithms. It would cost the WMF dearly in resources to build such a system themselves based on published research. In other words, as long as the output of the black box is CC-BY-SA the other factors aren't very important. In my mind if you consider using a corporation's semi-proprietary translation engine to be a violation of the WMF's principles then accepting visitors that come from Google in the first place would be an analogous violation. We have no idea how the search engine that is the single largest source of visitors to Wikipedia works, and yet we accept them graciously. On Sun, May 31, 2009 at 1:45 AM, Gerard Meijssen gerard.meijs...@gmail.comwrote: Hoi, Currently the translation engine by Goole works for some twenty languages. We have Wikipedias in over 250 languages and we localise in over 300. If we are to collaborate with Google on this, we should partner in the building of translation engines for our other languages. We could and we should consider this when the software was to be open source. Thanks, GerardM 2009/5/31 Foxy Loxy foxyloxy.wikime...@gmail.com I would guess a partership with Google would be a good idea because: 1) They are the best (according to Brian) and 2) If we were to go through with this proposal we'd want the translation technology now, not in X years when the technology catches up with google, if at all. And with many OSS/free projects, the X could be insanely high. On Sunday, 31 May 2009 2:50 pm, Fajro wrote: And why partner with Google? There are Free alternatives in development: http://www.apertium.org/ http://wiki.apertium.org/wiki/Main_Page -- △ ℱajro △ -- fl http://en.wikipedia.org/wiki/user_talk:fl ___ 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
Re: [Foundation-l] One Wikipedia Per Person (regarding the distribution of and the ability to read Wikipedia)
On Sat, May 30, 2009 at 7:38 PM, Brian brian.min...@colorado.edu wrote: snip The technical specifications of such a device allow for it to be extremely cheap. snip I think you are underestimating the size of Wikipedia. Even compressed a snapshot of the English articles with both text and low quality images would run you 20+ GB. At that storage capacity a handheld display device using modern technology would cost $200+, which is probably way too much to ask a third world person to pay. For comparison, OLPC has a total capacity of only 1GB. In another decade perhaps it would work, but I don't think it is currently an economical project to talk about giving large numbers of people static copies of Wikipedia. (There is perhaps something to be said for distributing a much smaller core subset of articles though.) -Robert ___ foundation-l mailing list foundation-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
Re: [Foundation-l] One Wikipedia Per Person (regarding the distribution of and the ability to read Wikipedia)
Hoi, The notion that this black box needs to use text that is licensed under the CC-by-sa is a folly. The data that is gathered by data mining strips the meaning of the text. Consequently it can be considered to be a completely and utterly separate work. Using text as the basis of a corpus is essentially less intrusive then using the same text for search engine purposes. I have never argued for the WMF to involve itself in machine translation. What I do argue is that the WMF might partner with organisations that are involved in machine translations. It is not just Google that comes to mind, Apertium is another project that has a different approach that is effective for certain language combinations. The legalities and practicalities of language technology are quite distinct from our standard considerations. Thanks, GerardM 2009/5/31 Brian brian.min...@colorado.edu Proprietary algorithms aren't what make their system better - it's that they have a larger corpus. Google has published a trillion token dataset for machine translation researchers but it's presumably just a subset of what they now have. The data that makes their system so good is already available public but it is not (yet) within the scope of the WMF to harvest all copyrighted information in order to increase the performance of already published machine translation algorithms. It would cost the WMF dearly in resources to build such a system themselves based on published research. In other words, as long as the output of the black box is CC-BY-SA the other factors aren't very important. In my mind if you consider using a corporation's semi-proprietary translation engine to be a violation of the WMF's principles then accepting visitors that come from Google in the first place would be an analogous violation. We have no idea how the search engine that is the single largest source of visitors to Wikipedia works, and yet we accept them graciously. On Sun, May 31, 2009 at 1:45 AM, Gerard Meijssen gerard.meijs...@gmail.comwrote: Hoi, Currently the translation engine by Goole works for some twenty languages. We have Wikipedias in over 250 languages and we localise in over 300. If we are to collaborate with Google on this, we should partner in the building of translation engines for our other languages. We could and we should consider this when the software was to be open source. Thanks, GerardM 2009/5/31 Foxy Loxy foxyloxy.wikime...@gmail.com I would guess a partership with Google would be a good idea because: 1) They are the best (according to Brian) and 2) If we were to go through with this proposal we'd want the translation technology now, not in X years when the technology catches up with google, if at all. And with many OSS/free projects, the X could be insanely high. On Sunday, 31 May 2009 2:50 pm, Fajro wrote: And why partner with Google? There are Free alternatives in development: http://www.apertium.org/ http://wiki.apertium.org/wiki/Main_Page -- △ ℱajro △ -- fl http://en.wikipedia.org/wiki/user_talk:fl ___ 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 ___ foundation-l mailing list foundation-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
Re: [Foundation-l] One Wikipedia Per Person (regarding the distribution of and the ability to read Wikipedia)
2009/5/31 Brian brian.min...@colorado.edu: Given currently existing technology, and technology that we can reasonably assume to be available within the next decade, how can the WMF best achieve its goal of giving every person free access to our current best summary of all human knowledge? Dead tree technology. Wikipedia based encyclopedias in the most widely used languages. Select the 40K most important articles (that will be fun). 40K was 2002 encarta and most people I knew who used it felt that that was a fairly complete encyclopedia. There are a number of languages with less than 40K articles. The problem ones are: Bengali (19K) Hindi (32K) Punjabi (1.4K) Javanese (19K) Tamil (18K) Marathi (23K) Sindhi (.3K) very low I'm not sure there is a Berber language wikipedia. Can't find it nor a Tamazight one. Anyone know what's going on here? Oriya (.5K) again very low Kannada (6K) Azeri (20K) Sundanese (14K) Hausa (.1K) very low Pashto (1.3K) although you might have a hard time finding volunteers to distribute anything in those areas. Uzbek (7K) Yoruba (6K) Amharic (3K) Strangely Telugu and Malayalam do break the 40K barrier. I've not included the various Chinese languages in this list because I don't understand how spoken languages map to written languages in china. Now a lot of those languages are Indian which since they tend to be fairly closely related and bilingualism is fairly common Bengali, Hindi, Punjabi and English should cover most cases. So how to fill the gaps? Auto translation is one option but not one I like.. Seeing if we can obtain funding to pay people to write articles is another. -- geni ___ foundation-l mailing list foundation-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
Re: [Foundation-l] One Wikipedia Per Person (regarding the distribution of and the ability to read Wikipedia)
2009/5/31 Foxy Loxy foxyloxy.wikime...@gmail.com: Assembling a chain of production that long, particularly for a non-profit foundation that doesn't have the best reputation (I'm not saying it's justified, but many people in high places will go 'ew, wikipedia'). [citation needed] People in high places appear to love us and/or respect our power, in general. - d. ___ foundation-l mailing list foundation-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
Re: [Foundation-l] One Wikipedia Per Person (regarding the distribution of and the ability to read Wikipedia)
On Sun, May 31, 2009 at 6:50 AM, geni geni...@gmail.com wrote: Dead tree technology. Wikipedia based encyclopedias in the most widely used languages. Select the 40K most important articles (that will be fun). Do you really think the 40K most important Wikipedia articles are more useful than a set of high school textbooks? Wikipedia is sometimes good for getting answers to specific questions, or as a place to find out what you don't know so you can then check other resources to learn it. But it can't replace a good textbook for learning something from scratch. Really, no encyclopedia can. ___ foundation-l mailing list foundation-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
Re: [Foundation-l] One Wikipedia Per Person (regarding the distribution of and the ability to read Wikipedia)
Hoi, May I remind you that the majority of our Wikipedia do not have 40K articles .. Thanks, GerardM 2009/5/31 Anthony wikim...@inbox.org On Sun, May 31, 2009 at 6:50 AM, geni geni...@gmail.com wrote: Dead tree technology. Wikipedia based encyclopedias in the most widely used languages. Select the 40K most important articles (that will be fun). Do you really think the 40K most important Wikipedia articles are more useful than a set of high school textbooks? Wikipedia is sometimes good for getting answers to specific questions, or as a place to find out what you don't know so you can then check other resources to learn it. But it can't replace a good textbook for learning something from scratch. Really, no encyclopedia can. ___ 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
Re: [Foundation-l] One Wikipedia Per Person (regarding the distribution of and the ability to read Wikipedia)
2009/5/31 Anthony wikim...@inbox.org: On Sat, May 30, 2009 at 10:38 PM, Brian brian.min...@colorado.edu wrote: I propose a cheap cellphone-sized device (OWPP) whose only purpose is to read Wikipedia. That's probably both the wrong form (too small) and the wrong content (too flighty) for people permanently without access to the Internet (who presumably also are without access to television - otherwise why not beam Wikipedia through whatever network carries the television signal?). Wikipedia over TV would never work. There isn't the bandwidth for it. TV is a broadcast medium, that means you have to be constantly sending everything anyone could want (or, at least, sending it fairly frequently, like teletext does). There is no way that is ever going to work for even a small portion of Wikipedia. You need to either give people the whole lot in one go (as the suggestion here is) or have it in a way they can request and article and promptly receive it (which is what the web does). ___ foundation-l mailing list foundation-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
Re: [Foundation-l] One Wikipedia Per Person (regarding the distribution of and the ability to read Wikipedia)
On Sun, May 31, 2009 at 8:52 AM, Thomas Dalton thomas.dal...@gmail.comwrote: 2009/5/31 Anthony wikim...@inbox.org: On Sat, May 30, 2009 at 10:38 PM, Brian brian.min...@colorado.edu wrote: I propose a cheap cellphone-sized device (OWPP) whose only purpose is to read Wikipedia. That's probably both the wrong form (too small) and the wrong content (too flighty) for people permanently without access to the Internet (who presumably also are without access to television - otherwise why not beam Wikipedia through whatever network carries the television signal?). Wikipedia over TV would never work. There isn't the bandwidth for it. So only broadcast a subset. TV is a broadcast medium, that means you have to be constantly sending everything anyone could want (or, at least, sending it fairly frequently, like teletext does). Presumably there's a hard drive at the other end. On one channel broadcast updates, on a second channel broadcast random articles weighted by relative importance. By the way, I'm not really sure what you mean by TV is a broadcast medium. But presumably anyone without Internet access but with TV access is receiving the TV signal through a broadcast, so I can safely ignore this nitpick. ___ foundation-l mailing list foundation-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
Re: [Foundation-l] One Wikipedia Per Person (regarding the distribution of and the ability to read Wikipedia)
2009/5/31 Anthony wikim...@inbox.org: Wikipedia over TV would never work. There isn't the bandwidth for it. So only broadcast a subset. A very small subset. TV is a broadcast medium, that means you have to be constantly sending everything anyone could want (or, at least, sending it fairly frequently, like teletext does). Presumably there's a hard drive at the other end. On one channel broadcast updates, on a second channel broadcast random articles weighted by relative importance. TV's with hard drives are a pretty new in the developed world and presumably all but non-existent in the developing world, I would be very surprised if many people have a TV with a hard drive and no internet access (or, at least, no ability to get internet access if they wanted it). So, you would have to give people these hard-drives, so you might as well fill them before you hand them out. So, what you are suggesting is the same idea as Brian suggested but with the ability to update articles over TV transmissions - not a bad extension to the idea, but it's the same basic idea. By the way, I'm not really sure what you mean by TV is a broadcast medium. But presumably anyone without Internet access but with TV access is receiving the TV signal through a broadcast, so I can safely ignore this nitpick. By broadcast medium I mean a one-way transmission of information. The TV people choose what you broadcast and you just choose to either pick up what they send or don't. You can't request specific information like you can online. That dramatically increases the bandwidth requirements, since you have to broadcast everything. ___ foundation-l mailing list foundation-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
Re: [Foundation-l] One Wikipedia Per Person (regarding the distribution of and the ability to read Wikipedia)
2009/5/31 Anthony wikim...@inbox.org: By broadcast medium I mean a one-way transmission of information. I don't know about yours, but my TV uses two-way transmission. So a statement that TV is a broadcast medium is just not correct. True, it's probably correct in the vast majority of situations, but, blah blah blah, I think you see what I'm getting at... The TV people choose what you broadcast and you just choose to either pick up what they send or don't. You can't request specific information like you can online. Umm, yes I can. But like I said, I was nitpicking. TV isn't a medium, and it isn't necessarily broadcast. Who has cable TV that can't get internet access? You mentioned TV in the context of a way of getting information to people without internet access, so I ignored the existence of cable since it doesn't apply. ___ foundation-l mailing list foundation-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
Re: [Foundation-l] One Wikipedia Per Person (regarding the distribution of and the ability to read Wikipedia)
On Sun, May 31, 2009 at 12:35 PM, Thomas Dalton thomas.dal...@gmail.comwrote: 2009/5/31 Anthony wikim...@inbox.org: On Sun, May 31, 2009 at 11:05 AM, Thomas Dalton thomas.dal...@gmail.com wrote: Who has cable TV that can't get internet access? I didn't say *cable* TV. What kind of TV do you have that can go two ways, then? The only types I know are cable, satellite and regular radio waves, only the first of which allows 2-way transmission. Radio and satellite (which is radio) are both capable of 2-way transmission. ___ foundation-l mailing list foundation-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
Re: [Foundation-l] One Wikipedia Per Person (regarding the distribution of and the ability to read Wikipedia)
On Sun, May 31, 2009 at 12:35 PM, Thomas Dalton thomas.dal...@gmail.comwrote: There is no such thing as one-way internet access. The internet is always 2-way. Perhaps so (depends on your definitions), but then, Wave probably isn't dependent on internet access in the first place. I see no reason it would be. ___ foundation-l mailing list foundation-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
Re: [Foundation-l] One Wikipedia Per Person (regarding the distribution of and the ability to read Wikipedia)
On Sun, May 31, 2009 at 2:51 PM, Thomas Dalton thomas.dal...@gmail.comwrote: 2009/5/31 Anthony wikim...@inbox.org: On Sun, May 31, 2009 at 12:35 PM, Thomas Dalton thomas.dal...@gmail.com wrote: There is no such thing as one-way internet access. The internet is always 2-way. Perhaps so (depends on your definitions), but then, Wave probably isn't dependent on internet access in the first place. I see no reason it would be. If it doesn't work over IP then it isn't the internet, and IP is a two-way protocol. That might work except it isn't true. UDP/IP is one-way. ___ foundation-l mailing list foundation-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
Re: [Foundation-l] One Wikipedia Per Person (regarding the distribution of and the ability to read Wikipedia)
2009/5/31 Anthony wikim...@inbox.org: On Sun, May 31, 2009 at 2:51 PM, Thomas Dalton thomas.dal...@gmail.comwrote: 2009/5/31 Anthony wikim...@inbox.org: On Sun, May 31, 2009 at 12:35 PM, Thomas Dalton thomas.dal...@gmail.com wrote: There is no such thing as one-way internet access. The internet is always 2-way. Perhaps so (depends on your definitions), but then, Wave probably isn't dependent on internet access in the first place. I see no reason it would be. If it doesn't work over IP then it isn't the internet, and IP is a two-way protocol. That might work except it isn't true. UDP/IP is one-way. It's not. How can you load a webpage without being able to send GETs and POSTs? ___ foundation-l mailing list foundation-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
Re: [Foundation-l] One Wikipedia Per Person (regarding the distribution of and the ability to read Wikipedia)
2009/5/31 Anthony wikim...@inbox.org: HTTP uses TCP/IP, not UDP/IP. Your comment was If it doesn't work over IP then it isn't the internet. If you'd like to change that to If it doesn't work over TCP then it isn't the internet, fine. But it probably wouldn't be difficult to run the Wave protocol over UDP. Then you could send one-way Wave updates through a one-way satellite feed, or a one-way OTA feed. Add in a low-bandwidth or intermittent connection to send in the other direction, and you can get an email account better than most people had in 1995. Remember when BBSes used to subscribe to UUCP email? However that has expensive upkeep costs and is reliant of functioning infrastructure. Books do not stop working just because somebody broke SAT-3/WASC. -- geni ___ foundation-l mailing list foundation-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
Re: [Foundation-l] One Wikipedia Per Person (regarding the distribution of and the ability to read Wikipedia)
2009/5/31 geni geni...@gmail.com: 2009/5/31 David Gerard dger...@gmail.com: For a practical example, the Schools Wikipedia is proving enormously popular with teachers in countries of all economic levels. Requires something that can read a DVD, or have said DVD dumped onto its hard disk somehow, and in print it'd be roughly 15 Britannica volumes. However it is english only as far as I'm aware. Yes indeed. It is an improvement on nothing, however, and shows how similar things could be done for other languages (e.g. Schools Wikipedia was done by a charity for use in its own schools). - d. ___ foundation-l mailing list foundation-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
Re: [Foundation-l] One Wikipedia Per Person (regarding the distribution of and the ability to read Wikipedia)
On Sun, May 31, 2009 at 3:20 PM, geni geni...@gmail.com wrote: 2009/5/31 Anthony wikim...@inbox.org: HTTP uses TCP/IP, not UDP/IP. Your comment was If it doesn't work over IP then it isn't the internet. If you'd like to change that to If it doesn't work over TCP then it isn't the internet, fine. But it probably wouldn't be difficult to run the Wave protocol over UDP. Then you could send one-way Wave updates through a one-way satellite feed, or a one-way OTA feed. Add in a low-bandwidth or intermittent connection to send in the other direction, and you can get an email account better than most people had in 1995. Remember when BBSes used to subscribe to UUCP email? However that has expensive upkeep costs and is reliant of functioning infrastructure. Books do not stop working just because somebody broke SAT-3/WASC. Depends on the topic. Some books go obsolete pretty quickly, and delivering new ones is quite expensive. I've got nothing against books, though. They're a big part of any solution. That said... On Sun, May 31, 2009 at 3:14 PM, geni geni...@gmail.com wrote: 2009/5/31 Anthony wikim...@inbox.org: On Sun, May 31, 2009 at 2:34 PM, geni geni...@gmail.com wrote: There are a number of existing projects to send out school text books. An encyclopedia however is a useful part of wider learning. I guess, but a print copy of some subset of Wikipedia doesn't seem like the best solution for someone who already has access to school textbooks. If you're talking about a major language, there are already encyclopedias written for them, and copies can probably be had for much less it would cost to publish a print edition of Wikipedia. Evidence? Remember there are rather a lot of major languages where any native speaker well educated and rich enough to actually buy an encyclopedia (even in the west britanica was a middle class symbol) is unlikely to want to buy one in that language. Brand new and for retail price, sure. But used and/or at cost (which surely there are publishers willing to provide for this sort of thing), I don't see how you could beat the established players. If you're talking about a minor language, I don't know. Are there languages for which Wikipedia is unarguably the best encyclopedia, with enough native speakers to make a print run feasible, and for which offering an encyclopedia in a non-native language wouldn't be more effective? Tagalog is the first example that comes to mind. Telugu perhaps but the pro English bias there would be an issue. Maybe. Want to start that focus group? It's not a focus group issue. It's a document what encyclopedia's actually exist in non European languages issue. I'm not sure we should waste everyone on this mailing list's time going through the details and formulating a plan. Let's take Tagalog. We've got 22 million native speakers, of which what % have internet access, and what % of those without it would be interested in a copy of Wikipedia? What kind of technology do these people have? What % have electricity? What % have access to a library? What are the schools like for them? Do the schools have computers? Do the libraries have computers? What topics would be most important? How big is the Tagalog Wikipedia? What are its strengths? What are its weaknesses? What is the government's role in education? How is the funding? What's the mean and median income? I'd love to put something into action here. But it's not something I see being worked out over foundation-l. ___ foundation-l mailing list foundation-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
Re: [Foundation-l] One Wikipedia Per Person (regarding the distribution of and the ability to read Wikipedia)
On Sun, May 31, 2009 at 4:42 PM, geni geni...@gmail.com wrote: 2009/5/31 Anthony wikim...@inbox.org: I'm not sure we should waste everyone on this mailing list's time going through the details and formulating a plan. Let's take Tagalog. We've got 22 million native speakers, of which what % have internet access, 15% maybe? http://www.internetworldstats.com/asia.htm That's Internet usage, not Internet access. Considering that 60% of the world uses cell phones (as in, has a subscription, those living in areas covered by cell phone signals is presumably much higher), and only 25% uses the Internet, apparently lack of Internet usage isn't due to an infrastructural problem in most cases. That's not to say that there isn't an infrastructural problem in some cases. In fact, those cases are probably the ones we should be focusing on. I've looked at bit and am having some trouble figuring out what percentage of the world lives in a location (other than by personal preference) where neither they nor someone nearby them can get Internet access. But schools would not be the target in this case since most of the pupils likely have a reasonable grasp of english since that is what some of the lessons are taught in: http://countrystudies.us/philippines/53.htm What would you suggest for the target? ___ foundation-l mailing list foundation-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
Re: [Foundation-l] One Wikipedia Per Person (regarding the distribution of and the ability to read Wikipedia)
geni wrote: Now a lot of those languages are Indian which since they tend to be fairly closely related and bilingualism is fairly common Bengali, Hindi, Punjabi and English should cover most cases. That's very generously European of you. The three Indian languages that you chose are all Indo-Aryan. Manipuri, Munda and Tamil are as different from these as they are from each other. Ec ___ foundation-l mailing list foundation-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
Re: [Foundation-l] One Wikipedia Per Person (regarding the distribution of and the ability to read Wikipedia)
Anthony wrote: On Sun, May 31, 2009 at 6:50 AM, geni geni...@gmail.com wrote: Dead tree technology. Wikipedia based encyclopedias in the most widely used languages. Select the 40K most important articles (that will be fun). Do you really think the 40K most important Wikipedia articles are more useful than a set of high school textbooks? Wikipedia is sometimes good for getting answers to specific questions, or as a place to find out what you don't know so you can then check other resources to learn it. But it can't replace a good textbook for learning something from scratch. Really, no encyclopedia can. Assuming that I were somewhere in rural Africa, and perfectly functioning hardware with Wikipedia software loaded in dropped in front of me from the sky like a magic Coke bottle from the Gods, how much would I then be able to use that gift to get a better yield from my little patch of poor farm-land? Ec ___ foundation-l mailing list foundation-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
Re: [Foundation-l] One Wikipedia Per Person (regarding the distribution of and the ability to read Wikipedia)
2009/6/1 Anthony wikim...@inbox.org: On Sun, May 31, 2009 at 6:52 PM, Thomas Dalton thomas.dal...@gmail.comwrote: 2009/5/31 Anthony wikim...@inbox.org: I just found another statistic. Mobile networks cover roughly 80-90% of the worlds population. For them, using that mobile network is probably the most cost effective solution. For the rest, giving them enough of an education to have the means to come live with the rest of us, is probably the most cost effective solution. Those are basic mobile phone networks, not internet phones. I don't think voice calls and SMS messages are going to be much help. It's mostly GSM. You're telling me these networks can't handle the use of a GSM modem? If it can carry voice, it can can carry data. Fair point. I guess I'm so used to broadband I forgot about the existence of dial up for a second! You would need to hand out phones, laptops, and network subscriptions, though - that's getting rather expensive just to give someone an up-to-date encyclopaedia. The network subscription could probably be heavily discounted if you were only able to phone one number and that was to a WMF phone line that handled the updates (so not strictly an internet connection). ___ foundation-l mailing list foundation-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
Re: [Foundation-l] One Wikipedia Per Person (regarding the distribution of and the ability to read Wikipedia)
On Sun, May 31, 2009 at 7:17 PM, Thomas Dalton thomas.dal...@gmail.comwrote: I guess I'm so used to broadband I forgot about the existence of dial up for a second! You would need to hand out phones, laptops, and network subscriptions, though - that's getting rather expensive just to give someone an up-to-date encyclopaedia. I guess I'm forgetting how cheap labor is in so many parts of the world. Here in the US we're talking about less than a week's work, but in an Indian call center we're talking about over a month. ___ foundation-l mailing list foundation-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
Re: [Foundation-l] One Wikipedia Per Person (regarding the distribution of and the ability to read Wikipedia)
2009/6/1 Anthony wikim...@inbox.org: On Sun, May 31, 2009 at 7:17 PM, Thomas Dalton thomas.dal...@gmail.comwrote: I guess I'm so used to broadband I forgot about the existence of dial up for a second! You would need to hand out phones, laptops, and network subscriptions, though - that's getting rather expensive just to give someone an up-to-date encyclopaedia. I guess I'm forgetting how cheap labor is in so many parts of the world. Here in the US we're talking about less than a week's work, but in an Indian call center we're talking about over a month. People working in Indian call centres probably already have internet access, or at least can access the internet somewhere (in a internet cafe, or something). They are generally quite highly educated (I believe many even have degrees, but can make more money in a call centre working for a foreign company than using their degree working for an Indian company). For people in rural areas, there is no way they could ever afford these things themselves, many have a subsistence lifestyle, there is no possibility to save up for stuff. ___ foundation-l mailing list foundation-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
Re: [Foundation-l] One Wikipedia Per Person (regarding the distribution of and the ability to read Wikipedia)
On Sun, May 31, 2009 at 9:08 PM, Thomas Dalton thomas.dal...@gmail.comwrote: 2009/6/1 Anthony wikim...@inbox.org: On Sun, May 31, 2009 at 7:17 PM, Thomas Dalton thomas.dal...@gmail.com wrote: I guess I'm so used to broadband I forgot about the existence of dial up for a second! You would need to hand out phones, laptops, and network subscriptions, though - that's getting rather expensive just to give someone an up-to-date encyclopaedia. I guess I'm forgetting how cheap labor is in so many parts of the world. Here in the US we're talking about less than a week's work, but in an Indian call center we're talking about over a month. People working in Indian call centres probably already have internet access, or at least can access the internet somewhere (in a internet cafe, or something). They are generally quite highly educated (I believe many even have degrees, but can make more money in a call centre working for a foreign company than using their degree working for an Indian company). For people in rural areas, there is no way they could ever afford these things themselves, many have a subsistence lifestyle, there is no possibility to save up for stuff. The educated people in rural areas generally get themselves out. If someone voluntarily chooses to live a subsistence lifestyle, there's no point in providing them with a free copy of Wikipedia in the first place. But still, over a month's salary is pretty steep, considering that there's no guarantee it'll help. I guess for now it's better to focus on providing access in schools and libraries. ___ foundation-l mailing list foundation-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
[Foundation-l] One Wikipedia Per Person (regarding the distribution of and the ability to read Wikipedia)
Given currently existing technology, and technology that we can reasonably assume to be available within the next decade, how can the WMF best achieve its goal of giving every person free access to our current best summary of all human knowledge? Consider that Google Translate has the best machine translation corpus, consisting not only of the Internet but also all United Nations translations and many other datasets. It is the closest existing thing to a Babelfish, now supporting 41 languages and winning all translation competitions for several years. It will continue to be the best for the foreseeable future. Consider that 75% of the world is not online and that there may be a way to beat market forces in the race to getting free Internet access to every person by literally giving Wikipedia to every person instead, offline. Our current micro-content distribution model would be sufficient if everyone had access to the Internet. They don't so it's not. Consider that the money the WMF could potentially raise through competitive market forces (the OLPC way) may lag behind the money they can raise through their idealistic goals, uncompromised values and principles, and smart ideas. This money can be used to give copies of the entirety of Wikipedia away. Consider that access to Wikipedia does not require readability proper (beautiful prose), just the ability to comprehend the information, and just barely. The human brain is the most powerful translator in existence, we just have to meet said brain halfway. We may see a meta language in our lifetimes but not within the next decade. The current best meta language is a set of fuzzy translations that are a function of the size of the source and target language corpuses. I propose a cheap cellphone-sized device (OWPP) whose only purpose is to read Wikipedia. The WMF teams up with Google to obtain CC-BY-SA translations from all supported source languages to all supported target languages. The device holds just one copy of all of the Wikipedia's in a single target language. The technical specifications of such a device allow for it to be extremely cheap. Let's let those of us fortunate enough to have access to the Internet write an encyclopedia and give it to those who are not, sooner rather than later. Brian ___ foundation-l mailing list foundation-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
Re: [Foundation-l] One Wikipedia Per Person (regarding the distribution of and the ability to read Wikipedia)
It does sound like an excellent idea, but it does appear to require us teaming up with Google, a hardware vendor, a software vendor (the OS of course), a distributor and various governments that may or may not wish they people having access to 'forbidden' information. Assembling a chain of production that long, particularly for a non-profit foundation that doesn't have the best reputation (I'm not saying it's justified, but many people in high places will go 'ew, wikipedia'). And then of course we have the money issue. On Sunday, 31 May 2009 10:38 am, Brian wrote: Given currently existing technology, and technology that we can reasonably assume to be available within the next decade, how can the WMF best achieve its goal of giving every person free access to our current best summary of all human knowledge? Consider that Google Translate has the best machine translation corpus, consisting not only of the Internet but also all United Nations translations and many other datasets. It is the closest existing thing to a Babelfish, now supporting 41 languages and winning all translation competitions for several years. It will continue to be the best for the foreseeable future. Consider that 75% of the world is not online and that there may be a way to beat market forces in the race to getting free Internet access to every person by literally giving Wikipedia to every person instead, offline. Our current micro-content distribution model would be sufficient if everyone had access to the Internet. They don't so it's not. Consider that the money the WMF could potentially raise through competitive market forces (the OLPC way) may lag behind the money they can raise through their idealistic goals, uncompromised values and principles, and smart ideas. This money can be used to give copies of the entirety of Wikipedia away. Consider that access to Wikipedia does not require readability proper (beautiful prose), just the ability to comprehend the information, and just barely. The human brain is the most powerful translator in existence, we just have to meet said brain halfway. We may see a meta language in our lifetimes but not within the next decade. The current best meta language is a set of fuzzy translations that are a function of the size of the source and target language corpuses. I propose a cheap cellphone-sized device (OWPP) whose only purpose is to read Wikipedia. The WMF teams up with Google to obtain CC-BY-SA translations from all supported source languages to all supported target languages. The device holds just one copy of all of the Wikipedia's in a single target language. The technical specifications of such a device allow for it to be extremely cheap. Let's let those of us fortunate enough to have access to the Internet write an encyclopedia and give it to those who are not, sooner rather than later. -- fl http://en.wikipedia.org/wiki/user_talk:fl ___ foundation-l mailing list foundation-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
