Talk:Reuleaux triangle/GA1

GA Review
The edit link for this section can be used to add comments to the review.''

Reviewer: Spinningspark (talk · contribs) 20:21, 30 October 2015 (UTC)

Looking... SpinningSpark 20:21, 30 October 2015 (UTC)


 * Lead
 * The information about manhole covers does not appear in the article body. Per WP:LEAD, the lead should be a summary of the body.
 * Why has the term circular disks been used instead of circles when the wikilink circular disk just goes to circle? Seems to be an easter egg.
 * Answer: mathematical pedantry. A circle is the outline; a disk is the shape inside it. If you intersect three circles, the result is just three points. You have to intersect the disks rather than the circles to get the Reuleaux triangle. —David Eppstein (talk) 20:05, 1 November 2015 (UTC)
 * "The name of Reuleaux triangles derives..." Better "The name Reuleaux triangle derives..." with terms referred italicised.
 * Reworded more simply, making this moot. —David Eppstein (talk) 20:05, 1 November 2015 (UTC)


 * Construction
 * The imperative mood of much of this section (First, use the compass...) is a WP:NOTMANUAL problem. It needs rewriting so that it is not instructional.
 * "...intersect three disks..." Why has the terminology changed from circles to disks? This is potentially confusing. [edit] Ah, perhaps that was meant to be the link in the lead


 * Mathematical properties
 * Could do with a wikilink to supporting line, or even explain in-article. I know it is already linked in the lead, but it is permissible to repeat links in the article body. A diagram with example supporting lines would make this so much clearer.
 * Linked, glossed, and illustrated. —David Eppstein (talk) 19:28, 1 November 2015 (UTC)
 * Is there a suitable article to wikilink curviliear, or curvilinear triangle?
 * I linked curve. We don't really have a general-purpose article on curved triangles, although there are some special cases linked from the "related figures" section. —David Eppstein (talk) 19:28, 1 November 2015 (UTC)


 * Reaching into corners
 * "The Harry Watt square is often used in mortising". The (reference #34) verifies neither that this tool is the Harry Watt square nor that its shape is a Reuleaux triangle, nor even that it cuts on the constant width principle. It also does not claim that this is a common method of mortising, and I seriously doubt that it is.
 * Sentence and source removed. —David Eppstein (talk) 02:01, 1 November 2015 (UTC)
 * "Panasonic robotic vacuum cleaner RULO has its shape based on the Reuleaux triangle".
 * The ref (#35) does not say this is a Reuleaux triangle, it just says "triangle". The video linked on Youtube does appear to say that, but it is in Japanese so I'm not sure what it does say, and relying on a manufacturer's promotional video is starting to get a bit spammy.  It strikes me that a plain triangle would actually reach into corners better, the curved shape only being for styling reasons and possibly smooth running along edges. If it's going to be kept, at least let's have a ref that verifies the fact - there are a few tech review sites out there that have covered it.
 * Another source added that specifically mentions the Reuleaux triangle basis for the shape. —David Eppstein (talk) 02:01, 1 November 2015 (UTC)
 * Contiguous wikilinks are deprecated.
 * Reworded to avoid contiguity. —David Eppstein (talk) 02:01, 1 November 2015 (UTC)


 * Rolling cylinders
 * "...with cross-sections that were Reuleaux triangles..." Should that be "...are Reuleaux triangles..."?
 * "Are" is correct, but I reworded this and in doing so made the point moot. —David Eppstein (talk) 22:56, 31 October 2015 (UTC)


 * Architecture
 * The bolding on "spherical triangle" should be removed. It is already bolded in the lead and that should not be repeated unless the section is the target of a redirect (MOS:BOLD).
 * "a triangle on the surface of a sphere". Could be wikilinked to spherical triangle which redirects to spherical trigonometry (the redirect is better in case that ever gets to be an article).
 * Both done. —David Eppstein (talk) 01:23, 1 November 2015 (UTC)


 * Other objects
 * "three equal tips also prevent wear". Surely what is meant is that three tips will give three times the lifespan compared with the single tip of an isoscelles triangle. The rate of wear depends more on the material than the shape.
 * Reworded, also to be less promotional. —David Eppstein (talk) 06:59, 1 November 2015 (UTC)
 * "...suggested as the shape of a reflecting mirror". This is referenced only to a patent, which are not always the best source of reliable information.  People can write all sorts of nonsense in a patent, as long as its not the same nonsense someone else has already claimed.  I'm not really following what that patent is driving at.  The claim for better optical properties revolves around the fact that the domain of the telescope, an inscribed ellipse, efficiently covers substantially less than 100% of a hexagonal tiling of mirrors.  The claim is that a Reuleaux triangle will be more efficient at covering a Reuleaux-shaped domain.  Well of course it will, but the goalposts have been moved!  This would be more acceptable if scholarly references could be found imparting reliability to this, or else news items covering it, which would raise it the level of notable even if it didn't work.  In a failed search for sources, I did come up with an alternative telescope related application, below, that you might consider including.
 * Removed. —David Eppstein (talk) 06:59, 1 November 2015 (UTC)
 * According to this paper, which appears to be a draft version of this, the antennae of the Submillimeter Array are arranged in Reuleaux triangles because "In 1994 Keto showed that interferometer layouts based on perturbed curves of constant width (in particular the Reuleaux triangle) offer the most complete sampling in the Fourier space of the image and hence the best image quality". Keto's original paper is available in full pdf here. Other papers discussing this are  (the last is paywalled except the abstract)
 * Added; thanks for the interesting application. I have subscription access to all of these but I ended up not using the Springer one because it said very little about the antenna configuration. I also used a photo of this to replace the removed Lancia logo. —David Eppstein (talk) 06:59, 1 November 2015 (UTC)


 * Signs and logos
 * I haven't checked them all, but the refs are not verifying that the logos are actually Reuleaux triangles. I looked at #57 and #61 (after machine translating)
 * Added a source specifically saying that the Petrofina logo is a Reuleaux triangle; removed some examples whose sources did not specifically describe their shape. —David Eppstein (talk) 01:03, 1 November 2015 (UTC)


 * Generalizations
 * "inradius". Please explain, gloss, or wikilink.
 * Added a gloss, within the subsection that explains this in more detail. —David Eppstein (talk) 22:55, 31 October 2015 (UTC)
 * "curved surface patches". What does patch mean in this context?
 * It's a standard but perhaps overly WP:TECHNICAL word for a piece of a surface. Expanded and reworded to avoid this word. —David Eppstein (talk) 22:55, 31 October 2015 (UTC)
 * Do the sources confirm that all this list of coins are Reuleaux heptagons? The "approximate shape" wording smacks of original research.  I know the British ones are and the Mauritius one looks likely, but the rest may just be guesswork by someone.
 * Cleared out the example farm and left only the ones that could be confirmed by sources as being Reuleaux polygons (the British and Canadian ones). —David Eppstein (talk) 22:55, 31 October 2015 (UTC)

SpinningSpark 18:13, 31 October 2015 (UTC)
 * References
 * Ref #19 (Gruber) refers to p. 78 but the convenience link goes to page 67. If there is not a link available to the correct passage, or the link must serve several references, it would be better to just link the front cover without a snippet search
 * I removed the link since there's no preview available for any page. —David Eppstein (talk) 18:14, 1 November 2015 (UTC)
 * Ref #70-75. The page number links fail to go to a specific page, so are useless. In fact, I am not getting preview from google at all for that book. The page number links should be remvoved leaving the page number unlinked.
 * Moot now since I removed all the coins sourced only to this coin catalog. —David Eppstein (talk) 18:14, 1 November 2015 (UTC)


 * I'm passing this now. The major outstanding issue was over WP:LEAD, which is explicitly a GA requirement.  However, on looking at that again, the manhole text is only one sentence and probably could not be summarised any further. SpinningSpark 19:12, 1 November 2015 (UTC)
 * Thanks! The other biggest thing I haven't done (yet?) was the issue with imperative in the description of the compass-and-straightedge construction. My general feeling is that NOTHOWTO is often taken too far; banning game playthroughs is a good thing but banning sequences of steps that are actually encyclopedic is a problem. It isn't clear to me whether changing to second person, or a more stilted form of third person (one first constructs a circle...etc) would really improve things. —David Eppstein (talk) 20:05, 1 November 2015 (UTC)
 * Yes, an arguable case, which is why I didn't mention it. Although WP:NOT is policy, and in theory a breach of it should be an automatic GA fail, in this case I would consider it more of a style matter. That is, it is not unwanted content, it just may not be in the form that everybody likes. SpinningSpark 20:12, 1 November 2015 (UTC)