Talk:History of Crayola crayons

Bot-listed on creation

Not a copy. The content here is being split from the other article, which I'm currently in the process of revising. I wanted to make sure it was saved here first. P Aculeius (talk) 01:26, 28 June 2015 (UTC)

Not sure why this page was bot-listed as a suspected copyright violation. The content was simply copied from another Wikipedia article (List of Crayola crayon colors), since the history section of that page and these tables were being removed in favour of a more straightforward list presentation. I wrote all of the text myself, and none of it is copied or directly quoted from any of the sources. P Aculeius (talk) 13:02, 29 June 2015 (UTC)

Torch Red: Renamed in 1998 or 1999?

Yesterday, I edited this page and the list of crayon colors to say that Torch Red was renamed "Scarlet" in 1999. Looking at the page now, I suspected P Aculeius would edit it to say that the color was renamed not long after it came out (1998), and sure enough, it was.

Here are some reasons why I thought my edit was correct:

• The source where this information came from, CrayonCollecting.com, listed the release year of the Wizard's Giant Box Crayons and the renamed Scarlet crayon as 1998, while another page there focusing around the Torch Red crayon's rarity lists the release year of the former as 1999, the same year the color Indigo was released.

In addition to all of this, the Crayola Color Chronology timeline says that Torch Red was renamed and Indigo was introduced in 2000 rather than either 1999 or 1998, thus expanding the confusion further; if a introduced-for-1998 crayon was renamed the exact same year it was introduced instead of a year or so afterward, then why does a different page on the source website say the box it was first available in was released in 1999 and the Color Chronology say it was released in 2000?

• In the "Retro" box of crayons, the true color for Torch Red was used for Shag Carpet Orange. However, on the crayon wrapper, it is clearly the same color wrapper as Torch Red's, suggesting that the latter name may have been used in 1999 as well as 1998.

What do you think about this situation? 2601:82:200:9E0B:F173:977B:EE9:17B4 (talk) 22:00, 20 April 2016 (UTC)

I'm glad I was able to fulfill your expectations. I've gone over the site very carefully to find basis for your edits. Part 25 identifies the 1997 "True Blue Heroes" color naming contest as the source of the Torch Red color, although the crayon did not receive a permanent name or become part of the regular assortment until the contest winners were announced in 1998, as described in Part 29. This article was written by Ed Welter, the site host, with thorough documentation, and is consistently presented in his chronological table.

The article specifically treating the Torch Red crayon is by another author, Kurt Baty, describing his quest to obtain the crayon. It consists largely of e-mails back and forth between the author and unnamed persons working for Crayola, all saying what they thought happened several years earlier, and apparently based on the best of their recollection, rather than anything documented. The Wizard's Giant Box of Crayons, which forms the backbone of Mr. Baty's story, was first dated 1997. Mr. Welter's history agrees, but states that the box did not appear on the market until early 1998. This is hardly surprising, as copyrights and trademarks for a particular design might be registered before the product is ready to be shipped to retailers.

Nowhere does this article say that the name "Scarlet" was introduced in 1999. It says that all of the 1999-dated boxes searched contained that crayon, which supports, rather than refutes, Mr. Welter's article with respect to the assertion that the crayon was renamed during its first year of general release. If the name had been changed in 1999, one would expect a significant number of 1999-dated boxes to contain "Torch Red". But none of them did; and despite Mr. Baty's conclusion that all the "Torch Red" crayons were destroyed, it seems highly unlikely that Crayola would do so if it could simply use up the existing stock in 1999-dated boxes; after all, the color wasn't different, only the name; and the name change wasn't publicized (why would it be?).

The 1997-dated box in which Mr. Baty actually found a "Torch Red" crayon was, presumably, released no earlier than 1998, since none of the boxes in question were marketed until 1998. References to the year 2000 in anonymous e-mails of unknown authority shouting in all caps really aren't very helpful, since we know that the crayon had already been renamed by the beginning of 1999. It tells us nothing that the occasional "Torch Red" crayon might be found in later boxes, or even whole batches of boxes, as Crayola could have found old stock in usable condition and simply shipped it at any point after renaming the color.

I cannot find anything stating that Indigo replaced Thistle in 2000. Part 30 specifically says that this occurred in 1999. The tables included agree with this, although between text and tables one should always trust text, since the text will be included in order, while a date in a table could easily be mistyped. There doesn't seem to be anything suggesting that this occurred in 1998, either. I assume that the timeline you're referring to is Wikipedia's Timeline of Crayola. This is not a helpful source; first we can't use other Wikipedia articles as sources; secondly that article has no citations whatever for its chronology (the only citations in the article are for claims that a senior moulder was in fact color-blind, and that Mr. Rogers poured the wax for the 100 billionth crayon). Since this article is thoroughly sourced, and Mr. Welter's article is clearly explained and documented, rather than a list of highlights cherry-picked for publicity purposes, there is no reason to regard the the Wikipedia Timeline of Crayola as authoritative by comparison.

If the timeline needs to be revised, feel free to take on that project; this article might be useful as a guide, but not a source. I can see there are other unsupported claims in it; for example, repeated references to the "retirement" of discontinued colors into the Crayola "Hall of Fame". This seems to have been a one-time publicity concept that has since been abandoned by Crayola; their current site contains no references to other colors being "retired into the Hall of Fame", and a guide to roadside attractions dated this year still describes it as containing the same eight colors "retired" in 1990. P Aculeius (talk) 03:42, 21 April 2016 (UTC)

Thanks P Aculeius for responding. I hadn't visited the crayon related wiki section for a few years now but you've cited my crayon research data well. The dates on Crayola items is at best an estimation in many cases. Sure, the containers contain dates on them starting in the 1990s but just because they printed a "1999" on a box might actually mean it didn't hit market until 2000, as an example. The crayons themselves are sometimes even less accurate given that they just "appear" in new assortments either as an introduction or "disappear" as a change in their strategy. Just because a container gets phased out of their catalog of offerings doesn't mean they necessarily used up all of the crayons slated for that container. Sometimes colors that they've changed have spillover until they use up the inventory unless there was a valid reason to destroy the remaining inventory (not cost efficient). Still, this Wiki page looks tremendously better than what was available several years ago. Thanks for that! Ed Welter (talk) 10:32, 19 May 2016 (UTC)

You hit the hammer in the nail there. While the Indigo color did not come around until circa September 1999 (when the renamed Indian Red (Chestnut) started to show up in crayon boxes (according to an archived news article about the latter color)), I didn't have a clue whether or not Torch Red was renamed Scarlet in either its debut year or 1999.

Furthermore, some 1997 or 1998 dated boxes with the original Torch Red color inside them could at least be available for the final quarter of 1998 and/or the first quarters of 1999 until the box is redesigned. This aspect might at least help anyone out a bit at best. 2601:82:203:9070:AD3E:606:F1B9:F92D (talk) 01:35, 28 July 2016 (UTC)

I see now, and thank you for helping me out with my mystery. Although I still think that the name Scarlet came about in 1999 (specifically because the Retro Shag Carpet Orange crayon's wrapper is the same color as that of Torch Red's), I appreciate your help and now know better. The Wizard's Giant Box of Crayons, however, was released in 1999, and the backbone of Kurt Baty's story that you were talking about is actually the original Giant Box of Crayons from 1998. 2601:82:200:9E0B:F1DB:EA7E:9CBB:5842 (talk) 23:51, 21 April 2016 (UTC)

After working on here, I thought about this talk and decided to come up with an observation:

Considering how often Crayola leaves their boxes hanging, Torch Red couldn't have been renamed in 1998. The idea that one (like P Aculeius, for example) would think that if it were renamed a year after, the box it was obtained in would be dated that particular year, can be considered largely false, as sometimes, Crayola would not update their boxes, leaving the current ones up for a certain amount of time before switching to a new design. This can be said for 1997 and 1998-dated larger boxes with the Torch Red crayon in them; the second version of the Giant Chest of Crayons, the Wizard's Giant Box of Crayons, is dated (and was apparently released in late) 1999. Thus, it can be said that the Giant Chest of Crayons and the first version of the Ultimate Crayon Bucket may have stuck around until 1999, and my point can be proven easily: the two boxes have Indian Red, Torch Red, and Thistle in them; the Wizard's Giant Box of Crayons contains Chestnut, Scarlet, and Indigo.

Continuing on, I received a 2010-dated 120-pack in 2013, and a 2014-dated version of the same size in 2016. Both boxes contain the same colors and names, and if we were to apply this to a early pack of 120 purchased or put on sale in early 1999, Torch Red would have been renamed a few months later, not 1998 as noted here. I had the Wizard's Giant Box of Crayons version for a long time (remains of it are scattered around the places I would not bother looking around in), and as expected, Scarlet is in that pack, not Torch Red.2601:82:203:9070:71F0:8A8F:9092:C7AA (talk) 19:11, 13 September 2016 (UTC)

I'm sorry, but you're misinterpreting what I said. I never suggested that boxes with a certain date on them had to have been manufactured (much less shipped) in that year. That's the exact opposite of what I said. It's completely irrelevant if torch red crayons appear in some boxes with dates after 1998, since those could represent old stock being used up. Mr. Baty's conclusions were based on the assumption that all of the remaining torch red crayons were destroyed when the color was renamed. But that assumption makes no sense, because there was no need to prevent crayons labeled "torch red" instead of "scarlet" from reaching the market.

While it's not impossible for 1999-dated boxes to have been shipped or sold after 1999, presumably many of them were shipped and sold in 1999. Mr. Baty didn't find any "torch red" crayons in 1999-dated boxes. If all of the boxes with that date contained "scarlet", then it's much less likely that the name change occurred in the course of the year, or one would have expected to find some 1999-dated boxes containing "torch red". Even if the change occurred in 1998, it wouldn't be that surprising if there were some "torch red" crayons in 1999-dated boxes, as old stock was used up. The fact that none of the 1999 boxes Mr. Baty received contained it strongly argues that the change occurred before the 1999 boxes were manufactured, which might actually have begun the previous year.

If you want to argue that the name change occurred after then, you really need to provide something that makes that seem likely, and explains why such a small number of 1999-dated boxes contained the crayon that none of those searched by Mr. Baty did. Something that doesn't merely contradict the current article, but shows some reason why that source should be believed and the current one should not. P Aculeius (talk) 00:51, 14 September 2016 (UTC)

Thanks for correcting me there. You made a perfect point there, and you're right, it is irrevelant for someone to find a Torch Red crayon in a post-1998 box, as that may indicate older stock being used.

There is another way this can work, however; while Kurt Baty didn't find any 1999-dated boxes with the Torch Red name in them (as you pointed out), a possible theory is that the change might have either happened before Indian Red was renamed Chestnut or during this event (where all the box graphics were changed). As I said before, the second version of the 120-pack was dated 1999, and there is a good possibility that the 1997-dated Giant Chest stuck around until the renaming process happened. If we were to go by standard color logic (going by what Crayola claims on their website), the true color was introduced in 1998, with Crayola suggesting the renaming to take place in 2000 (which is much less possible, as it couldn't have been renamed then) rather than 1999 (as I think) or 1998 (as CrayonCollecting.com thinks). As I write, there is an eBay listing for the Wizard's Giant Box, and it notes that the seller is not sure if Torch Red is in there. In reality, that name wouldn't be in that box, as the Scarlet name, apparently, wouldn't have come until almost right when the box in question was released. 2601:82:203:9070:E9B5:5275:8AE:2943 (talk) 21:40, 14 September 2016 (UTC)

I'm afraid that still comes under the heading of pure speculation. I can't absolutely prove that "torch red" didn't become "scarlet" until 1999. But the available evidence, and the best available source (Welter) indicate that it occurred or apparently occurred during the first year of production, and the boxes investigated by Mr. Baty support that hypothesis. What we don't have is evidence, as opposed to speculation, that it occurred later. The evidence wouldn't be the existence of "torch red" crayons in boxes dated after 1998, since that would be expected as old stock was used up (potentially turning up even after 1999). It'd be that a high proportion of boxes manufactured in or after 1999 would contain it. Like, 20%, 40%. Something that would suggest at least two or three months elapsed with only "torch red" being produced before "scarlet" went into production. It'd be pretty hard to find enough evidence that this is the case from independent sources, but if you can, then there'd be a basis for changing the date given in the article, or at least adding a note. P Aculeius (talk) 00:44, 15 September 2016 (UTC)

Thanks for correcting me there (again). You're right about the research, but (I may be wrong here, and this is bound to be taken as speculation than fact) the renaming supposedly more likely apparently occured in 1998 instead of fully. Why I think this is that older, 1997-98 dated boxes could have still waited to get purchased by consumers around late 1998 to early 1999, and some stores may not have switched to using the Scarlet wrappers by then. If your theory (I think) of 1999-manufactured boxes with Torch Red in them was correct, then (speculation again) we would consider the renaming to take place circa 1998-1999 (combining both the sources' and my information together). 2601:82:203:9070:215F:675A:F512:3AFB (talk) 23:54, 15 September 2016 (UTC)

"Apparent" means that it appears to have been so, not that the process was incomplete. Stores didn't change the wrappers; the manufacturer did. Since we know that boxes were dated each year of production at this time, and that the design might be prepared months ahead of time, it's probably safe to say that no 1998 boxes were being produced in 1999. At least, that's a reasonable presumption, and would need evidence to rebut it, of which there currently is none. The only evidence that would indicate a name change in 1999, apart from official company records or publications indicating when it occurred (also not known to exist), would be if a substantial proportion of 1999-dated boxes contained "torch red", indicating that wrappers bearing this name were still in production for a period of time that year. A small proportion would at best be ambiguous, since it could be accounted for by old stock being used up, or by 1999-dated boxes that were actually shipped in late 1998. What we can't do is "split the difference" between a statement from a reliable source and something else that is demonstrably no more than speculation. Mr. Baty didn't set out to establish a firm date, didn't find any official documents giving one (e-mails from anonymous sources within the company recalling what they thought happened years after the fact can't be considered official documents), and wasn't particularly interested in the exact date to begin with. He just wanted to obtain a "torch red" crayon. It makes no sense to change the date based on anything that's been described anywhere in this discussion. For the time being, it needs to stay as it is. P Aculeius (talk) 02:31, 16 September 2016 (UTC)

Thanks a lot for helping. However, when I said "stores may not have switched to using the newer wrappers", I meant that the stores could have still have some older boxes until getting renamed crayons in boxes sent to them by the manufacturer. 2601:82:203:9070:A986:E8A8:3DF1:77B4 (talk) 19:15, 16 September 2016 (UTC)

External links modified

Hello fellow Wikipedians,

I have just modified one external link on History of Crayola crayons. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:

• Added archive https://web.archive.org/web/20130117194713/http://www.crayolastore.com/custom_box.do to http://www.crayolastore.com/custom_box.do

When you have finished reviewing my changes, you may follow the instructions on the template below to fix any issues with the URLs.

Y An editor has reviewed this edit and fixed any errors that were found.

• If you have discovered URLs which were erroneously considered dead by the bot, you can report them with this tool.
• If you found an error with any archives or the URLs themselves, you can fix them with this tool.

Cheers.—InternetArchiveBot (Report bug) 16:25, 4 November 2017 (UTC)

Proposal

I am a little disappointed that the AfD outcome was "keep" and not "stub" but we do need to get the cruft out of this article. I'm going to proceed much more cautiously than if the result had been "stub".

Step 1

I am going to remove the unreferenced or blatantly improperly referenced Hex, RGB and HSV values. I think this is uncontroversial under WP:V so I am just going to do it. This will leave the tables with their remaining columns including the colours. That does not mean that I think the tables are defensible as they stand.

Step 2

I propose that we remove the tables and reduce the lists of colours to a plain comma separated list with no text or background colours. The notes with referenced would be kept and the ones without removed. (Maybe one or two could be worth keeping with a Citation Needed tag but certainly not many.) I am not going to do this unless there is a consensus for it. I would like to hear opinions on this. If anybody has counter-proposals then we can see which has the most support. We can make this an RFC if anybody wants to be official about it, or if we don't get enough different participants without it.

--DanielRigal (talk) 19:21, 30 July 2018 (UTC)

You shouldn't even have thought about doing this without first discussing it on the talk page and gaining consensus. But you went ahead and did it anyway, gutting the article as a backdoor way of accomplishing what you failed to achieve in the deletion discussion. It's been clearly explained more than once that the information here is properly and reliably sourced, yet you persist in asserting that it's not, even making patently false statements about your changes being "uncontroversial" under a policy that says nothing of the sort. The fact that you assert the uncontroversial nature of something that hasn't even been proposed on the article's talk page—after ignoring everything said about the same topic in the deletion debate—simply proves that you have no intention whatever of working collaboratively for the encyclopedia. You're just out to do whatever the heck you want, anyone else's opinion be darned. Your arguments were rejected in the deletion discussion: the article is cruft. The article is fancruft. The article is listcruft. The article is garbage. The valid contents of the article are obscured by garbage. If it were worth keeping, it wouldn't be nominated for deletion. Wikipedia is not a catalogue. The article is original research. The topic should be covered by a fan site, not Wikipedia. The topic is not notable. The contents are not verifiable. The sources are no good. One of the sources is commercial. One of the sources is published on a web site. A long litany of excuses all refuted and addressed by Wikipedia policy, and you're still trying to achieve the same thing you couldn't do in the deletion discussion. Your attempt to get rid of the article failed, so now you're just deleting as much of it as you can get away with. You should just acknowledge defeat and find something to work on where you can actually collaborate to improve Wikipedia, instead of ripping apart articles that annoy you with their very presence. P Aculeius (talk) 20:33, 30 July 2018 (UTC)

I'm not sure how far I am expected to stretch the assumption of good faith. I will try to stretch it a little further. You need to start talking on-topic. The stuff I removed is unverified and there is no way to verify it. Unverifiable stuff is to be removed. That is policy. There is no point trying to pretend that you didn't hear that. That stuff has to go if the article is to have a future as a genuine history. Are you willing to let this obvious improvement go through or do I need to start an RfC? Maybe you would prefer the dispute resolution process? Do you really want that?

Also, let me give you some advice. I know you think I am your enemy but I'm not. I've seen you remove vandalism and I know that you are trying to do something that you feel is worthwhile. The problem is that the walled garden of articles that you want to build is not appropriate to Wikipedia. I genuinely think that you should start a Crayola fan wiki. That is not me telling you to go away. If you would do that then you could contribute to both that and Wikipeda. You could put stuff appropriate to Wikipedia here and anything else you like in the fan wiki. That way everybody can have what they want without any problems. I'd support any request you made for access to deleted material if you wanted it for the fan wiki. This is a way forward. If you continue to try to turn Wikipedia into your personal fan wiki, ignoring policy when it doesn't suit you, then I fear that WP:NOTHERE may apply. There is no way forward in that. Please, step back from that. --DanielRigal (talk) 22:14, 30 July 2018 (UTC)

I still believe that my Step 1 above is uncontroversial but, to show good faith, I will wait for somebody else to endorse it before removing the unverifiable material again. I don't want to get bogged down in this though. It is Step 2 that needs discussion and where there may be a more justifiable diversity of opinion. --DanielRigal (talk) 22:32, 30 July 2018 (UTC)

Also, to further show good faith, I invite you to prove me wrong. Show me how to verify one set of the Hex, RGB and HSV values. Assume that I am really stupid and walk me through it step by step. Show me how to extract that data from a source that meets WP:RS in a valid way that does not constitute WP:OR. It may be that the scales fall from my eyes. Or, if I am an incurable idiot, maybe you can convince other people. Surely that is worth a try? I don't mind if you make me look stupid. I just want to see you make a meaningful engagement with the requirements of WP:V instead of waving it away like it is somebody else's problem. --DanielRigal (talk) 22:38, 30 July 2018 (UTC)

I'm not trying to make you look stupid. You just seem to be crusading against all topics involving Crayola colors. Three have already been deleted, and you've argued vociferously for deleting all or most of the remaining three, despite what I think—and what most of the other editors who've maintained the articles think—is adequate and verifiable documentation. Now, I've said more than once that simply describing the color of a crayon by eyeballing it is verifiable, just as if you said, "the book is about six inches long" or "the setting sun appears reddish". Wikipedia policy explicitly states that this is acceptable, as long as anyone can look at the same object and see whether the description is accurate. But this method isn't very precise, and people could easily describe the color of a crayon differently, even looking at the exact same swatch of color. That's why in nearly every case, the color codes are instead based on digital swatches that can be measured with great precision and little variability. That is, everyone who applies the same method should get either an identical reading, or a reading within a very small margin of error, which could be narrowed down in future if better samples become available. There's nothing wrong with presenting the greatest precision possible, and refining it if better sources or methods become available later. That's actually how Wikipedia tends to work.

There are different ways you can measure the digital color components of pixels on the computer. In theory they shouldn't vary at all based on your monitor settings, which don't affect the information about the components, although they can affect what each combination of components looks like to your eyes. These values, whether they be Red, Green, and Blue, expressed as 0–255 or as hex codes, or as Hue, Saturation, and Value, are precise mathematical concepts with exact equivalents. If the source image uses, for instance, 255 0 0 (solid red), you're also looking at hex code FF0000, and hue 0 100 100, and any method used to measure this should provide at least one of those values, no matter what your monitor is set to. It's the same for everyone, even though the monitors may look different. That's what makes a digital swatch verifiable: anyone can examine it to find out what the components are, and they'll get the same results, or in the case of, say, a photograph of wax from a crayon, approximately the same results. It's much better than eyeballing the color, although since anybody can color with a crayon and describe the hue, saturation, and value (I doubt most of our brains convert readily to RGB) as best we can, and someone with better information can correct it later. I consider that a last resort, but there were one or two colors for which no digital or digitized swatches were available.

I have a desk accessory called "Digital Color Meter" that comes with my operating system. You might have it too. But if you don't, this page discusses some equivalents. It doesn't matter which one you use, because you should get the same result no matter which you use. Everyone should see roughly the same RGB/HSV/Hex values if they measure a swatch's color. It's the digital equivalent to holding up a ruler to an object, except it's much more precise. If you don't have any of these applications, you can use Photoshop, Paint, Gimp, or other graphics editors. You don't actually have to edit anything, though. Just download or take a screenshot of the color swatch, and use the eyedropper tool to find out exactly what color it is. In the case of Crayola's current color swatches, the values are exact, because they're digitally generated and the website tells your browser exactly what RGB/HSV/Hex values to display for it, and every pixel of the swatch should be identical. Your screenshot might have fuzzy edges due to dithering, but anywhere in the middle should be good. If for some reason different pixels are slightly different colors, you can average the values. This is a "calculation", and defined as not original research, because anyone could do the exact same thing. It's just numbers; it's not editorializing. However, last I checked, Crayola's swatches were all pure color, with all pixels identical.

For most colors that are currently being produced, or which were still included in the assortment when this page was made, or which appear on archived versions of official releases from Crayola displaying what those colors are supposed to look like, this is the best possible source, at least for what the color is supposed to represent. Will you see a perfect duplicate of this when you color on a piece of paper? Probably not, because you won't be laying down wax with uniform thickness, with no paper visible through it, and what you see will be affected by the color of the paper, the light you view it with, and possibly manufacturing variations in the wax. But it's what the color is meant to look like. An official digital representation of the theoretical color. There could also be mistakes; two swatches with the same label, or one with the wrong color attached (not obvious if they're similar), just like you can find a typo in a book, or someone in a movie calls a character by the wrong name (like when Luke Skywalker shouts "Carrie!" after blowing up the Death Star). If there's an obvious mistake, you can look to other sources that appear to be correct.

The Digital Color Meter averages pixels for you, taking out another potential source of subjectivity. You just have to set the aperture to get the value of the most intense color in a swatch, such as one taken from an actual crayon. For older colors, ones not produced for many years, that's the best available source. I still have my old crayons, including several colors that aren't made anymore. But I found that my scanner couldn't reproduce swatches I made well. Some colors looked right, others didn't. The swatches used by Ed Welter to illustrate his history of Crayola Crayons seem to have digitized much better than that, and they include the best available samples of the oldest crayons, ones produced so long ago that very few people have them. If you measure the colors of those swatches, focusing on the most intensely-colored parts, you should get approximately the same values as the ones given in the tables. That's what makes them verifiable: anyone can use the same app or the same kind of app to measure a swatch of color, and they should all get roughly the same result if done carefully.

I don't imagine this will convince you at all, but I believe this is fully within both the letter and the spirit of Wikipedia's policy for verifiability. But I ask you to consider something else: if the colors as depicted appear to an editor, such as yourself, to be accurate representations of the colors as they appear on Crayola's website, Mr. Welter's swatches, or swatches you produce yourself with the crayons in question, then is the article in any way improved by removing the descriptions of those colors that do nothing more than define the exact colors depicted in the tables according to three different and precisely equivalent methods of describing them? As the verifiability policy states, contentious material, material likely to be challenged, is subject to verification. But if you or anyone else can see that the colors match up, then it's not contentious. It's readily apparent to anyone who looks at the same thing. If after all that, you find reason to believe that a particular swatch can be improved using a better source, or evidence that it's not quite accurate, then by all means, fix it. Improve it. But until you find that better source, or evidence of mistake, don't remove what appears to be substantially correct: a simple description of something that anyone can see for themselves, and improve if they're able to do so. That's the collaborative process. Many editors are perfectly satisfied with the sources provided and the way they're used. If you think they could be better, then help make them better. But please don't assume that these numbers come from nowhere, that they're not verifiable, that simply measuring something or carrying out a routine calculation constitutes original research, because that's not how that's defined by Wikipedia policy. P Aculeius (talk) 00:08, 31 July 2018 (UTC)

Thanks for confirming that you are using some sort of measurement software to derive the values yourself. That is WP:OR as I expected and precludes verification with reliable sources. I think you need to accept that this is contrary to policy and find somewhere else to publish your list of values. I think it is time for you to drop the WP:STICK on this. --DanielRigal (talk) 08:45, 31 July 2018 (UTC)

Right, so you're saying you've chosen to ignore everything I said about Wikipedia policy and verifiability, and apparently weren't actually interested in a discussion. Sorry I wasted my time trying to convince you of something when you'd already made up your mind, but claimed you could be convinced otherwise. You've already ignored the result of the deletion discussion on this article, but also the conclusion of the outside admin who closed the deletion discussion on "List", who explicitly stated that there was a strong argument that the sources used were adequate and appropriate for the purpose used: supplying descriptions of the colors. Your contention seems to be that the fact that the manufacturer of a crayon deliberately sets its RGB color as "346 96 93" isn't adequate to show that that's the color it's intended to represent, even if anyone can look at the same source and verify that it's "346 96 93". That being the case, there's no point in discussing any of the less precise measurements, even though anyone can check them for themselves, and determine whether the values used in the table here are accurate. But this isn't supported by any policy. Wikipedia policy says that things that are obvious, things that can be stated objectively and verified by anyone examining them, are valid sources for themselves. If anyone can hold a ruler up to a book and say, "it's approximately 11.25 centimeters tall", then you can cite the book for its own length, even if someone else might come along later and say, "I checked it, and it's 11.23 centimeters". I see no reason why measuring the RGB components of a swatch of color would be any different, as long as everyone who visits the cited source and checks it is going to get substantially the same results. Why are you determined to exclude verifiable information? Is there some benefit to users, apart from, "I don't want the information to be there"? P Aculeius (talk) 13:21, 31 July 2018 (UTC)

You can't post massive TL;DR screeds and expect that every word be read. I have parsed all the relevant parts. TBH, the only truly relevant part to your previous essay was the part where you admitted deriving the colour values yourself. That is enough to doom them on the spot.

The problem here is that you are misconstruing, or utterly ignoring, all policy that does not suit you. You can't publish your own Original Research in Wikipedia, even when your research might be perfectly OK if published elsewhere (Hint!). Your example of measuring a book is another example of improper OR. So your comparison is valid but it doesn't help your case. You can't seriously claim "obviousness" for Hex values. How is it "obvious" that a certain red crayon is, say, #F06030 and not, say, #F25E32? Please consider that you may have completely misunderstood what the policies actually mean. The level of WP:IDHT is becoming disruptive and you need to change your behaviour.

It was not me who put all those articles up for deletion, although I have been sorely tempted to do so in the past. Each article that has been deleted so far has been deleted on a clear consensus from several people that it was bad. It is not just me! I didn't even !vote Delete on all the AfDs. I have suggested merges and improvements on the ones that are capable of being saved. I am not trying to delete all coverage of Crayola and there is no justification for your accusations. You need to stop focusing on me and start looking at yourself.

Anyway. We are both wasting our time here unless we can get some more people in to settle this so I'm going to start an RFC shortly. That should bring some more people in. --DanielRigal (talk) 17:59, 31 July 2018 (UTC)

RfC about the tables of colour values (RGB, HSV and Hexadecimal)

The following discussion is an archived record of a request for comment. Please do not modify it. No further edits should be made to this discussion. A summary of the conclusions reached follows.

There is a consensus to remove unreliably or unsourced colours. This includes colours that were "estimated using a swatch" (as that is not a valid source).

There was no consensus to removed sourced material, though there was a substantial argument made (mostly based on the difference between subtractive and additive colours) to emphasize that the (sourced) Hex values should be in a column that denotes these are Crayola's hex approximations (wording can be hashed out in a separate discussion if necessary). The HSV/RGB values, being "calculated" and not directly referenced, will be removed (see North8000's argument regarding having both decimeters and meters in the same table). Note that this "compromise" falls under the "no consensus" regarding said hex values, and is open for re-negotiation and/or re-litigation of the appropriateness of including hex values in these tables. Primefac (talk) 21:34, 31 August 2018 (UTC)

---

Opinions are requested as to whether the numeric colour values (RGB, HSV and Hexadecimal) in the various tables in this article are appropriate? Is this excessively detailed overcoverage? If such values are deemed potentially appropriate, further opinions are requested as to how they can be validly referenced to provide verifiability. It has become clear (from discussion above) that these values were derived from colour swatches and sampled from websites in what I believe to be an act of original research. I believe the values to be unverified and also intrinsically unverifiable. As a glance at the discussion above will reveal, this is very strongly disputed with no consensus as the correct way to apply the policies, so opinions on this are requested. I have removed the value columns once and have been reverted. Ultimately, I would like to know whether to remove them again? If there is a consensus to remove them here, should similar columns of values also be removed from any of the other articles in the Crayola walled garden that survive their AfDs? --DanielRigal (talk) 18:52, 31 July 2018 (UTC)

Survey

• Support removal as proposer. --DanielRigal (talk) 18:52, 31 July 2018 (UTC)
• Oppose. The information is relevant, desirable, objectively determinable, and verifiable by anyone who checks the sources cited, and therefore does not constitute original research within the definition of the policy. If any format can be determined, all of them are known, since this is a routine calculation, and thus also not original research. P Aculeius (talk) 22:23, 31 July 2018 (UTC)
• Remove some, keep some. Summoned by the bot. Foreword: I believe this is valuable information and it is not excessive detail in my opinion. Now, going into the details, some colours have a reference to the Crayola page, the ones with a footnote referencing to "Crayola: Explore Colors, retrieved 5 May 2015.". I believe those should be kept. In my opinion, the official and ultimate reference for the RGB values of Crayola's named colours is and should be Crayola. I have tested the RGB values referenced in the list and coincide with the RGB code in Crayola's page. This vote/opinion should not be taken lightly, there is more to this RfC than meets the eye. The values were clearly datamined and are not explicitly referenced by Crayola. Still I believe that if Crayola's official page is assigning those RGB values to their own colours, they should be considered official. Going forward, it is my opinion that datamined information is and should be a valid source of information as long as the site is relevant to the data that has been datamined. This is certainly the case. Now, there are other colours whose value is estimated. Indeed they are referenced to a footnote that say "Color values estimated using swatch of original crayon." These values constitute original research and should be removed. Dryfee (talk) 17:07, 2 August 2018 (UTC)

As you moved my reply to this assertion, see below. P Aculeius (talk) 20:40, 2 August 2018 (UTC)

• Oppose per P Aculeius and others above. Also, the color values are extremely helpful in attempting to sort the page - they're pretty helpful to have. Paintspot Infez (talk) 22:00, 2 August 2018 (UTC)
• Oppose per the above - IDONTLIKEIT isn't a valid reason to delete. –Davey2010^Talk 13:39, 7 August 2018 (UTC)
• Support if well-sourced, delete any values sourced to "original color swatches." Per Dryfee, I agree that the Crayola website's color samples are an acceptable source for this data. It's definitely a grey area because the values are extracted from code instead of published text, but it's also a verifiable set of objective numbers with no possibility of measurement error.

The "Color values estimated using swatch of original crayon" source, on the other hand, is a clear case of OR. I would treat this the same way as a science article: The speed of light in a vacuum, for example, is a number that can be measured and verified using the appropriate equipment, but we would never use "estimated using Acme Light Speed Meter" as a source.

A single column (preferably Hex) is adequate to describe the color. Let's keep one and delete the other two.

Minor quibble while we're on the topic: Some of the color names are difficult to read, particularly the ones that use white text on a yellow background. For accessibility, the best practice would be to list the name in a separate column as black text on a white background. –dlthewave ☎ 16:34, 9 August 2018 (UTC)

If the equipment in question is something that anyone reading the article can use, and comes pre-installed on many people's computers, then it's clearly verifiable. Reading RGB/HSV/Hexadecimal from a digital color meter is nothing like measuring the speed of light in a vacuum. But more importantly, you don't even need to measure a swatch of wax for the swatches in the table to be verifiable, because they qualify as illustrations of the colors produced, and as such are acceptable provided that they appear to be accurate representations of said colors—again, something anyone can verify by examining the crayons themselves and drawing with them, or comparing them with extant swatches of those same crayons. Removing the RGB/HSV columns would make the tables unsortable by any useful criterion. Many readers will be familiar with only one of these sets of values, and the ability to sort by hue, saturation, or value as well as by separate channels of red, green, or blue, is desirable, while sorting by hex code isn't useful to anyone. As for the text colors and table arrangement, that's an entirely separate issue from deletion—if the proposed deletion of contents goes ahead, there might as well not be any tables, as half the contents would be permanently blank, and nearly everything left in the tables would simply duplicate the contents of another article. In effect, this proposal achieves by the back door what couldn't be done with the AfD: rendering most of the contents useless, and making merger into other articles necessary. But to address the text issue, black text on light colors affects the appearance of those colors, and on dark colors it's unreadable; when the current tables were designed, white was used uniformly in order to avoid a hodgepodge, as in previous tables. However, in order to ensure readability, olive text was used on the light yellows. P Aculeius (talk) 18:20, 9 August 2018 (UTC)

Perhaps I was unclear, my suggestion would be to have one regular black-and-white column for the names and a second, text-free column for the color swatches. Let's save it for a later discussion after this one is closed. –dlthewave ☎ 18:47, 9 August 2018 (UTC)

• Support removal Sampling colors from the Crayola website is a form of WP:SYNTHESIS and should be considered WP:OR unless the coordinates were explicitly provided. PaleAqua (talk) 03:50, 11 August 2018 (UTC)
• Remove Unsourced and unsourcable because it is faulty. A crayon (or any object which is not itself a light source) doesn't even have inherent RGB, HSV and "hexidecimal" values. One of those 9 numbers (H) is somewhat an inherent property of the object and measurable under certain protocols, and even then the results are protocol-specific. North8000 (talk) 13:43, 11 August 2018 (UTC)
• Keep " Light Venetian Red " and the background color, keep " 'Venetian Red, Light' on labels. Discontinued by 1910 " , Remove H=10 S=60 V=90 R=230 G=115 B=92 and Hexadecimal=#E6735C per WP:V, WP:RS, and Talk:History of Crayola crayons#Here comes the science. --Guy Macon (talk) 01:03, 12 August 2018 (UTC)

• Remove - Summoned by bot. Adding this information would be WP:OR unless we have a reliable source for verification. Meatsgains^(talk) 19:40, 22 August 2018 (UTC)

Threaded discussion

Because this discussion has many angles and this is a genuinely interesting RfC where both Daniel and Aculeius have very valid points, I propose to have 2 different sub threads corresponding to each of the issues raised in the RfC.

Discussion about whether the numeric colour values in the various tables are appropriate?

• User:DanielRigal, what do you mean when you say that numbers copied from other websites (such as [2], [3]) and even books ([4]) are "intrinsically unverifiable"? Usually, when you copy a number from a source, that fits squarely into our definition of "verifiable". Are you concerned that they aren't "official"? WhatamIdoing (talk) 15:27, 1 August 2018 (UTC)

I know it's against my interest to say so, but the first source, while it looks very professional, is probably copied from Wikipedia, so it really isn't good evidence. The second one might or might not be based on what we've already got here, but it gives no indication of source. The third one, the book, probably uses actual colors not copied from Wikipedia, although I think that Crayola itself, or a site that explains what the colors are/were and gives actual samples of them, are better sources. P Aculeius (talk) 16:34, 1 August 2018 (UTC)

The key point is that Crayola has never published numeric values for the colours and no reliable source has covered them. If Crayola had done so, or if some even semi-official standards body had published its values, then that would be OK for verification (although I'd still suggest that it would be overcoverage to include them). Sadly there is no ISO standard and no International Crayon Standards Committee to which we can refer. In so far as regulations and standards bodies care about crayons at all their only real concern is that the crayons are non-toxic and safe for kids, which seems fair enough really. Even top of the range professional artists crayons and pastels are not standardised to the degree suggested by a 24 bit colour value.

Lets go through the sources:

• w3schools claims that the colours are sourced to www.crayola.com/explore-colors but that does not have numeric values, only the colour names shown with a depiction of the colour.
• unicornomics.com is the personal blog of an author.
• A Colorful Kindergarten is a self-published book on lulu.com.

While this stuff is not changing my mind, the more I stare at it the more I begin to understand how it is possible that the values seem convincingly valid to some people. Please let me expand a little more on my reasons why they can not be. (I'm sorry if this is tl;dr.)

• We don't know how Crayola standardises its colours internally. They might use the Pantone system or something else completely different. It is incredibly unlikely that they have consistently used RGB, HSV or Hex given that they started making crayons in 1903 and nobody thought about colour that way back then. They are more likely to have started by defining their colours in terms of the chemical pigments used, essentially as recipes for making the crayons rather than defining the exact colours they produced. It is likely that they have moved on from that but we have no idea what they use now.
• We don't know how accurate the colours on the Color Explorer website are. Sure, they will have made an effort to make them close enough that nobody buying crayons based on the depictions is going to feel that they have been cheated or misled but that is not the same as saying that they can support precise 24 bit RGB, HSV or Hex values. 24 bit RGB or Hex values define a colour space of 16.7 million colours, although the human eye can not distinguish anything like that many.
• We don't know how accurate an estimation of a colour that is based on a crayon or a mark on paper can be unless it is produced using specialist measuring equipment. The appearance of a crayon is not always very helpful. Crayons often appear far darker than the marks they produce on paper. They need paper wrappers to indicate to children roughly what colour to actually expect, although I doubt that those are very accurate either. When a crayon is used to mark paper the perceived colour can be affected by the heaviness of the mark, by the exact colour of the paper (Remember that white paper is quite variable. It is incredibly unlikely to ever reflect exactly #FFFFFF relative to the light source) and the lighting conditions themselves. Of course, there are specialist machines that can illuminate a sample in a controlled and standardised way and measure its intrinsic colour accurately, independent of anything else. I am sure that Crayola own some of these. They may or may not produce results in terms of RGB, HSV or Hex. Even if they do, Crayola have never released those results and we have no access to alternative data of comparable quality.

So you see why I think this is a wild goose chase that can never lead to a truly verifiable set of values? It is not that people are not trying their best; It is that it just can't be done. --DanielRigal (talk) 17:53, 1 August 2018 (UTC)

Your current argument seems to be that if there's any possibility of even the slightest variation, then it's not verifiable. But that calls for a level of accuracy that's not required by other Wikipedia articles; approximate values are acceptable for other types of measurement. And while the human eye can occasionally tell the difference between two adjacent hues out of a 360° color wheel if placed side by side, or when being changed from one to the next, it's almost impossible to detect a 1% change in saturation or value. Very slight variations don't make a number that's clearly identified as an "approximation" inaccurate, particularly when the reasons why it can't be pinned down more precisely are provided. But with Crayola's digital swatches, it hardly matters, because the crayons have an "official" RGB/HSV/Hex equivalent, and even if the actual crayons don't produce a perfect replica of that notional color every time, anybody can verify what the official color is using their own browser tools or other software—and no matter which they use, the result will be the same, making it verifiable. No matter what your monitor is set to display, anyone who measures the color of a group of pixels set by the web host to 255 0 0, will measure 255 0 0, or HSV 0 100 100, or Hex #FF0000. If measuring actual wax instead of a digital swatch, you might get 253 2 3 or 255 1 2, but the difference would be so slight as to be indistinguishable, and thus meaningless for purposes of verifiability. If you find that the color provided bears no resemblance to the actual crayon, as the result of a typo or vandalism or any other reason, then fix it, like you would any other objective statement on Wikipedia that isn't supported by the source cited. Measure it yourself, and change the values to what they should be. Anyone else can come along later and verify the values using the same source. P Aculeius (talk) 20:55, 1 August 2018 (UTC)

I have made a little image showing how can you verify that the values posted in the page are correct. I understand this is most unorthodox, but in the end this is the Crayola official page. Crayola's official page showing RGB values Dryfee (talk) 17:36, 2 August 2018 (UTC)

Disagree about the colors obtained from color matching with swatches being original research. Anyone is capable of adjusting color sliders to match a particular color; as long as the color produced indicates that it's only approximate, rather than a precise measurement or an official value, and anyone can verify it by the same process, it's not original research. It would be different if someone said, "this is a better red than produced by Whitman crayons" or "this is a strikingly beautiful color", things that cannot be verified (although you could quote a published source making such statements, as long as it's presented as opinion). In the context of things that can be objectively verified, the concept of "original research" has a very limited application. P Aculeius (talk) 18:02, 2 August 2018 (UTC)

Verbatim from Wikipedia's definition of original research. "Wikipedia articles must not contain original research. The phrase "original research" (OR) is used on Wikipedia to refer to material—such as facts, allegations, and ideas— for which no reliable, published sources exist." I added the bold format. You derived the colours by yourself. This is not a reliable published source. Dryfee (talk) 18:40, 2 August 2018 (UTC)

And yet, you're ignoring the fact that objectively-determinable facts that anyone can tell by examining readily available objects are defined as not original research. A book is a valid and citable source for its contents, or its physical appearance. So is a crayon, or a swatch of crayon wax. There's no allegation or idea here. It's an objectively-determinable fact that anyone can verify. If correctly done, there's no chance of significant disagreement. P Aculeius (talk) 20:35, 2 August 2018 (UTC)

The crucial part of your response if "If correctly done". This is why these measurements must come from a reliable an published source. You are not a reliable or published source. I feel your are trying to argue against Wikipedia's rule of no original research and this is not the place. I'm not ignoring your aforementioned claimed fact, I am arguing that this is not a readily observable fact. Colorimetry is not something to be left to an editor using paintbrush or a browser add on. Patents have been filled over colours and millions of dollars have been dedicated into the research of these subjects. I made a simple illustration, following your procedure to measure the colour of a crayon swatch using Paintbrush's colour tool. Here are the results. As you can see, it is not an objectively determinable fact. Dryfee (talk) 23:20, 2 August 2018 (UTC)

No more so than typing a quotation correctly, or holding a ruler and reporting the lines on it correctly. Not sure what method you used, because it looks like you sampled individual pixels from a poorly-lighted photograph of a crayon in its wrapper, possibly the least useful method you could arrive at. What you should have done is taken a swatch colored with the crayon in question, and sampled multiple pixels from the thickest/most intensely colored part of the swatch. There should be very little variability, and the software used would be largely irrelevant, whether you use the Digital Color Meter desk accessory, which is standard equipment on many computers, and lets you set the aperture being sampled, or Paint, Photoshop, or Gimp eyedropper tools, or any other method. Any given pixel will have precisely one value across the board; it doesn't matter what kind of computer you use or what monitor settings you have—a pixel that reads #FD1108 (or 253 17 8 in RGB) to one person will be #FD1108 to everybody. You could also eyeball the color sliders to match a swatch, but not from a photograph of a crayon. As you should remember, you can't usually tell what color a crayon will color by looking at it. Many crayons look much darker than the wax they lay down. Think back to how easy it is to mistake the violet range for black just looking at the crayons. And it makes no sense to sample the label, since there are only just over a dozen label colors, and generally all of the colors in a family share the same label. Use a swatch of color, not a photograph of a crayon. P Aculeius (talk) 12:32, 3 August 2018 (UTC)

" What you should have done is taken a swatch colored with the crayon in question, and sampled multiple pixels from the thickest/most intensely colored part of the swatch. " No, what I should have done according to Wikipedia's Rules is find a reliable and reputable source. Dryfee (talk) 17:48, 3 August 2018 (UTC)

The crayon is the source for its own color, as per Wikipedia's guidelines on reliable sources. You don't need to cite any other source to say that "the book is red" or "the car is violet-blue". All you need to do is be able to look at it, and see pretty much the same thing that anyone else will see by looking at it. This is simply a little more precise: the wax is about HSV 280 90 90 when laid down on paper. After all, we're interested in the color the crayon is supposed to make when you color with it, not the color the crayon appears to have when it's sitting in the box. And anyone is capable of determining that within a reasonable degree of accuracy, either using an app or just adjusting the sliders to match. So it's not original research; it's an objective fact that anyone is able to verify by examining it. P Aculeius (talk) 21:27, 3 August 2018 (UTC)

"as per Wikipedia's guidelines on reliable sources" Quote them to me, please. I really don't see where you're getting this idea from. Dryfee (talk) 22:56, 3 August 2018 (UTC)

"After all, we're interested in the color the crayon is supposed to make when you color with it." Then the RGB/HSV/HEX values should be removed. "Violet-blue" should suffice. Dryfee (talk) 22:56, 3 August 2018 (UTC)

"And anyone is capable of determining that within a reasonable degree of accuracy" The reasonable degree of accuracy is the name of the colour, an RGB value is an unreasonable degree of accuracy, which is beyond your competence. Dryfee (talk) 22:56, 3 August 2018 (UTC)

"it's an objective fact that anyone is able to verify by examining it." Again Wikipedia's Rules applies to facts too! Dryfee (talk) 22:58, 3 August 2018 (UTC)

The crayons are a primary source for their contents, including the color that they make when the wax is laid down on paper. The relevant explanation is provided in WP:PRIMARYCARE, which is introduced by this language: "[m]aterial based on primary sources can be valuable and appropriate additions to articles. However, primary sources may only be used on Wikipedia to make straightforward, descriptive statements that any educated person—with access to the source but without specialist knowledge—will be able to verify are directly supported by the source." That's what this article does. The purpose of the tables is to illustrate what the crayon colors look like, and what the values of the colors used for illustrative purposes are. "This is what the crayon color looks like" is a straightforward, descriptive statement, and any educated person with access to the source, i.e. the crayon itself, a swatch of color from the crayon, a picture of a swatch, or the color as represented on the manufacturer's website, can verify either by examining it and seeing if it approximately matches the color shown in the table, or measuring it directly with free and easily available tools to determine the value of the pixels. No specialized knowledge is required to do so. Anyone—at least, anyone whose color vision isn't impaired—can compare the color produced by the crayon or represented on the manufacturer's website against the one displayed in the table, and decide if they match.

The explanation goes on to list a number of examples, which I've mentioned before: a book is an appropriate source for the names of the characters, the number of chapters, a basic summary of the plot. It is not a valid source for the interpretation of the symbolism, the author's intentions or motivations, or a critique of its literary qualities. I use this example because it's the easiest to understand, but not necessarily the most direct for this case. There's also an example of a painting, which expressly states that it's an appropriate source for the colors, shapes, and figures depicted. While the example simply refers to "red, white, and blue", there's no reason why descriptions have to be limited to the few dozen colors with names that most people will recognize. An editor is perfectly entitled to say, "this color is a very strong slightly orange red, with a little more white in it than that other color". Or, "the color of this crayon is approximately the same as 'Hermosa Pink' from Ridgway's Color Standards and Color Nomenclature." But why would we want to describe a color by reference to other colors and slight degrees and variations, when it can simply be depicted graphically, with notations as to the hexadecimal/rgb/hsv values of the depiction are? Showing someone what a crayon looks like has more visual impact, and more meaning for a reader, than describing it verbally—and it fits a lot more neatly in an article that contains a couple of hundred different colors.

It's pretty easy to approximate a color for illustrative purposes, even without measuring it, using the hue, saturation, and value sliders in any graphics editor, and then say, "it looks about like this", whatever values show up with the sliders, and however they translate into RGB/Hexadecimal. After all, even an untrained eye can tell the difference between hue 0 and hue 15, and say "it looks a little bit redder," or "a little bit more orange"; it's just as easy to say, "it's a little paler" or "a little darker", or "a little greyer", and eventually you can find a color that looks like a good match. No expert knowledge required to reproduce, and even less to verify: just look at the swatch and the color in the table, and ask, "do they match?"

Just as on point for using a crayon as a source for its own color is this language under "providing an original illustration": "[s]uppose that a Wikimedia contributor inserts a photograph or other media file to illustrate a Wikipedia article on a person, place, or other topic. Editors who do this routinely assert that the photograph depicts the subject of the article. The Wikimedia community assumes good faith that the illustration really depicts the thing. For example, it is not necessary to provide other pictures of a person or place as supporting evidence that a photo insertion into Wikipedia is what the content provider claims that it is, except in the case of a dispute." Now, a sample of color is much simpler than a photograph or a drawing. But it's still a representative picture of what a crayon color looks like. If we follow the community practice of assuming in good faith that the illustration really depicts the thing, then the only thing that really matters is, "is this what the crayon color looks like?" If it's not, then fix it! If it is, then it's fine for illustrative purposes. It's not a valid basis for deletion to argue that the color in the table isn't "official", when it's only described as "approximate"; nor does it matter if the source for the color is the crayon itself. P Aculeius (talk) 03:45, 4 August 2018 (UTC)

I think you are stretching the intention of WP:PRIMARYCARE. there is a difference between "that book is green" and that book is "0,255,0". Surely you must see that the latter claim is unnecessarily precise, and bound to be wrong. Dryfee (talk) 19:05, 10 August 2018 (UTC)

If that were what the tables said, that might be true. But what it actually says is, "that book is about 0, 255, 0* ... *for reasons A, B, and C it is impossible to render the exact color, so this is a close approximation" and that makes a significant difference. P Aculeius (talk) 19:32, 10 August 2018 (UTC)

But who made that approximation? Dryfee (talk) 22:48, 13 August 2018 (UTC)

Typically, that approximation was made by either some high-school student who sells "web design services" to businesses, a newly-hired intern who just finished reading web design for the complete idiot, or a busy salesman who bought one of those "build your website in minutes" software packages. --Guy Macon (talk) 11:53, 14 August 2018 (UTC)

It's worse that that; you can't approximate something (inherent color values of a crayon) that doesn't exist, implying such is an inherently erroneous (not reliably sourcable statement.). But I did propose a compromise under the "science" section below. North8000 (talk) 12:01, 14 August 2018 (UTC)

• Keep the values - Summoned by bot. I support keeping the color codes. The rendering of the color is important - it is after all a core parameter of the item. The different ways to represent them are in my mind useful to have, as the world goes more digital. For some reason, this strikes me as similar to the plot summaries for various television episodes on Wikipedia, which many have argued are unsourced and therefore should be removed. Yet, anyone can add or change a summary if they disagree with a plot point - the episodes themselves are the source of information, and the more fans there are, the faster the info is usually fixed. Likewise, anyone can print an image using a crayon color code, and if it's not right, argue to change it. But the colors are verifiable simply because they are in the public domain and are used by many - the people most interested in this info can police it. Again, not to belabor the point, but if a numerical representation is incorrect, the person can always change it and give their reasoning. There are enough eyes so that if one is wrong, it will be fixed soon enough. In the case of a color code being off and nobody noticing, then that's not really hurting anyone, but having something is better than nothing since it gives you a _target to work towards. TimTempleton ^{(talk) (cont)} 17:38, 7 August 2018 (UTC)

• This is less like a plot summary for a television episode than it is like a description of a character ("The Maurice Moss character is smarter than the Roy Trenneman character, but Trenneman is much better looking") which requires WP:OR. --Guy Macon (talk) 11:53, 14 August 2018 (UTC)

• Remove Unsourced and unsourcable because it is faulty. Invited by the bot. A crayon (or any object which is not itself a light source) doesn't even have inherent RGB, HSV and "hexidecimal" values. One of those 9 numbers (H) is somewhat an inherent property of the object and measurable under certain protocols, and even then the results are protocol-specific. North8000 (talk) 13:45, 11 August 2018 (UTC)

Discussion about whether is this excessively detailed coverage?

I believe it is not excessive, these are the entire point of the article. Dryfee (talk) 17:26, 2 August 2018 (UTC)

I'm not sure that it should be the entire point of the article though.

This is an article about the history of a commercial product (Which is why I put the RFC in the "Economy, trade, and companies" category). That history should include things like which markets they operate in or have operated in, where they manufacture, how many crayons they sell, any changes that they had to make for safety reasons, any markets that they failed in and withdrew from. You know, business stuff. Obviously colours will be mentioned too but I feel that at the moment we have a list of colours instead of the article that the title promises. A history is a narrative. This fails to tell the story of the product. It would be a bit like having an article on the history of the US presidents that was just a list of the height and weight of each US President and nothing else. It would tell you that Obama is tall and that Taft was fat but it would not provide a historic narrative that illuminates the history of the US presidency or any other encyclopaedic subject.

Perhaps it would help if I gave a couple of examples of what I think a good product history article looks like, rather than just griping about the things that I don't like? I quite like History of Lego and Meccano (not a separate history article but most of it is a history). They are mostly prose and mostly historical. They talk about which markets the product was sold in, the manufacturing of the products, market successes and failures, marketing and tie-ins. They track the major product lines but notice what they do not do. They do not attempt to catalogue each size, shape and colour of each Lego brick or Meccano piece. Maybe the Lego one is a little bit too austere and underillustrated but these are on the right track. I'd love to see the tables go completely from this article and just have descriptive prose with (several but not too many) good photos of the main past and current products in the Crayola range. That would give readers a much better intuitive understanding of what the product range has been without lots of complicated numeric tables to interpret and contextualise. --DanielRigal (talk) 19:41, 2 August 2018 (UTC)

I completely agree with you. Perhaps it is excessive. Perhaps all the tables could be merged into one with headers: colours Date introduced date discontinued (or vintage or whatever). instead of having a table for each epoch. I don't know sources to complement the article, but I also agree that other aspects of the history of the product, like the ones you mentioned should ideally be included. I think nobody is arguing against that. Dryfee (talk) 20:15, 2 August 2018 (UTC)

There's already an article that does that: list of Crayola Crayon colors. This article was split off from that one because it made no sense to duplicate everything in one article. But the user you're arguing with wants the same material to be deleted from both of them, and argued at length to delete this entire article, and the other one, and every other Crayola article, and when those proposed deletions failed, decided to delete all of the material he disagreed with, calling it "uncontroversial". There's a reason why this article exists: putting the different stages of the history in context. You want more history? Great. Go ahead and add some. As you said, nobody's arguing against it. This discussion was created for one purpose: to delete as much of the contents as possible. P Aculeius (talk) 20:39, 2 August 2018 (UTC)

• Not excessive - summoned by bot - see my argument above - I don't see these as two different arguments. TimTempleton ^{(talk) (cont)} 17:39, 7 August 2018 (UTC)

I'm going to disagree with two things in your argument:

1. "it is after all a core parameter of the item". Is it really? These are wax crayons. There is absolutely no reason to speculate that 24 bit hex values are used in their creation. These values come from original research done to mine them from the Crayola website's HTML source code. There is no intrinsic link to the wax crayon here at all.
2. "the colors are verifiable simply because they are in the public domain". Are they really? Crayola has never published these values. We are inferring a significance into numbers in their HTML source that Crayola is not explicitly asserting.

I do agree that there are two separate issues here. One is whether having the values is overcoverage and the other is whether they are verifiable. I tried to pitch it as two separate questions in the RFC.

--DanielRigal (talk) 18:52, 7 August 2018 (UTC)

I misunderstood your goal in splitting then. Verifiable is different than appropriate. If you want to change the wording then two threads make sense. I'm saying a color can be verified like a plot summary - you just have to do the work - watch the episode or put the color code into a computer and eyeball the output. TimTempleton ^{(talk) (cont)} 22:09, 7 August 2018 (UTC)

"Public domain" isn't really at issue here. Crayola has published their own current crayon values on their website, and it's not the first time they've done so, as they licensed their colors to other sites for illustrative purposes, and there have been multiple versions of their color pages over time. They've "published" their crayons by selling them to the general public since 1903. In this sense, the crayons are just as available as a painting hanging in a gallery or a building on the corner. Measuring something that anyone can examine isn't original research; checking a web site's coding and recording the contents isn't any different from opening a book and quoting a passage. But you don't have to do that to produce an illustration of a crayon color. All you need is to examine the crayon itself, and color with it on white paper. That's no more original research than an artist choosing the colors used to draw a picture of anything for an illustration on Wikipedia—which is clearly acceptable. Hex codes are a routine calculation made from HSV/RGB values, not some arcane, esoteric, and unverifiable data. Anyone can verify them just by looking at the crayon wax; but looking at it from another perspective, they represent the color used to illustrate the crayon wax, and are nothing more or less than the notations identifying the swatches in the table. It's also worth noting that you've never once argued that any of the colors are inaccurate; only that any data used to illustrate them must be excluded, so that it doesn't matter if the tables are accurate or not. How much longer is this going to continue?

Possible compromise?

I'm concerned that there seems to be quite a bit of WP:ILIKEIT going on here but I see that people do like the colours and want to keep them. How would everybody feel about a possible compromise as follows?

• The HSV and RGB values get removed as they are conversions of the Hex values and add no additional information. While the Hex is mined (in my view improperly) from a Crayola source, these HSV and RGB values are derived from that with no possibility to trace them back to any Crayola source at all.
• The table background colours and hex values stay but only when referenced to the source code of the Crayola Color Explorer site (which I still think is obvious and blatant OR but it seems futile to argue). The column heading should have a note to make it clear that the colours come from the source code of the Color Explorer website so that readers are not misled into thinking that they are more fundamental to the actual crayons than they are.
• If anybody comes along who is from Crayola (or anybody who legitimately represents them) complaining that we ripped off their data then we hang our heads in shame and remove it.

--DanielRigal (talk) 18:52, 7 August 2018 (UTC)

Sorry, but this is still the same set of arguments warmed over—anything it takes to get rid of as much of the article as possible. Calling it a compromise doesn't make it one. You still refuse to acknowledge that the sources are valid sources, the information verifiable, the values obtained by routine calculation. To the extent that any new reasons are presented here, I don't see how readers are going to be "misled" into thinking that anything is "more fundamental to the actual crayons than they are". Presumably, the color is the most fundamental aspect of any crayon. Are crayon colors variable? To a limited extent, but giving a value smack dab in the middle of what that color is claimed to be/appears to be is the best possible method for representing it, provided that the article plainly states that it's impossible to be any more exact than that—which this article has done ever since it was split off from the list of colors. Could Crayola "complain" about this data being proprietary? Possible, but highly unlikely. They can't copyright or trademark specific RGB/HSV/Hexadecimal values. Most of the color names are generic, and simply citing a number of details to a particular source isn't a copyright violation. This article doesn't reproduce Crayola's site, or any of its past or present versions; it doesn't make it less likely that Crayola will be able to sell crayons; it simply illustrates what the colors are supposed to/appear to look like, and gives the colors used to illustrate them in three different forms of notation, since while these can be obtained through nothing more than routine calculation, there's no reason to make people perform calculations in order to obtain them (and of course they also make the tables sortable in different ways). There's simply no valid reason to delete any of these colors; and citing ILIKEIT doesn't help when it's pretty obvious from the shifting arguments and reasons for deleting some or all of this article that WP:IDONTLIKEIT is a major motivating factor in this ongoing saga. P Aculeius (talk) 23:08, 7 August 2018 (UTC)

"You still refuse to acknowledge that the sources are valid sources" Correct! Mining the HTML code of a website is WP:OR. There are no valid sources here! Crayola has never published colour values for their crayons. Converting OR data to other formats by whatever off-Wikipedia means you choose is just OR on top of OR.

As for "anything it takes to get rid of as much of the article as possible": I don't appreciate the assumption of bad faith. I am not the person who put this walled garden up for AfD. I am trying to save what good content we can. --DanielRigal (talk) 23:27, 7 August 2018 (UTC)

Amazing how it's possible to argue repeatedly to delete vast quantities of information from an article that nobody had an issue with until the AfD, and present it as a heroic effort to "save" its content. Nobody's arguing to delete the things you want to "save", and all of your arguments have been for the deletion of all or most of this article. The AfD was closed as keep, and this is an entirely new discussion initiated by you, the main voice arguing to delete this and five other articles related to Crayola, two of which were in fact deleted, with no attempt whatever to "save" any content.

We have multiple valid sources here, including both the manufacturer's web site, and the history of an acknowledged expert in his field who sets forth his findings, the reasons for his conclusions, and illustrations to support the same. You keep saying "original research" over and over again, but you haven't said anything about the fact that describing an object, including the use of user-generated illustrations, is not original research, but valid use of an object as a primary source for its attributes. You've repeatedly complained about "data mining", but viewing the source code of a web page is something that anyone can do, as easily as opening the cover of a book and seeing what's in the index. You're also ignoring the fact that the information in these tables wasn't obtained through "data mining", but by measuring the value of individual pixels, just like using a ruler or a stopwatch to determine the objective characteristics of something; this is clearly allowed, because anyone can do it in order to verify that the description thereby generated is correct, without any specialized knowledge. But you don't even need to measure them; the swatches are illustrations of the colors represented, and anyone can make those by adjusting HSV sliders to match the approximate color of crayon wax, and anyone can verify the illustrations by comparing them to the crayon wax.

Wikipedia clearly states that routine calculations are not original research; yet you're still arguing that converting RGB to hexadecimal or HSV is, and urging that as a reason for exclusion. I'm not too sure of HSV, but I can prove that RGB to hexadecimal is routine: RGB uses a scale of 0 to 255, using digits from 0 to 9; hexadecimal is the same thing expressed with digits from 0 to 9 and then letters A to F, for sixteen digits instead of 10. FF is 16 16's, less one (since 00 is the same as RGB 0), or 255. FE is 254, FD is 253, etc. They're exact equivalents, not some kind of arcane mystery that only rocket scientists understand. HSV is the same kind of thing: a precise mathematical description of the same color. It's useful to have all three of them, because traditional artists often work by hue, digital artists by RGB, and web designers with hexadecimal codes; and while sorting by complete hex codes isn't very useful, you get different results when you sort by hue, saturation, or value, than when you sort by red, green, or blue. But it's no more original research to say that a swatch produced by FF0000 is the equivalent of 255 0 0, than it is to say that something two inches long is about five centimeters. The contents of this article meet Wikipedia's standards for verifiability, do not fall afoul of the policy concerning original research. Therefore, I can't see any justification for continuing to push for the deletion of this article, or any substantial part thereof, for the reasons being urged. P Aculeius (talk) 02:17, 8 August 2018 (UTC)

You are correct that the RGB and the Hex are entirely equivalent. This is a very good reason why we don't need both! It is literally just the same numbers listed twice. I guess I should withdraw the claim that converting Hex to RGB is OR. The conversion to HSV is still very questionable and still a castle built in the air if the hex is not validly sourced.

Anyway, I think we have both gone a bit too far with the TL;DR. How about we let some other people have their say?

Please? Anyone? Is there anybody out there reading this who has not lost the will to live yet? --DanielRigal (talk) 18:23, 8 August 2018 (UTC)

Here comes the science

It is impossible to match subtractive colors with additive colors. If Crayola says that a particular RGB additive color is a close match to the particular subtractive color of a crayon, we can include that information, but it really needs at least a footnote explaining that it is impossible to match subtractive colors with additive colors. There are a lot more subtleties to this; see Color. Color model, Color space and Gamut. --Guy Macon (talk) 05:43, 10 August 2018 (UTC)

Thanks. That is close to the point I have been incoherently grasping for when I keep on reminding people that these are wax crayons not 24 bit digital colours. Thanks again for making it much more clear.

You say "If Crayola says that a particular RGB additive color..." and I agree but I think that is a big "if". I think it hinges on what constitutes them actually "saying" that. They have a website, aimed mostly at children, where the kids can click on lists of colours and then click through to the products which have those colours in and buy them. It is not quite an e-commerce site itself but it is very tightly in orbit around one. It is borderline as RS but the real problem that I see is that nowhere on this site does it show these RGB/hex values at all. The only place they occur is in the HTML source code which is used to render the page. I don't see that as Crayola "saying" anything about the specific hex values.

If Crayola really was unambiguously saying anything equivalent to "This hex value approximates this crayon", by showing the hex value somewhere on the rendered page, then there would be no argument (apart from the separate ones over how we use/overuse them). We definitely could use the values. If that were the the case then I would agree with your proposed footnote. As far as I can tell Crayola are not making any public statement about the hex values, hence my very strong feeling that fishing the value out of the HTML source is OR. Do you have an opinion on that? --DanielRigal (talk) 10:46, 10 August 2018 (UTC)

Such a footnote already exists in the article, and has since it was created. It states unambiguously that the illustrative swatches merely approximate the actual crayon colors, as nearly as can be determined given the difference between additive and subtractive colors, the appearance of wax on different types of paper, and natural variability of the crayons themselves. The fact that an illustration may not be a perfect reproduction of the original doesn't preclude it from being a reasonably accurate representation, which is all it needs to be. Otherwise, we couldn't use digital photographs to illustrate solid paintings! P Aculeius (talk) 12:42, 10 August 2018 (UTC)

I agree that the HTML markup is not a reliable source. There is a significant probability that those RGB values were chosen by someone who makes web pages for a living and does not speak for Crayola. --Guy Macon (talk) 17:21, 10 August 2018 (UTC)

A crayon (or any object which is not itself a light source) doesn't even have inherent RGB, HSV and "hexidecimal" values. One of those 9 numbers (H) is somewhat an inherent property of the object and measurable under certain protocols, and even then the results are protocol-specific. North8000 (talk) 13:47, 11 August 2018 (UTC)

But it is possible to depict the color of an thing, such as crayon wax, in digital format, which necessarily involves RGB/HSV/hexadecimal codes. They represent the color of the illustration, so the question is whether the illustration reasonably approximates the color of the crayon. If so, then the RGB/HSV/Hexadecimal values are approximate digital representations of the color. I add that the three are precise mathematical equivalents of one another; if you have one, you can determine the other two with absolute certainty. P Aculeius (talk) 13:56, 11 August 2018 (UTC)

Aculeius, most of that is fundamentally wrong. In the context of common digital storage of color, your last sentence is right. I don't know how lengthy to get here. Color is a human perception, not an innate attribute. And it is the result of a host of variables, especially the spectrum of the illumination and the object in question. There is immense science, standards and protocols applied to, for example, make sure that the orange crayons or the color of house siding from different factories all look the same to typical observers under typical illumination sources. They all start by acknowledging that color space values are not innate to the object. For example, in testing "closeness of match" for crayons, they might start with one of the industry standard illuminants (EG D65 a standardized uneven spectrum roughy emulating noon sunlight) illuminate and image the crayon in a specified geometry, image it with a device with RGB sensors to create a set of RGB values (which record the output of the sensor) then mathematically recalibrate it primarily to adapt to the unique response curves of the imaged RGB values into a second different set of RGB values. Then in order to quantify the results in a way to match human-percieved differences they're transform it into a color space like CIE L*a*b* which has perceptual uniformity for a standardized typical observer. Color specifications (e.g. to keep color uniformity) are based on such a standard protocol, thus knowing that the results change with protocols. More elaborate methods record the spectrum of the light scattered by the object instead of an imaging system with RGB sensors. In Wikipedia the final arbiter is reliable sourcing, I put forth the opinion that it won't be reliably sourcable for those reasons. Yes, it is possible to get a "rough idea" of color. A websiter could take a picture in nice light of the crayons, being careful of some potential issues , look at the stored RGB values and put them on his website to give a rough idea. If a different web siter followed the same process, his numbers would be different but also give a rough idea. In each case the numbers also reflect his lighting etc. and are not innate to the crayon. Sincerely, North8000 (talk) 18:50, 11 August 2018 (UTC)

I never claimed that crayons had inherent hexadecimal codes. What I said was that color is a fundamental trait of crayons; the one attribute that distinguishes orange-red from orange-yellow is color. And it's certainly possible to depict the color of a solid object in a digital medium. Every picture of a physical object on Wikipedia does that, and for many of those images, it's important that the illustration match the actual color of the object as closely as possible, so that a reader will have a good idea what color it is. Could two reasonably accurate depictions of a crayon's color differ slightly? Of course. That's what makes any sample used to illustrate it "approximate". But the fact that it can't be definitive or invariable doesn't make it impossible to depict. An illustration that's too light, too dark, too strong, too dull, too reddish, too bluish, etc. is easily distinguished from one that looks like a close match, even if all crayons look the same in a dark room with the lights turned off. Crayola certainly didn't think it was impossible to depict its colors in a digital medium when it published collections of digital swatches identified with the names of its crayons. All the HSV/RGB/Hexadecimal values given on this page really do, is identify the colors used to illustrate the approximate color of the crayons. And that should be fine. P Aculeius (talk) 20:45, 11 August 2018 (UTC)

You left something out. The HSV/RGB/Hexadecimal values given on this page identify the colors used to illustrate the approximate color of the crayons, and cannot be traced to any reliable source. And that isn't "fine". --Guy Macon (talk) 01:06, 12 August 2018 (UTC)

According to WP:V we must be able to verify content by going to a reliable source that has published the information in question. Repeating the research used to produce the data does not meet this requirement. –dlthewave ☎ 01:33, 12 August 2018 (UTC)

You can see the colors by examining the crayons, and making a swatch, or from a picture of a swatch made from a crayon. The crayons are the source for the colors in the illustration, not for hex codes. The hex codes/RGB/HSV values are simply the digital equivalents of the colors used to illustrate what the crayons look like. Just as with other primary sources, it's not necessary for any other source to say what can be objectively determined by examining the object in question. It's not necessary for a third source to describe what anyone can see by examining something; you can describe the plot or name the characters in a novel by reading the book, and that's not original research; you can make a drawing of a building to illustrate what it looks like, and say "the building has 24 windows", which anyone can verify by examining the building (or for that matter, enough pictures of it to count all of the windows). And you don't need a third party source to say that "Jungle Green" is #29AB87 in order to illustrate it with a light, slightly bluish green that comes close to #29AB87, although since you can see that this is how Crayola depicts "Jungle Green" on its "Explore Colors" page, it's possible to be precise in a way that you couldn't if you were just coming as close as possible without knowing what they consider it to be. If you couldn't look at this, and just had a "Jungle Green" crayon to examine, and you illustrated it with #28AF85, that would still be a fair approximation of the color, and you could still say, "values of color: about #28AF85, or HSV 161 77 69, or RGB 40 175 133", because that's what the color used to depict "Jungle Green" in your table is. You don't need another source to say that's what the color you used is, any more than you need a source to say that "source" has three consonants and three vowels. P Aculeius (talk) 03:40, 12 August 2018 (UTC)

RGB is not simply a digital equivalent of colors, it is a specific color space based on a perceptual/subjective modeling of human version that approximates color by adding together three different wavelengths. This is not the same thing as counting the number of windows. At best it would be like looking at broken pieces of glass and saying that they came from approximately 24 windows assuming that they were double paned and came from a normal house. For example list of Crayola crayon colors gives red: 58, green: 166, blue: 85 as the color green. Yet from a sample drawn from that color crayon File:Crayola-Green.jpg I can get measurements of red: 39 green: 154 blue: 87 in one spot of solid color; red: 100 green: 201 blue: 136 in another spot of solid color and red: 59 green: 169 blue: 103 using a digital color meter with multiple pixel averaging aperture and avoiding spots that show through as white. This OR shows a value of a larger variation then that between some of the colors on the list. PaleAqua (talk) 04:13, 12 August 2018 (UTC)

I just tried looking at the references and I can't see where ANY reference gives any of those numerical values, not even that personal website. And for one "reference" that is used about 75 times is not even a reference, just a sentence written written by a Wikipedia editor. Another that is used 100+ times just goes to the Crayola web site home page. Did I miss something? North8000 (talk) 02:16, 12 August 2018 (UTC)

Again, you don't have to get specific values from the source. Just the color itself. You can easily find out what value each of Crayola's digital swatches has, because those are written into the source code, and can also be measured precisely using any number of tools, such as the digital color meter that comes standard on many computers. Since anybody can check a web page's source code using the browser menu, or measure a group of pixels with a digital color meter app, and expect to obtain the same results, it's not original research. But it's not necessary to find or measure these, because all you have to do is make an illustration that appears to accurately represent the color of the crayon. That's what the swatches of color in all of the Crayola tables do: they illustrate the color. The values given next to them indicate what digital color has been used to illustrate the crayon color. They don't need separate sources, since they simply identify the swatches they appear next to. In the case of the ones based on Crayola's own swatches, the colors used happen to be identical to those used on Crayola's web site. The others can be checked against swatches of crayon color to see if they appear to match. I add that the link you mention doesn't go to Crayola's home page. It goes to "Exploring Color", which has 248 distinct swatches of color displayed and identified if you scroll down, or click on any of the color families to see the colors of related crayons. With the exception of a few "metallic" colors that mix in flecks of white, they're all pure color, without variation if you measure them, or check the source code (shortcut: try right clicking, and choose "inspect element" in Firefox). But to reiterate: the colors are verifiable because you can look at the crayon wax and see whether the colors used to depict them are good matches. The RGB/HSV/Hexadecimal codes don't have to be proven by any other source, because they represent the colors of the swatches used to illustrate the crayon colors. The hex codes are the same ones used to put the swatches in the table, so by definition they have to match; the RGB and HSV values are simply mathematical equivalents of the hex codes, derived through routine calculation, and thus not original research. P Aculeius (talk) 03:17, 12 August 2018 (UTC)

Sorry but that's synthesis and hence original research. Just because those are the coordinates are obtainable from the source for the samples on Crayola's web site does not say that those are the coordinates for the colors. Just by giving out coordinates we are implying that the color of the crayons have exact RGB matches. This misleads our readers. Unless we have something that gives the coordinates directly we should not use them. PaleAqua (talk) 03:35, 12 August 2018 (UTC)

P Aculeius if I understand you, you are saying the the source is your looking at the color values they used for the swatches? If so, there is also a flaw in that argument. Again, this is in the context of my assertion that crayons don't even have an inherent RGB value. Use by Crayola of a set of RGB values to portray a crayon color is NOT a statement by them that the Crayon has an inherent RGB value and that that is it. I'm talking the logic part here, but not to split hairs.....it also arrives at a fundamentally technically false statement. Sincerely, North8000 (talk) 12:47, 12 August 2018 (UTC)

Maybe this will fix everything: Relabel as "Color values from Crayola's swatches on their web page" North8000 (talk) 13:45, 12 August 2018 (UTC)

There's a footnote at the top of the section: "Due to several factors, the values given should only be considered approximations. The apparent color of any crayon depends on the thickness with which the wax is laid down, the color and brightness of the surface being colored, and other considerations, such as the age and quality of individual crayons. In addition, crayons are produced using pigments, which are normally described using subtractive colors, with the primary colors of red, yellow, and blue; but electronic displays produce colors using the additive method, combining the primary colors of red, green, and blue."

I would consider the Crayola website to be a reliable source for a reasonably accurate representation of the color, as long as we clearly state how the colors used in the list were obtained and why they aren't an exact reproduction of the crayon. –dlthewave ☎ 15:57, 12 August 2018 (UTC)

While there is already a disclaimer the coordinates used in the table have been spread to other pages on Wikipedia without the disclaimer and even with this page the causal reader is going to see exact looking numbers and make the assumption that they are the values. Readers are typical not going to check the footnotes and are going to underweight the statement "The colors in the following table approximate each of the thirty-eight colors produced during this early period." before the first table. The tables are eye catching and will pull people straight towards them. This is kinda like a WP:WEIGHT issue, the articles take a small amount of space to disclaim exact coordinates and then a much larger and more vibrant space to give exact looking coordinates. PaleAqua (talk) 16:59, 12 August 2018 (UTC)

BTW to expand on the weight issues colors based on X11, HTML/CSS and Crayola crayons dominate many of our color articles. Further many of the colors on the X11 color terms originated from approximations of Crayola crayons[5]. Even further many of the early versions of our list of colors where cloned to make numerous encyclopedias and dictionaries of color on the web which are then used as references to using the coordinates on individual color and shades of... pages. PaleAqua (talk) 17:14, 12 August 2018 (UTC)

Yeah, I've come across serveral "every Crayola color ever made" color wheels that were sourced to Wikipedia. The arguments for including the coordinates are not compelling. Another problem is the fact that the "swatch of original crayon" values were apparently obtained by drawing on a piece of paper and somehow measuring the resulting color, with no indication of who or where the swatches came from.

Do you think we could remove the coordinates and leave the color samples on the page, or does all of it need to go? –dlthewave ☎ 18:04, 12 August 2018 (UTC)

My biggest concern is the coordinates themselves. I'm not as worried about the samples, though if we can do something like the samples in the Crayola#Colors it would be even better. I realize that might be difficult to do with colors no longer available. PaleAqua (talk) 18:25, 12 August 2018 (UTC)

I'm not a big fan of that approach, and I'm not sure how practical it is to obtain such images anyway, although I don't object to it if it can be done. One other alternative is to simply rely on photographs of the packs. After all, this is the visual information available to people buying the crayons in shops and it is good enough for them. It does not provide perfect information about the colours but it shows the overall size, shape, quantity and packaging of the product. --DanielRigal (talk) 18:44, 12 August 2018 (UTC)

What they really are is RGB etc. values of typical electronic depictions of the crayolas, not approximations of the Crayola's inherent RGB vales. The latter do not exist, a statement that implicitly says that they do is flat out wrong and not reliably sourcable. And words like "approximation of" do not solve that problem, they actually repeat the false statement.North8000 (talk) 12:30, 13 August 2018 (UTC)

Another Proposed Compromise And this is from possibly the strongest skeptic. Use only whatever form (Hex?) is actually used in the depictions on Crayola's web site. The conversion to others is redundant and some would argue wp:OR. (although RGB is a lot less cryptic and more useful....maybe you could get that from the display). For those less familiar, it is a straightforward conversion. Sort of like if you listed the length from a source in decimeters and made a second column with the length in meters. Relabel it as the "Color Values of the depiction on Crayola's web site", because it is NOT an inherent attribute of the crayon. North8000 (talk) 19:00, 13 August 2018 (UTC)

I think that anybody who understands RGB at all can look at the hex and see the same information there so the RGB is totally redundant. The conversion to HSV is caught in a situation where if it is not redundant then it must be OR so it fails either way. Removing RGB and HSV is a no-brainer. The hex is still somewhat dubious as it does not actually appear on the website at all as rendered, only in the HTML source, but I do not strongly object so long as we are clear where it it comes from. I think your proposed label for it would achieve that. Given that some other people have spoken in support of keeping some of the numbers I think this is the correct way to do it. The hex is the only set of numbers that trace back to Crayola at all. If we are keeping anything that is what we should keep. For any colours not on the Colour Explorer website we should not list any numbers at all. I don't strongly object to an approximate colour being shown in the background of a table cell for those but we should not show a number for it. Finally, we need to do something about the legibility of the colour names. I'm not that bothered which way that is solved so long as it is solved somehow. My first thought is to turn the text black for the lighter colours but if people prefer something else that is fine by me. --DanielRigal (talk) 20:02, 13 August 2018 (UTC)

Sorry but I disagree with this compromise. From experience this will still lead to cyclic coordinate problems in the future as online color dictionary sites will continue to mine Wikipedia for coordinates, thy converted values will then end up appearing on individual pages in color infoboxes as referenced back to those online color dictionaries and so forth. I don't have a problem with using the hex codes to color in boxes behind the colors or the like, but disagree with given the coordinates as visible values. PaleAqua (talk) 20:07, 13 August 2018 (UTC)

People taking the values from us and assuming that they mean more than they do is unfortunate. I agree that adding a better label on them now might be too little too late. Anyway, my preference is still for no hex at all but if we can't agree anything else then I'm OK with the compromise as a second preference. --DanielRigal (talk) 20:17, 13 August 2018 (UTC)

Well, the relabeling along the lines of "Color Values of the depiction on Crayola's web site" would resolve the only thing that I feel strongly about which is patently wrong information. And if mirror sites remove data from it's headings, they have a lot bigger problems that we can't worry about. I don't have any strong feelings about any of the other areas I was just trying to help. North8000 (talk) 22:17, 13 August 2018 (UTC)

So, to condense, it would be to include only hexadecimal values that came from the official Crayola web site, re-label that column as "Hex values of the depiction on the Crayola web site, and remove all other RGB, HEX HSV numbers. North8000 (talk) 19:40, 14 August 2018 (UTC)

Wait...what? At the top you wrote "Use only whatever form (Hex?) is actually used in the depictions on Crayola's web site", (The reality is that the form used in the depictions on Crayola's web site[6] looks like this: CHERRY Hue Family: Red) and now you are saying "include only hexadecimal values that came from the official Crayola web site" (ignoring the fact that the Crayola web site does not display any hexadecimal values for the colors). That's not a "compromise". That's continuing to use WP:OR to derive some hex values that no source has ever associated with crayon colors, simply because some web designer happened to use those values in the raw HTML. --Guy Macon (talk) 21:42, 14 August 2018 (UTC)

Personally, I'd like to see all of them removed. So now I'll play devil's advocate and argue against my preference. This is the official crayola website. And, with the revised heading, the statement in the article becomes (merely) that this was the numerical values chosen on the official Crayola website to depict the crayola colors.North8000 (talk) 02:10, 15 August 2018 (UTC)

I'll play devil's advocate too. I initially proposed keeping only the crayola website ones. However, after realising that assigning RGB (or any other colour coordinate values) to a crayon is like using litres to measure distance, as per Guy Macron's comments, I'm inclined to remove them all too. Moreover, removing them all would create a simple and easy to read table, as opposed a having a table with 90% of the values empty and the remaining ones sketchily sourced... Dryfee (talk) 16:26, 17 August 2018 (UTC)

Your "assigning RGB (or any other colour coordinate values) to a crayon is like using litres to measure distance" communicates what I was trying to communicate far better than I did. IMO the best thing would be to remove them all. I was just trying to craft a compromise......my wording removes the technical falsehood. North8000 (talk) 21:04, 17 August 2018 (UTC)

Han Solo: "I'm captain of the Millennium Falcon. Chewie says you're looking for a passage to the Alderaan system."

Ben Kenobi: "Yes indeed, if it's a fast ship."

Han Solo: "Fast ship? You've never heard of the Millennium Falcon?"

Ben Kenobi: "No. Should I have?"

Han Solo: "It's the ship that made the Kessel Run in less than twelve parsecs. I've outrun Imperial starships, not the local bulk-cruisers, mind you. I'm talking about the big Corellian ships now. She's fast enough for you, old man."

--Guy Macon (talk) 07:24, 18 August 2018 (UTC)

I just figured it out. Cool.  :-) North8000 (talk) 15:35, 31 August 2018 (UTC)

For anyone else who doesn't "get it" please attempt to convert twelve parsecs to hours. How many hours did it take for Han Solo to make the Kessel Run? Assigning RGB values to a crayon is like using liters to measure distance or parsecs to measure duration. --Guy Macon (talk) 16:30, 31 August 2018 (UTC)

The compromise is a viable option only when there are supporters with plausible arguments for both sides and if the compromise deals with most of each side's objections to the other. We don't need the compromise if there is a significant consensus for complete removal of all the numbers. It seems that we are swinging towards that position although the people who gave contrary opinions earlier can't be discounted just for not hanging around here to wade into 10 miles of ongoing discussion. The question is who should make that call and when? --DanielRigal (talk) 12:05, 18 August 2018 (UTC)

@DanielRigal: Are you familiar with the RfC process? Consensus is assessed by an uninvolved editor or admin when the RfC closes after 30 days. This is primarily based on the !votes in the Survey section since, as you pointed out, many editors simply give their opinion and move on. –dlthewave ☎ 19:38, 18 August 2018 (UTC)

@Dlthewave: Are you familiar with the the RfC process? Because you above assertion is incorrect. Please read WP:RFCEND. Note the part that starts with "there are several ways in which RfCs end". You just claimed that one of the five ways listed is the only way RfCs end. --Guy Macon (talk) 16:30, 31 August 2018 (UTC)

Not very familiar. It is ages since I did one of these and I had forgotten that it was 30 days. I should have looked it up rather than asked like that. Sorry. --DanielRigal (talk) 20:00, 18 August 2018 (UTC)

It's not simply votes, it includes consideration of the statements. North8000 (talk) 20:19, 18 August 2018 (UTC)

Also an RFC can't override wp:ver.North8000 (talk) 20:22, 18 August 2018 (UTC)

A bot just took the RFC tag off. Was that correct? Are we waiting for somebody to give a result or has that part been missed? --DanielRigal (talk) 21:21, 30 August 2018 (UTC)

I believe that it is based on the timestamp of RFC as it has been 30 days. Given the lack of new comments recently probably a good time to request a close at Wikipedia:Administrators' noticeboard/Requests for closure. PaleAqua (talk) 01:48, 31 August 2018 (UTC)

Thanks. I have done that now. --DanielRigal (talk) 13:29, 31 August 2018 (UTC)

The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

Table layout proposal

The color names superimposed on the color samples are hard to read and generally not good practice. I propose changing the tables to the following layout. I don't have a strong preference for the order or headings, however this style is very easy to create by moving each color name to the next line. –dlthewave ☎ 01:42, 10 September 2018 (UTC)

Color	Name	Hexadecimal in their website depiction	Notes
	Red	#ED0A3F
	Orange	#FF8833
	Yellow	#FBE870
	Olive Green	#B5B35C
	Light Chrome Yellow		On labels "Chrome Yellow, Light." Same color as "Light Yellow" (1903–1958) and "Lemon Yellow" (1903–1910, 1958–1990).
	Green
	Blue	#4997D0	Same color as "Celestial Blue" (1930–1949) and "Azure Blue" (1949–1958).
	Prussian Blue		Same color as "Midnight Blue" (1958–present).
	Celestial Blue		Discontinued by 1910.
	Purple		"Violet" from about 1914.
	Rose Pink	#FFA6C9	Same color as "Pink" (1903–1917) and "Carnation Pink" (1958–present).
	Burnt Sienna	#E97451
	Van Dyke Brown		Same color as "Brown" (1903–1935).
	Flesh Tint	#FFCBA4	Same color as "Flesh" (1949–1956, 1958–1962), "Pink Beige" (1956–1958), and "Peach" (1962–present).
	Burnt Umber
	Raw Sienna		Discontinued by 1910.
	Gold		Metallic; swatch represents nominal hue only. Available only in bulk after 1915.
	Silver	#C9C0BB	Metallic; swatch represents nominal hue only. Available only in bulk after 1915.
	Copper	#DA8A67	Metallic; swatch represents nominal hue only. Discontinued in 1915.
	Black	#000000
	Charcoal Gray		Discontinued in 1910.
	White	#FFFFFF

Should have the "in their website depiction" in the hex heading that I that I just added. Without that it is unsourced and unsourcable.North8000 (talk) 18:46, 10 September 2018 (UTC)

 Done –dlthewave ☎ 19:16, 10 September 2018 (UTC)

References