In light of the recent USA Today article:
In the same way that we are currently enforcing proper image tags using a bot, could we do the same with unsourced articles? Start out by placing {{unsourced}} in all the articles lacking sources, and then, if it is not sourced in a week, create something like the {{copvio}} page-replacer to hide the unsourced content (the entire article), explaining with a detailed message that the article must be thoroughly sourced.
In my mind, at least, it doesn't seem like there should be any difference between enforcing sources for images and sources for articles. If anything we should be enforcing the latter more, since articles form the basis of the encyclopedia. I know this won't solve everything, but I think it should be a vital part of Wikipedia; since we do not know who edits an article, we need to know that it is based on information that we can verify ourselves.
brian0918
On 11/30/05, Brian brian0918@gmail.com wrote:
In light of the recent USA Today article:
In the same way that we are currently enforcing proper image tags using a bot, could we do the same with unsourced articles? Start out by placing {{unsourced}} in all the articles lacking sources, and then, if it is not sourced in a week, create something like the {{copvio}} page-replacer to hide the unsourced content (the entire article), explaining with a detailed message that the article must be thoroughly sourced.
In my mind, at least, it doesn't seem like there should be any difference between enforcing sources for images and sources for articles. If anything we should be enforcing the latter more, since articles form the basis of the encyclopedia. I know this won't solve everything, but I think it should be a vital part of Wikipedia; since we do not know who edits an article, we need to know that it is based on information that we can verify ourselves.
It's an interesting idea, but it would apply to so many articles on such a coarse grained way it may wind up being ignored. Perhaps such a function could be tied to the TOC, so that each section would have a flag of whether it had been sourced to satisfaction.
-Andrew (User:Fuzheado)
brian0918 _______________________________________________ foundation-l mailing list foundation-l@wikimedia.org http://mail.wikipedia.org/mailman/listinfo/foundation-l
Andrew Lih wrote:
On 11/30/05, Brian brian0918@gmail.com wrote:
In light of the recent USA Today article:
In the same way that we are currently enforcing proper image tags using a bot, could we do the same with unsourced articles? Start out by placing {{unsourced}} in all the articles lacking sources, and then, if it is not sourced in a week, create something like the {{copvio}} page-replacer to hide the unsourced content (the entire article), explaining with a detailed message that the article must be thoroughly sourced.
In my mind, at least, it doesn't seem like there should be any difference between enforcing sources for images and sources for articles. If anything we should be enforcing the latter more, since articles form the basis of the encyclopedia. I know this won't solve everything, but I think it should be a vital part of Wikipedia; since we do not know who edits an article, we need to know that it is based on information that we can verify ourselves.
It's an interesting idea, but it would apply to so many articles on such a coarse grained way it may wind up being ignored. Perhaps such a function could be tied to the TOC, so that each section would have a flag of whether it had been sourced to satisfaction.
-Andrew (User:Fuzheado)
And who is going to have time to check all the sources. Also most of the material that I write comes from books as I hate the external link festivity some articles become. So who is going to check the books I use. Hint you can find my library I have at my disposal in Thailand here: http://nl.wikipedia.org/wiki/Gebruiker:Waerth/Mijn_bronnen (it is in Dutch but you get my drift).
Also I think it is wrong to play panick over one criticism. Yes things need to be done. We knew that for years already. To now start panicking and coming with emergency reactions we might not help the situation.
Walter / Waerth
Walter van Kalken wrote:
It's an interesting idea, but it would apply to so many articles on such a coarse grained way it may wind up being ignored. Perhaps such a function could be tied to the TOC, so that each section would have a flag of whether it had been sourced to satisfaction.
-Andrew (User:Fuzheado)
And who is going to have time to check all the sources. Also most of the material that I write comes from books as I hate the external link festivity some articles become. So who is going to check the books I use. Hint you can find my library I have at my disposal in Thailand here: http://nl.wikipedia.org/wiki/Gebruiker:Waerth/Mijn_bronnen (it is in Dutch but you get my drift).
Creating a "References" section or using a template to make a proper book citation will be just as detectable as using an external link. We are currently only talking about citing sources, not about verifying them. Articles need to be referenced before their contents can be verified.
Also I think it is wrong to play panick over one criticism. Yes things need to be done. We knew that for years already. To now start panicking and coming with emergency reactions we might not help the situation.
This isn't panicking. This is doing what we should have been doing all along, and treating our article content with the same importance that we treat our image content. This is a step in the right direction.
And who is going to have time to check all the sources. Also most of the material that I write comes from books as I hate the external link festivity some articles become. So who is going to check the books I use. Hint you can find my library I have at my disposal in Thailand here: http://nl.wikipedia.org/wiki/Gebruiker:Waerth/Mijn_bronnen (it is in Dutch but you get my drift).
Creating a "References" section or using a template to make a proper book citation will be just as detectable as using an external link. We are currently only talking about citing sources, not about verifying them. Articles need to be referenced before their contents can be verified.
Yes I agree with the referencing point. But in some articles on en: all sentences receive external links. Which is overdoing it imho. An article should have general references in the reference section. Like Books x,y,z, etc, websites a,b,c, etc. But not a reference for every sentence. That is overdoing it.
Also I think it is wrong to play panick over one criticism. Yes things need to be done. We knew that for years already. To now start panicking and coming with emergency reactions we might not help the situation.
This isn't panicking. This is doing what we should have been doing all along, and treating our article content with the same importance that we treat our image content. This is a step in the right direction.
It is panicking if I watch the deletion list on en: On which a perfectly legit article by a highly regarded nl: contributor was put up dor deletion by someone who doesn't know anything about the subject but Quotes CITE and some other policies. The user wasn't even explained on his own page why it was put up for deletion.
(URL: http://en.wikipedia.org/wiki/Wikipedia:Articles_for_deletion/North_Slavic_la... )
The real answer to this whole mess is (again imho) . Start becoming more aggressive on sources when articles are editted by anons. It was an anon that wrote the article that wasn't checked by anyone and made USA today. But what is happening now is that users with a good trackrecord are harassed more (this is happening on nl.wikipedia) while changes by anons are given the benefit off the doubt because we do not want to scare the anons (this is happening on nl.wikipedia). Sounds like something gone awry there. I personally still assume good faith with people who bothered to register. As 95% in my experience are serious and do not put garbage on there. The problems are mostly coming from anons. Or Vandals that seem to do it out of profession. Both of these groups should be scrutinised more. And not the regular contributors who will normally handle in good faith.
Waerth/Walter
The Soi case.
On the Dutch wikipedia I wrote the article [[soi]]. It was also written on the english wikipedia (http://en.wikipedia.org/wiki/Soi)
According to the CITE principal this article should be deleted right now as it is not from a book nor a website, but partly from my own and any Thais experience. Even though it contains pure facts which are the truth it doesn't cite anything as no sources are used. Note I didn't write the english article but it basicaly says the same as the Dutch article.
Facts: /*1) Soi*/ (Thai http://en.wikipedia.org/wiki/Thai_alphabet: *???*) is the term used in Thailand http://en.wikipedia.org/wiki/Thailand for a side-street branching off a major street (/thanon http://en.wikipedia.org/w/index.php?title=Thanon&action=edit/). An alley is called a /Trok http://en.wikipedia.org/w/index.php?title=Trok&action=edit/.
All true but do not have it from any book.
2) Sois are usually numbered, and are referred to by the name of the major street and the number, as in "Sukhumvit Soi 4",
Also please come and have a look but not from any book
3) All the even-numbered sois are on one side of the street, the odd-numbered ones on the other.
How do you want me to prove that from citations without flying you over here?
4) If for instance a soi is added between soi 7 and soi 9 it will get the name soi 7/1, the next one soi 7/2 etc.
Again fact new soi was added between 2 sois near my home got the 6/1 number as it was between sois 6 and 8. No sources sorry
5) Almost all sois also have a name.
I first wanted to be bolder as I have never seen a soi that doesn't also have a name. But I decided to be a bit less sure about it
6) On lower Sukhumvit road in Bangkok for instance the soi's are named after important landowners or families of landowners who had land in the area in the past.
Verified with Thais whose families actually come from these landowners and have been living in the area for a long time. Again no books.
7) Some sois become major thoroughfares and because of that get known by their name only. Examples are Thonglor (Sukhumvit soi 55), Asoke (Sukhumvit soi 21) etc.
Again plain fact. Books anyone?
So according to the CITE offensive this article and many like this should be deleted even though there is nothing that is not the truth in there. Is it original research? I do not think so. You see the problem is when you start this CITE offensive one day one of our "police" officers will find this article and put it up for deletion because there are no sources. People take these so called guidelines much to strict and cause damage with it.
This is why I would say be stricter on anons and newbies and maybe make some kind of ranking of wikipedians that are known to work in good faith or that are known IRL. Like me, anyone can always call me (see my phonenumbers) I even put my homeaddress up once. I might not always be able to quote directly where I got something from. But I am handling in good faith.
And with me there are many people like me.
Waerth/Walter
On 11/30/05, Walter van Kalken walter@vankalken.net wrote:
The Soi case.
On the Dutch wikipedia I wrote the article [[soi]]. It was also written on the english wikipedia (http://en.wikipedia.org/wiki/Soi)
According to the CITE principal this article should be deleted right now
Actually, a quick google search turned up a few relevant citations (one from Wikitravel, in fact). I added them to the English article. It is very very hard to find articles which are at once encyclopedic and verifiable, but have no suitable reference either online or in print. I cannot think of an example atm.
++SJ
SJ wrote:
On 11/30/05, Walter van Kalken walter@vankalken.net wrote:
The Soi case.
On the Dutch wikipedia I wrote the article [[soi]]. It was also written on the english wikipedia (http://en.wikipedia.org/wiki/Soi)
According to the CITE principal this article should be deleted right now
Actually, a quick google search turned up a few relevant citations (one from Wikitravel, in fact). I added them to the English article. It is very very hard to find articles which are at once encyclopedic and verifiable, but have no suitable reference either online or in print. I cannot think of an example atm.
++SJ
Hoi, I find it remarkable that this CITE thing may break one of our fundamental reasons for success. It will drive people away when it is used as a tool to justify deletions. Why will it drive people away, because it raises the entry level to a point where many people will just no longer bother to do something. Yes, there may be an EDIT button, but without a citation it will be gone.
Having citations is a good thing but please realise that it is best used when controversies arise. It is then part of the instrumentarium to come to a NPOV. When it used on its own I expect it to prove disastrous for the influx of new people into our community. Certainly if this is deemed to be necessary to be followed by all the projects that do not have the maturity of the English language wikipedia.
So, when you in the en.wikipedia decide to have this new deletion policy, consider the fallout it may have elsewhere.
Thanks, GerardM
On Fri, 2 Dec 2005, Gerard Meijssen wrote:
SJ wrote:
On 11/30/05, Walter van Kalken walter@vankalken.net wrote:
The Soi case.
On the Dutch wikipedia I wrote the article [[soi]]. It was also written on the english wikipedia (http://en.wikipedia.org/wiki/Soi)
According to the CITE principal this article should be deleted right now
Actually, a quick google search turned up a few relevant citations (one from Wikitravel, in fact). I added them to the English article. It is very very hard to find articles which are at once encyclopedic and verifiable, but have no suitable reference either online or in print. I cannot think of an example atm.
++SJ
Hoi, I find it remarkable that this CITE thing may break one of our fundamental reasons for success. It will drive people away when it is used as a tool to justify deletions. Why will it drive people away, because it raises the entry
This should *never* be used to justify deletions. Ever. But it should be used to justify tagging articles as unsourced; or marking them for cleanup... just as we tag articles as stubs, or as unwikified blobs of text, when that is what they are. Of course, writing more than 2 paragrapyhs, or wikifying raw text, 'raises the entry level' -- so we do not require it.
But we *do* require it for 'good style' and acceptance into the circle of enlightened articlehood.
Having citations is a good thing but please realise that it is best used when controversies arise.
Having citations is fundamental to being a good reference work.
This is *not* ( I say it again ) related to deletion policy.
SJ
Having citations is a good thing but please realise that it is best used when controversies arise.
Having citations is fundamental to being a good reference work.
This is *not* ( I say it again ) related to deletion policy.
SJ
Yes SJ you will be able to handle it in a correct matter and so will many other wikipedians. But you must admit we have a group of users who will read this and take it as a golden rule and start upsetting people. There are lots of wikipedians who simply do not read something as a whole. They just read MUST CITE and that is what they'll remember and they go on causing havoc. And it will be worse in other languages. More than once I had to stop somebody who was citing english wikipedia rules which are not applicable to nl.wikipedia.
Now you and me unfortunately do not have the time to watch the respective wikis we work on 24 hours a day. So before you start writing this down please thank about it.
Waerth/Walter
Walter van Kalken wrote:
Having citations is a good thing but please realise that it is best used when controversies arise.
Having citations is fundamental to being a good reference work.
This is *not* ( I say it again ) related to deletion policy.
SJ
Yes SJ you will be able to handle it in a correct matter and so will many other wikipedians. But you must admit we have a group of users who will read this and take it as a golden rule and start upsetting people. There are lots of wikipedians who simply do not read something as a whole. They just read MUST CITE and that is what they'll remember and they go on causing havoc. And it will be worse in other languages. More than once I had to stop somebody who was citing english wikipedia rules which are not applicable to nl.wikipedia.
Now you and me unfortunately do not have the time to watch the respective wikis we work on 24 hours a day. So before you start writing this down please thank about it.
It's always comforting to hear voices of reason. :-)
Unfortunately those who fixate on "MUST CITE" are also prone to ignore <big><bold>MUST USE COMMON SENSE</bold></big>. Sigh! Is a four word rule so much more difficult than a two word rule? ... even when you point out that there is an identical word in both?
Ec
Ray Saintonge wrote:
Walter van Kalken wrote:
Having citations is a good thing but please realise that it is best used when controversies arise.
Having citations is fundamental to being a good reference work.
This is *not* ( I say it again ) related to deletion policy.
SJ
Yes SJ you will be able to handle it in a correct matter and so will many other wikipedians. But you must admit we have a group of users who will read this and take it as a golden rule and start upsetting people. There are lots of wikipedians who simply do not read something as a whole. They just read MUST CITE and that is what they'll remember and they go on causing havoc. And it will be worse in other languages. More than once I had to stop somebody who was citing english wikipedia rules which are not applicable to nl.wikipedia.
Now you and me unfortunately do not have the time to watch the respective wikis we work on 24 hours a day. So before you start writing this down please thank about it.
It's always comforting to hear voices of reason. :-)
Unfortunately those who fixate on "MUST CITE" are also prone to ignore <big><bold>MUST USE COMMON SENSE</bold></big>. Sigh! Is a four word rule so much more difficult than a two word rule? ... even when you point out that there is an identical word in both?
Ec
My experience is that we have a group of users for whom using common sense unfortunately is beyond their grasp. You should know that by now. Many people want security in life and rules give security to them so they unfortunately fixate more on the rule part than common sense part :(
Waerth/Walter
On 12/2/05, Ray Saintonge saintonge@telus.net wrote:
On Fri, 2 Dec 2005, Gerard Meijssen wrote: Having citations is a good thing but please realise that it is best used when controversies arise.
SJ wrote: Having citations is fundamental to being a good reference work.
This is *not* ( I say it again ) related to deletion policy.
Walter van Kalken wrote:
Yes SJ you will be able to handle it in a correct matter and so will many other wikipedians.
The policy says: "The goal of Wikipedia is to become a complete and reliable encyclopedia. Verifiability is the key to becoming a reliable resource, so editors should cite credible sourceshttp://en.wikipedia.org/wiki/Wikipedia:Cite_sourcesso that their edits can be easily verified by readers and other editors."
I see no problem with citing to "the experience of anyone in the society." And although a controversy may arise - usually something so common will have citations available somewhere and smart people will be able to find them - as they did here. If not, as someone else said - appeal to common sense.
Jim
As Danny has repeatedly mentioned, normal published textbooks, including encyclopedias, have every single fact cited and checked before the publisher will go on with printing the book. These citations aren't made public, but they are done, nonetheless. Why should we be any different? This doesn't necessarily mean putting 1000 sources in the reference section. There are other options we can consider, or new ways of citing content online, that are different from the methods used in printed books.
Jim wrote:
On 12/2/05, Ray Saintonge saintonge@telus.net wrote:
On Fri, 2 Dec 2005, Gerard Meijssen wrote: Having citations is a good thing but please realise that it is best used when controversies arise.
SJ wrote: Having citations is fundamental to being a good reference work.
This is *not* ( I say it again ) related to deletion policy.
Walter van Kalken wrote:
Yes SJ you will be able to handle it in a correct matter and so will many other wikipedians.
The policy says: "The goal of Wikipedia is to become a complete and reliable encyclopedia. Verifiability is the key to becoming a reliable resource, so editors should cite credible sourceshttp://en.wikipedia.org/wiki/Wikipedia:Cite_sourcesso that their edits can be easily verified by readers and other editors."
I see no problem with citing to "the experience of anyone in the society." And although a controversy may arise - usually something so common will have citations available somewhere and smart people will be able to find them - as they did here. If not, as someone else said - appeal to common sense.
Jim _______________________________________________ foundation-l mailing list foundation-l@wikimedia.org http://mail.wikipedia.org/mailman/listinfo/foundation-l
Brian wrote:
As Danny has repeatedly mentioned, normal published textbooks, including encyclopedias, have every single fact cited and checked before the publisher will go on with printing the book. These citations aren't made public, but they are done, nonetheless. Why should we be any different? This doesn't necessarily mean putting 1000 sources in the reference section. There are other options we can consider, or new ways of citing content online, that are different from the methods used in printed books.
Like instead of having the references in and under the article have a seperate page like a talk page? And we just make "notes" which link to the references on that "references"page?
Waerth/Walter
On Sat, 3 Dec 2005, Walter van Kalken wrote:
section. There are other options we can consider, or new ways of citing content online, that are different from the methods used in printed books.
Like instead of having the references in and under the article have a seperate page like a talk page? And we just make "notes" which link to the references on that "references"page?
Precisely. An entire "references" page for each article, and an entire article for/about each reference. This will revolutionize collaborative research and analysis.
SJ
SJ wrote:
On Sat, 3 Dec 2005, Walter van Kalken wrote:
section. There are other options we can consider, or new ways of citing content online, that are different from the methods used in printed books.
Like instead of having the references in and under the article have a seperate page like a talk page? And we just make "notes" which link to the references on that "references"page?
Precisely. An entire "references" page for each article, and an entire article for/about each reference. This will revolutionize collaborative research and analysis.
Radical!
Ec
Walter van Kalken wrote:
Brian wrote:
As Danny has repeatedly mentioned, normal published textbooks, including encyclopedias, have every single fact cited and checked before the publisher will go on with printing the book. These citations aren't made public, but they are done, nonetheless. Why should we be any different? This doesn't necessarily mean putting 1000 sources in the reference section. There are other options we can consider, or new ways of citing content online, that are different from the methods used in printed books.
Like instead of having the references in and under the article have a seperate page like a talk page? And we just make "notes" which link to the references on that "references"page?
The page on which one chooses to put the quotes is only an aesthetic function. The important thing is that they are findable and public. The purpose of citations is to give the reader the opportunity to verify the data for himself. He can't do that if the citations are not public.
Ec
Ray Saintonge wrote:
Walter van Kalken wrote:
Brian wrote:
As Danny has repeatedly mentioned, normal published textbooks, including encyclopedias, have every single fact cited and checked before the publisher will go on with printing the book. These citations aren't made public, but they are done, nonetheless. Why should we be any different? This doesn't necessarily mean putting 1000 sources in the reference section. There are other options we can consider, or new ways of citing content online, that are different from the methods used in printed books.
Like instead of having the references in and under the article have a seperate page like a talk page? And we just make "notes" which link to the references on that "references"page?
Well, I think there are better options to consider. One post that was made here that has been pretty much ignored is linked below. I talked with brion about this, and he said that he thought it would be a big step in the right direction, although we should consider this option as more of a starting point for branching off ideas, rather than the final way it should be:
http://mail.wikimedia.org/pipermail/foundation-l/2005-December/005312.html
I especially like the option to include "cited text" and "paraphrase". So, as his sample image shows, we can essentially cite every bit of an article, thus becoming as sourced as any published book or encyclopedia, but better!.. because our sources are public, whereas with Britannica, you have to trust the word of the contributers (and the typing/proofreading abilities of their staff). I'm not sure if I like his "red box enclosing uncited text" scheme. Another possibility using this method would be to lightly highlight text that is not sourced in this way.
The page on which one chooses to put the quotes is only an aesthetic function. The important thing is that they are findable and public. The purpose of citations is to give the reader the opportunity to verify the data for himself. He can't do that if the citations are not public.
Ec
Agreed completely. By working to have every bit of our text not only cited, but to have their sources public, we would be moving beyond the verifiability of other encyclopedias.
brian0918
On 12/2/05, Brian brian0918@gmail.com wrote:
http://mail.wikimedia.org/pipermail/foundation-l/2005-December/005312.html
I especially like the option to include "cited text" and "paraphrase". So, as his sample image shows, we can essentially cite every bit of an article, thus becoming as sourced as any published book or encyclopedia, but better!.. because our sources are public, whereas with Britannica,
That would be more sourced than most published books or encyclos.
We can also track and show (as mouseover text?) the contributed-date and contributing-user for every word/set of words in an article. There are a few nuances in getting the attribution right, but we can at least offer a first-approximation without much work. Figuring out how to effectively store and render that information, is another question...
++SJ
SJ wrote:
On 12/2/05, Brian brian0918@gmail.com wrote:
http://mail.wikimedia.org/pipermail/foundation-l/2005-December/005312.html
I especially like the option to include "cited text" and "paraphrase". So, as his sample image shows, we can essentially cite every bit of an article, thus becoming as sourced as any published book or encyclopedia, but better!.. because our sources are public, whereas with Britannica,
That would be more sourced than most published books or encyclos.
We can also track and show (as mouseover text?) the contributed-date and contributing-user for every word/set of words in an article. There are a few nuances in getting the attribution right, but we can at least offer a first-approximation without much work. Figuring out how to effectively store and render that information, is another question...
This idea is way way overdoing it in my opinion and would surely chase people away. I feel for having citations moved elsewhere. I do not feel every letter needs to be referenced only the most important facts.
On the "references" for the soi article. I read them, both pages. And the article on wikitravel is : written by people like ourselves so how can we use it as a reference? And second the article on wikitravel definately contains wrong info. So I am going to remove that source.
The other source just shows how the word SOI is written in Thai. I do not really feel the need that such a thing would need to be referenced. But I will leave it in the article
Waerth/Walter
On Sat, 3 Dec 2005, Walter van Kalken wrote:
SJ wrote:
We can also track and show (as mouseover text?) the contributed-date and contributing-user for every word/set of words in an article. There are a few nuances in getting the attribution right, but we can at least offer a first-approximation without much work. Figuring out how to effectively store and render that information, is another question...
This idea is way way overdoing it in my opinion and would surely chase people away. I feel for having citations moved elsewhere. I do not feel every letter needs to be referenced only the most important facts.
It's not a matter of 'chasing people away' - leave such a feature off by default; it wil reduce the processing problems. But tracking who added particular chunks of text is quite literally providing the latest reference (and the only reference we can automatically guarantee to be accurate within the site).
On the "references" for the soi article. I read them, both pages. And the article on wikitravel is : written by people like ourselves so how can we use it as a reference? And second the article on wikitravel definately contains wrong info. So I am going to remove that source.
"People like ourselvs" -- I love it. And what other kind of people are there? So you're going to remove the source because you disagree with it? Even though you have no other source?
The Wikitravel page is another world-readable source discussing the same topic in English -- by people who may be just as 'qualified' to opine about Sois as you are. If you disagree with that source, you should at least note that people are talking about this subject on other sites.
The other source just shows how the word SOI is written in Thai. I do not
It helps demonstrate that the word exists; which is a start (since we don't have a wiktionary entry about it yet).
SJ
SJ wrote:
On 12/2/05, Brian brian0918@gmail.com wrote:
http://mail.wikimedia.org/pipermail/foundation-l/2005-December/005312.html
I especially like the option to include "cited text" and "paraphrase". So, as his sample image shows, we can essentially cite every bit of an article, thus becoming as sourced as any published book or encyclopedia, but better!.. because our sources are public, whereas with Britannica,
That would be more sourced than most published books or encyclos.
We can also track and show (as mouseover text?) the contributed-date and contributing-user for every word/set of words in an article. There are a few nuances in getting the attribution right, but we can at least offer a first-approximation without much work. Figuring out how to effectively store and render that information, is another question...
"King Disrupters" do not thrive by shying away from ambitious projects. :-)
I haven't a clue about the technical feasibility of this idea. Since this seems like a move from the presentation of the world's knowledge to the integration of the world's knowledge. That would certainly represent a fulfilment of freeing the world's knowledge in the sense that Jimbo expressed in his Wikimania speech. Every quantum of knowledge would be connected with every other quantum of knowledge.
The first reference approximation involves nothing more than adding broad references in a separate section near the end of an article. Annotations or "foot"notes whcich relate to discrete segments of a text are really a second approximation. The main text must be readable to the average reader whose interests do not require in depth study; having unwanted pop-up mouse-overs does not endear the reader to any project.
As Gerard pointed out we also need to keep connected with the average contributor, to who we already owe much. He still has a lot to say and a lot to offer. We would be delighted if he could provide full bibliographical support for his own material, but a more realistic expectation would be a single link. we don't want to fall into soome new form of academic elitism.
I can now understand why some material may need to be kept unavailable to the public through us. If we are going to connect everything that must include copyright material which can only be caged in inaccesible archives until it can be set free.
Ec
As Gerard pointed out we also need to keep connected with the average contributor, to who we already owe much. He still has a lot to say and a lot to offer. We would be delighted if he could provide full bibliographical support for his own material, but a more realistic expectation would be a single link. we don't want to fall into soome new form of academic elitism.
Ec, thanks for that one ... I suppose you understood also what I mean - I am not against giving references (I add them when I have them here), but I am not willing (or better I really do not have the time) to re-research whatever I learnt ... and also: many things come out of stuff I translated - sometimes when I translate things they are not for the public in that moment (new machinery developments, sometimes press releases that are prepared beforehand etc. - this is impossible to be followed up and also to be cited as source - these, in the moment I work on it, must be kept secret and I would never give that information out before you can't read about it in other places) and that after years are just "normal" things people talk about every day. This is valid for many of us.
I can now understand why some material may need to be kept unavailable to the public through us. If we are going to connect everything that must include copyright material which can only be caged in inaccesible archives until it can be set free.
Well, this is one of those things ... some poeple would give parts of their copyrighted and non GFDL material to a portal if they may decide any license for it. This means: somehwere on the web, we should have such a place. There are for example authors that allow for publication of their work, but they don't want it under GFDL - so what would be wrong with such a place? It would make sure that things are collected in "one library" and the when things become copyright free it can simply be transferred to the right projects. And: it would be much easier to connect things.
Ciao, Sabine
___________________________________ Yahoo! Messenger: chiamate gratuite in tutto il mondo http://it.messenger.yahoo.com
Brian wrote:
Ray Saintonge wrote:
Walter van Kalken wrote:
Brian wrote:
As Danny has repeatedly mentioned, normal published textbooks, including encyclopedias, have every single fact cited and checked before the publisher will go on with printing the book. These citations aren't made public, but they are done, nonetheless. Why should we be any different? This doesn't necessarily mean putting 1000 sources in the reference section. There are other options we can consider, or new ways of citing content online, that are different from the methods used in printed books.
Like instead of having the references in and under the article have a seperate page like a talk page? And we just make "notes" which link to the references on that "references"page?
Well, I think there are better options to consider. One post that was made here that has been pretty much ignored is linked below. I talked with brion about this, and he said that he thought it would be a big step in the right direction, although we should consider this option as more of a starting point for branching off ideas, rather than the final way it should be:
http://mail.wikimedia.org/pipermail/foundation-l/2005-December/005312.html
I especially like the option to include "cited text" and "paraphrase". So, as his sample image shows, we can essentially cite every bit of an article, thus becoming as sourced as any published book or encyclopedia, but better!.. because our sources are public, whereas with Britannica, you have to trust the word of the contributers (and the typing/proofreading abilities of their staff). I'm not sure if I like his "red box enclosing uncited text" scheme. Another possibility using this method would be to lightly highlight text that is not sourced in this way.
The page on which one chooses to put the quotes is only an aesthetic function. The important thing is that they are findable and public. The purpose of citations is to give the reader the opportunity to verify the data for himself. He can't do that if the citations are not public.
Ec
Agreed completely. By working to have every bit of our text not only cited, but to have their sources public, we would be moving beyond the verifiability of other encyclopedias.
brian0918
Hoi, And indeed, we would drive many people . I understand a wish for sources but by putting red boxes and what have you around what has not been sourced you make us into something what we are not. We would be as expert as what Larry Sanger wants in his new project. Where is the difference between Nupedia and Wikipedia ?
Have a separate page for your sources, have a talk page for discussion but leave the text proper alone. If you want to keep Wikipedia accessible this is the entry level where people are to be welcome. When something is not sourced, it does not give you the right to delete without prior verification. That is; you have to prove what is wrong as in you have to assume the good faith of the contributor. When you do you have the sources to prove your point you can change it like you always could but you now do it with the conviction of having sources to back it up.
At some stage there was a BIG row in the Dutch wikipedia; there was a guy, who got banned and everything, who pushed a particular point of view. He quoted sources. His sources were in a language where he was the only one who could read it, it was in Tamazingh. His sources were refused because of it and because of being extremely different from accepted knowledge. Now given that history is written by the victor, and given that this is particularly POV, having sources written by the victor make many historical sources suspect. So how are you going to square what you think you know about history an what can be "proven" by English language sources and what can be "proven" by sources of a different origin? By the way the Tamazingh POV was certainly not a victors POV.
This blind rush into trusting sources.. Please think what you try to do and be aware of the fall out. The attitude that only sourced information is good and reliable assumes no good faith. It assumes that a professional approach is best. It makes us into a Nupedia where we were once open and free.
Thanks, GerardM
Hoi, Re-iterating what has been said does not make for understanding. By being inflexible and apparently not understanding what concerns there are on the other end of an issue. You entrench your position.
I will tell you why we are different. We are different because we are not a professionally produced publication. We do not have the pretence that we know it all. We invite everyone to contribute and these other publications do not. When you, and here I repeat myself, do not care for the people who have never cited anything and make them out to be wrong as a consequence, you do not understand why we are different.
Again, citations have their place but do not quote your sources relating subjects like religion because they are all hopelessly point of view. The only thing you can achieve there is that you show that someone published a point of view. Citations and sources are a double edged sword, they either allow you to clear up disputes or they allow you to destroy what differentiates Wikipedia from these "normal published textbooks, including encyclopaedias".
So "obviously" you must be right because you quotes your source.. Then again if this is the case, what did you add to the discussion ? And what makes your source the authority that makes the difference ?
Thanks, GerardM
Brian wrote:
As Danny has repeatedly mentioned, normal published textbooks, including encyclopedias, have every single fact cited and checked before the publisher will go on with printing the book. These citations aren't made public, but they are done, nonetheless. Why should we be any different? This doesn't necessarily mean putting 1000 sources in the reference section. There are other options we can consider, or new ways of citing content online, that are different from the methods used in printed books.
Jim wrote:
On 12/2/05, Ray Saintonge saintonge@telus.net wrote:
On Fri, 2 Dec 2005, Gerard Meijssen wrote: Having citations is a good thing but please realise that it is best used when controversies arise.
SJ wrote: Having citations is fundamental to being a good reference work.
This is *not* ( I say it again ) related to deletion policy.
Walter van Kalken wrote:
Yes SJ you will be able to handle it in a correct matter and so will many other wikipedians.
The policy says: "The goal of Wikipedia is to become a complete and reliable encyclopedia. Verifiability is the key to becoming a reliable resource, so editors should cite credible sourceshttp://en.wikipedia.org/wiki/Wikipedia:Cite_sourcesso that their edits can be easily verified by readers and other editors."
I see no problem with citing to "the experience of anyone in the society." And although a controversy may arise - usually something so common will have citations available somewhere and smart people will be able to find them - as they did here. If not, as someone else said - appeal to common sense.
Jim
Gerard Meijssen wrote:
I will tell you why we are different. We are different because we are not a professionally produced publication. We do not have the pretence that we know it all. We invite everyone to contribute and these other publications do not. When you, and here I repeat myself, do not care for the people who have never cited anything and make them out to be wrong as a consequence, you do not understand why we are different.
This is absolutely fundamental to what we are doing!
Again, citations have their place but do not quote your sources relating subjects like religion because they are all hopelessly point of view. The only thing you can achieve there is that you show that someone published a point of view. Citations and sources are a double edged sword, they either allow you to clear up disputes or they allow you to destroy what differentiates Wikipedia from these "normal published textbooks, including encyclopaedias".
In other words citations should be used to help the reader make up his own mind, not as tools for making up his mind for him.
Jim wrote:
The policy says: "The goal of Wikipedia is to become a complete and reliable encyclopedia. Verifiability is the key to becoming a reliable resource, so editors should cite credible sourceshttp://en.wikipedia.org/wiki/Wikipedia:Cite_sourcesso that their edits can be easily verified by readers and other editors."
Sometimes this kind of debate seems to be between those who understand rules without reading them and those who read rules without understanding them.
Ec
Gerard Meijssen wrote:
I find it remarkable that this CITE thing may break one of our fundamental reasons for success. It will drive people away when it is used as a tool to justify deletions. Why will it drive people away, because it raises the entry level to a point where many people will just no longer bother to do something. Yes, there may be an EDIT button, but without a citation it will be gone.
Having citations is a good thing but please realise that it is best used when controversies arise. It is then part of the instrumentarium to come to a NPOV. When it used on its own I expect it to prove disastrous for the influx of new people into our community. Certainly if this is deemed to be necessary to be followed by all the projects that do not have the maturity of the English language wikipedia.
So, when you in the en.wikipedia decide to have this new deletion policy, consider the fallout it may have elsewhere.
This is the back to basics argument. It's amazing how well it works in so many different circumstances.
Success is a poison pill. It can be deadly if you use it without understanding the instructions.
Ec
SJ wrote:
On 11/30/05, Walter van Kalken walter@vankalken.net wrote:
The Soi case.
On the Dutch wikipedia I wrote the article [[soi]]. It was also written on the english wikipedia (http://en.wikipedia.org/wiki/Soi)
According to the CITE principal this article should be deleted right now
Actually, a quick google search turned up a few relevant citations (one from Wikitravel, in fact). I added them to the English article. It is very very hard to find articles which are at once encyclopedic and verifiable, but have no suitable reference either online or in print. I cannot think of an example atm.
This is the things-we-take-for-granted category of articles. We may dispute why they were done that way, but it's unquestionable that it's the way it's done.
In North America house numbering is even on one side of the road and odd on the other. In some places the numbers start from 1 where the road begins, and in others it conforms to a predetermined grid of "100" blocks. It took a while to understand when I was looking for an address in London that the numbers went up one side to the end and down the other. Tokyo took a different kind of adjustment when I found the numbers going around the block.
We become accustomed to the way things are done in our home towns, and never think of needing citations to prove it.
Ec
Walter van Kalken wrote:
Yes I agree with the referencing point. But in some articles on en: all sentences receive external links. Which is overdoing it imho. An article should have general references in the reference section. Like Books x,y,z, etc, websites a,b,c, etc. But not a reference for every sentence. That is overdoing it.
It depends on the article. Mostly those will be highly contentious articles where the content of the sentences was in fact questioned. See the talk page.
(I have occasionally written defensively on matters of controversy between the participants, referencing each phrase.)
This isn't panicking. This is doing what we should have been doing all along, and treating our article content with the same importance that we treat our image content. This is a step in the right direction.
It is panicking if I watch the deletion list on en: On which a perfectly legit article by a highly regarded nl: contributor was put up dor deletion by someone who doesn't know anything about the subject but Quotes CITE and some other policies.
Another nomination from ignorance, apparently. I'm not surprised there are now webcomics about the way deletion works on en: in practice.
The user wasn't even explained on
his own page why it was put up for deletion.
I floated this idea on AFD's talk page. Apparently the regulars consider it too much effort.
- d.
On 11/30/05 12:19 AM, "Andrew Lih" andrew.lih@gmail.com wrote:
On 11/30/05, Brian brian0918@gmail.com wrote:
In light of the recent USA Today article:
In the same way that we are currently enforcing proper image tags using a bot, could we do the same with unsourced articles? Start out by placing {{unsourced}} in all the articles lacking sources, and then, if it is not sourced in a week, create something like the {{copvio}} page-replacer to hide the unsourced content (the entire article), explaining with a detailed message that the article must be thoroughly sourced.
In my mind, at least, it doesn't seem like there should be any difference between enforcing sources for images and sources for articles. If anything we should be enforcing the latter more, since articles form the basis of the encyclopedia. I know this won't solve everything, but I think it should be a vital part of Wikipedia; since we do not know who edits an article, we need to know that it is based on information that we can verify ourselves.
It's an interesting idea, but it would apply to so many articles on such a coarse grained way it may wind up being ignored. Perhaps such a function could be tied to the TOC, so that each section would have a flag of whether it had been sourced to satisfaction.
-Andrew (User:Fuzheado)
Doubt it would be ignored, you can cite and then be accused of doing original research, how fun.
I had a cited source for John Dean's article and the minute John Dean cried about it in his column on Salon.com it was immediately changed anyway. I think CITE is a start but it needs to be fleshed out more because even amongst sources you have varying degrees of reliability.
--Guy (User:Wgfinley, still on Wikibreak because of junk like this)
Brian wrote:
In the same way that we are currently enforcing proper image tags using a bot, could we do the same with unsourced articles? Start out by placing {{unsourced}} in all the articles lacking sources, and then, if it is not sourced in a week, create something like the {{copvio}} page-replacer to hide the unsourced content (the entire article), explaining with a detailed message that the article must be thoroughly sourced.
I don't know about enforcing it ... but, on en: at least, the {{unreferenced}} tag (which {{unsourced}} redirects to) is there for anyone who wants to use them. I put it on articles I happen across that are (a) not a stub (b) don't have even an external link somewhere.
In my mind, at least, it doesn't seem like there should be any difference between enforcing sources for images and sources for articles. If anything we should be enforcing the latter more, since articles form the basis of the encyclopedia. I know this won't solve everything, but I think it should be a vital part of Wikipedia; since we do not know who edits an article, we need to know that it is based on information that we can verify ourselves.
It's hard to *enforce*. It's already good practice per [[:en:Wikipedia:Verifiability]].
- d.
David Gerard wrote:
Brian wrote:
In the same way that we are currently enforcing proper image tags using a bot, could we do the same with unsourced articles? Start out by placing {{unsourced}} in all the articles lacking sources, and then, if it is not sourced in a week, create something like the {{copvio}} page-replacer to hide the unsourced content (the entire article), explaining with a detailed message that the article must be thoroughly sourced.
I don't know about enforcing it ... but, on en: at least, the {{unreferenced}} tag (which {{unsourced}} redirects to) is there for anyone who wants to use them. I put it on articles I happen across that are (a) not a stub (b) don't have even an external link somewhere.
I agree about that, but I think the templates currently aren't taken too seriously. People will be motivated to cite/reference with this proposed system, or something like it. And it seems like something that we should automatically be requiring. Peer-reviewed encyclopedias do not require this, because they have trusted, known contributors. Our contributors are not known in anything like the same sense, so we have to show the public that our content is based on verifiable information.
In my mind, at least, it doesn't seem like there should be any difference between enforcing sources for images and sources for articles. If anything we should be enforcing the latter more, since articles form the basis of the encyclopedia. I know this won't solve everything, but I think it should be a vital part of Wikipedia; since we do not know who edits an article, we need to know that it is based on information that we can verify ourselves.
It's hard to *enforce*. It's already good practice per [[:en:Wikipedia:Verifiability]].
I don't think it would be too hard to enforce. To start, a bot can look for any external links in the article, or for use of one of the common reference templates. Either that, or search for a section called "References", "Sources" or "External links". To enforce image tagging, we have appropriate image templates which are flagged/ignored by a bot. We need only do the same thing with article templates.
Brian wrote:
I don't think it would be too hard to enforce. To start, a bot can look for any external links in the article, or for use of one of the common reference templates. Either that, or search for a section called "References", "Sources" or "External links". To enforce image tagging, we have appropriate image templates which are flagged/ignored by a bot. We need only do the same thing with article templates.
OK, a semi-automatic bot-assisted {{unreferenced}}-tagging. That'll find them. We can then run "find the sources" drives. Sound workable?
- d.
David Gerard wrote:
Brian wrote:
I don't think it would be too hard to enforce. To start, a bot can look for any external links in the article, or for use of one of the common reference templates. Either that, or search for a section called "References", "Sources" or "External links". To enforce image tagging, we have appropriate image templates which are flagged/ignored by a bot. We need only do the same thing with article templates.
OK, a semi-automatic bot-assisted {{unreferenced}}-tagging. That'll find them. We can then run "find the sources" drives. Sound workable?
I like the idea very much. It will encourage people to not only provide citations for unsourced articles, but make it a general part of everyday activity.
Brian wrote:
David Gerard wrote:
Brian wrote:
In the same way that we are currently enforcing proper image tags using a bot, could we do the same with unsourced articles? Start out by placing {{unsourced}} in all the articles lacking sources, and then, if it is not sourced in a week, create something like the {{copvio}} page-replacer to hide the unsourced content (the entire article), explaining with a detailed message that the article must be thoroughly sourced.
I don't know about enforcing it ... but, on en: at least, the {{unreferenced}} tag (which {{unsourced}} redirects to) is there for anyone who wants to use them. I put it on articles I happen across that are (a) not a stub (b) don't have even an external link somewhere.
I agree about that, but I think the templates currently aren't taken too seriously. People will be motivated to cite/reference with this proposed system, or something like it. And it seems like something that we should automatically be requiring. Peer-reviewed encyclopedias do not require this, because they have trusted, known contributors. Our contributors are not known in anything like the same sense, so we have to show the public that our content is based on verifiable information.
Templates aren't taken too seriously because there are too many of them. There is no motivation in your scheme. If your {{unsourced}} template is only going on the articles most people won't notice it, and won't notice anything is wrong until they have need to visit the article only to find the information is gone.
Why shouldn't peer reviewed encyclopedias be put to the same standard. What makes our long standing contributors any less trusted? Many of us do not accept this kind of elitism.
Ec
The problem is that the bulk of our content has no explicit source. It may be correct and unchallenged, but no source is listed. Additionally, a slightly more sophisticated level of vandalism would simply list a spurious source. Beyond that lies referencing a source which is known to be in error.
Fred
On Nov 29, 2005, at 10:58 PM, Brian wrote:
In light of the recent USA Today article:
In the same way that we are currently enforcing proper image tags using a bot, could we do the same with unsourced articles? Start out by placing {{unsourced}} in all the articles lacking sources, and then, if it is not sourced in a week, create something like the {{copvio}} page-replacer to hide the unsourced content (the entire article), explaining with a detailed message that the article must be thoroughly sourced.
In my mind, at least, it doesn't seem like there should be any difference between enforcing sources for images and sources for articles. If anything we should be enforcing the latter more, since articles form the basis of the encyclopedia. I know this won't solve everything, but I think it should be a vital part of Wikipedia; since we do not know who edits an article, we need to know that it is based on information that we can verify ourselves.
brian0918 _______________________________________________ foundation-l mailing list foundation-l@wikimedia.org http://mail.wikipedia.org/mailman/listinfo/foundation-l
Fred Bauder wrote:
The problem is that the bulk of our content has no explicit source. It may be correct and unchallenged, but no source is listed. Additionally, a slightly more sophisticated level of vandalism would simply list a spurious source. Beyond that lies referencing a source which is known to be in error.
It's not necessary to source *everything*. Paris is the capital of France, everyone knows that. We don't need to cite some obscure French governmental listing which proves this.
What does need to be cited are facts which are not common knowledge (like statistics) and opinions which must be attributed.
I've now started removing any additions to pages I have on my watchlist which I do not think are common knowledge and have no source for the claims, and I have asked contributors to cite where they are getting this information from.
Chris
Good move, but take it easy.
Fred
On Nov 30, 2005, at 11:07 AM, Chris Jenkinson wrote:
I've now started removing any additions to pages I have on my watchlist which I do not think are common knowledge and have no source for the claims, and I have asked contributors to cite where they are getting this information from.
Chris
Brian wrote:
In light of the recent USA Today article:
What article? Would it not be better if you followed your own rules about citing something before you go around spreading panic.
In the same way that we are currently enforcing proper image tags using a bot, could we do the same with unsourced articles? Start out by placing {{unsourced}} in all the articles lacking sources, and then, if it is not sourced in a week, create something like the {{copvio}} page-replacer to hide the unsourced content (the entire article), explaining with a detailed message that the article must be thoroughly sourced.
This is absolutely unrealistic. The image tagging project has been going on for perhaps a year (?), and it's gradually getting to where we want it. No reliable editor opposes the idea of having pages properly referenced, and ever since having referenced pages became the norm there has been great progress in that area. Now because of one alleged newspaper article you expect several hundred thousand articles to be repaired in a week. If not you will on your own initiative have a bot go through and delete any material that YOU find inappropriate.
Perhaps you should start by showing a little trust in your fellow editors rather than begin with a series of hostile enforcement actions.
In my mind, at least, it doesn't seem like there should be any difference between enforcing sources for images and sources for articles. If anything we should be enforcing the latter more, since articles form the basis of the encyclopedia. I know this won't solve everything, but I think it should be a vital part of Wikipedia; since we do not know who edits an article, we need to know that it is based on information that we can verify ourselves.
Although I agree that the text in articles is more important than the images, which are mostly only there to make things look pretty, I also think that two years might be a more realistic time frame for cleaning this up.
Ec
I think there is a good deal of un-necessary panic going around about this USA Today thing. Did it really tell us anything new? No. Misinformation in rarely visited articles has ALWAYS been a problem.
It's also not nearly as amenable to any kind of automated process as image tagging.
Let's not run around like headless chickens because some journalist found that conspiracy-theorist things had been put into his Wikipedia article.
Fact is: Wikipedia's improving. At quite a tremendous rate, in fact. It's easy to forget (a) the magnitude of the task, and (b) how bad we were even a year ago.
-Matt
I insist that this problem is not different from the problem of Newspapers or Magazines.
Wherever there is mass-diffusion, there is a risk of disinformation.
Matt Brown wrote:
I think there is a good deal of un-necessary panic going around about this USA Today thing. Did it really tell us anything new? No. Misinformation in rarely visited articles has ALWAYS been a problem.
It's also not nearly as amenable to any kind of automated process as image tagging.
Let's not run around like headless chickens because some journalist found that conspiracy-theorist things had been put into his Wikipedia article.
Fact is: Wikipedia's improving. At quite a tremendous rate, in fact. It's easy to forget (a) the magnitude of the task, and (b) how bad we were even a year ago.
-Matt _______________________________________________ foundation-l mailing list foundation-l@wikimedia.org http://mail.wikipedia.org/mailman/listinfo/foundation-l
Matt Brown wrote:
I think there is a good deal of un-necessary panic going around about this USA Today thing. Did it really tell us anything new? No. Misinformation in rarely visited articles has ALWAYS been a problem.
Yes, but it doesn't mean that we shouldn't do anything about it. We know about it, now let's fix it.
It's also not nearly as amenable to any kind of automated process as image tagging.
What is wrong with my suggestion? Just because it doesn't work as well as image tagging doesn't mean that we shouldn't bother trying it at all.
Let's not run around like headless chickens because some journalist found that conspiracy-theorist things had been put into his Wikipedia article.
That journalist writes for the most read paper in the US. He tells everyone that Wikipedia is useless. People read that and believe that Wikipedia is useless. This is a fact that we have to face, not ignore out of pride.
Fact is: Wikipedia's improving. At quite a tremendous rate, in fact. It's easy to forget (a) the magnitude of the task, and (b) how bad we were even a year ago.
We can easily get to a point in time when we will be overwhelmed by the number of editors and new content, and are wishing that we had taken some sort of measures early on, rather than sit back and boast about improvements.
Brian wrote:
Matt Brown wrote:
I think there is a good deal of un-necessary panic going around about this USA Today thing. Did it really tell us anything new? No. Misinformation in rarely visited articles has ALWAYS been a problem.
Yes, but it doesn't mean that we shouldn't do anything about it. We know about it, now let's fix it.
Fixing it is not the same as burying it.
Let's not run around like headless chickens because some journalist found that conspiracy-theorist things had been put into his Wikipedia article.
That journalist writes for the most read paper in the US. He tells everyone that Wikipedia is useless. People read that and believe that Wikipedia is useless. This is a fact that we have to face, not ignore out of pride.
He's not the first one to say that Wikipedia is "useless", and I'm sure he won't be the last. If the critic makes specific observations enough will notice to do something about those points. We will never silence the ones who just make general comments no matter how good we get. As long as we threaten to undermine their vested interests they will continue the criticism.
Fact is: Wikipedia's improving. At quite a tremendous rate, in fact. It's easy to forget (a) the magnitude of the task, and (b) how bad we were even a year ago.
We can easily get to a point in time when we will be overwhelmed by the number of editors and new content, and are wishing that we had taken some sort of measures early on, rather than sit back and boast about improvements.
So who's sitting back and boasting? Those improvements have happened. Accepting that does not mean that everybody stops making improvements. Editors just need to stick to what they're good at instead of worrying about what they feel others are not doing.
Ec
On Wed, 30 Nov 2005, Matt Brown wrote:
I think there is a good deal of un-necessary panic going around about this USA Today thing. Did it really tell us anything new? No. Misinformation in rarely visited articles has ALWAYS been a problem.
I beg to differ : this highlighted one or two new things.
1) Wikipedia is important enough to significantly injure the reputations of people/organizations that are known elsewhere on the web. It may provide 2 or more of the top 10 search results for content about them. An old problem, but getting worse with increased popularity. This is the most egregious case I have seen.
2) While misinformation in rarely-visited articles has always been a problem, yes, and seems almost characterizable -- at least, WP editors are able to pick out potentially troublesome articles at a glance. Truly offensive and dangerous misinformation should be easiest of all to characterize. Nevertheless, the biggest and most serious reusers of static copies of WP content do not currently differentiate between possibly-problematic and probably-unproblematic material. This magnifies the impact of the problem, and can confuse readers who are used to looking for more than one independent sources, as 'independence' gets harder to identify.
3) This article was linked to from other pages. It was noticed and wikified. At this stage, something could have been done about it. If the first page of our style guide included not only "how to bold the topic" and "how to link out", but also "how to flag emotional or controversial content" and "how to mark short, unsourced articles" -- this might have been noticed and fixed during the ensuing three months.
It's also not nearly as amenable to any kind of automated process as image tagging.
Why not? Requiring a 'references' section for every article (thanking the heavens that WP is not paper), and reminding every editor that *every* new article should come with at least one reference, seems a responsible thing to do. Can you offer a reason not to have such a section for any article?
If you're writing about one of those topics that is a) not private research/analysis of your own, but b) has never been written about anywhere else [that you know of], then we need a new class of references : "personal observation by [user]", with a relevant tag not unlike the original-reporting templates used on Wikinews. Then it will be crystal clear that readers should visit your page, and see whether they trust you as the primary/original observer/author.
Fact is: Wikipedia's improving. At quite a tremendous rate, in fact. It's easy to forget (a) the magnitude of the task, and (b) how bad we were even a year ago.
The responsibilities of being widely-read are growing at the same rate, if not faster.
SJ
On 12/1/05, SJ 2.718281828@gmail.com wrote:
- Wikipedia is important enough to significantly injure the reputations
of people/organizations that are known elsewhere on the web. It may provide 2 or more of the top 10 search results for content about them. An old problem, but getting worse with increased popularity. This is the most egregious case I have seen.
Perhaps I simply considered this 'old news' more than many.
- This article was linked to from other pages. It was noticed and
wikified. At this stage, something could have been done about it. If the first page of our style guide included not only "how to bold the topic" and "how to link out", but also "how to flag emotional or controversial content" and "how to mark short, unsourced articles" -- this might have been noticed and fixed during the ensuing three months.
Good point. Wikification should start including such things, as well as formatting fixes.
I think it's also the case that we should be especially careful about unattributed claims about living persons. It's the topic where misinformation can be most hurtful.
Why not? Requiring a 'references' section for every article (thanking the heavens that WP is not paper), and reminding every editor that *every* new article should come with at least one reference, seems a responsible thing to do. Can you offer a reason not to have such a section for any article?
I meant that more than that is hard to automate. You're right that the bare minimals can be easily checked. However, I can't see that we can automate much beyond that with ease.
If you're writing about one of those topics that is a) not private research/analysis of your own, but b) has never been written about anywhere else [that you know of], then we need a new class of references : "personal observation by [user]", with a relevant tag not unlike the original-reporting templates used on Wikinews. Then it will be crystal clear that readers should visit your page, and see whether they trust you as the primary/original observer/author.
Much of this falls under 'original research', doesn't it? Or are you talking about the cases where someone believes that something is true but doesn't have the references to hand?
-Matt
If you're writing about one of those topics that is a) not private research/analysis of your own, but b) has never been written about anywhere else [that you know of], then we need a new class of references : "personal observation by [user]", with a relevant tag not unlike the original-reporting templates used on Wikinews. Then it will be crystal clear that readers should visit your page, and see whether they trust you as the primary/original observer/author.
Much of this falls under 'original research', doesn't it? Or are you talking about the cases where someone believes that something is true but doesn't have the references to hand?
I am surprised that both you SJ and Mart are saying this. Obviously both off you didn't read my case about the soi article on english wikipedia. Which is an article like many which is almost impossibly referenced. Since the two of you never read it I'll post it again here. And according to the 2 of you this article which is perfectly well the truth should be deleted because it is either not referenced or "personal research" ?
Waerth/Walter
The Soi case.
On the Dutch wikipedia I wrote the article [[soi]]. It was also written on the english wikipedia (http://en.wikipedia.org/wiki/Soi)
According to the CITE principal this article should be deleted right now as it is not from a book nor a website, but partly from my own and any Thais experience. Even though it contains pure facts which are the truth it doesn't cite anything as no sources are used. Note I didn't write the english article but it basicaly says the same as the Dutch article.
Facts: /*1) Soi*/ (Thai http://en.wikipedia.org/wiki/Thai_alphabet: *???*) is the term used in Thailand http://en.wikipedia.org/wiki/Thailand for a side-street branching off a major street (/thanon http://en.wikipedia.org/w/index.php?title=Thanon&action=edit/). An alley is called a /Trok http://en.wikipedia.org/w/index.php?title=Trok&action=edit/.
All true but do not have it from any book.
2) Sois are usually numbered, and are referred to by the name of the major street and the number, as in "Sukhumvit Soi 4",
Also please come and have a look but not from any book
3) All the even-numbered sois are on one side of the street, the odd-numbered ones on the other.
How do you want me to prove that from citations without flying you over here?
4) If for instance a soi is added between soi 7 and soi 9 it will get the name soi 7/1, the next one soi 7/2 etc.
Again fact new soi was added between 2 sois near my home got the 6/1 number as it was between sois 6 and 8. No sources sorry
5) Almost all sois also have a name.
I first wanted to be bolder as I have never seen a soi that doesn't also have a name. But I decided to be a bit less sure about it
6) On lower Sukhumvit road in Bangkok for instance the soi's are named after important landowners or families of landowners who had land in the area in the past.
Verified with Thais whose families actually come from these landowners and have been living in the area for a long time. Again no books.
7) Some sois become major thoroughfares and because of that get known by their name only. Examples are Thonglor (Sukhumvit soi 55), Asoke (Sukhumvit soi 21) etc.
Again plain fact. Books anyone?
So according to the CITE offensive this article and many like this should be deleted even though there is nothing that is not the truth in there. Is it original research? I do not think so. You see the problem is when you start this CITE offensive one day one of our "police" officers will find this article and put it up for deletion because there are no sources. People take these so called guidelines much to strict and cause damage with it.
This is why I would say be stricter on anons and newbies and maybe make some kind of ranking of wikipedians that are known to work in good faith or that are known IRL. Like me, anyone can always call me (see my phonenumbers) I even put my homeaddress up once. I might not always be able to quote directly where I got something from. But I am handling in good faith.
And with me there are many people like me.
On 12/1/05, Walter van Kalken walter@vankalken.net wrote:
If you're writing about one of those topics that is a) not private research/analysis of your own, but b) has never been written about anywhere else [that you know of], then we need a new class of references : "personal observation by [user]", with a relevant tag not unlike the original-reporting templates used on Wikinews. Then it will be crystal clear that readers should visit your page, and see whether they trust you as the primary/original observer/author.
Walter, please read what I wrote carefully. The kind of article you wrote should surely be kept in Wikipedia, but it should also clearly identify you, Walter van Kalken, Wikipedia editor, as the source -- in a template that clarifies which of various reasons led to the lack of external links. From least to most like OR : 1) printed references exist, but the author couldn't find a cite at the moment 2) the author knows of no printed references; personal communication with (named source/authority) 3) the author doesn't know of any printed references; direct personal observation (simple observation, similar to a photo; not 'research') 4) the author doesn't know of any printed references; 'common knowledge' or 'common sense'
Much of this falls under 'original research', doesn't it? Or are you talking about the cases where someone believes that something is true but doesn't have the references to hand?
Not necessarily. It is important to distinguish between 'research' and 'observation'. When I add a photo I took of a U2 concert, and say "U2 concert in Boston, December 2005", that's personal observation. The image page tells you that [[User:Sj]] took the photo and uploaded it on a given date. Likewise for articles, in the very-rare case that you've learned about a fact not noted elsewhere online (part of an article); or a topic which is /referenced/ in many places but not /explained/ anywhere online (a new article), perhaps first-hand from an expert. In practice, I think the number of such topics is miniscule. For instance...
I am surprised that both you SJ and Mart are saying this. Obviously both off you didn't read my case about the soi article on english wikipedia. Which is an article like many which is almost impossibly referenced.
A minute's googling turned up at least two suitable references; I added them to the English article. If I knew any Thai, I'm certain I could come up with more in that language.
Since the two of you never read it I'll post it again here.
Don't get your undies in a bundle.
- All the even-numbered sois are on one side of the street, the
odd-numbered ones on the other.
How do you want me to prove that from citations without flying you over here?
That's tough to find by searching online. Nvertheless, I'm certain there are references that discuss simple street numbering, in all manner of languages; even guidelines for numbering new streets in Thai.
- If for instance a soi is added between soi 7 and soi 9 it will get
the name soi 7/1, the next one soi 7/2 etc.
Ditto.
- On lower Sukhumvit road in Bangkok for instance the soi's are named
after important landowners or families of landowners who had land in the area in the past.
Verified with Thais whose families actually come from these landowners and have been living in the area for a long time. Again no books.
Just name individual named-sois, and perhaps link to references that mention that name for one or two of the major ones.
- Some sois become major thoroughfares and because of that get known by
their name only. Examples are Thonglor (Sukhumvit soi 55), Asoke (Sukhumvit soi 21) etc.
Again plain fact. Books anyone?
Again, just link to sources that refer to these by their name (ideally also mentioning their soi number as well)
or that are known IRL. Like me, anyone can always call me (see my phonenumbers) I even put my homeaddress up once. I might not always be
Right. Which is why, if the author can't find a source (generally one will exist, /somewhere/), there should be a standard way to add a 'reference' bullet to an article that explains the author should be treated as the source until a better one is found... and listing a reason that no other source is provided.
++SJ
Matt Brown wrote:
On 12/1/05, SJ 2.718281828@gmail.com wrote:
Why not? Requiring a 'references' section for every article (thanking the heavens that WP is not paper), and reminding every editor that *every* new article should come with at least one reference, seems a responsible thing to do. Can you offer a reason not to have such a section for any article?
I meant that more than that is hard to automate. You're right that the bare minimals can be easily checked. However, I can't see that we can automate much beyond that with ease.
On en:, even that will help a *great* deal.
If you're writing about one of those topics that is a) not private research/analysis of your own, but b) has never been written about anywhere else [that you know of], then we need a new class of references : "personal observation by [user]", with a relevant tag not unlike the original-reporting templates used on Wikinews. Then it will be crystal clear that readers should visit your page, and see whether they trust you as the primary/original observer/author.
Much of this falls under 'original research', doesn't it? Or are you talking about the cases where someone believes that something is true but doesn't have the references to hand?
Either original research or that case. In that case, the personal observation should go on the talk page for others to find a good reference for.
- d.
wikimedia-l@lists.wikimedia.org