Wikipedia:No original research/Noticeboard

From Wikipedia, the free encyclopedia
Jump to: navigation, search
Welcome to the no original research noticeboard
This page is for requesting input on possible original research. Ask for advice here regarding material that might be original research or original synthesis.
  • Include links to the relevant article(s).
  • Make an attempt to familiarize yourself with the no original research policy before reporting issues here.
  • You can also post here if you are unsure whether the content is considered original research.
Sections older than 14 days archived by MiszaBot II.
Click here to purge this page
(For help, see Wikipedia:Purge)
Shortcuts:
If you mention specific editors, please notify them. You may use {{subst:NORN-notice}} to do so.

Additional notes:

  • "Original research" includes unpublished facts, arguments, speculation, and ideas; and any unpublished analysis or synthesis of published material that serves to advance a position. Such content is prohibited on Wikipedia.
  • For volunteers wishing to mark a discussion resolved, use {{Resolved|Your reason here ~~~~}} at the top of the section.
To start a new request, enter a name (section header) for your request below:


Search this noticeboard & archives

"Summary Tables" for Fatal Dog Attacks article[edit]

This is a request for help determining if the Summary Tables in the article "Fatal dog attacks in the United States" violate WP:NOR. To give you the background, this article is an incomplete list of some of the people who have died after being bitten by dogs in the US going back to 1887. The list is composed almost entirely of fatalities that were reported in the news media, not in any other more reliable primary or secondary sources. At the end of the article, there are Summary Tables that attempt to report the percentage of bites per year attributed to different breeds of dogs. For example, in 2013, it says "Other mastiff-type (3) (9%)." I feel this is original research and not the type of thing we should be publishing on Wikipedia. However, after long discussion on the talk page, another editor has insisted that it is exempted based on WP:CALC. These are the reasons I feel that this is Original Research:

  1. WP:NOR states "All interpretive claims, analyses, or synthetic claims about primary sources must be referenced to a secondary source, rather than to the original analysis of the primary-source material by Wikipedia editors." In this article we are calculating statistics from sketchy primary sources, ie news media.
  2. According to WP:CALC, "Routine calculations do not count as original research. Basic arithmetic, such as adding numbers, converting units, or calculating a person's age, is allowed provided there is consensus among editors that the calculation is an obvious, correct, and meaningful reflection of the sources." However, in the case of this article the primary sources are already vague and contradictory about breeds of dogs. For example, the same dog might be referred to as a "Lab mix," a "mutt," a "bulldog mix," or a "mastiff mix." Thus, there is no way to ensure that the summary tables are "an obvious, correct, and meaningful reflection of the sources." Furthermore, even if the primary sources were not contradictory, there would be no "obvious, correct, and meaningful" way to summarize the data on all the different breeds of dogs. For example, there is no way to know if a purebred Labrador and a Labrador-mix should be combined together into the same percentage.

The extensive discussion on this topic can be found on the Talk page. Any input in resolving this question would be greatly appreciated. Onefireuser (talk) 20:35, 1 August 2014 (UTC)Onefireuser

Talk page link is here (you accidentally linked to main namespace. As an aside, I had never heard of this article before today, but earlier today I wanted to know how many fatal dog attacks there had been in the US each year and ended up there. Weird coincidence that it would show up on NORB the same day. 0x0077BE [talk/contrib] 23:11, 1 August 2014 (UTC)
Thanks for correcting the link! Glad you've found the article helpful. Just know that our article is likely catching less than 75% of all the incidents that actually take place. There is a lot of variation in the newsworthiness of different dog-bite-related fatalities and many victims' families understandably don't want to be all over the evening news so they don't get reported. If you want the most accurate possible estimates, I would direct you to the scientific references (from the CDC and JAVMA) at the beginning of the Wikipedia article.Onefireuser (talk) 01:25, 2 August 2014 (UTC)Onefireuser
Wait: you know of other fatal dog attacks in the USA yet haven't added them to the list? Please, hold back no longer! Chrisrus (talk) 04:49, 2 August 2014 (UTC)
Yeah, I saw what it was and that it wasn't appropriate for assessing the total number, but it also gave me I think a sense of the order of magnitude on the number of fatal dog attacks. I (and I think all other humans) are kinda bad at assessing the magnitude of very small risks, so I was curious if this was one of those things that kills a surprising number of people or some small amount. I just took the number I saw and figured it was that plus up to about 1 order of magnitude bigger, which it sounds like you'd agree is probably accurate. 0x0077BE [talk/contrib] 02:01, 2 August 2014 (UTC)

My position is that summary is not original research: see WP:SUMMARYISNOTOR. We just count and give the readers our best description of the top two types of dog that we've collected in a year. If all we can say is "mixed breed dog", we should say that. If the list included one Welsh spaniel, one field spaniel, a cocker/springer mix, and a King Charles Cavelier, one "spaniel type", and one "spaniel mix" the obvious way to summarize that for the people would be "Spaniels, spaniel-types, and spaniel mixes: Six fatal attacks". If there are ten known such attacks by packs of stray mongrels on Indian reservatins (yes, that is a thing that happens regularly, albeit not six times in one year, as far as we know) we should say "Six attacks by packs of mongrels on or near Indian reservations." Just saying "ten fatal attacks by mixed breed dogs" is what we should say if that's the best we can do, but we should always try to do the best we can so if we don't have to use such a vague and problematic term as "mixed breed", then we should be more helpful to the reader than just saying "ten mixed breed". Chrisrus (talk) 05:04, 2 August 2014 (UTC)

So Wikipedia editors searched through all the U.S. newspapers since the Revolution and the first fatal dog attack they found was in an obituary for a woman who died in 1887. The next mention was from a 1901 article. That is original research. In order to present a list we would need to find a source that had already compiled a list. The fact that no reliable sources have done so indicates that the list meets notability, and should be deleted. TFD (talk) 01:49, 3 August 2014 (UTC)
The issue I meant to raise was with the Summary Tables. Are you suggesting the entire article should be removed? Onefireuser (talk) 01:55, 3 August 2014 (UTC)Onefireuser
The article is fine, its just the "Media reports of fatal dog attacks in the United States." Since according to the article there have been epidemiological studies of dog-bite fatalities since 1977, they could be used to develop summary tables. TFD (talk) 02:01, 3 August 2014 (UTC)
Yes This whole section is research into reports tallied to come to a conclusion. It's obviously inappropriate and (as others have explained) misleading. Mangoe (talk) 02:56, 3 August 2014 (UTC)
I agree - it's Wikipedia editors doing the research. We need to use proper epidemiological studies. Dougweller (talk) 12:08, 4 August 2014 (UTC)

Pardon me for not putting together a better explanation here earlier. I will remove its deletion tonight because it the ground on which it was deleted do not apply in this case because the summary is not WP:OR. Also its deletion does not constitute article improvement. Let me explain:

The summary (Section Three) cannot be removed on original research grounds because summary is not original research. Section Three is merely a summary of Section Two, and summary, as is well known and well established, is not original research. Therefore I will revert its deletion tonight because it was removed on demonstrably incorrect grounds.

The summary is needed to improve the article for the reader and its deletion does not improve the article. Very little of Section Two is visible at any one time to the reader, and any one screenful might give a biased sample of attack types. If you will please just look at it for a moment, click the scroll bar at different places. What is the reader going to make of this? One gets a certain impression of the frequency of different types of attack from any one screenful, and scrolling up and down to get a wider impression is of limited help to the readers looking for a wider perspective. The reader can't see the whole list at once and run their eye down the different columns. It is helpful to the reader to have a summary to see exactly how common different attack types are in Section Two. For the reader, I revert its deletion.

Chrisrus (talk) 02:04, 5 August 2014 (UTC)

I think part of the issue/confusion here might be that a large part of Section Two is also original research. For many of the incidents, we have had to do a lot of sleuthing/interpretation of various, conflicting primary sources (poor primary sources: online news articles) to determine breed of dog and whether or not the incident even qualified as an attack (e.g. people in their 90s who died of renal failure weeks after receiving a dog bite on the leg). For example, see the discussion here Talk:Fatal_dog_attacks_in_the_United_States#Clear_Case_of_misIdentification regarding an incident in which media sources identified the dog as a Golden Retriever mix, but editors argued that we should call it a "Duck Tolling Retriever" because of their own interpretation of photos of the dog. This type of thing is a recurrent theme with many of the incidents on the page.Onefireuser (talk) 12:26, 5 August 2014 (UTC)Onefireuser
ALL of section 2 is OR. It's a systematic tabulation of incidents, exactly the kind of thing that researchers do. Even if we don't summarize it, we invite anone who comes along to do so. The whole section should go. Mangoe (talk) 14:34, 5 August 2014 (UTC)
I weakly agree with this. It might be a good idea to fork this into a list article. "List of notable fatal dog attacks in the US", maybe. There's some precedent for incomplete lists of this type (List of unusual deaths, List of inventors killed by their own inventions, List of political self-immolations, etc.). Their inclusion in the encyclopedia is somewhat controversial it seems, but there are some cases where the consensus seems to be in favor of keeping. I think maybe forking the content and de-emphasizing the stuff about the breed of dog would be a step in the right direction there. 0x0077BE [talk/contrib] 15:28, 5 August 2014 (UTC)
That could be a good option. If we do go that route, I would drop the word "notable" from the title. With the exception of Diane Whipple, and perhaps a few others, these dog-bite-related fatalities are not really notable, although they are sad and unfortunate.Onefireuser (talk) 16:16, 5 August 2014 (UTC)Onefireuser
It is still OR. In the 1980s for example Rottweiler attacks became a public issue, so the papers began to report them. All a list would tell us is that the media covered Rottweiler attacks, and leave the reader with a misrepresentation of which dogs carried out more attacks. That's why we need someone familiar with dog attacks and how the media reports them to compile a list and comment on media coverage and which dogs are most likely to attack. We need secondary sources to do this, since OR prevents us from doing it ourselves. TFD (talk) 20:03, 5 August 2014 (UTC)

First of all, Section Two is not under discussion. The question here is, may we summarize Section Two into Section Three. That is all - off topic statements are not going to be applicable to this proposal to delete the summary. Different topics are being discussed here, but there is nothing in this discussion that explains why summarizing the list is somehow original research. Unless this is done, I will restore Section Three tomorrow. Second, Section Two is not original research. It simply reports what is found in the available WP:RSes of each attack. If Section Two is original research, then why isn't the corresponding section of or the entire article List of fatal alligator attacks in the United States by decade, Coyote attacks on humans, Shark attacks in South Australia, List of fatal bear attacks in North America, Killer whale attacks on humans, Dingo attacks on humans, and many more such articles. Not to mention ever list of discrete items taken from different sources all over Wikipedia, including List of famous dogs. Why are you singling out this particular article for such censorship? This is wrong. Chrisrus (talk) 03:44, 6 August 2014 (UTC)

For one thing, section 3 was a summary of section 2. If Section 2 is original research (the topic of this board), it moots the (related) question of whether or not section 3 is original research as well. The issue regarding section 3 is that compiling a list ourselves out of various samples is original research (see WP:SYNTH - creating an original research statement out of the juxtaposition of reliably sourced statements). My suggestion above was to fork Section 2 into its own article similar to the ones you've listed. Some of those you've listed have what I would consider to have the same problems, but you've got to start somewhere. See WP:OTHERSTUFF.
Randomly picking one to illustrate the difference between a list that is original research and one that is reliably sourced, I noticed that Shark attacks in South Australia references SharkAttackFile.info. Assuming that's a reliable source (whether or not it is is a different discussion, because it's plausible to have a reliable source that looks exactly like this), you'll see that that claims to be an exhaustive and complete list of all shark attacks worldwide, broken down by location and type of shark. We don't have anything similar to that in this, we're cobbling together a list of dog attacks reported in the media - essentially doing ourselves what a secondary source is supposed to do - that's pretty much the definition of original research.
Oh, and to clarify a point, under no circumstances should you add the summary tables back into the articles without first building a consensus for your position. As it stands now, a number of other editors have weighed in on this topic, and all of them agree that this is a clear violation of WP:NOR. That's a pretty clear indication that there's a consensus for removal. Repeatedly adding it back in against the consensus would be effectively edit warring. 0x0077BE [talk/contrib] 13:01, 6 August 2014 (UTC)
I agree completely. This is the kind of work that social statistics people do, and a number of the similar articles ought to be deleted for having the same fault. Mangoe (talk) 18:33, 6 August 2014 (UTC)
I'm sorry, it's just not true that Section 2 is the topic of this section. It is not. Section Three is the topic, specifically its recent deletion on WP:OR grounds. The arguments for the deletion on those grounds may be seen here above, if we tease it out from the irrelevancies. That argument is that putting numerical totals at the end of a list which has been compiled from discrete WP:RSes constitutes original research. That is wrong for Wikipedians to put totals at the bottom of columns on such lists. That it's wrong for the List of fatal bear attacks in North America to include these maps List_of_fatal_bear_attacks_in_North_America#Maps. There is nothing wrong with adding those maps or totals or text summaries after discretely sourced lists. It is long established and hugely well precedented project-wide consensus that that WP:SUMMARYISNOTOR. This has been argued before and always lost. This consensus is the grounds on which I restore the summary tables at the bottom of that list, as soon as appropriate.
Another topic on this page can be that Section 2 be deleted on WP:OR grounds. If it succeeds, Section Three will have to be deleted at that time because, as you rightly say, it "moots the (related) question of whether or not section 3 is original research as well".
I think you will agree that here is not the place to discuss splitting that article unless it has something to do with WP:OR. As you know, this place is for discussions of cases of possible WP:OR problems, only. How would splitting Section Two off address any WP:OR problems? TALK:List of fatal dog attacks in the United States is a better place to discuss splitting. It's off topic here. The topic in this thread is, should Section 3 be restored, or not not be restored on OR grounds alone. This is the topic here. Because I'm planning to restore it as soon as it is appropriate because WP:SUMMARYISNOTOR. Chrisrus (talk) 20:18, 6 August 2014 (UTC)
Regarding Section 2, obviously we can discuss whatever we want here, and it was suggested that Section 2 be removed as original research. I was trying to say that I'm not sure how I feel about that, but I was suggesting that at the very least it be split off into another article - I think that framing it as "List of notable fatal dog attacks in the US" or something similar would help with the OR problem by making it very clear that it's not an exhaustive list. Honestly, I'm fine with discussing this elsewhere.
That said, I think you make a good point about the bear attacks article - there are a million sources on that page, so I'd probably want to get someone more familiar with that material to address whether there's some secondary source tying all those reports together and/or indicating that it's an exhaustive list. Frankly, I'm thinking we might want to aim for a wide-audience RFC on the rules about these types of list in general. Many of them basically seem like they fail either WP:SYNTH or WP:INDISCRIMINATE.0x0077BE [talk/contrib] 03:54, 8 August 2014 (UTC)
While anyone can discuss anything they here, this place may only be rightly used to discuss whether particular edits are OR or not OR.
The only thing relevant to me restoring Section Three found here is whether or not discretely sourced list summaries violate WP:OR. That the list itself - please understand this important distinction - should be deleted or not deleted on OR grounds has nothing to do with my restoring the summary of the list while the list itself still stands.
You seem to agree that whether the article should be split or not split has nothing to do with whether I should restore Section Three today. I am not going to split or unsplit the article today, I'm going to restore the Summary Tables to the list, that is all. Me restoring the Summary Tables does not effect whether it is split or not split, and seems off-topic here unless it has something to do with WP:OR.
Whether discretely sourced lists in general violate WP:OR, and therefore should all be deleted, is a separate issue from me restoring the Summary Tables to that discretely sourced list today. If we decide to delete all discretely sourced lists, the list of fatal dog attacks in the USA will be one of them, and the summary tables will likely be deleted at that time. As it is not, at this point, clear that that is going to happen, it is no argument for not restoring the Summary Tables today. Chrisrus (talk) 14:25, 8 August 2014 (UTC)
So, is there some argument why I should not restore the Summary Tables today, and you all can just delete it later when you delete the list that it is a summary of? Chrisrus (talk) 14:25, 8 August 2014 (UTC)
I think that it is clear to me that the summary tables at the very least are OR because as I've said over and over again, the only way they would be a "straightforward calculation" is if the information we were trying to convey was "how many media reports have been included in the Wikipedia article", which is just random information. Add to that that it actually looks a whole lot like something that you'd put in some sort of meta-analysis of case studies or an epidemiological study and the summary tables are clearly OR independent of whether section 2 is OR. Finally, the breakdown by breed of dog is the most inappropriate part of the article, as that seems to imply something about how likely various breeds of dog are to kill someone, which is borderline POV.
Even ignoring these arguments, I'd say that basically everyone here (except for you) at this point seems to support the removal of section 3 - Onefireuser and I do, even though I'm not sure that either of us is 100% solid on the removal of Section 2. The other editors have indicated that section 2 needs to be removed in its entirety, and Mangoe's reasoning specifically mentioned that Section 2 is OR because it invites users to draw the conclusions explicitly drawn in the text of Section 3. Given that everyone here (except you) has advocated removal of section 3, either because it is mooted by the removal of section 2 or as OR itself, I think a reasonable place to start is to remove section 3, then move on to the discussion of section 2, at least until anyone expresses that if we keep section 2 we should also keep section 3.0x0077BE [talk/contrib] 15:44, 8 August 2014 (UTC)
Here, first, you seem to be saying that all incomplete list summaries on Wikipedia should be deleted because they are all by definition WP:INDISCRIMINATE. Is that correct? Because we summarize incomplete lists all the time, sticking to the same example of the maps on Bear Attacks in North America. Are you asserting a general rule that would apply there as well?
Second, there is no analysis in the summaries. It's just the totals of the ages dog types totals by year, that is all. We can probably improve it by adding more and better totals and clear and obvious categories vs. iffy cases. Why just list the two top kinds of dogs, why not list them all? We could use pie charts or something. I invite you to participate in improving the summary tables. That would constitute article improvement, while deleting them would not because just because when scrolling through you might get the impression that, for example, it looked like mostly mostly children, but they were actually a minority in a particular year. The list is too long and the reader needs a summary, actually more summary would be better, such as including maps of the locations of each of them, how many were spayed and neutered, how many had attacked before, how many were chained up and left alone, etc, to the extent that doing so is possible. Then people researching the topic will be even better served than they will be when I put that summary back, but putting it back will definitely help the reader. Who knows what they might use the information for? For all we know, it could be very important.
Third, you seem to be saying that it's not WP:OR but WP:NPOV that is the most important thing to you. You worry about the effect on the reader. Readers will look at that and come to the conclusion that you assert to be wrong: that certain types of dog are more likely to kill someone than others. You know that is not true: no type of dog kills more often than any other type of dog, so you don't want to give the readers information that contradicts this knowledge you have. You are worried that, (correct me if I'm wrong) if the readers are provided with this information, they will come to the wrong conclusion, in your point of view, on this emotional topic. The problem is, you assert, that there are no types of dogs that consistently kill more people than others in the United States. That is your position, is that correct, this is what you're trying to say? If so, even if you are right, the reason for the deletion was not NPOV, but OR, and as such would not stand in the way of me restoring Section Three today. You could then re-delete it with the new NPOV grounds, and we could maybe move this discussion to the NPOV discussion page because it would belong there, not here, because it's not original research, but the effect on the reader would be misleading.
If I understand you correctly here, at the end, you are saying that Section Three should be removed because Section Two is going to be removed some time in the future. Is that correct? Because I wouldn't be too sure about that. That definately remains to be seen, please agree. People have tried to delete this list many times before, as to be expected with an emotional topic like this one with facts that people want or don't want to be true or if true not widely known. They've not succeeded thus far. The prudent thing here is clearly to wait and see. If you are right, the summary will also be deleted at that time. For now, I'm putting it back to help the readers get the totals and percentages.
Finally, will you please stop personalizing this and counting heads and address yourself solely to evaluating the arguments being made regardless of how many or which people made them? And focus on the wider consensus represented by precedent and policy and so on, please, that's the real consensus that matters: see WP:NOTDEMOCRACY. Chrisrus (talk) 18:34, 8 August 2014 (UTC)

────────────────────────────────────────────────────────────────────────────────────────────────────Your first two counter-points indicate that you've completely missed my point. Answer this question - why would anyone care, at all about the number of dog attacks listed in this Wikipedia page? What is the purpose of the section? I'll respond to the other points separately so that you can respond to this question directly, because it's at the heart of the question of why these summary tables are inappropriate, independent of whether Section 2 is appropriate.0x0077BE [talk/contrib] 20:05, 8 August 2014 (UTC)

We collect all reliably sourced fatal dog attacks in the USA on that list. What leads you to believe that we are missing any? If you are aware of other reliably sourced fatal dog attacks in the USA, please let us know so that we can add them to the list.
Basically, the purpose of the list of fatal dog attacks in the USA is the same as the purpose of, say, for example, List of fatal cougar attacks in North America: to inform readers who are interested in the subject, that is all. Why do you ask me this question? If it's because you think we're out to malign certain types of dog, rest assured, it is not. If one type of dog shows up more often it's because they show up more often, not because we wanted them to. I personally am interested in all types of dog articles and animal attack articles for no particular reason, so you might as well have asked me why people would want to write or read the article Klingon starships: some people are interested, that's why. However, there may be a difference between animal attack articles and articles like Klingon starships. Someone might read animal attack lists and articles and notice something in our summaries and data that might keep someone from getting hurt. You never know, it could happen. They might decide to do or not do something based on what they'd learned and that could help them not get attacked by a bear, cougar, dog, coyote, and so on. But we don't tell them what to do or think; unless it's just to pass on recommendations from the CDC or some such. We just give them the mere facts as we know them in a neutral point of view manner,that is all.
Where are you going with this? Why did you ask? i assure you, nothing is being hidden or held back or distorted to try to lead the readers to any conclusion. If the majority of victims were full grown men, or if the majority of such dog attacks that we can cite well enough to include were spaniels or bloodhounds, we would simply pass that information along to the readers. I cannot say the same for the opposition to the article, however. Much of it is motivated by the desire to push their personal points of view that no dog is more likely to be involved in a fatal attack than any other, and any evidence to the contrary should be censored lest the reader come to a different conclusion.
You still have not addressed my points just above. Chrisrus (talk) 21:12, 8 August 2014 (UTC)

The Four Deuces, Dodo Bird, Onefireuser, Mangoe, Dougweller: Chrisrus is making some noises to the effect that he's going to restore section 3 of this article, based on his interpretation of Wikipedia policies. I have discussed this with him at his talk page (warning: talk page is huge and he for whatever reason refuses to archive), and he currently suggests that there is no consensus for the removal of section 3, because some of you have advocated for the removal of section 2 and section 3, and that there is currently no consensus to remove section 2. Frankly, I think this is a misrepresentation of the situation at hand. I think there is a broad consensus to remove section 3, and a weaker movement towards removal of both section 2 and section 3. Can you guys weigh in and clarify so that we can assess the current consensus on the matter? 0x0077BE [talk/contrib] 14:23, 8 August 2014 (UTC)

Agree the table should not be added per the discussion above. TFD (talk) 15:50, 8 August 2014 (UTC)
Just so you know, we're discussing whether the long-standing summary tables should have been recently removed on OR grounds alone, not adding anything that hasn't stood for so long enough that it constituted the tacit existence of a general consensus to keep.
Now, you refer to what was stated earlier. Your argument above (correct me if I'm wrong) was in support of the removal of Section Two and Section Three both, not whether such as summaries as List_of_fatal_bear_attacks_in_North_America#Maps all constitute WP:OR, across the board, as a general principle. Please understand that this is a separate question from the deletion of the summary, only, on the grounds that it's OR to summarize any such list. Please understand that me restoring the Summary Tables while the List stands does not affect any future deletion of the list itself. If later we delete Section Two, Section three will be deleted then and at that time, only, with it because there will be nothing for it to summarize . So please clarify. Are you saying that all discrete item list summaries, such as List_of_fatal_bear_attacks_in_North_America#Maps, for example, always constitute WP:OR, across the board, as a general principle, or that only this particular summary constitutes one does so? If so, why? Chrisrus (talk) 16:47, 8 August 2014 (UTC)
Support removal of sections 2 and 3 per above arguments. PearlSt82 (talk) 19:01, 8 August 2014 (UTC)
Can you clarify whether you'd support the removal of section 3 if section 2 were to remain in place, or does the justification for removal, in your estimate, flow entirely from Section 2? 0x0077BE [talk/contrib] 19:48, 8 August 2014 (UTC)
I would support the removal of section 3 if section 2 stays in place. In my opinion, section 3 is more problematic because as noted is summary table of randomly collected data through media reports. I think the biggest problematic issues with section 2 are the breed section (media reports are usually inaccurate and visual identification of dog breed by anyone is only about 25% accurate, and most dogs that are listed as "pit bulls" are some form of mutt with a large head), the fact that data is really only present for the last 20 years or so, and is a collection of media mentions, rather than drawing on a list compiled by a third party organization. PearlSt82 (talk) 20:11, 8 August 2014 (UTC)
First, it's not breed of dog, it's the type of dog which it lists. This is only a breed in some cases. Second, where are you getting this "only about 25% accurate" information? Do you have an WP:RS for it? Third, how are these randomly collected? We collect all the reports on fatal dog attacks in the USA, not just random ones here and there, all of them. We do not discriminate, and there is nothing random about it. As long as a WP:RS reports it, we do too. If you are aware of attacks that go unreported, how are you aware of them? There are extensive disclaimers on that list about the possible limits of news reports, but news reports are WP:RS, too, and countless citations prove. Do not censor this information on the grounds that the facts might lead the readers to a point of view you don't share. Let the facts be as they may, and the readers may conclude whatever they conclude, that's not for us to worry about. If the facts show that most of the fatal attacks are by, for example, German Shepherds, then people are free to use that information however they like, even if it is to support a position that you don't agree with. Chrisrus (talk) 20:37, 8 August 2014 (UTC)
"Category of dog" is currently being used interchangably with "breed". Notice it doesn't say "molosser" or "shepherd", but outright states breed like "Rottweiler" or "Pomeranian". The 25% accurate comes from the Victoria Voith studies, which can be viewed here: #1 #2. These lists are randomly assembled because they are drawn from media reports, primarily from the last 10-15 years. If something is not reported on, like the majority of data pre-2000, it doesn't appear in the list. Its just poor methodology for being of any practical use, I'm not trying to "censor" anything. PearlSt82 (talk) 20:48, 8 August 2014 (UTC)
You are wrong about that. On that article, "category of dog" is not being used interchangeably with "breed" at all. It includes many things including "sled dog", "shepherd", "pack of feral dogs", and many other non-breed descriptions, including "pit bull itself, if you would click on it that's what it is. We are aware of that paper by Voith and have regularly consider it's lessons about identifying mixed breed dogs. She did no testing as to how often they could identify purebreds or packs of feral dogs or sled dogs or dogs of a general phenotype regardless of ancestry. All the dogs in her study were mixed breeds. Everyone knows it's nigh on impossible to guess what made up a random mixed breed dog, but you can say if it conforms generally to general types without regard to lineage. So if WP:RSes indicate that it's some kind of scenthound-type-mix, we say "scenthound-type-mix" or whatever is the safest. So you have nothing to worry about, although I invite you to participate in overseeing the categorizing of the dogs in the article.
Its extremely unlikely that the majority of dogs responsible for fatalities are purebreds, and the studies clearly demonstrate that respondents can incorrectly identify mixed breed dogs as purebreds. PearlSt82 (talk) 12:45, 9 August 2014 (UTC)
Can I vote for renaming the article to "Owners who failed their dogs and caused them to fatally attack"? ;) JMJimmy (talk) 21:20, 8 August 2014 (UTC)

I am not going to fall into the trap of commenting on Section 2 at this time. Regarding Section 3, it seems that we have a very clear consensus that it is Original Research. It is original research for many, many reasons, but I feel that the least ambiguous of those reasons are the ones I stated months ago on the Talk page. I will repeat those reasons here: These are quotes from WP:NOR:

  1. "All interpretive claims, analyses, or synthetic claims about primary sources must be referenced to a secondary source, rather than to the original analysis of the primary-source material by Wikipedia editors."
  2. There is an exemption for simple calculation: "Routine calculations do not count as original research. Basic arithmetic, such as adding numbers, converting units, or calculating a person's age, is allowed..." However, this exemption only applies "provided there is consensus among editors that the calculation is an obvious, correct, and meaningful reflection of the sources."

Regarding point 1, the summary tables are clearly an analysis and are not referenced to any secondary source. They are not exempted by WP:CALC because they are not a clear, obvious, and meaningful reflection of the sources. There is simply no way to arrive at clear, meaningful totals of "dog type" based on the news reports. This is because the "dog types" reported are at times inconsistent, contradictory, vague, or meaningless. For example, if in one year three incidents were attributed to a "Labrador retriever," a "Lab-bulldog mix," and a "Golden retriever-Collie mix," would we summarize this as "3 attacks due to retriever type dogs," "2 attacks due to Labrador retrievers and Lab mixes," or 1, 1, and 1?

There are numerous other reasons that Section 3 is Original Research, but I feel that the reasons I have just described are sufficient to justify the removal of the tables.Onefireuser (talk) 23:30, 8 August 2014 (UTC)Onefireuser

I will make one additional point. Our study in Section 2 is far from a complete survey of all dog-bite related fatalities. Another study based on CDC WONDER data found that there were at least 26 deaths in 2000. Our study on this Wikipedia page identified only 5 in 2000. Since this is a slightly different topic than the appropriateness of the Summary Tables, please see the Talk page for additional discussion.Onefireuser (talk) 01:11, 9 August 2014 (UTC)
I hope you will agree that looking at one source about, for example, one fatal polar bear attack in Alaska in 1980, and then looking at another source about another fatal polar bear attack in Alaska in 1980, and then putting them together and calculating and arriving at the conclusion that there were at least two polar bear attacks in Alaska in 1980, doing that is hardly the kind of "interpretive claim" that is being referred to there.
I think you will agree that, to the extent that summarizing the list entails simply counting how many fatal attacks were attributed to, say, German shepherd dogs, in a particular year, that is as clear, obvious, and meaningful a calculation as can be. You are right, however, that Wikipedians might disagree about how best to deal with, to use one of your examples, a "Lab-bulldog mix". However, we should have tried, in good faith, to arrive at the consensus among us editors of that article that the guideline you quoted refers to as which would be the best way to summarize them. Counting it as "one Mixed breed" for example is pretty clear and obvious, but there might be an even better way that we could count it that would improve the article even more; or maybe "other". As Wikipedians, we are supposed to try to work things out together based on principles and such with service to the reader at the main goal.
My reply to this last additional point is in the place you directed any readers to look, but as you mentioned it, I will just summarize my reply here by saying that of course there is no real danger that the reader will think that the list or its summary is intended to be exhaustive because the reader is told repeatedly that it is not, (not to mention that it is common sense that it couldn't be). I will also note that neither are the attacks studied by the sources in Section One, nor any such study, and that there's nothing unusual or problematic about that because it's an incomplete list like countless others on Wikipedia. Chrisrus (talk) 02:08, 10 August 2014 (UTC)

I had marked this issue Resolved a while back because there was a consensus that included 0x0077BE, TFD, Mangoe, JMJimmy, Dougweller, PearlSt82, and myself Onefireuser. However, the tag was removed. There has only been one dissenting voice in the discussion and the Summary Tables have been removed for a while now from the article. Does anyone wish to continue debating this question or can we mark it resolved? Onefireuser (talk) 18:58, 24 August 2014 (UTC)

Much of the above discussion was off the topic of specifically may or may not such lists may be deleted or blocked on OR grounds. We cannot rightly block any restoration of the summary tables on OR grounds pointing to this discussion without first including here a substantive reply to the following points in favor of not removing/restoring the summary tables to the article. The above tally of usernames is immaterial.
Please respond substantively to this:
It has been pointed out here above, and not substantively objected to as yet, that summarizing such lists cannot not constitute original research by synthesis, citing WP:CALC and many precedents all over Wikipedia, so many that it points to a clear long standing Wikipedia-wide consensus that we may summarize such lists, with many examples given. Since that time, an attempt was made which can be seen as a test case, to delete the summary of Fatal bear attacks in North America on WP:OR grounds. This blocked citing WP:SUMMARYISNOTOR and WP:CALC grounds, and pointed to widespread precedent and practice. So the summary removal on that article failed, setting yet another precedent. And there are many, many more such examples of such summaries all over Wikipedia, so there can not be Wikipedia consensus that to summarize such a list is WP:OR.
Please work out in good faith cooperation with other editors the best summary tables for that list. Please do not block the planned restoration of the summary tables on OR grounds again, because it's been established that such summaries are not OR violations. If you do undo the summary table restoration, please do so on some other grounds, not WP:SYN or WP:OR, because it has been substantively demonstrated and not substantively refuted that summarizing such a list per se does not constitute any OR violation. Consider if you will undoing the restoration on other grounds, perhaps notability or list project guideline violations. But please understand that you may not rightly do so on OR grounds citing this discussion. Chrisrus (talk) 02:58, 30 August 2014 (UTC)
It's been some time we've been waiting for a substantive reply to these points. I'll wait a bit before adding the resolved tag. Chrisrus (talk) 05:29, 5 September 2014 (UTC)
I agree. There has been a strong consensus here and on the article's own Talk page that the Summary Tables constitute OR. Most of the arguments have been very articulately stated by 0x0077BE. Feel free to add the resolved tag. Thanks. Onefireuser (talk) 18:09, 9 September 2014 (UTC)
You may disagree, but you have to do so substantively. Please address this below, quoting above:

1. Summarizing such lists cannot not constitute original research by synthesis, citing WP:CALC and many precedents all over Wikipedia, so many that it points to a clear long standing Wikipedia-wide consensus that we may summarize such lists, with many examples given. Since that time, an attempt was made which can be seen as a test case, to delete the summary of Fatal bear attacks in North America on WP:OR grounds. This blocked citing WP:SUMMARYISNOTOR and WP:CALC grounds, and pointed to widespread precedent and practice. So the summary removal on that article failed, setting yet another precedent. And there are many, many more such examples of such summaries all over Wikipedia, so there can not be Wikipedia consensus that to summarize such a list is WP:OR. Chrisrus (talk) 17:44, 13 September 2014 (UTC)

Tag removed. There was a long discussion here and a CONSENSUS that the summary tables in the Fatal Dog attacks article are Original Research. Onefireuser (talk) 02:19, 14 September 2014 (UTC)
This is not a substantive reply to the points. According to Wikipedia:Stonewalling, avoiding substantive discussion of the issues related to an edit while engaging in behavior that is typical of disputes simply creates the appearance of a real substantive dealing with the evidence and reason when none (or little) exists. When no substantive objection to a change exists, stonewalling is not required, so stonewalling is used when those opposed to the change don't actually have a substantive objection to the edit, or when they know whatever argument they have can be easily refuted, or is contrary to long-standing Wikipedia-wide consensus that such lists may be summarized. Therefore, absent some evidence or reason why summarizing that list constitutes WP:OR which has not been already refuted, the summary may be restored. Also, please read WP:NOTDEMOCRACY. We need a substantive reply, not repeated reference to a sort of head count of points that have already been refuted. Chrisrus (talk) 06:46, 14 September 2014 (UTC)
Chrisrus and I couldn't see eye to eye on the talk page, so I brought the discussion to this noticeboard. Here, everyone else has agreed that the tables are original research because they calculate summary statistics in exactly the way that researchers creating original research for publication do every day. They are not simple calculation like 2+2=4 or calculating the age of someone born on 2/4/1977. They are statistics that require numerous assumptions to derive the numerators and denominators. There is no one, simple, clear, or correct way to to create these statistics. Therefore they are no exempted by WP:CALC and they are OR. We've been over this a number of times. I'm not sure what else Chrisrus would like to get out of the discussion. 0x0077BE also made a number of other articulate points about why this is OR.
Can any additional editors weigh in on how we can resolve this issue? Onefireuser (talk) 21:21, 15 September 2014 (UTC)
The issue is resolved. If he doesn't like the outcome of this discussion, the onus is on him to get more people in in the hope that consensus will turn the other way. He doesn't get to repeat himself again and again until no one cares and then claim his superior logic trumps consensus. --Dodo bird (talk) 21:55, 15 September 2014 (UTC)
Nothing has changed. Seeking out incidents and tallying them is obviously research, exactly the sort of thing that shouldn't be done. If this continue to be belabored, things may be headed off on the road to ARBCOM. Mangoe (talk) 16:13, 16 September 2014 (UTC)

@USER:0x0077BE:

As you know, and can see if you scroll up, much of the above, much of the above discussion is irrelevant to whether summarizing that list into those tables is covered by WP:CALC or not. You and I will agree that it is covered by CALC to do if, as you say, a simple calculation like 2+2=4. So look here: https://en.wikipedia.org/w/index.php?title=Fatal_dog_attacks_in_the_United_States&oldid=619383147#Summary_tables See? One child under the age of one in 2005, plus one child under the age of one in 2005, equals two children under the age of five in 2005. One plus one equals two.
Your previous concern above, not repeated by you here, AND not having anything to do with whether it's WP:OR to summarize that list, was noted and appreciated, thank you very much, because that concern of yours, that readers will think that total is complete, can be dealt with by following the incomplete list procedure and adding "incomplete list" tag, see here: https://en.wikipedia.org/wiki/Wikipedia:WikiProject_Lists#Incomplete_lists once again, not only at the start of the main, Part II list, should reader somehow miss the introduction, which by the way can be improved to address that concern even better than it already does.
So there'll be no reason for you to worry that the reader will misunderstand that this is merely a summary of the article's list of items we have a citation for, not a total of all fatal dog attacks in the US, including those we don't have a citation for or which haven't been added. This is the way we always do such things all over Wikipedia, this is a wide, broad, and long-standing consensus that you and others here should know about and understand and respect.
Your other previous concern, that the reader might come to a conclusion that you think is wrong, that looking at the information on that list and summary, that the reader might, based on the attacks we've been able to get on the list thus far, come to a conclusion that you don't think is warranted, that concern is also irrelevant to whether this it's covered by CALC to summarize this list, and therefore not OR. Furthermore, that concern of yours, that a reader looking at that might become unfairly prejudiced against certain types of dogs, was enough to that you, and anyone else who thinks it's the business of Wikipedia to censor information that might make certain categories of animals look bad, is classic POV editing. The fact is, one fatal poodle attack, plus one fatal poodle attack, equals two fatal poodle attacks. This is what that summary is and therefore clearly covered by CALC. We are not trying to lead the reader to any conclusion, we are merely giving them the facts! That you or any reader wouldn't like what conclusion those facts might seem to lead the reader to on their own, and therefore want to delete or destroy the list is the type of thing we've seen before at Wolf attack, for example, editors coming in with the preconceived notion that a wolf would never attack a person, that's not in their nature, and then go about trying to destroy the article or censor or distort the article to reflect that notion, no matter what a fair summary of the WP:RSes say about that subject, for the greater good of rehabilitating the reputation of a particular type of canid that has been unfairly maligned by facts by republicans or hunters or ranchers or others who maligning the poor innocent canid, the vast majority of which never hurt anybody. We've seen it at Coyote attack as well. Each time, these accusations are not founded: we are not out to draw people to any conclusion, we are just Wikipedians who work on all kinds of animal articles and animal attack articles, just when it comes to canid attacks like dingo attack for some reason it's suddenly wrong, but no one seems to care if it's a bear attack or elephant attack or wild boar attack.
Be that as it may. That concern as well is also completely irrelevant and never should have been stated above or motivated anything that was stated above. As to whether that summary is OR or covered by CALC, it is. One teenager plus one teenage equals two teenagers; one sled dog plus one sled dog equals two sled dogs; WP:CALC applies. Chrisrus (talk) 07:01, 19 September 2014 (UTC)
I've heard your arguments and you've seen my counter-arguments. So far, we haven't seen any support for your position, the only question about which the consensus is less clear is whether the list itself is original research. I tend to believe that, in the absence of a source containing the comprehensive list, then it iis OR. I think this is a common belief, though since there are many such articles, I would recommend that if we're going to have that discussion we centralize it somewhere and advertise it on the various centralized-discussion places to get cast a very wide net. So far, editors here again seem to believe these individually-sourced lists are OR, but I suspect that - based on their prevalence - there would be more pushback on that if we had a wider net, which is why I haven't personally pushed for the removal of the list until we can have that discussion. 0x0077BE [talk/contrib] 17:07, 19 September 2014 (UTC)
@Chrisrus. We may all be able to agree that, as you say, "one fatal poodle attack, plus one fatal poodle attack, equals two fatal poodle attacks." Unfortunately, that is not the type of data we have collected in our research. For example, how would you apply WP:CALC to one incident where "police said they were 'relatively sure' they were Rottweilers or Doberman Pinschers" and one incident by "Mixed breed dogs (possibly including Rottweiler, Pit bull, Beagle, and Labrador retriever)" and another involving "Rottweiler and 2 Rottweiler-Pug mixes"? As you well know, those are the type of data points we have collected. If we wanted to know how many Rottweilers that is, we could add it up in a number of different ways to arrive at anywhere from 1 Rottweiler all the way up to 6 or more. It is exceedingly clear that this is not a trivial calculation such as 2+2=4. Onefireuser (talk) 17:52, 19 September 2014 (UTC)
I would like to say that even if 100% of the breeds of dogs were positively identified, I'd still be uncomfortable with these summary tables because, as I've said many times before, they represent irrelevant information, i.e. the number of dog attacks identified by Wikipedians. They also have an element of SYNTH (again, well-trodden ground) to them, since we're taking disparate sources and synthesizing a new fact (i.e. "we've listed 20 dog attacks"). Identifying the specific problems like inconsistent identification of breeds and improper sampling is effectively just criticizing the research methodology; it's valuable because it helps editors to understand why there's the policy against original research, but in the end even if all the problems with the research methodology were fixed, it's still research. In fact, if the problems with the methodology were completely fixed, this whole problem would be moot, because you could compile this information into a scientific paper, submit it to a peer-reviewed journal and then have a valid consolidated source for the tables.0x0077BE [talk/contrib] 18:06, 19 September 2014 (UTC)
@USER:Onefireuser. First, based on everything you've said here, it seems you do/would not oppose restoration of the summary table of victim ages, just the dog category summary table, is that correct? It seems you don't have a problem with summarizing the ages being summarized in the table. Chrisrus (talk) 04:32, 21 September 2014 (UTC)
@USER:0x0077BE Are you saying that all summaries of disparately sourceddynamic incomplete lists constitute original research? After all, they all summarize they count "just those we've collected thus far" from different sources. Take, for example, List_of_fatal_bear_attacks_in_North_America#Maps, is that SYN?
You know very well the points made by myself and many others on this topic, as I have expressed it many times. Yes, my current position is, weakly, that disparately sourced incomplete collections which are not essentially navigational aids are most likely either original research or violations of WP:INDISCRIMINATE. As I mentioned in the talk page on List of fatal bear attacks in North America, a conversation in which you took part, I am only OK with that article because there is a comprehensive source for everything up to 2009, and I oppose the addition of bear attacks not in that comprehensive list to the maps on that page. So yes, your incessant repetition of the exact same points, which has convinced no one, has not somehow changed my mind on this matter. 0x0077BE [talk/contrib] 15:34, 22 September 2014 (UTC)
Ok, what about this list, Leopard_attack, and countless others like it? Chrisrus (talk) 18:49, 23 September 2014 (UTC)
Have you actually read the article Leopard attack? It does not contain a summary table. It does not even contain a list of non-notable attacks. It just has a sampling of some notable attacks. Many of those attacks have their own entire articles. They all seem to be referenced to reliable secondary sources, not news reports. A better example to make your point might be List_of_fatal_snake_bites_in_the_United_States. However, this article also does not have a summary table. And this article has way more problems even than the dog attack article. That snake bite article should probably be deleted in its entirety. Onefireuser (talk) 00:43, 24 September 2014 (UTC)
Please let's look together at the summary table of leopard attacks in the article Leopard attack. There is a disparately-sourced, WP:INCOMPLETE summary table on the right if you scroll down a bit, called "Number of Human Deaths due to Leopard Attacks". Do you see it? Please note that, right after the title, there's a dagger and a corresponding note. Let's read the note at the bottom. You will agree that it, in few words, clearly communicates the idea that these totals are simply those that some random Wikipedians have been able to confirm so far in WP:RSes as of last edit, that is all. The note implies to the reader that the numbers are maybe good for giving people a sort of rough idea, that is all, for whatever it's worth to the reader, take it or leave it, we did this ourselves; we didn't get it from one source; we compiled it from many. And that's OK. The article Leopard attack may, can, should, and does have that table there, so please don't delete it because it improves that article and does not violate or constitute WP:OR. Chrisrus (talk) 03:37, 24 September 2014 (UTC)
First, no one involved in this discussion has suggested making any changes to Leopard attack. Second, the table you are referencing in that article is a list of numbers that come from secondary sources. Reporting totals from secondary sources is very different than collecting raw data and analyzing it yourself. One is Original Research and the other is not. Onefireuser (talk) 05:46, 24 September 2014 (UTC)
Onefireuser: By saying that about leopard attack, I understand you agree that such disparately sourced, dynamic, incomplete lists are not WP:SYN. Is that correct? Because if it weren't, you'd advocate its removal.
Next, imagine we added a total to the bottom of it, that list of leopard attacks, and clearly marked that total as simply a summary of that list, not of all reliably sourced leopard attacks, much less all leopard attacks, just the total of those Wikipedia has collected so far on that list, would that constitute us analyzing raw data and publishing our analysis?
What about that summary table of the victim ages from the US fatal dog attack list? Does that part, just that part, constitute analysis of raw data, to your mind? Let's hold that separate from the totals of the types of dogs involved in the attacks. Just the table of victim ages. It's my impression based on everything you've personally said that the table of victim ages is covered by WP:CALC. Is that correct?
I'm not suggesting this but, just imagine, please, if we had maps of the locations of the fatal dog attacks as a summary of list, kind of like the bear attack maps. Would that constitute us doing our own "analysis" of the raw data, to your mind? Chrisrus (talk) 06:09, 26 September 2014 (UTC)
1. The leopard attack article is not a list.
2. Original research is something that could be published in the scientific literature.
3. Do you feel this conversation needs to continue to take place on the Noticeboard? Onefireuser (talk) 13:25, 26 September 2014 (UTC)
I'm sorry, that's not really a substantive reply. Please reply to the following relevant point: There is consensus support for list of leopard attacks within the article leopard attack, is that correct?
Next, please answer this: if we put a total at the bottom of that list, would it be covered by WP:CALC or not? I mean, so long as it's clear that that total would simply be a summary of that WP:INCOMPLETE leopard attack list, nothing more. Is that correct? Chrisrus (talk) 19:46, 26 September 2014 (UTC)
Ok, so I went ahead and added the total to the list within Leopard attack. I hope you will agree that doing so is covered under WP:CALC and explain. Next, I plan to restore and update the summary tables of victim ages on the dog attack article, which I haven't seen a clear objection to. Then we'll see how that goes, and then look at the rest. Chrisrus (talk) 06:09, 1 October 2014 (UTC)
Ok, because the leopard attack list totals, that summarization is allowed, we clearly have consensus for summarizing incomplete, desperately sourced lists. Therefore, barring any explanation to the contrary, we may restore the dog attack list totals at any time. Chrisrus (talk) 20:53, 12 October 2014 (UTC)

──────────────────────────────────────────────────────────────────────────────────────────────────── It should be obvious by now that you should not do that. The consensus is clear that the summary tables are inappropriate for the dog attacks. This is not an endurance game where whoever can ignore the other side the longest wins. The consensus was resolved quite some time ago against inclusion of the summary tables. Just because we've stopped responding to you once the consensus was made very, very clear does not mean that you can take our silence as consent. I've stopped responding because you haven't presented any new arguments, and everyone else agrees that the arguments you have presented are not valid. 0x0077BE [talk/contrib] 21:18, 12 October 2014 (UTC)

Oh, also, I've reverted your edits on the leopard page. Per your own reasoning, that's an equivalent situation, and per the overwhelming and clear consensus on this page and on the list of fatal dog attacks page, summary tables for disparately sourced lists is OR. This is getting POINTY, honestly. 0x0077BE [talk/contrib] 21:21, 12 October 2014 (UTC)

@USER:Dodo bird:

Very well. If you insist, I will go and get more people. Do you have any concerns about how I should and should not go about doing that? Chrisrus (talk) 07:01, 19 September 2014 (UTC)
Obviously you should not canvas for new people who support your position, you should just bring new eyes to the discussion by posting neutral notices in appropriate places (like Wikiproject Lists, Village Pump, etc). At this point, I doubt you'll find many takers, honestly, because these huge blocks of text you're fond of posting are a real barrier to entry. 0x0077BE [talk/contrib] 17:07, 19 September 2014 (UTC)
I don't agree that this should be necessary, as one should pay any attention to how many people are saying what, but rather what has been said. So I'm holding off for the moment on that. It may come to that, however, but I hope not. This is an important matter. Chrisrus (talk) 07:13, 21 September 2014 (UTC)

@USER:Mangoe:

"Seeking out incidents" is not what is being discussed here. "Tallying them" is. According to WP:CALC, tallying them is not WP:OR. Please read it and stop believing that tallying is OR. It's not. If you want to object to "seeking incidents out" and reporting them in Wikipedia, please start another section or something. Chrisrus (talk) 07:01, 19 September 2014 (UTC)
This is a very touchy issue as I have noticed POV warriors ruining WP articles on this topic in the past such as converting an article about a particular breed into a dissertation on dog bites. That was fixed but there is definitely an agenda-driven cohort on this topic I would also point out that to some degree all of WP is a form of synthesis but there are secondaries on this topic such as the CDC report and the free downloadable PDF book "Pitbull Placebo" and any information based on those should be presentable as a WP article with only the normal amount of "synthesis". Thanks. Wikidgood (talk) 00:17, 17 October 2014 (UTC)
I agree with the argument that the tables are OR and also probably the list itself. As per my above post there are plenty of secondary sources. Even then there is a problem of WP:SYN but with these random lists there is insufficient secondary sourcing and a real tempation for POV warriors to doctor the results. Wikidgood (talk) 23:38, 17 October 2014 (UTC)

Open Office[edit]

As background, Oracle used to have two products: OpenOffice.org, an open-source office suite, and Oracle Open Office, a commercial version of the suite.

In April 2011, Oracle announced:

"Oracle Corporation (NASDAQ: ORCL) today is announcing its intention to move OpenOffice.org to a purely community-based open source project and to no longer offer a commercial version of Open Office. ... we believe the OpenOffice.org project would be best managed by an organization focused on serving that broad constituency on a non-commercial basis." (source)

In June 2011, Oracle announced it would donate the open-source project, OpenOffice.org, to the Apache Foundation: "Donating OpenOffice.org to Apache gives this popular consumer software a mature, open, and well established infrastructure to continue well into the future." (source).

Q1. Do these sources represent original research (as primary sources) on the question of the status of OpenOffice.org and Oracle Open Office?

Next, we have secondary sources from April 2011 that say Oracle decided to discontinue Oracle Open Office (or "commercial development of the OpenOffice.org (OOo) office suite") and that "Oracle says that it is ready to hand over control of the project to the community". For example this, Ars Technica article.

Q2. Do these sources support a statement that the open-source project, OpenOffice.org, is discontinued or was discontinued at the same time as Oracle Open Office? That the commercial version, Oracle Open Office, is disconnected is not disputed.

Lastly, if the answer to Q2 is 'yes', how should we interpret these sources in light of later sources — after OpenOffice.org was donated to the Apache Foundation — that say the open-source project continues (renamed) as Apache OpenOffice?

--Tóraí (talk) 21:47, 10 October 2014 (UTC)

This is the wrong place for this discussion. There is no original research in stating the sources say something. The original research question is trying to state that the two sources I have provided do not matter.
The question is whether WP:UNDUE is being violated. Walter Görlitz (talk) 03:45, 11 October 2014 (UTC)
OK. For clarity, what are the two sources you're providing and what do you claim they say about (a) OpenOffice.org, the open-source project; and (b) Oracle Open Office, the commercial version of the suite? I ask because I believe your interpretation of the sources is OR. --Tóraí (talk) 11:57, 11 October 2014 (UTC)

Top Gear Controversies[edit]

During the controversial recent visit by Top Gear to Argentina, in addition to the controversy over the plate on Clarkson's car, more tenuous links were alleged to the other cars in the party. Officials in Ushaia have claimed the name of the Gurkha knife is the EKH (it is not its the Kukri) and one of the cars had the plate EKH 46 J.

An editor has found an American website selling Gurkha collectibles[1], which uses it product designations eg EKH-GACI-19. He has made the leap that this is proof that this is the makers' designation for the weapon and edited accordingly. He has then edit warred a comment to this effect on the article Top Gear controversies [2]. In army service the Kukri is designated the "The British Army Issue Kukri, Service Number One".

I would appreciate some help in explaining the multiple policy violations of WP:OR and WP:SYN to this editor. WCMemail 22:19, 10 October 2014 (UTC)

That link is not sufficient for the statement that "'EKH' is a product code fur kukri knives with the manufacturer". For one thing, the link is to a retailer, not the manufacturer. Secondly, it's a synthesis to blend it with the rest of the sentence. We need once source that says the whole thing. --Tóraí (talk) 22:51, 10 October 2014 (UTC)
Agreed that it is OR, if the difference requires careful visual comparison. If someone on the shot called a gun "a knife", yea, it's no OR to say otherwise, but the difference between two very similar knifes requires a source to back that. --MASEM (t) 23:28, 10 October 2014 (UTC)

Regional powers map[edit]

"Major regional powers in teal, and minor regional powers in light teal"

This map on the right can be seen in Regional power, but I fear all the countries were shaded arbitrarily into "major" and "minor". First: is this map really needed? Second: if it is, can someone help me find the source(s) that categorizes these countries like that? Fitzcarmalan (talk) 09:54, 18 October 2014 (UTC)

  • Oh my, what a mess. I think the problems stem from the article itself, in that the image simply reflects (in general terms) the list of "regional powers" though the sub-list of coloured "minor regional powers" would seem to be complete original research, being unsupported by the assertions in the article itself. The problem, then, isn't the image but the list. Some of the sources there are horrible. I don't the the image is necessary, even if we could get the list right. But the list should be the priority. Stlwart111 11:55, 21 October 2014 (UTC)
The list itself seems to have some problems as they come from arbitrary discrete sources like speeches that may be using varying definitions of "regional power", but it looks to me like the dark green countries are the G20 countries, and the light green countries are other countries that have been described as "regional powers", but are not in the G20. Nope - I'm wrong, looking at some of the sourcing, it seems like all the "major" powers are ones that have been described as either "regional superpowers" or "major regional powers". The others are just referred to as regional powers. It's not clear to me that "major" and "minor" are terms of art, or that they are being used this way in the sourcing, however. 0x0077BE [talk/contrib] 12:48, 21 October 2014 (UTC)
Yeah, Canada is in the G-20 but isn't coloured at all. And Argentina (also G-20) is a "minor" power. Stlwart111 12:56, 21 October 2014 (UTC)

Mainstream scientific assessment of climate change[edit]

Opinions of neutral uninvolved eds eagerly sought!
We have a

Discussion of the latter article is often chaotic, as many editors talk about diverse issues in the same breath. However, the issue I'm trying to present is laser-focused on the leads of the two articles.
The lead of the main article tries to summarize the mainstream scientific perspective. To comply with WP:FRINGE's requirement to establish the context for fringe statements, the lead of the latter article does that too. However, for a long time they have been out-of-synch, using overlapping but different text and sources. A poll question has been posted asking

Given that the mainstream assessment is summarized on the basis of the RSs with greatest WP:WEIGHT at the main article "Scientific opinion on climate change", would a neutral uninvolved editor reasonably expect the same sources to be used to present the same summary [at the sub-article "List of scientists opposing..."] unless there was a really good RS-based reason to do something different?

Please offer your thoughts in the thread located at the subarticle via this link. NewsAndEventsGuy (talk) 13:11, 21 October 2014 (UTC)

Are you sure this is a NOR issue? Incidentally, the title "opposing the mainstream assessment" seems to me to be a possible NPOV issue. It suggests that the list of scientists are somehow mavericks who refuse to go along with the unthinking "mainstream." Perhaps "opposing the scientific consensus" would be more accurate and neutral. Onefireuser (talk) 01:42, 22 October 2014 (UTC)

Is it WP:SYNTH to make an article on the Indo-Canadian population in Greater Vancouver separate from that of the Indo-Canadian population of British Columbia?[edit]

Related to the RFC at: Talk:Indo-Canadians#Merge_discussion: On whether Indo-Canadians in British Columbia and Indo-Canadians in Greater Vancouver be separate or should the latter be merged into the former...

Would it count as WP:SYNTH to have a dedicated article on the Indo-Canadian population of Metro Vancouver? (Vancouver, Surrey, and other Vancouver suburbs). There are books, articles, etc. focusing on Indo-Canadians in Metro Vancouver and there are books, articles, etc. focusing on Indo-Canadians in British Columbia. A Wikipedia editor believes that the Indo-Canadian communities in Metro Vancouver cannot be separated from those elsewhere in the province and therefore it's not appropriate to have a separate article focusing on those from Vancouver. WhisperToMe (talk) 06:19, 23 October 2014 (UTC)

That the query here can't even get the proper usage re Greater Vancouver correct for starters, "A wikipedia editor" who's from British Columbia and had lived in Greater Vancouver and the Lower Mainland for 40 years is who is saying that the history and society of Indo-Canadians and any other ethnic group in Vancouver can NOT be treated separately - as is being touted here by someone who didn't know anything about BC, or even the right term to use for Indo-Canadians, less than two weeks ago. \
Claiming that they can and should be treated separately is completely original research/synth and is being advanced by somebody without any deep knowledge of BC, of Indo-Canadian history/society, based apparently on cursory reading of titles only. The "in BC" title was seemingly created to prevent me moving the "in GV" title there asI I had done with another of his creations, formerly Germans in Vancouver.
Right now, though my original opposition to a separate "Asian Indians in Vancouver" title from the national Indo-Canadians title was the foundation of the merge discussion in question, with the huge amount of work this earnest young editor has amassed in the course of a mere few scant days, the more obvious POV fork separating/claiming that writing about "in BC" so as to include Greater Vancouver would be "original research" is the opposite of the case; in reality writing about such topics so as to limit them to Vancouver or Greater Vancouver as though they were separate or separable from BC-at-large contexts titling is where the original research lies. "but but but but" by pointing at and adding up titles to provide justification for this sophomoric separation is clear evidence of SYNTH;' not listening to but repeatedly rejecting the advice of a local editor of long standing is AGF, pure and simple; not listening to reason while scurrying around to find/fit guidelines and misuses of sources to justify bad ideas....well, that's just purely wikipedian. Skookum1 (talk) 07:07, 23 October 2014 (UTC)

Skookum has a point does he not?♦ Dr. Blofeld 13:46, 23 October 2014 (UTC)

I understand what his point is, but I believe his assertions are not supported by the literature. Take a look at these statistics on the Indo-Canadians in Vancouver:
  • International Journal of Punjab Studies, Volume 2. Sage Publications, 1995. p. 178. "[...]and also in the two largest populations of Sikhs outside of India — in Britain, in London, and in Canada, in Vancouver."
  • Tucker, Alan. The Penguin Guide to Canada. Penguin Books, 1991. p. 539. "Vancouver has the largest overseas community of South Asians (from India, Pakistan, and Sri Lanka — many by way of Uganda or Fiji) outside of London, and the largest Sikh community outside of India." - See search page
  • Todd, Douglas. "Mapping our ethnicity Part 1: South Asia in Surrey" (Archive). Vancouver Sun. May 2, 2012. Retrieved on October 23, 2014. "West Newton is where Metro Vancouver’s main annual Vaisakhi parade draws hundreds of thousands of Sikh and Hindu celebrants. It’s among the largest South Asian diaspora communities on the planet — second only to enclaves in London."
Second largest South Asian/Sikh diaspora after London... Those are some serious assertions of independent notability, aren't they?
When discussing the Irish immigration to New York state, would it be fair to say that New York City has its own distinct aspects of Irish immigration? Surely Irish people went elsewhere in the state. Surely Irish immigration elsewhere in the state can be discussed. That doesn't mean we should trash Irish in New York City which survived AFD back in 2007.
Wikipedia is not based not on the personal experiences of its editors. Let's take a look at WP:V: "Its content is determined by previously published information rather than the beliefs or experiences of its editors." - That's the reason why Skookum's argument is flawed. It is based on his experiences and beliefs and not what the literature actually says.
WhisperToMe (talk) 14:33, 23 October 2014 (UTC)
Geezus "notability" is NOT the issue here, the context of the article is the issue; I've made that point over and over again and been buried by presumptions and dismissals and "refusing to listen" to an informed local. Indo-Canadian Surrey and Indo-Canadian Abbotsford, so to speak, cannot be separated as topics, for example. But tell that to someone who less than ten days had never heard of either place, much less knew the right term to use; plus condescending and patronizing lines like placate Skookum to show I'm not ignoring the rest of the province. (that was an early objection from him)..."early objection" being less than a few days ago, and in fact I'd commented on his ignorance of the rest of the province. More condescension: "why Skookum's argument is flawed. It is based on his experiences and beliefs and not what the literature actually says." My almost-59 years of experience but what I'm talking about is not BELIEFS which is utterly ludicrous to state; I'm talking facts; if the literature and its titling focus on the province's main metropolis is one thing, pretending that my informed advice is "flawed" when you're a complete neophyte on (a) Indo-Canadians and (b) British Columbia is insulting rubbish. I advised you that a province-wide article was the way to go rather than one limited to Greater Vancouver (actually your original title used "Vancouver" only) and you not only ignored me and went and created a POV fork and now are stonewalling and forum-shopping to maintain your agenda...which seems to be WP:OWN, over and and over again. Given your userpage says you are a "Young Adult" it seems that I am around FORTY YEARS OLDER THAN YOU, grew up in a town and went to a high school with many Indo-Canadians in it (Mission) your comments about my "experience and beliefs" being admissible is immature and pretentious beyond belief. You have commandeered a major topic and presumed to author it yourself, so rapidly I must wonder, in fact, about COPYVIO.....Skookum1 (talk) 15:00, 23 October 2014 (UTC)
More and more I'm of the opinion there should be a "knowledge test" and "logic test" for becoming an admin.... as well as basic courtesy rather than fake wikiquette...such as "respecting your elders".Skookum1 (talk) 15:04, 23 October 2014 (UTC)
If an article concept has independent notability then it should get its own article period.
Sir, your experiences are also not what determines the content of the articles: WP:V: "Its content is determined by previously published information rather than the beliefs or experiences of its editors." I think it's very regrettable that your argument is that your view is better because of who you are, not because of any reliable sources that have presented. This goes against the very principle of WP:V where who you are does not matter.
If you think there's a copyvio, please look at the sources and compare them to the article, and you tell me what they are. You can ask any people good at checking the sources and comparing them to the articles. Why not ask User:Moonriddengirl? She's very good at checking for these things so have her check Indo-Canadians in Greater Vancouver and Indo-Canadians in British Columbia. If you can't find any and if she can't find any, then that's that. I am very cautious and careful about avoiding close paraphrasing and I try my best to avoid it. If you cannot find any instances of COPYVIO, kindly stop the accusations. @Moonriddengirl:
WhisperToMe (talk) 15:14, 23 October 2014 (UTC)
I find it "very disturbing" that someone who only found out about the proper term to use had commandeered a major topic by loading up on sources he's found without fully reading them, and ignoring advice on where and how to find other sources as to where to look for items on IC's beyond Greater Vancouver, and pompously lecturing me and patronizing me over and over again. I am not in range of any BC libraries (he knows I'm in Cambodia) and he's spent day and night for week "advancing his obstinacy" and making demands on my time that I do what he says WHY should I do that? If he wanted helpdoing the research he should have done before starting these articles at all he should have been more open to input from somebody actually FROM the places he's talking about. There are many aspects to this subject you have no clue about yet, some of the very politically volatile (e.g. Indo-Canadian crime for starters) ' IO referred you to local histories (meaning those of BC's other large centres, many of which have notable Indo-Canadian populations intrinsically connected to those in the Lower Mainland (Greater Vancouver/Fraser Valley); that I can't respond to your snap-of-the-fingers "show it to me now [or I'll file a requsted move it back where I created it at) is yet more wiki-arrogance.Skookum1 (talk) 15:24, 23 October 2014 (UTC)
WP:RX and http://reddit.com/r/scholar (no, that's not using Reddit as a source) can help you get whatever you need.
Now, I must defend WP:V as a core principle of this project. Thank you.
WhisperToMe (talk) 15:35, 23 October 2014 (UTC)

Staying on topic, I think using the books of Kamala Elizabeth Nayar can find non-OR ways of distinguishing rural BC Indo-Canadians and Vancouver Indo-Canadians. This is one piece of evidence I have:

The Punjabis in British Columbia: p. [3]: "There is a striking difference between Skeena Punjabis and urban Punjabis with respect to the fourth stage[...]For Punjabis living in large Canadian urban centres like Toronto and Vancouver,[...]On the other hand, for those living in the Skeena region,[...]"
She talks about the differences between urban and rural Indo-Canadians. She hasn't said "Vancouver" in isolation (she says it as an example of a large Canadian urban center) but I'll comb her works with a fine brush and see what I can find.

WhisperToMe (talk) 15:47, 23 October 2014 (UTC)

Pictogram voting comment.svg Comment As I thought: Nayar's book makes a point of specifically comparing Indo-Canadians in Vancouver to Indo-Canadians in Skeena.

  • The Punjabis in British Columbia: p. xx.
    • "Not only do these internal migrants now face the hererogeneity of the Vancouver Punjabi community, but they encounter a different experience of multiculturalism from the one they encountered in Skeena."
    • "[...]this study should prove useful not only to members of the Punjabi community in Canada, but also to Canadians in general because it provides a more comprehensive understanding of the socio-cultural and econoic dynamics of the Punjabi community in the Skeena region of northwestern BC and Metro Vancouver."

Now we have a book which intentionally discusses Vancouver in isolation and compares it to another BC community. The same person wrote a Vancouver-specific book, The Sikh Diaspora in Vancouver: Three Generations Amid Tradition, Modernity, and Multiculturalism. You can see a preview of the book. In fact:

  • The preface in p. xi says: "This study examines the Sikh community's process of adaptation to Canadian society in Vancouver." and "Especially so, because the Sikh community in Vancouver is unique among South Asian communities in that many of its members hail from an agricultural society."

WhisperToMe (talk) 15:52, 23 October 2014 (UTC)

Pictogram voting comment.svg Comment Nayar, The Sikh Diaspora in Vancouver, p. 201: "In contrast to Sikhs in Vancouver, which has a large Sikh community, [sic] Sikhs in small towns throughout British Columbia interact far more with other communities." and "The Vancouver Sikh community is more insulated from the mainstream and is networked according to village and clan ties (partic-[...]" (don't have the preview for p. 202-203) - If Nayar is making Vancouver Sikhs to be distinct, then it's not SYNTH to write an article specifically about the habits of Vancouver Sikhs. WhisperToMe (talk) 16:44, 23 October 2014 (UTC)

Pictogram voting comment.svg Comment Here is something interesting:

  • Nayar, The Sikh Diaspora in Vancouver, p. 211. "Those who have been raised in Vancouver, who have experienced the world basically confined to that city, view multiculturalism as 'how things are' and 'necessarily good' because it allows people to keep their culture[...]In contrast, those who have lived elsewhere - be it in England, Singapore, Hong Kong, small B.C. towns, or the United States - assess multiculturalism more critically." - Again, making Vancouver Sikhs a distinct topic
  • Now here: Nayar, The Punjabis in British Columbia, p. 286-287. "Nayar's social-anthropological study - on the multifaceted process of the Vancouver Sikh community's adaptation[...]the Canadian-born generation living in Vancouver whose antecedents had originally settled in rural BC tended to assess Canada's policy of multiculturalism more critically than those born and raised in BC's Lower Mainland." - This is referring to the previous book The Sikh Diaspora in Vancouver and is in the footnotes.

WhisperToMe (talk) 17:19, 23 October 2014 (UTC)

  • there does not appear to be SYN issue - multiple reliable sources specifically discussing population distinctions. The question is: "Can this aspect be covered appropriately in the 'parent level article' without creating an WP:UNDUE weight to the distinctions, or are there sufficient sources to create a valid spin out?" Wikipedia generally favors the spin out where there are sufficient sources to create content more than a stub paragraph, which appears to be the case here. -- TRPoD aka The Red Pen of Doom 17:32, 23 October 2014 (UTC)
I agree that this isn't a synth/OR problem. I think that Skookum1 is not being very civil, and that his arguments from personal experience are obviously spurious, at its core the dispute isn't about original research, just demarcation. 0x0077BE [talk/contrib] 19:40, 23 October 2014 (UTC)

Present Perfect[edit]

I have been reading the article Present perfect, generally, it seems to lack verifiable sources but my main concern is connected with the table present which lists examples of present perfect being used. The very first point is not a clear example of present perfect because it is dependant on the type of verb used. In most situations, to express continuation up to now it would be preferable to use present perfect continuous, however certain verbs are unable to be expressed in a continuous form (have, know, for example) and therefore have to be expressed in Present perfect. This should probably be made clear in the article and definitely shouldn't be used as the first example of present perfect in use.

This can all be supported in 'Practical English Usage' by Swan (Oxford 2005) page 440.

I'm not sure this is really OR, beyond the fact that at some point most examples are going to be essentially some form of original content. That said, I can see how it can be confusing, I'm going to substitute in one of the examples from Uses of English verb forms. 0x0077BE [talk/contrib] 20:01, 23 October 2014 (UTC)