Jump to content

User talk:JMF

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia
(Redirected from User talk:Concrete Cowboy)
History of Milton Keynes Milton Keynes Calendar (New Style) Act 1750 John Ogilby Robert Hooke

The Signpost: 14 August 2024

[edit]
* Read this Signpost in full * Single-page * Unsubscribe * MediaWiki message delivery (talk) 22:48, 14 August 2024 (UTC)[reply]

Nomination of A508 road for deletion

[edit]
A discussion is taking place as to whether the article A508 road is suitable for inclusion in Wikipedia according to Wikipedia's policies and guidelines or whether it should be deleted.

The article will be discussed at Wikipedia:Articles for deletion/A508 road until a consensus is reached, and anyone, including you, is welcome to contribute to the discussion. The nomination will explain the policies and guidelines which are of concern. The discussion focuses on high-quality evidence and our policies and guidelines.

Users may edit the article during the discussion, including to improve the article to address concerns raised in the discussion. However, do not remove the article-for-deletion notice from the top of the article until the discussion has finished.

10mmsocket (talk) 09:24, 19 August 2024 (UTC)[reply]

The Signpost: 4 September 2024

[edit]
* Read this Signpost in full * Single-page * Unsubscribe * MediaWiki message delivery (talk) 13:26, 4 September 2024 (UTC)[reply]

Swanbourne picture

[edit]

Hi! You asked about the location of the photo of Swanbourne that I uploaded some years back. It's actually closer to Newton Longville. It's looking towards the bridge carrying the road from NL to Bottledump Roundabout, from the Bletchley side. Tom walker (talk) 11:35, 15 September 2024 (UTC)[reply]

ta, I have updated the caption at Varsity Line accordingly. 𝕁𝕄𝔽 (talk) 13:04, 15 September 2024 (UTC)[reply]

Testosterone levels in male and female

[edit]

Glad that we fixed the thing from 1/4 at lest to 1/15, obviously we could find many other laboratory sources with somewhat different normal ranges. Good how wiki works! John Krisinger (talk) 01:38, 20 September 2024 (UTC)[reply]

The Signpost: 26 September 2024

[edit]
* Read this Signpost in full * Single-page * Unsubscribe * MediaWiki message delivery (talk) 20:09, 26 September 2024 (UTC)[reply]

Fire

[edit]

In response to [1] see for example: [2] [3] [4] and [5] but there is more when you google "Castlethorpe 1905 fire". Polygnotus (talk) 16:24, 30 September 2024 (UTC)[reply]

I was kinda hoping that the cn and dubious tags might persuade the new editor to do the work on their own contribution . If not, I guess we can do it. --𝕁𝕄𝔽 (talk) 17:08, 30 September 2024 (UTC)[reply]
You are a far more optimistic person than I am. Polygnotus (talk) 17:09, 30 September 2024 (UTC)[reply]
See The Signpost above: we must encourage the newbies. 𝕁𝕄𝔽 (talk) 17:25, 30 September 2024 (UTC)[reply]
It is encouraging for a newbie to see how their addition is improved and turned into a well-sourced part of the article.
It is (obviously) discouraging to see "[better source needed][dubious – discuss]".
If we want to increase editor retention we should focus on where the problem actually lies, which is not the carrot; it is the stick. Polygnotus (talk) 17:37, 30 September 2024 (UTC)[reply]
True, I hadn't appreciated that perspective. --𝕁𝕄𝔽 (talk) 18:10, 30 September 2024 (UTC)[reply]
Thank you! Polygnotus (talk) 05:00, 1 October 2024 (UTC)[reply]

New towns and counties

[edit]

Hi, you reversed one of my edits, where I changed the counties in which the new towns are located from ceremonial ones to historic ones.

The explanation you gave was "Peterborough WAS in Northamptonshire". Peterborough still is in Northamptonshire; the historic counties were not abolished by the LGA 74, nor were they abolished by the introduction of ceremonial counties in 1997. It just isn't in the ceremonial county of Northamptonshire (being in that of Cambridgeshire, which is legally distinct from the historic county Cambridgeshire of which Peterborough isn't a part), nor either of the Northamptonshire council areas. JoeyofScotia (talk) 12:21, 10 October 2024 (UTC)[reply]

Per long established and frequently re-iterated policy agreed at Wikipedia talk:WikiProject UK geography, historic counties are just that – historic. Washington is in Tyne and Wear; Peterborough is in Cambridgeshire just as Milton Keynes is in Buckinghamshire (that there are other UAs with those names is irrelevant). It is not at all helpful to modern readers to have the article clogged with WP:SOAPBOXing about the world as it was 50 years ago.
If you remain unconvinced, please open a new topic at the WikiProject, where a wider variety of views can assess your arguments. 𝕁𝕄𝔽 (talk) 13:42, 10 October 2024 (UTC)[reply]
They're historic, not historical. They still exist, they just have no formal function.
Peterborough is in the ceremonial county of Cambridgeshire. This being the case does not mean it ever ceased to be in Northamptonshire. The historic counties are not the world as it was 50 years ago, they're the world as it is now in all but administrative and lieutenancy (as an aside, Peterborough administratively was not under Northamptonshire Council after the 1880s). Peterborough is in Northants, Wolverhampton is in Staffordshire, Washington is in Co Durham. Your text implies that these towns stopped being in these counties, which is simply incorrect. JoeyofScotia (talk) 15:14, 10 October 2024 (UTC)[reply]
Take it to Wikipedia talk:WikiProject UK geography, we are not going to resolve it here. My reversion of your edit is consistent with the policy as I understand it. If the consensus of the discussion supports your interpretation, I will give way. --𝕁𝕄𝔽 (talk) 15:29, 10 October 2024 (UTC)[reply]

The Signpost: 19 October 2024

[edit]
* Read this Signpost in full * Single-page * Unsubscribe * MediaWiki message delivery (talk) 11:14, 19 October 2024 (UTC)[reply]

Your thread has been archived

[edit]
Teahouse logo

Hello JMF! The thread you created at the Teahouse, "Simple math is not SYNTH", has been archived because there was no discussion for a few days.

You can still read the archived discussion. If you have follow-up questions, please create a new thread.

See also the help page about the archival process. The archival was done by lowercase sigmabot III, and this notification was delivered by KiranBOT, both automated accounts. You can opt out of future notifications by placing {{bots|deny=KiranBOT}} on top of the current page (your user talk page). —KiranBOT (talk) 03:12, 27 October 2024 (UTC)[reply]

Document calculator formulas with Template:Math?

[edit]

Please have a look at the input related formulas at the second from bottom level in the information hierarchy: Template_talk:Body_roundness_index#informationHierarchyInputRelatedCalculations

I've tried to use Template:Math but failed.

  1. It took me hours of struggling and finally gave up, reverted to formulas in plain text.
  2. The Math template generates an image, no links possible to explain variables?

To me, the current text formula's looks fine, but some Math enthusiasts may prefer template:Math based version. That template is not for me, gives me headaches.

Are you comfortable with that Template? Would you like to step in and convert my text version to Math versions? Is it possible to keep hyperlinks?

While struggling with the Math template, I found the following BRI formula in the sandbox version of the calculator which is the same as the live version:

formula=364.2-365.5×pow(1-(pow(waistcm/6.28318,2)/pow(0.5×heightcm,2)),0.5)

Several issues with it:

  1. It is different version than documented on the BRI page. That is no big worry as it gives the same results. Still it would be easier to understand the code if it matched the version on the BRI page.
  2. It does not use the pi constant but a hardcoded 6.28318 for 2pi with a limited number of decimals. That is not a big deal, until close to the very crisp NICE boundary of 0.4, where every decimal matters. 0.39999 is out of range, 0.4 is healthy. I think that we should use maximum available accuracy for the computations. The WHtR has 2 versions, one accurate for computations and one to display. The BRI result does not matter, is a dead end in 4.0, does not propagate to other formulas.
  3. It could be optimised and reuse the WHtR result. I think the calculator is hitting formula propagation limits at the moment, so the less propagation the better. The current formula recomputes if waistcm or heightcm changes, which gives 2 propagation rules. It could reuse the WHtR result, which would be one formula propagation definition less.
And tricky, not confirmed by a reliable secondary source, but I think the WHtR - BRI connection would be clearer with showing how they are one and the same family:
grandparents
waist and height
parents
WHtR
child
BRI

Or in math:

It does simplify the complex BRI formula, good for the rest of us.


Can the WHtR formula be a hyperlinked image to WHtR at the BRI page to make the connection super clear? Uwappa (talk) 10:41, 28 October 2024 (UTC)[reply]

@Uwappa: I have to admit that I've never learned the syntax for <math>. Presumably you have searched Help:Displaying a formula so maybe someone at Help talk:Displaying a formula might help?
  1. Where do you consider that wlinks would be needed? We can assume that everyone knows what waist and stature are.
  2. We absolutely should use pi as a variable for as much precision as the system offers. But if a little judicious rounding gave you boundaries at the right place, I can't see that anyone would complain provided that you document it using the hidden note technique.
  3. Yes, I'm afraid we do need a secondary source. Although it is permissible in Wikipedia per WP:2+2=4 to recognise that they are mathematically equivalent because the constant (pi) can be ignored. But it is a WP:SYNTH violation to assert that the raft of citations for WHtR can be recycled as equally applicable to BRI.
  4. the formula is independent of the units used, so long as they are used consistently. So there is no need to specify "in cm".
  5. the source uses c/2π (=radius) and h/2 to calculate the eccentricity and it seems that custom and practice (as shown at Ellipse is to use half the long axis and half the short axis. Which is why Thomas et al. have shown it like that. The mathematical simplification that you have done is trivial and doesn't compensate for the loss of information on how it got there. So no, I wouldn't.
Have I covered everything? 𝕁𝕄𝔽 (talk) 14:23, 28 October 2024 (UTC)[reply]
Yes you covered everything.
  • As far as I understand, Math generates a picture. Now it is possible to create hyperlinks on part of a picture, but that is too time consuming for me. The text versions are good enough for me. If anybody wants to see Math there, go for it, but leave me out.
  • Agree, go for max precision with pi. It feels a bit silly as the whole pi concept applies to circular waists only and sub cm waist differences won't matter much anyway in real life. And the margin of error for measuring waist and height is happily ignored. But well, no need for additional inaccuracies.
  • The formulas caused enough turmoil, I won't touch them for now.
It will probably be a matter of time before more people realise that WHTR is the smaller brother of BRI, even part of the BRI formula. You could argue that BRI has only one input parameter: WHtR.
The calculator itself will help, calculating both WHtR and BRI from the same input. Amazing that the current live version does not convince.
People will see that it is silly to wait for a secondary source on 2 formulas using the same 2 input variables.
And if not in version 4.0, then the penny may drop when they see an idea for version 5.0
  • No, no, even a simple concept like waist raises questions:
    • where is the waist? Is that the smallest bit, the hips, the belly button?
    • English skills may be limited, is height waist the same as high waste?
    • Will the width measures from a picture do? (no, that is diameter) What does Circumference word mean? It sounds like something very painful.
    • Today I added instructional photos, go and see Template_talk:Body_roundness_index#informationHierarchyInput
    • You can't aim to low here. The unconscious competent will just ignore the hyperlinks. Pictures and hyperlinks will help the conscious incompetent.
  • Yes and no on the independency of unit. Yes you could input mm iso cm and it would still work. The nasty detail here is that the imperial height is two input variables, which is not the same as one for a simple computer without a human brain. The calculator formulas just use cm, being the smallest unit. An imperial input will still be accurate, as it produces decimals in cm.
  • A strong no to WP:SYNTH. WHtR results apply to ratios of height and waist. BRI results apply to ratios of height and waist. Spot the difference. I can't. There is nothing to SYNTH here. It like freezing water would yield different conclusions dependent for Celcius than Fahrenheit. They don't. Water does not even know about those two scales when it freezes. Similar for the a combination of height and waist. One will be conclusions for height 178 waist 60 and the other will be conclusions for height 178 and waist 60. Spot the difference. A coloured WHtR graph will look exactly the same as a BRI graph, the only difference will be the black lines for a WHtR scale or a BRI scale.
A BRI chart without the black BRI lines will be exactly the same as a WHtR chart without WHTR lines. Both re health levels for length waist combinations. In fact, such a lineless chart could serve as a paper body roundness calculator for risklevel. Just look at the (height, waist) coordinates for the colour.
  • Would you like to dive into the 4.0 sandbox wikicode and have a go at the changes?
    • It would be good if some more people understand the calculator concepts.
    • I am really impressed by the calculator. Please go and see the source of User:Doc_James/BRI the very first version.
    • Other simple examples: Inch#Equivalents, Centimetre#Equivalence_to_other_units_of_length. Please inspect the wikicode. It is ridiculously easy. It really nicely hides all the complex formula propagation behind it.
    • For me Template:Calculator is a joy. Template:Math gives me a headache.
    • My expectation is that the calculator will be very popular indeed.
    • I am not exactly sure how the template handles rounding. I expected an internal value for computing and a rounded value for displaying. But I saw some strange things after rounding WHtR to 2 decimals. So I had to create a hidden, unrounded version of WHtR for calculations which now propagates to the rounded one being displayed, a dead end of the propagation chain.
Your "No I prefer to stay away from the calculator code" is fine with me, if I don't see any changes by tomorrow, I'll make the changes myself, make two versions on BRI, one that uses height and weight as input, the other one WHtR. It would surprise me if they produced different results, but I'll check to be sure before ditching the current formula.
I went through your previous concerns. I expect quite some turmoil when 4.0 goes live. 4.0 seems to be so much more than just a calculator. For many it will be hard to believe that it is just a calculator. It might even make it to the top 100 of medical calculators. Well having seen the pity state of the commercial BRI calculator, that would not even surprise me. I think 4.0 is already way beyond its commercial equivalent. Funny that the wikipedia one will be free available, including source code. Ha ha ha.
See your concerns addressed in:
  1. "Is the body roundness calculator really just a calculator? Isn't it AI?" in Template_talk:Body_roundness_index#Information_hierarchy
  2. the list of WP guidelines
What you can do to help:
  • Fill the question marks for NICE based risk level. Any source for WHtR values below 0.4. I'd say that is a very rapid down from "no increased health risks" to "further increased health risks" and even "no living subjects to study". Any mortality known based on WHtR research? Any WHtR values known for emaciated people? There must me.
  • specify WHtR value specific colours at: Template_talk:Body_roundness_index#Colours_for_Body_Roundness_4.0. Key colours will do, I will compute the gradients, just like last time
  • Have a look at how the sub WHtR 0.4 colour gradient works out now in the sandbox. To me it looks good, very good.
  • Should colours at 0.4 to 0.7 also have a gradient? My suggestion: yes, because things rapidly go down there!
  • Have a peek into the future at: Template_talk:Body_roundness_index#Moving_dot_on_graphic_for_version_5.0?
Uwappa (talk) 16:30, 28 October 2024 (UTC)[reply]
  • Where exactly is the waist: see Waist#Waist measurement.
  • SYNTH: before you spend a load more time on it, I urge you to test the question at Wikipedia talk:Identifying reliable sources (medicine)
  • What's happening about hip circumference? Thomas mentions it as significant but doesn't take it any further. It's a real pity that her calculator has been taken down. You should ask why.
  • Question marks for NICE based risk level: well above my pay grade, I'm afraid. You need at least a qualified medic (who will probably tell you to read the citations at WHtR. Ditto the colours.
  • "Is the body roundness calculator really just a calculator? Isn't it AI?": No. It is not even an expert system. It is just grade school division in a pretty package.
  • Imperial units: if people want to put in half-inches (as n.5), how is that a problem. If it matters, then there is some false precision at play. You can just blur the edge-cases. Most people who use imperial cam give their height in inches without tech support.
  • Should colours at 0.4 to 0.7 also have a gradient? See WP:NOR. Sorry.
  • Moving dot is a good idea.
𝕁𝕄𝔽 (talk) 20:05, 28 October 2024 (UTC)[reply]
Ha ha ha, spot on. You are absolutely right.
  • Correct! It is not an expert system.
  • Yes! it is just a calculator in a pretty package.

Yet, I expect a lot of turmoil when 4.0 goes live.
Panic, This is just too complex for me so it must be dangerous! This must be AI! This is something that can not be maintained by most Wikipideans, completely incomprehensible! Now of course, I myself is not stupid, I is just toooooo busy doing really important things, have no time. I don't understand it, can't maintain it, so it must be removed.
Pfft, this is nothing, I am not even impressed, can't be bothered. Remove it anyway, I'll come up with some guideline violations that sound serious!
  • This is an expert system giving medical advice!
  • It violates many Wikipedia guidelines!
  • This must be Javascript, a serious security problem!
  • Panic, panic, panic!
Please click show at Template_talk:Body_roundness_index#AI_or_not_AI? and see the plain English answer to all of the expected panic.
Should that text be always visible to avoid turmoil? Uwappa (talk) 22:58, 28 October 2024 (UTC)[reply]
I am thinking about wrapping the calculator in 3 tabs.
  1. the calculator itself, the default first tab
  2. AI or not AI? With the plain English answers to avoid turmoil questions
  3. the information hierarchy, the formulas and propagation explained to the rest of us, next to Wikipedia guidelines.
The documentation could then be for the lucky few that do know the difference between a calculator, a spreadsheet, an expert system and AI. Those are likely the ones that can help to maintain it, while keeping the user interface as simple as it is.
I'll think about your remarks. To be continued... Uwappa (talk) 23:17, 28 October 2024 (UTC)[reply]
Done:
  1. went back to the drawing board for unitless input. That was a major puzzle for imperial feet and inches for the current calculator template.
  2. bold decision: just one input field for height, one for waist, let the computer compute all conversions
  3. You can really use any unit now, use pixels if you want (and do not forget to multiply waist diameter by pi)
  4. use unitless height and weight for WHtR. The formula in the coding is now really sweet: formula=waist/height
  5. compute BRR using pi with max available decimals, use formula as on BRI page and reuse whtr to minimise formula propagation rules. Current sandbox version: formula=364.2-365.5*sqrt(1-pow((whtr/pi),2)) Yes, that is real code that I can understand, unlike the Math template
  6. checked new BRI formula against old one, same results, even better decimals. Removed old BRI formula
  7. reverted idea to truncate WHtR and BRI. Just show max decimals there too. Let the user do the rounding to own desire.
  8. created feet to inches and yard to feet calculator at Foot_(unit), ha ha, no that calculator is not AI, not an expert system either.
  9. updated Template_talk:Body_roundness_index#informationHierarchyInputRelatedCalculations formulas, simplified conversions, linked to conversion calculators at pages for cm and inch
  10. updated formula on BRI page, removed cm as unit from text. Fine-tuned text. Moved calculator close to formula.
  11. to be continued for your other comments...
My design doubt for now: Is the one input field for height OK for people using feet and inches?
There are quite a lot of conversion rules involved for height as the feet and inches make it complicated. It is fun to see though how the calculator works like a spreadsheet. And it won't do much harm on formula propagation as all those conversions are on the end of a chain. Only the unit-less height and waist go on to update whtr and the whole bunch that is derived from whtr.
Your user page shows Milton Keynes, England, so you are probably surrounded by people used to feet and inches.
And you look quite enthusiastic about getting the calculator right.
Would you like to do a real world user test with the Sandbox Calculator?
See you tube video by Jakob Nielsen on YouTube
Please avoid a focus to the input field. Shut up about that. Don't tell them what you are testing. Do not influence the subjects.
  • Tell the subject that you are testing a new user interface. It is not them who are the subject of testing. No worries. You want to test a brand new user interface that is a candidate for the top 100 of medical calculators. It needs to be tested and it must be tested fast. There is no money involved here, nothing to gain, just volunteer work for Wikipedia. The benefit for them is knowledge about their own health which they are welcome to share with friends and family they care about the most.
  • Ask them a question at the highest possible level: What would be a healthy waist size for you? For that question they won't need to measure their own waist. Knowing their height suffices to answer the question. Do not give any other instructions. Shut up. Answer 'I don't know' to any question. In real life there won't be a calculator helpdesk either.
  • Most people will know their height. If not will they be creative and use an alternative? Like standing against a wall counting bricks? Or use a piece of string/towel as a unit?
  • Take notes of what happens when. Note time to the second accuracy, what they are doing and if it is a success. Videoing may not be a good idea, as people act differently when being filmed.
  • Shut the f*** up and take notes. What actions do they take? What difficulties do they encounter? What goes well?
  • Shut up. Don't help if they struggle. Let them figure it own on their own.
  • A failure is a very useful test result too. Note where they failed and when.
  • Do not discuss solutions. Every body has an opinion. Just takes notes, even if they talk complete nonsense as their wrong of train of thought may be valuable. Don't argue, don't discuss, just shut the f*** up and take notes. If they want green socks with pepperoni, just note that down. Shut up, you are not discussing a solution, you are collecting valuable real time experiences.
  • Even shut up if they go on a completely wrong path. Shut up and take notes of the path they take. Others may do the same.
  • What method did they choose to enter feet and height? Note the time and what they did, both unsuccessful and successful.
  • Did they find the optimum waist size? At what time did they finish? How long did it take them in total?
  • And please publish the results on Template_talk:Body_roundness_index. I'm curious!
  • You'll probably love the experience. But slow down. Don't test any more than 5 people, save 'fresh' ones for next versions.
And yes, it is OK if that is too much trouble, too difficult or too time consuming and I'll find another way of testing. Uwappa (talk) 10:43, 29 October 2024 (UTC)[reply]
  • I won't be testing anybody, I am not a medic.
  • Habits in the UK vary. Although the majority still use feet and inches, a substantial minority (including me) use cm.
  • Most people who use feet and inches know the inches-only equivalent (most people are ≤ 5'10" and since 5' = 60", it is easy to work it out.
  • So long as your calculator doesn't conflict with MEDRs and WP:NOR, there should be no complaints about adding it to articles provided that it is unobtrusive.
𝕁𝕄𝔽 (talk) 11:51, 29 October 2024 (UTC)[reply]
OK, no worries, I'll ask Zefr.
Breaking news, now in sandbox: input WHtR to compute waist size.
And yes, that works for healthy ranges values too, try 0.4 and 0.5.
Please go and try to find a healthy waist size for some one your height.
:D Uwappa (talk) 11:59, 29 October 2024 (UTC)[reply]

中国房间 = AI OR NOT(AI)

[edit]

How about the following answer:

中国房间 = AI OR NOT(AI)

Ha ha ha, I like the Chinese characters there! And the AI OR NOT(AI) could be a Math formula.

The 2 of them combined could be shown as a label for a radio button, which can be styled as a clickable tab. Inspired by: user:Cmglee.

It will be a little inside joke for

  • people who can read Math formula's
  • IT experts who understand the boolean joke
  • see the link with To_be,_or_not_to_be
  • people who can read Chinese characters
  • people who are familiar with the Chinese_room

That will be just a the lucky few.

For the rest of us, that will be a incomprehensible puzzle, which is explained on Template_talk:Body_roundness_index#AI_or_not_AI? which will be the contents of that tab.

People who understand the explanation will understand that calculator = NOT(AI) and will become part of the lucky few.

Uwappa (talk) 04:02, 30 October 2024 (UTC)[reply]

I was thinking recently that Chinese room would be a cute GA target, given my other onwiki pursuits. Remsense ‥ 论 04:11, 30 October 2024 (UTC)[reply]
谢谢。你回复得真快！
你能用 Template:Math 帮我写一下那个数学题吗？
Template:Math 对我来说太难了。

English translation for the rest of us, using the Chinese room called Google translate :D

Uwappa (talk) 04:39, 30 October 2024 (UTC)[reply]

WP rules and regulations OK?

[edit]

Please visit: Template_talk:Body_roundness_index#Information_hierarchy and have a look at the WP rules and guidelines.

Those are based on your and Zefr's earlier worries.

  1. Anything missing from previous talks?
  2. Is it correct?
  3. Is it WP:PLAIN, understandable for IP users?

Uwappa (talk) 11:17, 4 November 2024 (UTC)[reply]

Suggestion

[edit]
  1. Gain some experience with Template:Calculator
  2. Start with a simple one in your own sandbox, e.g. input a, calculate b=2*a
  3. Look at current calculators for conversions between metric and imperial, e.g. at inch

Uwappa (talk) 20:24, 4 November 2024 (UTC)[reply]

@Uwappa: Why? But taking a guess (forgive me if I'm wrong):
  1. Yes, I know that it calculates 80 / 178 = 0.449438202247191 but there is a huge difference between the result of a calculation and the judgement that must be made on what is displayed, taking the precision of the input into account. That judgement is critical and must be made.
  2. no comment.
  3. Yes, the conversion between inches and cm is well defined (international yard and pound agreement) BUT (and it is a very big but) people do not measure their waist and height in microns or thousandths of an inch. The input measures are crude so the output must be equally crude, that is the fundamental point of false precision. It is one of the most basic principles of doing science and engineering. --𝕁𝕄𝔽 (talk) 22:50, 4 November 2024 (UTC)[reply]

[dubious – discuss] I can not read minds. The calculator can not read minds. The calculator does not know the goal people are after.

  • Yes, people may use it for adjudicating the body roundness of real people. Excellent, that may save a few lives.
  • But just as well, they may be just playing games, hackers trying to let it crash with silly input. E.g. is SQL_injection possible to hack a database?
  • Wikipedian JMF could use it to support a "2 decimals suffice" point of view.
  • In the usability test today, the paramedic had to input dimensions of this fake extremely obese "patient":
Patients_for_bri_calculator_usability_test.jpg

That did not go down well, it was too ridiculous, so I'll update the test. Ha ha ha, the usability test revealed an error in the usability test.

  • People may use it to verify the credibility of the calculator. What WHtR value does it show if a height = 2x waist?
  • What if I enter data measurements from a picture? Will it be a factor pi off, because diameter != circumference?
  • How does it deal with negative numbers, zeroes?
  • How well can it cope with insane large numbers, like 1E88?
  • I myself used it during development. I needed precision. Was 0.3999999999999999 classified correctly as below 0.4?
  • I used it to map WHtR values to silhouette index numbers (1, 2, 3, ..., 20). With BRI values as a nasty intermediate as the current silhouettes are still BRI based.
  • I used it to compare results of the old BRI formula (with rounded 2 pi hardcoded) against the new formula (using a pi with max decimals and just WHtR as input)

So I firmly disagree with the rounding. You do not know what people use it for.

  • when you round, you can't go back to precision. What will you do with 0.399999? Display as 0.40 with an error message "Unspecified for WHtR below 0.4"???
  • when you keep precision, you can leave the rounding to the human.

On point 2: How about a calculator that is art related? E.g. something for a Golden_ratio? Enter a number and the calculator will

  • take it as a width and compute height
  • and... take it as a height and compute width

On point 3: Any unit will do. People may may use a piece of string or cloth as a unit. E.g. a towel the size of the waist. And it will work just fine. Uwappa (talk) 00:01, 5 November 2024 (UTC)[reply]

The Signpost: 6 November 2024

[edit]
* Read this Signpost in full * Single-page * Unsubscribe * MediaWiki message delivery (talk) 08:04, 6 November 2024 (UTC)[reply]
Retrieved from "https://en.wikipedia.org/w/index.php?title=User_talk:JMF&oldid=1256521747"