Wikipedia talk:WikiProject UK Railways/Archive 43

This is an archive of past discussions. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.

Archive 40

Archive 41

Archive 42

→

Chains RFC

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 were seven possible options that were discussed in this RfC. We can dismiss a number of them immediately; proposals 2 (2/14), 4 (8/24), 5 (7/15) and 6 (3/9) were comprehensively rejected with more than 2/3 opposition. A late proposal, Proposal 7, had a few supports but did not have enough participation to be considered in this close.

This leaves Proposals 1 and 3. Of these, Proposal 3 was clearly the most supported, with over 75% support; Proposal 1 had around 60% support. However, these two Proposals are actually quite similar.

Proposal 1: "The use of chains as a unit of measurement in articles on railway lines and railway stations is appropriate, where chains (or miles and chains) are given in a quoted source. Where chains are not given in a quoted source, they should not be introduced by way of conversion, but decimal miles and decimal kilometres (to two decimal points) shall be used."
Proposal 3: "Where chains are given as a measurement, then conversion shall be from miles and chains to both decimal miles and kilometres (two decimal places)."

Given the common use of chains in railway-related sources, then WP:V obviously suggests (and the majority of commenters agree) that if the source uses that measurement then it should be noted (and WP:MOSNUM appears to agree on this point ("UK engineering-related articles, including those on bridges and tunnels, generally use the system of units that the topic was drawn up in")) which of course in many such cases is miles and chains, and will thus appear in many of the sources. There is general agreement that if the miles & chains measurement is to be used, it should be double-converted into fractional miles and metric distances (indeed, this was even mentioned by some opposers, i.e. "I oppose using just chains and km, I am ok with using fractional (decimal) miles, chains, and km, in any order.").

So analysing the general consensus here, I am going to close this as a combination of the most popular options. "The use of chains as a unit of measurement in articles on railway lines and railway stations is appropriate, only where chains (or miles and chains) are given in the quoted source. Where chains are given as a measurement, then conversion shall be from miles and chains to both decimal miles and kilometres (two decimal places)". Black Kite (talk) 20:10, 7 August 2018 (UTC)

The issue of the use of chains has again reared its head. It is now time to put this to bed by having a discussion and reaching consensus on the use of chains in railway articles.

Articles likely affected

Articles on railway stations and railway lines in the British Isles.

Articles possibly affected

Artcles on railway stations and railway lines where the British had strong influence in their construction. Basically the British Empire and other areas, such as Argentina.

The dispute

Some editors claim that the use of chains should be deprecated as they are a generally archaic measurement. Other editors claim that the use of chains is appropriate as long as external sources use chains, and that chains are a unit of measurement still used by the modern railway in the United Kingdom today.

Previous discussions

Talk:Darlington railway station#Distance from London (January 2018)
Talk:East Croydon station#Chains (July 2018)

Policies affecting this dispute

WP:V

Guidelines affecting this dispute

Enforcement proceedures affecting this dispute

WP:GS/UKU

I therefore submit the following two proposals for adoption. Mjroots (talk) 19:29, 8 July 2018 (UTC)

Proposal for adoption (I)

The use of chains as a unit of measurement in articles on railway lines and railway stations is appropriate, where chains (or miles and chains) are given in a quoted source. Where chains are not given in a quoted source, they should not be introduced by way of conversion, but decimal miles and decimal kilometres (to two decimal points) shall be used. For example High Speed 1 was built in metric units.

Support

Mjroots (talk) 19:32, 8 July 2018 (UTC)
Andy Dingley (talk) 19:42, 8 July 2018 (UTC)
~~Optimist on the run (talk) 21:30, 8 July 2018 (UTC)~~ switching support to option III
Jc86035 (talk) 00:09, 9 July 2018 (UTC)
SovalValtos (talk) 02:12, 9 July 2018 (UTC)
Dr Sludge (talk) 03:55, 9 July 2018 (UTC)
Serial Number 54129 11:18, 9 July 2018 (UTC)
-mattbuck (Talk) 16:17, 9 July 2018 (UTC)
Markbassett (talk) 03:00, 11 July 2018 (UTC) (prefer proposal 3)
Salix alba (talk): 07:53, 11 July 2018 (UTC)
scope_creep (talk) 08:34, 12 July 2018 (UTC)
--Redrose64 🌹 (talk) 07:20, 13 July 2018 (UTC)
Kerberous (talk) 13:26, 13 July 2018 (UTC)
Johnuniq (talk) 03:51, 14 July 2018 (UTC)
The chain was a very familiar unit back in the days when the old measurements were used. For the millennial, we have metric. Hawkeye7 (discuss) 03:55, 14 July 2018 (UTC)
The metric system has nothing to do with millennials (people born in the 1980s or 1990s, depending on definition); British metrication started in the mid-1960s and was in place for most purposes within a decade. — SMcCandlish ☏ ¢ 😼 06:32, 14 July 2018 (UTC)
Weighed food only went over to metric in the 1990s, and then with protest, see metric martyrs. Many people of 50+ think in imperial units and then convert when forced to do so. Martin of Sheffield (talk) 09:34, 14 July 2018 (UTC)
When I was there last the road signs were still in miles. Hawkeye7 (discuss) 22:19, 14 July 2018 (UTC)
Still are. The vast majority of British millennials (let alone older generations) would instinctively use miles for geographical distances. But I suspect most would struggle if you asked them how many yards or feet there are in a mile, let alone how many chains. Kahastok talk 22:48, 14 July 2018 (UTC)
Support where, as the proposal says, the original sources are in imperial. Martin of Sheffield (talk) 09:34, 14 July 2018 (UTC)
Useddenim (talk) 10:08, 14 July 2018 (UTC)
Support in cases where WP:UNIT's instruction that UK engineering-related articles use "the system of units that the topic was drawn up in" applies. --Ahecht (TALK
PAGE) 22:41, 25 July 2018 (UTC)
Certes (talk) 21:23, 28 July 2018 (UTC)

Oppose

Fish+Karate 09:23, 9 July 2018 (UTC) - I oppose using just chains and km, I am ok with using fractional (decimal) miles, chains, and km, in any order. Fish+Karate 10:14, 9 July 2018 (UTC)
Mangoe (talk) 12:19, 9 July 2018 (UTC)
Mark999 (talk) Mark999 (talk) 15:47, 9 July 2018 (UTC)
Carrite (talk) 06:14, 10 July 2018 (UTC)
The chain should never be the primary unit. A conversion (from mi or km) to chains is acceptable if sourced. Dondervogel 2 (talk) 23:24, 12 July 2018 (UTC)
What is appropriate for technical use within the industry is very different from what is appropriate for encyclopedia readers. I don't think we can expect readers to know about chains, so we should not present distances to them in those units. —David Eppstein (talk) 00:31, 13 July 2018 (UTC)
No. This unit is meaningless to almost all our readers. Worse, it has multiple meanings even in the UK (different dimensions in Scotland). See also Chain (unit)#Contemporary use: "In Britain, the chain is no longer used for practical survey work ...[and]... is no longer taught in British schools". — SMcCandlish ☏ ¢ 😼 03:38, 14 July 2018 (UTC); revised: 06:32, 14 July 2018 (UTC)
Definitely not, for reasons clearly stated above. Tony (talk) 09:25, 14 July 2018 (UTC)
No, for many reasons, most notably the potential inconsistency in sources (the distance between the Station A and Station B was three miles [source 1], and the distance between Station B and Station C was 5 kilometres [source 2]), and the longstanding problem of POV pushers choosing sources to match the system of units they prefer, rather than the other way around. Note that such a system has been proposed by said POV pushers on dozens of occasions at WT:MOSNUM, and has been rejected every time. Separately, I'm also opposed because it is not clear that chains measurements require a conversion to decimal/fractional miles for users of imperial units that are unfamiliar with chains. Kahastok talk 11:44, 14 July 2018 (UTC)
- To add to this, note that if it were proposed to replace where chains (or miles and chains) are given in a quoted source with where the line was originally drawn up in chains (or miles and chains) (wording based on the exception in WP:UNITS, no objection to changing to correct terminology if required), and if point III or IV below were also adopted, I would withdraw my opposition. Kahastok talk 12:27, 14 July 2018 (UTC)
Oppose. Wikipedia should not be used by rail workers to build or maintain rail lines, and no one else wants or needs to know about chains. Jc3s5h (talk) 12:00, 14 July 2018 (UTC)
Oppose, same reasons as Jc35sh, except nobody needs to know about the obsolete imperial chain. Surveying chains are still used in India, but they are metric, so the continued use of something nearly but not the same dimension is misleading to Wikipedia's worldwide readership.--John Maynard Friedman (talk) 13:50, 17 July 2018 (UTC)
Oppose. I doubt that 95% of the population has any idea what it means; every time the word is used, it would need to be explained. Why bother?? Use measurements that people understand. Peter K Burian (talk) 22:57, 28 July 2018 (UTC)

Proposal for adoption (II)

Where chains are given as a measurement, then conversion shall be from miles and chains to kilometres (two decimal places). Where the conversion is 49 chains (1078yds or 985.7m) or less, conversion shall be to metres. Conversion from miles and chains to decimal miles and kilometres shall be deprecated.

Support

Mjroots (talk) 19:33, 8 July 2018 (UTC)
~~Andy Dingley (talk) 19:41, 8 July 2018 (UTC)~~ In favour of Option III below
Support, because the converse would prohibit use of source material. --Redrose64 🌹 (talk) 10:52, 26 July 2018 (UTC)

Oppose

Jc86035 (talk) 00:09, 9 July 2018 (UTC)
Dr Sludge (talk) 04:08, 9 July 2018 (UTC)
ianmacm (talk) 05:16, 9 July 2018 (UTC)
Fish+Karate 09:24, 9 July 2018 (UTC)
Mangoe (talk) 12:20, 9 July 2018 (UTC)
Mark999 (talk) Mark999 (talk) 15:47, 9 July 2018 (UTC)
Markbassett (talk) 02:54, 11 July 2018 (UTC)
The chain should never be the primary unit. A conversion (from mi or km) to chains is acceptable if sourced. Dondervogel 2 (talk) 23:26, 12 July 2018 (UTC)
No. This unit is meaningless to almost all our readers. Worse, it has multiple meanings even in the UK (different dimensions in Scotland). See also Chain (unit)#Contemporary use: "In Britain, the chain is no longer used for practical survey work ...[and]... is no longer taught in British schools". — SMcCandlish ☏ ¢ 😼 03:39, 14 July 2018 (UTC); revised: 06:33, 14 July 2018 (UTC)
Oppose—as above. Tony (talk) 09:26, 14 July 2018 (UTC)
Useddenim (talk) 10:08, 14 July 2018 (UTC)
Oppose because most readers using imperial units (including in the UK) won't know what a chain is. Kahastok talk 11:44, 14 July 2018 (UTC)
Oppose. Wikipedia should not be used by rail workers to build or maintain rail lines, and no one else wants or needs to know about chains. Jc3s5h (talk) 12:01, 14 July 2018 (UTC)
Oppose, same reasons as Jc35sh, except nobody needs to know about the obsolete imperial chain. Surveying chains are still used in India, but they are metric, so the continued use of something nearly but not the same dimension is misleading to Wikipedia's worldwide readership. --John Maynard Friedman (talk) 13:58, 17 July 2018 (UTC)

Proposal for adoption (III)

(Per Ianmacm below) Where chains are given as a measurement, then conversion shall be from miles and chains to both decimal miles and kilometres (two decimal places).

e.g. "It is 10 miles 28 chains (10.35 mi; 16.66 km) measured from London Bridge."

Support

Andy Dingley (talk) 09:41, 9 July 2018 (UTC)
Fish+Karate 09:50, 9 July 2018 (UTC)
ianmacm (talk) 10:59, 9 July 2018 (UTC)
Jc86035 (talk) 11:50, 9 July 2018 (UTC)
Mangoe (talk) 12:21, 9 July 2018 (UTC) but prefer IV below
Dr Sludge (talk) 15:35, 9 July 2018 (UTC)
Mark999 (talk) Mark999 (talk) 15:47, 9 July 2018 (UTC)
· · · Peter (Southwood) ^(talk): 06:49, 10 July 2018 (UTC)
GrindtXX (talk) 10:58, 10 July 2018 (UTC)
Serial Number 54129 (talk) 17:22, 9 July 2018 (UTC)
Optimist on the run (talk) 12:42, 10 July 2018 (UTC)
Markbassett (talk) 03:02, 11 July 2018 (UTC)
Salix alba (talk): 07:54, 11 July 2018 (UTC)
Thryduulf (talk) 00:44, 12 July 2018 (UTC)
Maproom (talk) 07:16, 13 July 2018 (UTC)
Johnuniq (talk) 03:53, 14 July 2018 (UTC)
Useddenim (talk) 10:08, 14 July 2018 (UTC)
Support if it is decided that chains are an appropriate topic-specific unit (a point that I have no opinion on). Kahastok talk 11:44, 14 July 2018 (UTC)
TBM10 (talk) 21:24, 14 July 2018 (UTC)
Support if chains are used (which is separate from this proposal) --Ahecht (TALK
PAGE) 22:33, 25 July 2018 (UTC)
Certes (talk) 21:26, 28 July 2018 (UTC) (equally happy with II, but III seems more likely to fly)
Wikid77 (talk) 12:58, 1 August 2018 (UTC) (chains are official, but U.S. cars show decimal miles [not fractions], and so also show both mi/km)

Oppose

Mjroots (talk) 09:47, 9 July 2018 (UTC)
scope_creep (talk) 08:35, 12 July 2018 (UTC)
The chain should never be the primary unit. A conversion (from mi or km) to chains is acceptable if sourced. Dondervogel 2 (talk) 23:27, 12 July 2018 (UTC)
No. This unit is meaningless to almost all our readers. Worse, it has multiple meanings even in the UK (different dimensions in Scotland). See also Chain (unit)#Contemporary use: "In Britain, the chain is no longer used for practical survey work ...[and]... is no longer taught in British schools".
This proposal would be less awful that some of the others; prefer Prop V first, Prop VI second. — SMcCandlish ☏ ¢ 😼 03:40, 14 July 2018 (UTC); revised: 06:33, 14 July 2018 (UTC)
Agreed. If they know what a mile is, they know what a chain is. Hawkeye7 (discuss) 04:54, 14 July 2018 (UTC)
Oppose. Wikipedia should not be used by rail workers to build or maintain rail lines, and no one else wants or needs to know about chains. Jc3s5h (talk) 12:02, 14 July 2018 (UTC)
Oppose, same reasons as Jc35sh, except nobody needs to know about the obsolete imperial chain. Surveying chains are still used in India, but they are metric, so the continued use of something nearly but not the same dimension is misleading to Wikipedia's worldwide readership. But if we must include it, then Dondervogel 2's proposal is the least worst option. --John Maynard Friedman (talk) 14:00, 17 July 2018 (UTC)

Proposal for adoption (IV)

Distances shall be shown in miles and/or kilometres. Chains shall not be used.

e.g. "It is 10.35 miles (16.66 km) measured from London Bridge."

Support

This or III, which I think has more chance of succeeding. Fish+Karate 12:19, 9 July 2018 (UTC)
Mangoe (talk) 12:21, 9 July 2018 (UTC)
Carrite (talk) 06:13, 10 July 2018 (UTC)
G-13114 (talk) 22:10, 13 July 2018 (UTC)
Second choice, after my added Prop. V, way below. The chains unit is meaningless to almost all our readers. — SMcCandlish ☏ ¢ 😼 03:39, 14 July 2018 (UTC); revised: 06:14, 14 July 2018 (UTC)
I guess. Tony (talk) 09:27, 14 July 2018 (UTC)
Support if it is decided that chains are not an appropriate topic-specific unit (a point that I have no opinion on). Kahastok talk 11:44, 14 July 2018 (UTC)
Support per wp:worldwide. --John Maynard Friedman (talk) 14:03, 17 July 2018 (UTC)

Oppose

Optimist on the run (talk) 11:48, 9 July 2018 (UTC)
Serial Number 54129 (talk) 13:09, 9 July 2018 (UTC)
Mjroots (talk) 12:51, 9 July 2018 (UTC)
The joy of all things (talk) 13:14, 9 July 2018 (UTC)
Andy Dingley (talk) 13:38, 9 July 2018 (UTC)
Dr Sludge (talk) 15:35, 9 July 2018 (UTC)
Mark999 (talk) Mark999 (talk) 15:47, 9 July 2018 (UTC)
-mattbuck (Talk) 16:18, 9 July 2018 (UTC)
GrindtXX (talk) 12:07, 10 July 2018 (UTC)
ClemRutter (talk) 22:11, 10 July 2018 (UTC)
Markbassett (talk) 02:53, 11 July 2018 (UTC)
This would require misrepresentation of at least some sources. Thryduulf (talk) 00:43, 12 July 2018 (UTC)
scope_creep (talk) 08:36, 12 July 2018 (UTC)
I see no need to deprecate a conversion (from mi or km) to chains if sourced. Dondervogel 2 (talk) 23:28, 12 July 2018 (UTC)
--Redrose64 🌹 (talk) 07:22, 13 July 2018 (UTC)
Education is Wikipedia's mission and letting people know that chains are used by sources (given conversions are provided) is helpful. Johnuniq (talk) 03:50, 14 July 2018 (UTC)
Education is our mission. Hawkeye7 (discuss) 04:53, 14 July 2018 (UTC)
SovalValtos (talk) 08:45, 14 July 2018 (UTC)
Jc86035 (talk) 08:58, 14 July 2018 (UTC)
Useddenim (talk) 10:08, 14 July 2018 (UTC)
TBM10 (talk) 21:24, 14 July 2018 (UTC)
Ahecht (TALK
PAGE) 22:34, 25 July 2018 (UTC)
Certes (talk) 21:27, 28 July 2018 (UTC)
Wikid77 (talk) 13:04, 1 August 2018 (UTC)

Discussion

Just to note that if chains are an archaic measurement (which I dispute), the so are furlongs, but that doesn't stop horse races being measured in them. Optimist on the run (talk) 21:33, 8 July 2018 (UTC)

Also in horseback riding, the size of dressage rings seems rooted in chains, as 1x2 chains or 1x3 chains, now metric 20 m × 60 m (65.6 ft × 196.9 ft). -Wikid77 (talk) 14:39, 1 August 2018 (UTC)

I'm opposing proposal II primarily because chains are nowadays a fairly rare unit of measurement and it may help some readers in the US, UK and a few other countries to specify decimal miles. If it were common there wouldn't be multiple editors mocking the article for containing an archaic unit of measurement out of context and/or editing the article to replace said unit. Jc86035 (talk) 00:09, 9 July 2018 (UTC)

I've supported a compromise option, which is "It is 10 miles 28 chains (10.35 mi; 16.66 km) measured from London Bridge." This is enough to keep the railway purists happy while avoiding the MOS:JARGON problem.--♦IanMacM♦ ^{(talk to me)} 04:59, 9 July 2018 (UTC)

I'd agree with that, so I'll add it as a formal option. Although it's unclear what the best imperial unit would then be for shorter distances - yards? Andy Dingley (talk) 09:37, 9 July 2018 (UTC)

Fractional miles, maybe. 0.5 miles = 40 chains, and so on. Maybe switch to yards for really short measurements (10 chains or less?), as 3 chains is a lot easier to parse as 66 yards / 60.3 meters than it is as 0.0375 miles. Fish+Karate 10:22, 9 July 2018 (UTC)

Neither of the two options are satisfactory to me. My preference would be to deprecate the use of chains entirely, following the MOS:JARGON principle that states Do not introduce new and specialized words simply to teach them to the reader when more common alternatives will do. Failing a complete deprecation, then at the very least the distance should be given in fractional miles / kilometers for the benefit of the reader, with chains also provided if they must be. We should not expect our readers to be expert in archaic railway measurements, and we should not expect them to have to stop and work out what "1 mile 48 chains" means, when "1.6 miles" is much easier for the average person who thinks in miles to comprehend (I exclude those who prefer their distances in kilometers, as I think everyone is in agreement that a kilometer conversion is appropriate). I would also like to nip in the bud the idea that just because a source gives a distance in miles and chains, it would be original research or some weird violation of WP:V to convert it to fractional miles. Converting a distance for the convenience of the reader is not breaking WP:V. Fish+Karate 09:16, 9 July 2018 (UTC)

If neither option is acceptable, then you are at liberty to oppose both. I honestly won't think any the less of you for doing so. You are as entitled to your opinion as I am to mine, and a civil discussion is taking place. That is how it should be. If the use of decimal miles is not deprecated, then these conversions may be included. You will notice that I have not edited the East Croydon station article to remove the decimal miles conversion. The issue is being discussed and that means the status quo is maintained until a decision is made.

Chains are hardly "new" or "specialised". They have been a standard unit of measurement for hundreds of years in the British Isles. It is accepted that they are not a "common" measurement nowadays outside of the rail network, although I do believe some horse race distances may be quoted in miles, furlongs and chains. The use of furlongs in the horse racing world does not seem to be challenged. For the benefit of those unfamiliar 10 chains = 1 furlong, 8 furlongs = 1 mile (80 chains). Mjroots (talk) 09:29, 9 July 2018 (UTC)

@Mjroots: Please can you insert a third option, proposing IanMacM's format, which I would support? Or would you be happy for me to do so? This RFC would have been much easier for editors to parse by just proposing the possible different formats, and asking people to pick their preferred option(s). Like so:

It is 10 miles 28 chains (10.35 mi; 16.66 km) measured from London Bridge
It is 10 miles 28 chains (16.66 km) measured from London Bridge
It is 10.35 miles (16.66 kilometres; 10 miles 28 chains) measured from London Bridge
It is 10.35 miles (10 mi 28 ch; 16.66 km) measured from London Bridge
It is 10.35 miles (16.66 km) measured from London Bridge

And so on. I would add that nobody has said chains are a new measure. I don't think it's right, though, to say that chains are not a specialised measure, if they are exceedingly uncommon outside of the rail network or horse racing. And in horseracing, race distances in the United Kingdom are measured in miles, furlongs and yards. Fish+Karate 09:39, 9 July 2018 (UTC)

Looks like I was beaten to it. A third proposal has already been inserted. Interesting re horse racing being in miles, furlongs and yards. For some strange reason, tunnels are only measured in yards or miles and yards, not in chains. Again, I'm in favour with going with what the source says, and not introducing other measurements except conversion to metric. Mjroots (talk) 09:50, 9 July 2018 (UTC)

The third proposal is much better, as the two options you proposed are 1) We use chains and km if there is a source or 2) We use chains and km whether there is a source or not. That is either deliberately or inadvertently missing the point of people's objections in the first place, which was "why are we using chains at all". The third proposal still uses chains, so it's a compromise, but at least helps the reader by translating them into fractional miles and km. Fish+Karate 10:17, 9 July 2018 (UTC)

The reason why we use chains at all is because sources use them. So we are adering to a policy (WP:V), which trumps a guideline (WP:MOS). It may have been better for the last sentence of my Proposal II to have been listed separately as Proposal III, but it's too late now to change it. Mjroots (talk) 11:11, 9 July 2018 (UTC)

But the source being in chains doesn't mean we can't convert them instead, or as well. And just a note that while I don't, and I don't think anyone does, really want to go looking to change the RFC now, it's never too late to change, or re-start, a bad RFC. Fish+Karate 12:18, 9 July 2018 (UTC)

WP:UNIT states In non-scientific articles relating to the United Kingdom, the primary units for most quantities are metric or other internationally used units, except that: [copying bullet point one only] UK engineering-related articles, including those on bridges and tunnels, generally use the system of units that the topic was drawn up in (but road distances are given in imperial units, with a metric conversion – see next bullet); If the sources use miles, chains and yards then this MOS guidance specifically allows for their use in articles as long as conversions are added assuming it is accepted that railway lines/station articles etc are engineering topics. Incidentally WP:UNIT also allows for the use of archaic measurements like hands for the height of horses. Nthep (talk) 10:25, 9 July 2018 (UTC)

I'm reluctantly ok with chains providing we also provide a conversion into decimal miles and km. I've been convinced that far. Fish+Karate 12:18, 9 July 2018 (UTC)

Regardless of the antiquity of chains, and their lingering relevance to issues of surveying and real estate, their use in British rail articles is obscure to outsiders, even the American rail-savvy. American employee timetables prefer tenths of a mile, for comparison. Since we can and are even allowed to do simple arithmetic, I don't see verification as an issue here, even if decimal amounts are somewhat less precise, as it's easy enough to verify that they reflect the value in chains. Mangoe (talk) 13:34, 9 July 2018 (UTC)

@Mangoe: - That the chain is obscure presents an opportunity for Wikipedia to educate the reader, by clicking on the link and going to the article. It is part of Wikipedia's remit to educate, is it not? Mjroots (talk) 16:26, 9 July 2018 (UTC)

As I believe was mentioned above, "do not introduce new and specialized words simply to teach them to the reader when more common alternatives will do" already answers that. Mangoe (talk) 17:03, 9 July 2018 (UTC)

@Mangoe: Why do you feel that the chain is a "new word"? --Redrose64 🌹 (talk) 20:12, 10 July 2018 (UTC)

@Redrose64: Pretty sure Mangoe meant (and that the MOS:JARGON guidance means) it was "new to the reader", not a new word as in a word that has only existed for a short time. For instance, until I arrived here, chains were an unknown unit of measurement to me, and thus a "new word". —Joeyconnick (talk) 20:31, 10 July 2018 (UTC)

As far as furlongs in the horse racing articles are concerned, the North American articles all give race lengths in miles and fractions, then miles and furlongs (and meters if they remember, which they frequently don't); the Aussies and Kimis use meters. It's only the British and Irish articles that omit fractional/decimal miles. SO presumably the corresponding project has institutionalized the same exceptionalism. Mangoe (talk) 17:03, 9 July 2018 (UTC)

I'm surprised no-one has mentioned that a chain is the distance between the wickets on a cricket pitch – a sight that will be familiar to many English speakers outside North America. Maproom (talk) 07:35, 13 July 2018 (UTC)

@Maproom: They did. --Redrose64 🌹 (talk) 22:48, 13 July 2018 (UTC)

I'm pro-chain as the main unit of measurement. The UK rail network is mostly measured in miles and chains (hello there LU, HS1 and all you other crazy exceptions), and while we are allowed to convert I think it's sensible to keep the original unit of measurement. I certainly don't think we should go with miles and yards, as that would be giving unjustified accuracy. I don't think however that chains are appropriate for the lead of the article. I've written a number of Good Articles about stations, and experience there has given me the following standard:

The lead should include distance from the local terminus, in integer miles and kilometres (or 1 decimal place if it's a distance under say 3 miles).
The description section should include distance from the local and national termini (plus adjacent stations) in miles and chains with a conversion to 2dp kilometres.
When first introduced, chains should be linked and a note left stating that UK railways are, for historical reasons, measured in miles and chains, and that there are 80 chains for each mile.
Distances should be measured as appropriate - platform lengths are generally in yards (convert to metres), distances if a station is resited would be in chains, and whatever unit the source states should go first.

It's not perfect, and I welcome the clarity this RfC will hopefully bring. -mattbuck (Talk) 17:18, 9 July 2018 (UTC)

Miles and chains is how they are measured. We must retain the unit or we lose accuracy. When we do the convert, there are some interesting points to consider.

Internal consistency: once it is decided to do miles chains =>(miles centimiles, km m) once, all conversions in that article should be of that format. The test should not be whether the each geotag has a source quoting chains, but any.
Centimiles are a made up unit, miles to 2dp are units for expressing a rounded approximate distance.
I believe that in railway uses such as on the HS1, distances are measured in km to 3dp- again km to 2dp would be seen as a rounded approximate distance.ClemRutter (talk) 23:28, 10 July 2018 (UTC)

I'm thinking proposal 1 to show chains if used in cite is decent, except that one probably could always find the track atlas that shows use of chains or one could always google up something that does not use chains so it becomes a useless guide. I'd think it depends more on the context but that too is debatable and it would perhaps confuse folks to sometimes show one way and sometimes show another so for now I'm thinking proposal 3 (just show both) is simpler to use and understand. I'd suggest that Imperial measures and Gunter's chain should also be shown as a background topic, just as one might expect Gauge to address variations in measure. Cheers Markbassett (talk) 03:10, 11 July 2018 (UTC)

@Markbassett: - the intention of Proposal 1 is that if a source uses chains, then the article can. That another source doesn't use chains does not nullify the validity of the source that does use chains. However, should it be the case that no sources can be found that use chains, then they shall not be introduced by way of conversion. Mjroots (talk) 20:13, 11 July 2018 (UTC)

User:Mjroots - Understood. But I'm dubious on how much if any such actually exist. I'm thinking that almost all lines are old enough to be in a track atlas that uses chains and the Engineer's Line Reference refers to it, such as this. Even if a new line is made that was not made to imperial measure, I would not be surprised to find external references provide a translation -- so I'm dubious that 1 has any realistic utility. Cheers Markbassett (talk) 02:56, 12 July 2018 (UTC)

Miles/chain information will always be available because it's what Network Rail uses. You don't need to go searching obscure books to find the data. NR has released its definitive source into the public domain (the Sectional Appendix - check out our wikipage about it). The SA comes in 7 flavours and currently lives at http://archive.nr.co.uk/browseDirectory.aspx?root=&dir=%5csectional%20appendix%5cSectional%20Appendix%20full%20PDF%20copies. I suggest you download them all for safe keeping, since NR may take them down eventually. It's not the end of the world if they do though, as there are railway professionals here who own some printed versions. If you need info, or help deciphering them all you have to do is ask. Dr Sludge (talk) 11:33, 12 July 2018 (UTC)

This book uses miles and chains:

Kelman, Leanne (December 2017) [1987]. Brailsford, Martyn (ed.). Railway Track Diagrams 1: Scotland & Isle of Man (6th ed.). Frome: Trackmaps. ISBN 978-0-9549866-9-8.

It was published seven months ago and depicts the railway network as of November 2017. --Redrose64 🌹 (talk) 11:37, 12 July 2018 (UTC)

And that has nothing whatsoever to do with what a general work like Wiki should show. The question is not what is used within the industry, it's whether it is for obsessives or the general public. The general public neither know, nor ought to know, what a chain is. — Preceding unsigned comment added by 86.180.239.153 (talk) 12:51, 12 July 2018 (UTC)

@86.180.239.153: Could you express support/oppose for the various options given above? Thanks--♦IanMacM♦ ^{(talk to me)} 15:40, 12 July 2018 (UTC)

Speaking as an historian, I do know what a chain, rod, perch, rood, and acre is. I'm also quite happy with pounds, shillings and pence too. gallons, bushels, pecks, firkins, barrels, tuns? Yep, I know about them as well. That a reader doesn't know what a chain is, gives Wikipedia the chance to enlighten that reader, by means of a wikilink and article. Mjroots (talk) 17:01, 12 July 2018 (UTC)

I don't know what modular forms are, and our article about them enlightens me not one bit. But I don't cry out for its removal. --Redrose64 🌹 (talk) 19:53, 12 July 2018 (UTC)

Terrible analogy. The average person can understand an amount given in chains through the application of a bit of 2nd grade arithmetic, but it's going to take a lot of math education to get anywhere with modular forms. Mangoe (talk) 20:12, 12 July 2018 (UTC)

Speaking as another historian, I also know about chains, furlongs, miles and acres. I had to. And I'm here because we use them, with the convert template, in military history articles. The readers are searching for information. So give it to them. Hawkeye7 (discuss) 05:07, 14 July 2018 (UTC)

@Dondervogel 2: None of the proposals suggest to use chains as the primary unit. Two of them mention miles and chains, that is, a measurement such as 10 miles 28 chains. Think of it like feet and inches, or pounds and ounces. For proposals I, II and III, you wrote "A conversion (from mi or km) to chains is acceptable if sourced." - but none of these proposals are to do that conversion but the opposite, indeed, prop I says where chains are given in a quoted source (original italics). --Redrose64 🌹 (talk) 07:29, 13 July 2018 (UTC)

@Redrose64: The unit chain is poorly defined and unfamiliar to most readers. Mixing it with the mile does not change that, but just complicates the conversion to well defined and familiar units. There is always merit in converting to the units of the original source, and even stating the exact value in those units where precision is an issue, but not be at the expense of clarity. Dondervogel 2 (talk) 11:12, 13 July 2018 (UTC)

@Dondervogel 2: The chain is not "poorly defined", it is an exact measurement, equal to 1/80th of a mile (22 yards, 66 feet, or precisely 20.1168m). As for being unfamilar, I'm unfamilar with US gallons, knots and nautical miles, but that shouldn't prevent them being used as units. Optimist on the run (talk) 11:42, 13 July 2018 (UTC)

@Optimist on the run: I don't object to use of unfamiliar units where it is necessary to do so to remove ambiguity, but the mile, yard and foot are all defined by modern standards organizations. The chain is not (I have seen the chain defined variously as 66 feet, 74 feet, 100 feet and 20 metres- why should I trust your assurance that it means 66 feet?), and is therefore both unfamiliar AND ambiguous - a double whammy from which it should not be permitted to recover. Dondervogel 2 (talk) 11:54, 13 July 2018 (UTC)

@Dondervogel 2: It is defined as 66 feet - see chain (unit) - about 20 metres. There was also a Scottish Chain, which was 74 feet, and Americans in certain circumstances used 100 feet (except in Texas where it was 55.5ft). But it is only the English version we're referring to here, and articles can explicitly define that as 1/80 of a mile if needed. In a UK rail context, chains are always 66ft. Unless it's a physical chain anyway.

Converting to the original source unit is backwards - why would anyone ever take a source measured in pounds and ounces, but provide the weight instead in kilograms with a conversion back to pounds? It's wasted effort and introduces inaccuracies. If a source is measured in miles and chains, you state the distance in miles and chains, with appropriate conversion into more standard units like kilometres. -mattbuck (Talk) 18:18, 13 July 2018 (UTC)

@Dondervogel 2: I don't expect you to trust my word (that would be WP:OR). Instead, trust a reliable source - e.g. {{Quail-2}} states "80 chains = 1 mile; 1 chain = 22yards/20.11m Optimist on the run (talk) 19:46, 13 July 2018 (UTC)

I prefer to trust my own reliable source which informs me that a chain is "a unit of distance used or formerly used by surveyors. Although the unit is not often used today, measured distance along a road or railroad is commonly called chainage regardless of the units used. The traditional British surveyor's chain, also called Gunter's chain because it was introduced by the English mathematician Edmund Gunter (1581-1626) in 1620, is 4 rods [1] long: that's equal to exactly 1/80 mile, 1/10 furlong, 22 yards, or 66 feet (20.1168 meters). The traditional length of a cricket pitch is 1 chain. Gunter's chain has the useful property that an acre is exactly 10 square chains. The chain was divided into 100 links. American surveyors sometimes used a longer chain of 100 feet, known as the engineer's chain or Ramsden's chain. (However, Gunter's chain is also used in the U.S.; in fact, it is an important unit in the Public Lands Survey System.) In Texas, the vara chain of 2 varas (55.556 ft) was used in surveying Spanish land grants. In the metric world, surveyors often use a chain of 20 meters (65.617 ft). See also shackle and shot [2] for anchor chain lengths." In other words, it is sometimes 66 feet as you claim, and sometimes not. Dondervogel 2 (talk) 20:23, 13 July 2018 (UTC)

@Dondervogel 2: so you're just ignoring that the source of measurements explicitly defines the measure it uses. -mattbuck (Talk) 21:53, 13 July 2018 (UTC)

When considering the railways of the United Kingdom (which is what we are doing here, this being the talk page of WikiProject UK Railways), the chain is always 66 feet. We shouldn't need to bring in any of these other measures also called "chain", since these railways are not in America, Texas or Spain, nor on board ship or otherwise at sea (I'm surprised that the 10-yard chain used in gridiron football hasn't been mentioned yet). The 20-metre chain apparently used in the metric world need not cause confusion either, since in writing "10 miles 28 chains", we are implicitly not in the metric world. Unless there is some "metric mile" of which I am unaware. --Redrose64 🌹 (talk) 09:38, 14 July 2018 (UTC)

The point you appear to be missing is that the readership of WP articles are not limited to UK railway fans, all of whom are no doubt comfortable with the use and definition of the UK railway chain, which I am happy to believe is 66 feet. The other 99.9 % are not. Dondervogel 2 (talk) 09:55, 14 July 2018 (UTC)

But what about the readership of WP articles on British railways? I suspect it is a different demographic. Hawkeye7 (discuss) 19:38, 15 July 2018 (UTC)

That could be, but I thought the idea was to avoid jargon so that articles can be read by a wide audience. Dondervogel 2 (talk) 19:57, 15 July 2018 (UTC)

@Mjroots: Please clarify proposal I. Currently the first sentence reads The use of chains as a unit of measurement in articles on railway lines and railway stations is appropriate, where chains are given in a quoted source. Unlike proposals II and III, this one does not mention miles and chains - and I think that some people may be reading it as if it were chains alone, that is, it is suggesting that we would write 828 chains (16.66 km). I don't think that was your intention. --Redrose64 🌹 (talk) 07:35, 13 July 2018 (UTC)

I've clarified the proposal to say "chains (or miles and chains)". The latter will generally be the case in sources. Mjroots (talk) 07:46, 13 July 2018 (UTC)

This entire thing is a big WP:CONLEVEL failure. Wikiprojects can't make up their own rules against side-wide consensus. If some support for use of chains emerges from this echo chamber, then make a proposal at WT:MOSNUM to change our handling of the chain unit. MoS often makes topical variants, but they're site-wide MoS consensuses, not specialized-style fallacies picked by a handful of fans of some topic at their wikiproject page. — SMcCandlish ☏ ¢ 😼 03:46, 14 July 2018 (UTC)

That's great—Wikiprojects can be shut down and the Masters of the Encyclopedia at MOS can make all the rules. Let the argument with the greatest number of blue links win! Johnuniq (talk) 03:57, 14 July 2018 (UTC)

That's a really strange thing to say. Please show me at WP:WIKIPROJECT or WP:PROJGUIDE where is suggests that wikiprojects' purpose is to make up rules? Turns out we have a guideline saying the opposite: WP:PROJPAGE, and a policy, WP:CONLEVEL making the point even more broadly (you don't get to make up your own rules because you're a faction or tagteam that doesn't happen to be in a wikiproject). Let's not be silly. WP's rules are written by the people, site-wide, who care to work on them, at site-wide policies and guidelines, centralized places where they can be vetted. They are not written in wikiproject echo chambers full of people already like-minded about "their" topic and how magically special it is. — SMcCandlish ☏ ¢ 😼 04:59, 14 July 2018 (UTC)

@SMcCandlish: Apart from railways in the UK, and Cricket, where else might you consistently expect to find chains in use nowadays? I put it to you that this is the one area where chains are regularly used, and that means that this is not a CONLEVEL failure. This issue needs thrashing out, and out it is being thrashed. Mjroots (talk) 04:34, 14 July 2018 (UTC)

Chains not being in broad use, just surviving in two slivers of human activity, is a reason to not use them on WP, which has precious few readers who are cricket-playing British railroad employees. — SMcCandlish ☏ ¢ 😼 04:59, 14 July 2018 (UTC)

You are wrong. It is a reason not to impose their use across the whole of Wikipeda. We are not seeking to do that. For example, we're not introducing the use of chains into an American railroad articles, because no sources use chains when discussing that subject. What we are doing is following what sources say. Therefore if a source uses chains, we use chains with the necessary conversion to metric. Mjroots (talk) 05:37, 14 July 2018 (UTC)

I'm not at all wrong; you're simply engaging in a transparent straw man and responding to argument that doesn't exist. Neither I nor anyone else I'm aware of has ever suggested this had anything to do with or any effect on American railroad articles. Let me spell it out really clearly: it is not helpful to readers for WP to measure things in chains, even in a subject where some specialist sources have done so, because our readers are not specialists. The rest of this I already addressed below. — SMcCandlish ☏ ¢ 😼 06:04, 14 July 2018 (UTC)

(edit conflict) @SMcCandlish: Would it surprise you to know that basically no one is aware of all the intricacies of policy? I knew it wasn't in the right place (I think it's only because you've said this before about another discussion I've participated in) but it wouldn't have been worth the fuss to move it somewhere else.

Regarding your point that "no one knows what it is", that probably isn't a valid criterion for not using chains. Probably not a lot of people remember how long a light year really is or know how much a candlepower… candles, but they're somewhat common in topic-specific articles where they are the base unit of a conversion into a more common unit. Obscure units are also similarly common in articles about old things (and the railway station which made this RfC happen arguably qualifies as one such old thing, as it was opened 177 years ago). Eratosthenes contains two similar units called stadia, and Great Pyramid of Giza mentions the Egyptian cubit, each of which is converted into the standard SI length units. Even for modern units, the original unit is usually given preference: New Horizons has kilograms and metres first. Jc86035 (talk) 04:44, 14 July 2018 (UTC)

Mentioning the unit at an especially relevant article is nothing like imposing it on everyone across a whole category of them. Railways are not ancient subjects. Some units are international standards (ISO, etc.) and used thereby per MOS:NUM. Chain's isn't. Some units don't translate well to conversion (the average person isn't apt to conceive a clear difference between 12.7 trillion miles and 83.4 trillion miles). Chains are not such a case; everyone understands at least one of miles/feet and kilometers/meters, and what such a distance means in a human-experience sense, at the scale of a railway. — SMcCandlish ☏ ¢ 😼 04:59, 14 July 2018 (UTC)

@SMcCandlish: - where is this proposing to impose the use of chains on articles? The proposal is specifically against doing that. Chains are not going to be inserted into articles such as High Speed 1, High Speed 2 etc because no sources use them. What we are seeking to do is to back the use of chains where sources use them per WP:V, which is a policy. MOSUNIT, which is a guideline already backs their use in British engineering articles, although it doesn't mention chains by name. SFF, which you referred to earlier, is only an essay and as has been shown above, has little relevance here. Mjroots (talk) 05:06, 14 July 2018 (UTC)

Yet you would insert them anywhere you can find sources who did use chains. This is a misunderstanding of what "follow the sources" means. We have no need to account for chains except as they appear in directly quoted material, or except in a special context, e.g. "[Topic] was original specified in chains, at [amount] chains ([conversion here])", or whatever. Once in the article. But WP should not itself be measuring things in chains, any more than we'd switch to cubits and fractional cubits at misc. articles about Ancient Egypt. Speaking of which, "Great Pyramid of Giza uses cubits a bunch of times" is not an excuse or a reason; it's an example of an article that needs cleanup. It was delisted as a GA almost ten years ago for real reasons and it's not gotten any closer to being a GA again. — SMcCandlish ☏ ¢ 😼 06:01, 14 July 2018 (UTC)

@SMcCandlish: The railways are signposted in chains, and distances along them are apparently still measured in chains because they originally were measured in such a way (the source used at East Croydon station is in chains, and the discussion at Talk:East Croydon station mentions recent publications which still use chains). Furthermore, WP:MOSCONVERSIONS says Where an imperial unit is not part of the US customary system, or vice versa […] a double conversion may be appropriate. Jc86035 (talk) 05:14, 14 July 2018 (UTC)

Been over this already, too, here. This is the classic WP:SSF: "WP must write like specialized sources do just because specialized sources do it." Nope. — SMcCandlish ☏ ¢ 😼 06:06, 14 July 2018 (UTC)

Since when does an essay trump policy? And don't reply with WP:IAR, because that won't wash. Mjroots (talk) 06:29, 14 July 2018 (UTC)

What on earth are you talking about? No IAR rationale has been presented. Ignoring a guideline you don't like in this case is not going to objectively improve the encyclopedia, just pander to a small sliver of the railfan scene. Second, when people refer to essays, it means "see the rationales presented in this page. They have been saved here because it's tedious to re-state them over and over again". Referring to one is a time-saver for everyone. It doesn't not amount to any sort of argument that the essay "trumps policy"; everyone who edits Wikipedia for more than about a month understands this. I surmise, then, that you also understand this, and are thus trotting out this even sillier straw man than the one above in an effort to WP:WIN. It's poor debate, sorry. No one buys that kind of nonsense. — SMcCandlish ☏ ¢ 😼 06:41, 14 July 2018 (UTC)

MOS:UNIT already has a usecase for chains (or any other uncommon unit of measurement) so if anything this RFC (which I believe was completely unnecessary) is about reaffirming current policy/practice, not changing it. Nthep (talk) 12:36, 14 July 2018 (UTC)

Correct. Mjroots (talk) 13:24, 14 July 2018 (UTC)

MOS:UNIT is explicit that in UK engineering-related articles - which I would imagine includes all the articles we're discussing - we use the units of design.

Where the proposal in proposal (I) deviates from MOS:UNIT is where it proposes to use the units of the source and not the units of the design. This system - source-based units - was proposed (generally by the same person) at WT:MOSNUM over and over and over and over and over and over and over and over and over and over and over and over and over and over and over and over and over and over and over and over and over and over and over for many years and rejected every time in large part because of the disruption and scope for POV pushing that it caused. In other topics we have seen major problems with editors picking less-reliable (and sometimes less accurate) sources solely because they preferred the system of measurement.

There are lots of sources out there and lots of them use lots of different units. Saying the unit of design instead is a sensible approach, most likely has a very similar result to a good-faith implementation of the rule in proposal (I), is clearly and umambiguously in line with MOS:UNIT, and reduces the scope for disruption. Kahastok talk 13:02, 14 July 2018 (UTC)

It is not the case that reliable secondary sources are merely following the primary sources and using the units of design in their works. It should be taken as read that WP:RS applies to "sources" where mentioned in any of the proposals I to V inclusive. Mjroots (talk) 13:24, 14 July 2018 (UTC)

Let me put this another way. If I can find a reliable source that gives all these distances in kilometres, does that mean that we can get rid of chains entirely from Wikipedia and make miles secondary in all of these contexts?

While it may not be intended, by the current wording of proposal (I) the answer is "yes". I have seen this mistake made far too many times to believe otherwise. I have a problem with that because I do not believe it reflects actual modern usage or the spirit or letter of WP:MOSNUM. Kahastok talk 13:27, 14 July 2018 (UTC)

@Kahastok: - the answer would be a firm "no", because your finding a source in km would not negate the other sources in miles an chains. The intent of Proposal I is to prevent the introduction of chains where sources don't use them, for example in articles on European railways, but to allow the established practice of using chains in UK railway articles where it is appropriate to do so and can be reliably sourced. Mjroots (talk) 14:03, 14 July 2018 (UTC)

That may be the intention but it's not what the proposal actually says. And you seem to be saying now that if I can find a source that describes a continental European railway that uses miles (and chains) first, we have to use miles and chains first. Which is no less problematic. Kahastok talk 15:58, 14 July 2018 (UTC)

@Kahastok: Hmm, let's see. Areas of Europe formerly under British control, where chains may possibly be found in sources? Gibraltar, Heligoland? No railways. Malta? possibly. Cyprus? Possibly, but the Middleton Press book on the subject only goes to the nearest ¼ mile. Republic of Ireland? Probably. But yes, if a source can be found that uses chains, the proposal supports their use in an article. Mjroots (talk) 16:16, 14 July 2018 (UTC)

Which would be even harder to justify per WP:MOSNUM. You can get away with chains in the UK, on articles on engineering projects drawn up in chains. But the fact remains, that's not what the proposal actually says.

The proposal says that you do what the quoted source says. If someone comes up with a source that puts the distance from London Bridge to East Croydon in kilometres and replaces the one in the article, then under this proposal they have licence to get rid of the chains and put kilometres first. And if they do the same with every article on a station or railway line on the network, then there will be no more distances measured in miles first, let alone miles and chains first.

And before you dismiss this as unlikely or OTT, note that we have editors active on Wikipedia who have made changes of that kind of scale on precisely this logic. Kahastok talk 16:48, 14 July 2018 (UTC)

Jc86035 re your comment in proposal V, it is obvious that you, and others, have not read both linked discussions at the top of this RFC. It is possible to link chains only in a conversion - 232 miles 50 chains (232.63 miles, 374.37 km). Mjroots (talk) 07:26, 14 July 2018 (UTC)

Already corrected in that comment. Jc86035 (talk) 15:48, 14 July 2018 (UTC)

I browsed through this discussion. The sources that use chains appear to be oriented toward those who design, operate, or maintain UK railways (and also of interest to enthusiasts, in the same way that certain geocachers hunt benchmarks for recreation rather than utility). I submit that Wikipedia articles about railways should be oriented to rail travelers, and thus should not use chains. Jc3s5h (talk) 13:55, 14 July 2018 (UTC)

{[replyto|Jc3s5h}} Wikipedia is not a travel guide. Your argument seems to be "I don't like chains, so they shouldn't be used". As you repeatedly pointed out, "Wikipedia should not be used by rail workers to build or maintain rail lines", but no one is claiming that it is used. Hell, most official diagrams are not used by rail maintainers - in fact they specifically say "not to be used for safety-critical decisions". What is being argued is that the official unit of measurement for the railway is miles and chains - that's what appears in the Sectional Appendix - and that that therefore articles should state those official measurements, as well as appropriate conversions. No one is saying we should use only chainage, they're just saying that, since the measurements are in chains, we shouldn't hide that behind a conversion which might imply greater accuracy than is given. -mattbuck (Talk) 14:50, 14 July 2018 (UTC)

Repinging Jc3s5h, as unfortunately Mattbuck's ping didn't work—and we certainly would not want to deprive Jc3s5h of the opportunity of meeting a cogent argument when one is presented :) —SerialNumber54129^{paranoia /cheap sh*t room} 14:56, 14 July 2018 (UTC)

Mattbuck, do any of these official diagrams give any statement of who their intended audience is? Does the nature of the website, or portion of a website, indicate who the intended audience is for these diagrams? Is their any indication that the intended audience of these documents is anything like the Wikipedia audience?

As for Wikipedia not being a travel guide, it would have been better if I had indicated our target audience should be travelers and other outside observers, as opposed to railroad workers. Certainly we wouldn't exclude historical information about a railroad merely because the history wouldn't help a traveler get from London Paddington station to Berwick-upon-Tweed railway station (which I had cause to look into a few years ago). Jc3s5h (talk) 15:19, 14 July 2018 (UTC)

I think this RFC misses the point. The problem isn't chains themselves, so long as appropriate conversions are provided. But I do have a problem with the level of precision they provide. That level of detail isn't desirable for the lead section, which is simply an overview. If a station is recorded as 79 miles 2 chains from the London terminus, that should just be rounded to 79 miles for the lead.

Moreover, the use of chains appears to result in excessive precision. A station is not a point object, but a linear feature that covers a significant distance. A platform is bound to be several chains long. That means saying a station is an exact number of chains from another location is far too precise. If the exact miles/chain figure is quoted, it should state what the measurement actually relates to. So not "the station is X from London Bridge", but "the station's ticket office is X from London Bridge". The use of chains makes much more sense in other contexts, like describing the point of an derailment or the location of a (short) bridge.--Nilf anion (talk) 22:23, 14 July 2018 (UTC)

What about the use of chains as the customary unit for radius of railway curves, which is clearly not purely topographical information, but a matter of railway engineering? "The town of Morpeth in Northumberland, England has what is reputed to be the most severe curve (17 chains or 340 metres radius) of any main railway line in Britain." says Rail accidents at Morpeth (the number given repeatedly by HMRI was 14 chains, but that's another matter). The WP article on Minimum railway curve radius has no mention of chains (perhaps because there is no discussion of GB mainline railway practice) but instead uses feet or metres (so for example the radius of the curve at Beddgelert on the original Welsh Highland Railway (laid out c 1922) is said to have been 40 m (131 ft) Rjccumbria (talk) 20:00, 15 July 2018 (UTC)

@Rjccumbria: For the curve at Morpeth, I've checked the radius using OpenStreetMap data (with the JOSM editor) and the curve does appear to have a minimum radius of about 340 metres (17 ch) and a maximum radius of about 390 metres (19 ch). Jc86035 (talk) 17:50, 16 July 2018 (UTC)

@Jc86035: No offence meant, but I would tend to take the inspectorate's word for it. Were your measurements on the Down line (where the 1969 accident report says 17 ch. nominal but notes a short section at 15.4 ch at the point of derailment ?) or the Up line, where the 1984 accident report gives significantly lower radii? I will put fuller details in the talk for Rail accidents at Morpeth, which is a more appropriate place to pursue the matter further. Rjccumbria (talk) 19:07, 16 July 2018 (UTC)

@Rjccumbria: The tracks are less than 10 m (0.50 ch) apart on the map, so I don't think there would be a significant difference. The curve being slightly elliptical is probably why the accident reports have varying values, although since the tracks on the OSM map were drawn directly from satellite imagery they should be accurate with an error of about ±6 metres at maximum. Jc86035 (talk) 19:17, 16 July 2018 (UTC)

Re-measuring by drawing a line from just outside the Sainsbury's on the map (which is roughly perpendicular to both end points of the arc) over the whole curve, I get 340–375 metres, with the radius increasing smoothly towards the station. This is all original research, obviously, but I think it's more likely that if the original reports had such varying values that they had systematic measurement errors of some kind. Jc86035 (talk) 19:25, 16 July 2018 (UTC)

@Wikid77: Re this edit: what are "U.S. cars" and how do they relate to UK Railways? --Redrose64 🌹 (talk) 19:20, 1 August 2018 (UTC)

Compromise proposal for adoption (V)

What we really should be doing is 1) giving this in metric, 2) converted to US customary so everyone understands, then 3) in the specific cases where chains were (maybe sometimes still are) also used in some sources, then also providing the chains measurement to keep the specialists happy, and to account for some signage still using chains.

e.g. "It is 16.66 km (10.35 mi, or 10 mi 28 chains) measured from London Bridge."

Support

Easy-peasy. The problem with the earlier proposals is they all either put chains first, or they just ban chains. However, this discussion has run for so long without this obvious compromise being present that the whole thing is a false-dichotomizing [wait for it] train wreck. — SMcCandlish ☏ ¢ 😼 06:13, 14 July 2018 (UTC)
Dondervogel 2 (talk) 06:44, 14 July 2018 (UTC) This one I can support. SI units provide an international unifying language. An alternative I would also feel comfortable with in this context is starting with miles and yards, and converting these to decimal miles, decimal kilometres and (optionally) to miles and chains.
Weak support, with WP:MOSCONVERSIONS text something like "If an uncommon, non-metric unit is used in a source, use a conversion to a more common unit before the unit used in the source. Use |order=out in {{Convert}}: {{Convert|10|mi|28|ch|mi michlk km}} 10.35 mi (10 mi 28 ch; 16.7 km)". @SMcCandlish: Unfortunately it looks like it's currently impossible to ~~only link chains or~~ have two decimal places for miles and kilometres while not adding trailing zeroes to the chains measurement, so this might be problematic. Jc86035 (talk) 07:18, 14 July 2018 (UTC)
Not everything has to be in a template. If we demand that it be in a template, then we need to fix the template code to handle it. — SMcCandlish ☏ ¢ 😼 07:24, 14 July 2018 (UTC)
Some such compromise seems like the right direction here. Most of the world won't know what to make of chains, but decimal miles and km are widely understood, so making the chains primary doesn't seem like a good thing for anyone but British rail fans. And there's no reason to outlaw chains altogether, either. The original proposal (The use of chains as a unit of measurement in articles on railway lines and railway stations is appropriate, where chains (or miles and chains) are given in a quoted source.) has a couple of issues: (1) "in articles on railway lines and railway stations" is a weird carve-out; let's not make style rules that depend on which type of article something is in; and (2) "is appropriate, where chains are given in a quoted source" has unclear intent -- does this mean make chains the primary unit if any source does? This seems to go to far. That's why a compromise is needed, where the source with chains can be quoted, but not drive the whole article's style. Dicklyon (talk) 17:18, 14 July 2018 (UTC)
Not quite sure how this differs significantly from option 3 but yeah, anything that avoids just miles + chains / km is fine. Fish+Karate 20:41, 14 July 2018 (UTC)
Support—Tony (talk) 01:14, 15 July 2018 (UTC)
Support as a reasonable compromise: metric first, then USCM, then perches and other archaic fish. --John Maynard Friedman (talk) 14:16, 17 July 2018 (UTC)

Oppose

Mjroots (talk) 06:24, 14 July 2018 (UTC) With the exception on new lines such as HS1, the British railway network was built in miles, chains and yards. Why do we need to convert British units to US units? Mjroots (talk) 06:24, 14 July 2018 (UTC)
See MOS:CONVERT. This doesn't appear to be a valid oppose rationale (i.e. a desire to avoid conversion to US units in this case when it is our standard to do so, site-wide). — SMcCandlish ☏ ¢ 😼 07:30, 14 July 2018 (UTC)
SMcCandlish - this objection to editors comments because WP:IDONTLIKEIT in heading into the realms of WP:DISRUPT. Please stop. There is a discussion section which really should be used for commenting. Mjroots (talk) 07:56, 14 July 2018 (UTC)
There's nothing disruptive in calmly pointing out when a rationale conflicts with WP:P&G. — SMcCandlish ☏ ¢ 😼 00:35, 15 July 2018 (UTC)
Don't see what we get from mandating three measurements every time apart from cluttering the articles. The old measurements (miles and chains, per the source) and the new (kilometres) should be enough. Hawkeye7 (discuss) 07:26, 14 July 2018 (UTC)
Also fails MOS:CONVERT since chains isn't a US measure. — SMcCandlish ☏ ¢ 😼 07:31, 14 July 2018 (UTC)
But chains are listed as United States customary units. And MOS:CONVERT says: "Where an imperial unit is not part of the US customary system, or vice versa – and in particular, where those systems give a single term different definitions – a double conversion may be appropriate" Hawkeye7 (discuss) 21:37, 14 July 2018 (UTC)
Read Chain (unit). It had some regional usage in the US (e.g. in historical Texas) but with a different length value. — SMcCandlish ☏ ¢ 😼 00:34, 15 July 2018 (UTC)
If you have a source which measures in weebles, but people are more familiar with grozzits, you put the weebles first because that's the sourced measurement. The conversion is there to help people, if you put it first then it appears to be the actual measurement (and I think we all agree that you wouldn't randomly convert rail measurements to chains if they weren't already the primary unit). You want to put in km and decimal miles, sure, whatever, but it needs to be clear that they are conversions, not the original measurement. Furthermore, regardless of chains, distances in the UK are not measured in kilometres. The UK is decimal for a lot of things, but miles are still the primary units for long distance. Therefore any measurement of distance in the UK would have primary measurement in miles, which is what the MOS says to do. -mattbuck (Talk) 08:23, 14 July 2018 (UTC)
The statement "you put the weebles first because that's the sourced measurement" is not consistent with the statement "any measurement of distance in the UK would have primary measurement in miles". Because it allows a POV pusher to replace every source with the one they've found that happens to do everything in kilometres and so flips all the distances to kilometres. It's happened before. Measurements of British football players in the Premier League who were active in 2011 are given in metric first. Measurements of other British football players are generally given in imperial first (per MOS:UNIT). Why? Because a single editor spent the most of 2011 flipping the units in all the articles claiming consistency with a source chosen because it was metric. Kahastok talk 13:14, 14 July 2018 (UTC)
@Kahastok: Sorry, you're right on that, my point was that UK distances are measured in miles, not km, and thus the sources will generally say miles, and so putting km first seems very unlikely (except on HS1, LU, etc).

Of course, that being said, UK roads are a bit weird, because while the signs are in miles they are maintained in kilometres (the small blue signs with yellow numbers you see on the motorway are engineering markers in km). What I think we're coming towards here is that there is no single system which can handle all cases. I personally believe that because chains are still used in the UK rail industry for distance measurement, and that is how track diagrams are measured, that's what we should use. But I'd be very surprised if Network Rail didn't do their engineering in metres, because imperial measures are just plain stupid. -mattbuck (Talk) 13:43, 14 July 2018 (UTC)
SovalValtos (talk) 08:44, 14 July 2018 (UTC)
As Mattbuck. Andy Dingley (talk) 09:11, 14 July 2018 (UTC)
First, the reliable sources are not metric. Second. U.S. customary is irrelevant since the United Kingdom is not in the U.S. Third, we convert from the sourced value to another value in one operation - we don't use the sourced value, convert that (offline presumably) to another unit and then back-convert to the first units. --Redrose64 🌹 (talk) 09:42, 14 July 2018 (UTC)
Oppose where it goes against the MOS rule that engineering-related article should use the units of design. Kahastok talk 11:44, 14 July 2018 (UTC)
Dr Sludge (talk) 12:58, 14 July 2018 (UTC)
Mark999 (talk) Mark999 (talk) 14:57, 14 July 2018 (UTC)
TBM10 (talk) 21:28, 14 July 2018 (UTC)
Oppose. Once again SMcCandlish is trying to mess things up. Useddenim (talk) 21:13, 14 July 2018 (UTC)
Nonsense. SMcCandlish made a constructive proposal, intended to help 99.9 % of WP readers and you respond with WP:IDLI Dondervogel 2 (talk) 21:45, 14 July 2018 (UTC)
I'm used to it. — SMcCandlish ☏ ¢ 😼 00:34, 15 July 2018 (UTC)
Optimist on the run (talk) 22:07, 14 July 2018 (UTC)
per Redrose64. Thryduulf (talk) 09:05, 15 July 2018 (UTC)
My experience of reliable sources for citations of British railways invariably default to either miles and quarters or miles and chains. Geof Sheppard (talk) 12:42, 16 July 2018 (UTC)
The way I see it the important point of this proposal is not which familiar unit to start with (km or mi both work), but NOT to start with an unfamiliar one. Dondervogel 2 (talk) 14:01, 16 July 2018 (UTC)
Exactly. It's "reader-hateful" to do that. — SMcCandlish ☏ ¢ 😼 23:09, 16 July 2018 (UTC)
Certes (talk) 21:29, 28 July 2018 (UTC)

Compromise proposal for adoption (VI)

It doesn't matter whether we start with km or mi. Either will do, so 1) give in km or mi as appropriate, 2) convert to mi or km so everyone understands, then 3) in the specific cases where chains were (maybe sometimes still are) also used in some sources, then also provide the value in miles and chains.

e.g. "It is 16.66 km (10.35 mi, or 10 mi 28 chains) measured from London Bridge."

or "It is 10.35 mi (16.66 km, or 10 mi 28 chains) measured from London Bridge."

~~or "It is 10 mi 616 yd (10.35 mi, 16.66 km, or 10 mi 28 chains) measured from London Bridge."~~

Support

Dondervogel 2 (talk) 14:20, 16 July 2018 (UTC) same reasoning as for proposal V, which I also support.
Sure, but this is just proposal V in different wording. I notice that yard was struck above, and support its striking; it's not useful in the context. — SMcCandlish ☏ ¢ 😼 01:58, 17 July 2018 (UTC)
Yes, but @Geof Sheppard: appeared to !vote against V on the grounds that it was converting from km to mi, so I'm hoping he will approve of VI. Dondervogel 2 (talk) 12:01, 17 July 2018 (UTC)
--John Maynard Friedman (talk) 14:18, 17 July 2018 (UTC)

Oppose

Optimist on the run (talk) 14:35, 16 July 2018 (UTC)
Mark999 (talk) Mark999 (talk) 15:53, 16 July 2018 (UTC)
Mjroots (talk) 16:58, 16 July 2018 (UTC)
Jc86035 (talk) 17:52, 16 July 2018 (UTC)
-mattbuck (Talk) 21:36, 16 July 2018 (UTC)
--Redrose64 🌹 (talk) 22:00, 16 July 2018 (UTC)
Dr Sludge (talk) 11:23, 17 July 2018 (UTC)
Andy Dingley (talk) 13:57, 17 July 2018 (UTC)
Certes (talk) 21:30, 28 July 2018 (UTC)

Discussion of proposals V and VI

Familiarity

It really doesn't matter whether the primary unit is metric or customary. The point is to start with a familiar unit without deprecating a conversion to chains, where appropriate to do so. Dondervogel 2 (talk) 14:10, 16 July 2018 (UTC)

No one else has advocated a conversion to chains. Andy Dingley (talk) 15:36, 16 July 2018 (UTC)

In general the first used unit should be the source's unit with conversions in parenthesis to both aid the reader and alert them to the fact that this is an editorial change. For instance: the school's pool was 22 yards (20 m) whereas at the public one was 25 metres (27 yd). To express the school's pool in metres might not be accurate, and there is no way anyone builds to 27 m! Martin of Sheffield (talk) 16:25, 16 July 2018 (UTC)

Per Martin, you give the measurement used in the source, then convert, unless there are very strong reasons not to do so. That some people don't understand chains is not a very strong reason. Mjroots (talk) 17:01, 16 July 2018 (UTC)

Martin of Sheffield is incorrect. The order of units should be consistent within any Wikipedia article. Since various sources will use various units, the unit used by the source will not always agree with the source used first in a Wikipedia article. Jc3s5h (talk) 20:24, 16 July 2018 (UTC)

@Jc3s5h: The order of presentation need not have the sourced unit first. For example, {{convert|10|mi|28|chain|km|lk=out|order=flip}} produces 16.7 kilometres (10 mi 28 chains). --Redrose64 🌹 (talk) 23:06, 16 July 2018 (UTC)

In some articles, the reviewers have asked for this to be used to make the order consistent throughout. Hawkeye7 (discuss) 02:32, 17 July 2018 (UTC)

I'm sorry Jc3s5h but that reads that "the distance is 16.7 km, oh and by the way, for reader convenience it is 10 miles 28 chains". Consider a non-numerical example which ought to clarify things: if quoting from Sir Gawayn and þe Grene Knyȝt one would write "and oueral enker grene (And over all bright green)" rather than "And over all bright green (and oueral enker grene"). To do otherwise might be considered WP:SYNTH. Martin of Sheffield (talk) 07:54, 17 July 2018 (UTC)

It depends on what kind of article one is writing. If one were writing in a journal that analyzes Old English literature, one would probably put the Old English first and the modern translation in parenthesis. If one were writing an encyclopedia article about Sir Gawain and the Green Knight one would probably omit the Old English altogether. In Wikipedia WP:MOSNUM indicates in the following bullet point that an article has one primary unit of measure for each type of quantity, which is given first, followed by a conversion (if necessary):

Where the article's primary units differ from the units given in the source, the {{convert}} template's |order=flip flag can be used; this causes the original unit to be shown as secondary in the article, and the converted unit to be shown as primary: {{convert|200|mi|km|order=flip}} → The two cities are 320 kilometres (200 mi) apart.

Jc3s5h (talk) 11:44, 17 July 2018 (UTC)

The three preceding bullet points appear to prefer source (conversion) and the one you quote says "the ... flag can be used", not "the flag ... must be used". Incidentaly the quoted middle English (not old English, that was half a millennium earlier) came straight from the Wiki page. Martin of Sheffield (talk) 12:55, 17 July 2018 (UTC)

What measures should be used in the UK is a matter of law. See http://www.legislation.gov.uk/uksi/1995/1804/made.

There are exceptions from metric for various items (beer, speed limits) but railway station locations aren't there.

Distances are explicitly covered: "the use of the mile, yard, foot or inch for road traffic signs, distance and speed measurement;"

Chains do not make it. There seems absolutely no rational reason to diverge from what parliament and common practice say. — Preceding unsigned comment added by 86.181.162.134 (talk) 17:01, 16 July 2018 (UTC) Sock of blocked User:86.183.81.7

See exception 3 of above act. Incidentally that act does give another source for the precise definition of a chain if there are still any doubters. Optimist on the run (talk) 19:41, 16 July 2018 (UTC)

<s<(1) It's an SI (secondary legislation not an Act (primary). (2) section 3 refers to 'an existing provision [that] ... has legal effect' If you rely on that you have to cite which provision you're talking about. Hand waving doesn't count. — Preceding unsigned comment added by 86.181.162.134 (talk) 06:44, 17 July 2018 (UTC)Sock of blocked User:86.183.81.7

To quote from that document: "The Directives provide for the phasing out of the use, for economic, public health, public safety or administrative purposes". To assert that "what measures should be used in the UK is a matter of law" is a simplification, particularly if attempting to apply commercial directives to non-commercial purposes. Martin of Sheffield (talk) 20:21, 16 July 2018 (UTC)

Of course it's a simplification - if you in normal conversation convert everything to cubits nobody will lock you up. But for Wikipedia to knowingly set a policy that goes against it would be remarkably stupid. — Preceding unsigned comment added by 86.181.162.134 (talk) 06:44, 17 July 2018 (UTC)Sock of blocked User:86.183.81.7

Agreed with the first sentence, but that's not how IP 86.181.162.134 was presenting it in his bland assertions. Wiki is free to use any units it wants, EU regulation is irrelevant. Martin of Sheffield (talk) 07:54, 17 July 2018 (UTC)

~~You agree that for Wikipedia to knowingly set a policy that goes against it would be remarkably stupid? — Preceding unsigned comment added by 86.181.162.134 (talk) 10:16, 17 July 2018 (UTC)~~ Sock of blocked User:86.183.81.7

How can the law of the country be irrelevant to a decision? The argument is what units to show, and the law tells use what is standard. What good reason is there to ignore it? — Preceding unsigned comment added by 86.181.162.134 (talk) 10:56, 17 July 2018 (UTC) Sock of blocked User:86.183.81.7

"Standard" is meaningless in this context. A popular saying in engineering is "The nice thing about standards is there are so many to choose from." Legislation is not a reliable indicator of what people reading an encyclopedia article will understand. As an example, reading US legislation would indicate the US mostly uses SI. But no physician has ever measured my blood pressure in kilopascals. Jc3s5h (talk) 11:50, 17 July 2018 (UTC)

The debate is about establishing a standard practice, and to back your side you argue that standards are pointless? Dear god ... — Preceding unsigned comment added by 86.181.162.134 (talk) 12:43, 17 July 2018 (UTC) Sock of blocked User:86.183.81.7

Please be aware that user Jc3s5h has moved this debate to the MOS talk page. Martin of Sheffield (talk) 14:08, 17 July 2018 (UTC)

Censoring rational and well stated comments in a discusion of this kind does not quite seem the thing to do. — Preceding unsigned comment added by 86.183.80.97 (talk) 16:07, 17 July 2018 (UTC)

Oppose per my previous comment on V. The sourced unit should be first, conversions afterwards. In particular we should not include yards unless they are explicitly used because that then implies we are measuring to the nearest yard, and we're not, we're measuring to the nearest 22. -mattbuck (Talk) 21:38, 16 July 2018 (UTC)

The yards are not important and I have struck out that part of the proposal. The order is important, and as pointed out by Jc3s5h, needs to be consistent throughout an article, which is incompatible with following sources in the way you suggest. Dondervogel 2 (talk) 22:18, 16 July 2018 (UTC)

Dondervogel 2, thanks for getting rid of the yards. However I would argue that this is an exception to the ordering consistency, simply because while you can get colloquial sources stating that a bridge is about 7ft or 2m high, or that it's 300yds to the shops or 300m; we are dealing with the official measurement of the railway. It's not "someone put a ruler on Google Maps and rounded it off". -mattbuck (Talk) 16:12, 17 July 2018 (UTC)

I was looking for current use and found this in a 2013 Network Rail document.

Worksite marker boards are to be placed 100m from each end of the worksite. They are to be positioned in the 4 foot with the red lights facing away from the worksite. The Engineering Supervisor may give you a mileage/chainage to place the marker boards e.g. 35 miles 20 chains.
Yards – Chains – Miles

22 yards = 1 chain

20 chains =¼ mile

40 chains = ½ mile

60 chains = ¾ mile

80 chains = 1 mile

Where there is less than 100m between the worksite and the detonators then the marker boards must be placed at those detonators.
Multiple worksites must have a minimum of 100 between the marker boards and which means there must be a minimum of 300 yards between worksites

Don't you just love the mix of units? It looks like the author couldn't decide if that last 100 should be yards or metres.Cavrdg (talk) 06:32, 19 July 2018 (UTC)

Sources in kilometres or decimal miles

@Dondervogel 2: There are more than 2,500 railway stations in Great Britain that are presently open and which are part of the National Rail network. Assuming that you have one or more reliable sources that gives the distances of stations in either kilometres or decimal miles (to a precision better than a quarter mile), roughly how many of those 2,500 are listed in such sources? If you don't want to count them, just list the sources here and I'll investigate. --Redrose64 🌹 (talk) 22:20, 16 July 2018 (UTC)

I don't know how to interpret "distances of stations", but of course I can find sources that mention distances to and from stations in decimal miles. The distance from Clapham Junction to Victoria is 3.1 miles by car, 3.0 miles by foot and 4.1 miles by bicycle. And there's no reason to limit ourselves to distance between stations. It's also legitimate to think of the distance between a station and nearby landmarks (eg 0.2 mi from Hampton Court station to Hampton Court Palace). The list is almost endless. Dondervogel 2 (talk) 08:28, 17 July 2018 (UTC)

Dondervogel 2 Google is not a reliable source, and none of those are rail distances. No one, anywhere, is arguing that we ought to say that the distance from the station to the town centre is 40 chains rather than 0.5 miles, they are saying that the distance between two points on the railway, which is measured in chains, is Y chains (Z km). If I have five oranges, and then later I have two oranges, the difference may be 420cm² of peel,^[1] but you damn well measure the difference in oranges. If I ask for the difference between pi radians and 2pi radians, the difference may be 180 degrees, but it doesn't make sense to measure it that way. -mattbuck (Talk) 16:06, 17 July 2018 (UTC)

I can find sources other than Google if you like, but doing so would miss the point by 80 chains. We seem to agree that the distance to Hampton Court Palace should be expressed in miles (or yards). The way I see it, consistency then demands that all distances in Hampton Court railway station need to be in miles (or yards), not chains, and THIS is where we differ. It is also where the RfC fails, because the only option that ensures consistency of units also bans the use of the chain. Dondervogel 2 (talk) 16:51, 17 July 2018 (UTC)

But our article on Hampton Court railway station does not give a distance to Hampton Court Palace, and only mentions it once - in the second paragraph (beginning "Across the river the station serves ..."). The only distances given in that article are measured along the railway itself, such as the distance from Waterloo - which I can source very easily. --Redrose64 🌹 (talk) 21:12, 17 July 2018 (UTC)

That was a hypothetical example. If you want a real one try Victoria Station, which apparently has "2.25 miles (3.62 km) of platform". Consistency would require that distance to be written as "2 miles 20 chains (2.25 mi, or 3.62 km)". Dondervogel 2 (talk) 21:33, 17 July 2018 (UTC)

[Miles and] chains are only used for distances along a railway. Industry practice is to use metres for train and platform lengths. One chain is roughly a carriage length, so it is important to use a unit short enough to distinguish a platform half a chain longer than a train from one half a chain shorter. Certes (talk) 22:11, 17 July 2018 (UTC)

I question your assertion that chains are only used for distance along a railway (is radius of curvature not sometimes measured in chains?) but you are missing the point. We are discussing not how chains are used by sources but how they are to be used on Wikipedia. Wikipedia tries to reach a wide audience by using familiar unit systems and avoiding specialist jargon. The chain is an example of specialist jargon. Dondervogel 2 (talk) 07:52, 18 July 2018 (UTC)

Historically, the radius of curves was not just measured, but specified in chains. Chains were the primary unit, to the extent that I recently saw an accident report of the 1860s which whilst happy to give station distances in fractional miles and the length of track features in yards, talked of a curve of 80 chains radius followed by one in the opposite sense of 60 chains radius. So - as far as curve radius is concerned - the chain was the unit used in UK engineering design. As I understand it, that should lead to its retention (with a conversion to metres). If it does not, two arguments for suppressing chains in favour of yards seem to me to have become intermingled - (1) that chains are a dumb Brit unit and (better) practice elsewhere should be followed instead, and (2) that chains are a dumb Brit unit whereas everybody understands yards so they should be used instead. Given that US railroad practice seems to be to measure/specify curve radius in feet: which argument is preferred in this context? Rjccumbria (talk) 08:32, 18 July 2018 (UTC)

@Dondervogel 2: Wikipedia tries to reach a wide audience by using familiar unit systems and avoiding specialist jargon I refer you to my post of 20:40, 17 July 2018 at WT:MOSNUM. --Redrose64 🌹 (talk) 12:00, 18 July 2018 (UTC)

Miles and chains were used in road construction as well, at least as late as thge Second World War. Distances on the Bulldog-Wau road were measured in miles and chains. Hawkeye7 (discuss) 09:19, 18 July 2018 (UTC)

^[2] Andy Dingley (talk) 12:47, 18 July 2018 (UTC)

(edit conflict) @Dondervogel 2: In which case, it is clear that you are misunderstanding the reasoning behind this whole RfC.

Brief version: on articles about GB railway stations, some of us have added distances using miles and chains; but an unregistered user does not like this and has been either reverting the edits, or reducing them to miles only (example). Various registered editors, myself included, have reinstated what the IP removed. A discussion was started at Talk:East Croydon station#Chains. Once it became clear that the IP was not concentrating on a single station, an RfC was started here.

Longer version: Some of us have reliable sources (I know of three sets of sources: Trackmaps (aka Quail), the Sectional Appendix, and the PSL Guides) that show the mileage of each railway station on the Network Rail system in miles and chains. A few railway stations also have the distance in kilometres, but they are very much in the minority, and in some of these cases, they use a different zero point: Amersham station is quoted as 84.85 km (this is from Ongar), and 23 miles 70 chains (from Baker Street). We have been adding these distances to various station articles, and some of us (not all, admittedly) have been showing our sources in accordance with the policy on verifiability. We have not carried out any calculations using Google maps, that would be against the policy on original research. Why are such distances so easy to source? The railways of the United Kingdom are required by law to accurately measure the distances along the line, primarily in order to calculate charges for carriage. Take a trip along any Network Rail route, and you will see on one side of the line or the other, every quarter mile, a post which might be yellow, white or blue. These posts indicate the distance from the zero point for that line. Besides the quarter-mile posts, fixed structures also bear their distances, such as this bridge. Since these measurements are the distances along the railway line, they are naturally not going to be the distances along public roads, nor as the crow flies. Landmarks are completely irrelevant. --Redrose64 🌹 (talk) 16:35, 17 July 2018 (UTC)

@Redrose64: Nearby landmarks, including the distance to and from, are a perfectly legitimate subject for an article on a railway station. I agree with mentioning distances in chains where sourced, but just not as the primary unit (on grounds of wtf). I accept your point about OR (see my reply to [[User:mattbuck|mattbuck]] addressing this). Dondervogel 2 (talk) 17:03, 17 July 2018 (UTC)

No-one (except you) is questioning whether nearby landmarks are a legitimate subject for an article on a railway station. However they do not (and aren't going to start) define the distance between stations along the railway. Andy Dingley (talk) 12:45, 18 July 2018 (UTC)

References

^ According to this chart the average orange has a surface area of around 140cm²
^ 3.5 µacres

Appropriate place in articles for this level of detail?

I wonder if the friction arises because of [what seems to me!] inappropriate positioning of this detailed information? For the article Clapham Junction to begin with "is 2 miles 57 chains (4.4 km) from London Victoria" seems [to me] intended to stop most people from reading further. The lead should summarise the article, pettifogging detail like that should be a long way towards the bottom. I for one would withdraw my objection to chainage being given if it were appropriately placed in the article. The lead should say no more than "about 2.5 miles (about 4.5 km) from Victoria" and leave the details for the body. Comments? --John Maynard Friedman (talk) 17:41, 18 July 2018 (UTC)

John Maynard Friedman, I encountered the same issue at GA review, and they're right. The lead shouldn't mention chains, because it's not something people can understand readily. I personally go for integer miles in the lead, though if it's below about 3 then go for 1 decimal place (before anyone says I'm a hypocrit, I don't object to the idea of decimal miles, I just don't think they should replace chains completely). The lead section is meant to summarise at high-level, and chains are a much more intricate detail. They belong in the article, but not the lead. It's like education - you learn about a subject at school, but if you keep going you find out that what you learnt was a deliberate simplification.

As I stated earlier, the lead should have the distance in integer miles from the local major station. Then in the description section you state distances in chains to the local major, the national terminus, and adjacent stations, with conversion to 2dp km and a note explaining why you're using such a weird unit. Other lengths should be in decimal miles, yards or metres as appropriate. -mattbuck (Talk) 17:54, 18 July 2018 (UTC)

That's fine, so long as the body does have the accurate mileage (with source). Most of them don't, particularly when TBM10 (talk · contribs) added it to the lead alone. In the cases where I've added a chainage to those articles which did not previously have one, I've done it in the body (example). The problem that we have - and the whole raison d'être of this RfC - is those people who wish chains to be expunged regardless of where they occur in the article or whether they're sourced or not. --Redrose64 🌹 (talk) 20:16, 18 July 2018 (UTC)

I'm not convinced that they do, that is why I raised the question. I was one of those who sought their removal: it got right up my nose to be confronted in the second sentence of an important article by a distance given in a measure that, to almost all readers, is deliberately opaque. If this proposal is accepted, I think both perspectives will be satisfied. --John Maynard Friedman (talk) 22:29, 18 July 2018 (UTC)

I don't understand your attitude at all. Chains and miles are archaic. Sure. So why don't you want to remove the miles from the lead entirely and just use kilometres? The use of decimals came in with metrification; decimal miles was not the way the old measurements were/are rendered, and accuracy is lost. The purpose of the use of miles and chains is for WP:V, to allow the text to be checked against the source, which in an old source (and in the UK, apparently still many current ones), and therefore absolutely required by our policy. Now, if the reader is unfamiliar with miles, then the link is available, in line with out educational mission. If they are already familiar with miles, then they know how long two miles and 57 chains is, but that is not the purpose, as everyone knows how long 4.5 kilometres is. Hawkeye7 (discuss) 23:32, 18 July 2018 (UTC)

I am sorry I haven't had time to give this debate the level of attention it deserves. But to compensate I will just make a few dogmatic statements. These arise from standing on platforms waiting for delayed and cancelled commuter trains into both London and (Manchester long ago) and pondering the mysteries of planet British Rail. Firstly, chains are only used to express the distance between the station/bridge and the terminus. Secondly, chains are currently the only ways that distances between the station/bridge and the terminus.(illustrations above). Running distance for the locomotives along the track are all in km. Thirdly, in Wikipedia obscure measurements are converted into common units to aid understanding. The common unit, the SI unit is km to 3dp. In wikipedia we do not convert one obscure unit into SI, and then continue to translate it in another obscure unit. In Wikipedia primaracy is given to the reference source unit, which is then converted in SI. In Wikipedia we have a problem with some editors that did not into learn SI in school; in certain areas of WP, exceptions have been made. Fifthly, the convention for alternative spellings (UK AUS, etc) is based vary between the location of the artifact, the language of the first contributor, language of major contributor. Fly-in language switchers are shot down. Sixthly, there is a parallel here. Writing a manual of style is valuable and onerous work- but the job is to document the method of handling a source not rather than to change existng conventions. Sixthly the lede is a précis of the body- regretfully the distance measure needs to be there, so the distance in chains from say London Victoria needs to be there needs to be cp'd from the body. ClemRutter (talk) 00:07, 19 July 2018 (UTC)

ClemRutter, I don't understand you data, where does it come from? There are no Kilometres or Kph on the UK railway. Lineside speed boards and cab speedos are all in mph. Distance markers are in miles from a given datum. We're not metric here (HS1 is a strange anomaly of the CTRL). Also it's worth mentioning purely for the sake of dogma that British Rail stopped running the infrastructure back in 1994 so perhaps it's a little misplaced to be blaming them for the current chaos? (see http://www.rail.co.uk/british-railway-history/british-rail/). Dr Sludge (talk) 11:48, 19 July 2018 (UTC)

Hawkeye7, if you reread, I am willing to concede that the articles should give the precise chainage or metreage where provided by an RS. The compromise that I am proposing is that [a] the the lead should give only an approximate distance in miles and km and [b] there should be a Location section in the article, giving the precise distance as given in the citation (and perhaps a local map too, like this). That way, both interests are satisfied [or at least have equality of misery]. --John Maynard Friedman (talk) 14:28, 19 July 2018 (UTC)

For anyone interested, work is underway at talk: Bletchley railway station to produce a worked example for the proposal above. My thanks to those who have discovered it already and contributed. --John Maynard Friedman (talk) 22:13, 19 July 2018 (UTC)

Pre-existing examples of the two-level detail can be found at Patchway. -mattbuck (Talk) 18:22, 20 July 2018 (UTC)

Yes, that's what I had in mind. It seems to that this model meets the purists' demand for “as built“ historical measures and equally had a lead section that is not a turn-off to a world wide readership [and people in the UK younger than about 50 I guess?].

I propose formally that this model be adopted as policy. Do we need another round of discussion? --John Maynard Friedman (talk) 20:09, 20 July 2018 (UTC)

@John Maynard Friedman: I don't understand the proposal being made. Can you spell it out here? Dondervogel 2 (talk) 07:14, 21 July 2018 (UTC)

Dondervogel 2 if we're judging by Patchway and other articles I wrote, it's the following:

The lead section should contain a distance from the regional major station (along the railway) in integer miles, or to 1 decimal place if under 1-3 miles, with conversion to kilometres at the same number of decimal places.
Within the main body, distances along the railway should be measured in miles and chains, with a conversion to kilometres at 2 decimal places.
When chains are first mentioned, there should be a note explaining that UK railways are measured in chains. It is also useful to specify there are 80 chains to the mile.
Distances to other local landmarks should be measured in yards or decimal miles, with conversion to metres/km, with an appropriate degree of precision.
Civils aspects, such as tunnels, bridges and platforms, should be measured in whatever they were designed in. That's probably yards and feet, so add appropriate conversion to metres/km.

-mattbuck (Talk) 08:24, 21 July 2018 (UTC)

I see. Bullets 1-3 address most of my concerns. I don't agree with bullets 4 and 5 though, as these will lead to people switching units halfway through articles - a recipe for confusion and not needed to address the problem with the use of the chain. If bullets 4 and 5 are deleted, only the ambiguity of the chain unit remains unaddressed. One solution is to state explicitly that in the context of UK railways, the term "chain" is used to mean "Gunter's chain", equal to 22 yd. This point could be merged with bullet 3 or added as a new bullet. Dondervogel 2 (talk) 08:47, 21 July 2018 (UTC)

Thank you, @mattbuck, that is exactly what I had in mind.

@Dondervogel, yes, I ought to have spelled it out. Would you be content with [[Gunter's chain|chain]] being used in the article at first mention (excluding convert template, to avoid overlinking) i.e., that every article doesn't need to include an explicit history lesson, wanted or not. --John Maynard Friedman (talk) 09:23, 21 July 2018 (UTC)

The measuring device Gunter's chain and the unit of length Chain (unit) seem to be being muddled. Is a wikilink to Chain (unit) what is intended to suffice?SovalValtos (talk) 09:34, 21 July 2018 (UTC)

Yes, the term "Gunter's chain" seems to have 2 different meanings. I think that both Chain (unit) and Gunter's chain can be improved by explaining this. Do I think that's enough? Not really. If it's Gunter's chain we mean then we should say so (just like we don't say "ton" but "long ton" or "short ton"). Dondervogel 2 (talk) 09:52, 21 July 2018 (UTC)

I would point out that the note at Patchway specifies there are 80 chains to the mile. -mattbuck (Talk) 13:25, 21 July 2018 (UTC)

@Dondervogel 2: But there is no need to confuse- in WP we try to be inclusive and put all sort of wierd exceptions and historical dead ends into our articles. A chain is a chain (to quote a minor UK politician), 80 to the mile- I didn't know that Gunter had one, that is trivial off focus information. You dont see a reference to Gunter on bridge plaques, and the UK is many nautical miles from Texas! If my daughter looked at this article see wouldn't have a clue about chains so would just use the SI measurement- if she was being pedantic she would back convert- anyone who had been to primary school before the dinosaurs- would have to recite an acre is 10 square chains.

I do have multiple issues with the WP Chain (unit) article, the article needs a good shaking- which is possibly the source of the confusion in this article. But any discussion of the source or length of the standard chain needs to be exercised there not here. I think you will agree.ClemRutter (talk) 09:55, 21 July 2018 (UTC)

We have multiple sources stating that a chain is 20 m, and one that says 30 m. Where the term is used, the various meanings need disambiguation. Dondervogel 2 (talk) 10:05, 21 July 2018 (UTC)

Discuss that over on Chain (unit)- I have put a new section on the talk page.ClemRutter (talk) 11:04, 21 July 2018 (UTC)

A chain is most emphatically NOT 20 metres. It is precisely 22 yards. Mjroots (talk) 12:25, 21 July 2018 (UTC)

An English chain is precisely 22 yards. An Indian surveyor's chain is 20m [30m also available]. A Scottish chain is different still. That is how we started this mess and we are starting to round in circles, so if you must continue this topic, please take it to talk:chain (unit). --John Maynard Friedman (talk) 13:02, 21 July 2018 (UTC)

The proposal itself seems sensible, but some of the circles are spurious. There is no English chain, nor a Scottish chain; both sides of the border the chain - where used - is of 22 imperial yards. The Scottish chain was a customary unit (an eightieth of a Scottish mile) in its day, but that was before the adoption/imposition of a uniform ('imperial') system of weights and measures throughout the UK. It has not been in use for over 150 years. By 1863, Rankine ("Regius Professor of Civil Engineering and Mechanics at Glasgow University") was clear that the chain used for purposes of civil engineering in Britain was of 22 yards. He described the old Scots customary units of length as "obsolete" but mentioned them "as they occasionally occur in old plans".Rankine, William John Macquorn (1863). A Manual of Civil Engineering (2nd ed.). London: Griffin Bohn & Co. p. 3. Rjccumbria (talk) 19:04, 21 July 2018 (UTC)

@Dondervogel 2: the only reason we're talking about chains here is because the railway uses them. Therefore, because that's how the railway is measured, we should use those units. But you seem to now be saying that because one area uses a special unit the rest of the article has to as well. That doesn't make sense. Platforms and tunnels are not measured in chains, they're measured in yards. So on the precedent of bullets 1-3, we should use the units they are measured in. Yes, this means we will have two different units used - it's unfortunate, but we are still using imperial as the first unit and metric as the second. -mattbuck (Talk) 13:37, 21 July 2018 (UTC)

@Mattbuck: No. My point here is that we should avoid confusing the reader with ambiguous or unfamiliar units. The chain is both ambiguous and unfamiliar. Dondervogel 2 (talk) 14:50, 21 July 2018 (UTC)

Much as i agree with Dondervogel 2 on the principle, i have to agree with mattbuck on the practice: we have to deal with the world as it is, not as we would like it to be. Virtually all the sources we have are in miles and chains. until the railway is resurveyed in metric (and the data published), that is what we have to use. It is wp:syn and wp:or to change the data to another measure and lose the original. Provided we don't turn most readers off by slapping them in the face with obscure jargon in the first few sentences, we've done what we should. Uf technical measures are used in the body, so long as modern equivalents are provided too, then we've done our job. --John Maynard Friedman (talk) 15:34, 21 July 2018 (UTC)

Routine conversion between units is not OR. Dondervogel 2 (talk) 16:12, 21 July 2018 (UTC)

As stated above, if you don't accept that a chain is 1/80 miles, then define it as such in the article. The sources are quite clear that in the context of British railways that is what a chain is. Are we also to require a source to say that we are using the modern definition of a mile, not a nautical mile, a Welsh mile, a Roman mile or a mile the length of which is my height times a number I just pulled out of my backside? -mattbuck (Talk) 18:54, 21 July 2018 (UTC)

I can't speak for others but my own position on this has just changed. A google search did not reveal a single independent confirmation of Rowlett's assertion that a unit of distance equal to 20 m is called a chain. In the absence of that confirmation I'm inclined to think either Rowlett got it wrong, or is simply referring to a measurement instrument (called a metric chain) of length 20 m. Unless/until someone can find a source to back up Rowlett on this (20 m chain as a unit) I withdraw my assertion that the term 'chain' is ambiguous in modern use (as a unit). I stand by my assertion that it is an unfamiliar unit, and should not be used without first telling the reader what the length is km or mi. Dondervogel 2 (talk) 20:40, 21 July 2018 (UTC)

I think we should define it in a note as 1/80 mile the first time it comes up, which should not be in the lead. It should be the main unit for measuring rail distances, but it should never be used without a conversion to metric. -mattbuck (Talk) 21:04, 21 July 2018 (UTC)

Yes we should include a conversion to metric, and this is why I always use the {{convert}} template. I also supply this with the param |lk=in so that the word "chains" is linked to our article Chain (unit). But when the unit is linked, there should not be any need for additional explanation. --Redrose64 🌹 (talk) 14:38, 22 July 2018 (UTC)

A point in favour of keeping chains where used (at least in the article body) is that a lot of articles include Railway Clearing House maps and junction diagrams, which means that if the reader wants to be able to fit those maps into the overall information in the article. to me, it seems preferable to use a less common unit consistently than to mix units for purely stylistic reasons. Of course, if someone wants to re-draw those maps… 130.220.8.157 (talk) 00:14, 23 July 2018 (UTC)

Formalisation of the vague consensus everyone mildly peeved but none truly unhappy propsal (VII)

I thought I'd formalise the proposal I have outlined several times here, for the benefit of all.

The following applies only to railways measured in miles and chains (no HS1, LU, etc).
Chains should not appear in the lead. Any mileages in the lead (eg from regional hub) should be in integer miles with conversion to kilometres
Distances along the railway within the body should be given in miles and chains, with conversion to kilometres or metres as appropriate.
Lengths of tunnels, bridges, platforms, people, should be in whatever unit is appropriate and referenced (probably miles and yards with conversion to metres)
Where they first appear, the chain should be linked or defined and, ideally, a note given as to why the article is using that unit
Do not convert to chains if a source does not use them already.

-mattbuck (Talk) 17:19, 1 August 2018 (UTC)

Support

-mattbuck (Talk) 17:19, 1 August 2018 (UTC)
Support if a few details are ironed out. (Does point 6 trump point 3? What about curve radii?) Certes (talk) 17:43, 1 August 2018 (UTC)
No one has yet given any authoritative sources using kilometres or decimal miles on Network Rail tracks. I'd say curve radii should be in whatever a curve radius is measured in. If the source says chains then use chains. -mattbuck (Talk) 18:51, 1 August 2018 (UTC)
No. The unit used should not depend on the unit used by any particular source. That is a route to chaos, and if indeed this is the intent of this proposal I withdraw my support. Please clarify. Dondervogel 2 (talk) 19:46, 1 August 2018 (UTC)
Oh for christ sake... What I'm saying is I do not know what curves are measured in. If they're measured in chains then use a chain. If they're measured in multiples of Lucy Collett's left tit, use multiples of Lucy Collett's left tit. But if it's measured in multiples of Lucy Collett's left tit, don't convert it to chains. The railway length is measured in chains, they're a really fucking stupid measurement but there we go. Because there are no authoritative sources which use decimal miles or kilometres, we use what the authoritative sources use - which is chains. -mattbuck (Talk) 22:30, 1 August 2018 (UTC)
With this interpretation I'm still on board. Dondervogel 2 (talk)
I can live with this (~~subject to above request for clarification~~). Bullet 5 is the key, as it addresses the readers' lack of familiarity. I don't think a link is enough on its own, but a footnote would do it. Dondervogel 2 (talk) 17:52, 1 August 2018 (UTC)

Oppose

Argument

Closing

This RFC has run for three weeks now. Are there any objections to my closing it in the next few days? Mjroots (talk) 18:24, 28 July 2018 (UTC)

It would be helpful if we could at least agree some kind of conclusion. So may I propose this as a summary:

there is no consensus to remove any reference to chains, indeed there is a strong consensus that they should be retained.
there is a majority in favour of using only familiar units [miles, km] in the lead provided that the distance in chains is given in the body. I would like to say that there is also a consensus for this but as I am one of the proponents, would someone else say aye or nay [only! we don't need to debate it all over again]
the chain should be wlinked in the same section of the body.

Is that a fair summary? --John Maynard Friedman (talk) 18:47, 28 July 2018 (UTC)

I would add that there is no consensus for using miles and chains as the primary unit anywhere, but I think there is consensus for converting to these units where sourced. Dondervogel 2 (talk) 19:11, 28 July 2018 (UTC)

@Dondervogel 2: How can you convert to miles and chains when those are the units used by the sources? Or have you finally found sources in kilometres (or decimal miles) without noting them above, despite my repeated requests? --Redrose64 🌹 (talk) 20:38, 28 July 2018 (UTC)

Distances in the UK are almost universally expressed either in kilometres and metres or in some mixture of miles, yards and feet. I do not need a reference to show that because it is immediately apparent to anyone spending (say) a week there. I have lived in the UK for approximately 25 years, and in that time I have not once encountered the chain. The onus is on you to show that it is used widely enough for the reader to have a faint inkling of what it means, not on me to show that familiar alternatives exist.
I do not see a problem converting from miles to chains. It is straightforward.

Dondervogel 2 (talk) 21:43, 28 July 2018 (UTC)

"lived in the UK for approximately 25 years", so I'm guessing you went through the educational system elsewhere. I was born and brought up in the UK and have lived here all my life. I can assure you when I was at school we did learn about chains (aged about 10) and I know that for all practical purposes I convert metric to Imperial to get a "feel" for it. 2.7 km is - let me see - about a 1.5 miles, so say 30 minutes walk. Martin of Sheffield (talk) 22:30, 29 July 2018 (UTC)

I no longer live in the UK, but my primary and secondary education was spent entirely in the British education system (always in British schools, although not all in the UK). I am old enough to have GCE 'O levels' (not GCSEs), and I spent a total of 7 years at British universities. In that time I encountered cups, gallons, furlongs, hands, Frog 'n Firkins and snakebites, but I have not once encountered the unit chain. You tell me it was once common and I believe you. It's just not common today. Dondervogel 2 (talk) 22:50, 29 July 2018 (UTC)

Fair enough, thanks for clarifying the ambiguity. We must be of a similar age (O-levels). I can actually remember being shown the chain that the groundsman used to lay out sports pitches and the running track - all in imperial units. That year seemed to consist of learning conversion tables and then doing arithmetic (long division and long multiplication) on imperial units. The following year (IIRC) we were introduced to the system "used in Europe" - ie the metric system. Martin of Sheffield (talk) 08:55, 30 July 2018 (UTC)

@Dondervogel 2:Distances in the UK are almost universally expressed either in kilometres and metres or in some mixture of miles, yards and feet except on the railways where they are expressed in miles and chains. Weird but thats the UK. I have spent a few days taking photos of bridge plaques, and become aware of just how ubiquitous they are. Every footbridge at every station will have one of these oval plaques- you will have stood next to them and not seen them! Every time your train passes over a bridge, the driver will see a larger oval sign marked in miles and chains, but you probably haven't looked up from the road beneath. The primary measurement on railway is miles and chains, all our sources are written that way, we then {{convert}} to kilometres/metres for the target general reader. We never need to convert the other way, you don't convert universal to niche for the fun of it. We never need to use yards as they don't exist in this context, and WP is not attempting to promote the Imperial system we are just reporting what the source tells us in the system it uses, then providing the translation into the universal SI units for the world wide readership. ClemRutter (talk) 00:46, 29 July 2018 (UTC)

Well put. Another exception is UK horse racing, which uses miles and furlongs. If a race is over 1 m 2 f then it's sensible to give 2 km as a conversion but neither Wikipedia nor the general public would describe it as a "two kilometre race". Certes (talk) 01:15, 29 July 2018 (UTC)

We are editing articles according to Wikipedia norms, which are a group of pages collectively known as policies and guidelines. Two of these policies concern original research and verifiability. When I edit an article, I do not make stuff up: I add information based upon the sources to hand. Let's consider the original point of contention, which is the distance of East Croydon station. I have three sources to hand for that:

Yonge, John (November 2008) [1994]. Jacobs, Gerald (ed.). Railway Track Diagrams 5: Southern & TfL (3rd ed.). Bradford on Avon: Trackmaps. map 14C. ISBN 978-0-9549866-4-3.
"Kent / Sussex / Wessex Routes Sectional Appendix Module KSW2" (PDF). Network Rail. 2 June 2018. p. 182.
"VTB2 Windmill Bridge Junction to Earlswood mileage change". Railway Codes. 11 July 2018.

all of which (be warned: the second one takes a long time to load, being over 43 MB in size) show that this station is 10 miles 28 chains from London Bridge. None of these sources use decimal miles; miles and yards; or kilometres. The problem in your last sentence is this: if the source is in miles and chains, as these all are, why convert it to something else and then convert it back again? Let me say this once more: we have a policy on verifiability and if you expect me to go against that policy, you should seriously consider your future as a Wikipedia editor. --Redrose64 🌹 (talk) 23:38, 28 July 2018 (UTC)

Distance in the UK are still measured in miles. There are a very few exceptions, such as HS1 which is measured in kilometres. On the roads, it's miles and ¼ miles, on the railways it's miles and chains, except tunnels which are measured in miles and yards. Mjroots (talk) 05:36, 29 July 2018 (UTC)

I see no new points being made and therefore no need to respond, except to point out that furlongs are more familiar to the great British public than chains. There are pub names with furlong in the title and the term is also used in news describing horse races, so the reader is not surprised to see the unit used (I at least have heard of a furlong for these and related reasons).

Perhaps there should be a similar discussion about the use of furlongs on Wikipedia, but not here.

Dondervogel 2 (talk) 07:36, 29 July 2018 (UTC)

@Mjroots: There is already a closure request at WP:AN/RFC#Wikipedia talk:WikiProject UK Railways#Chains RFC. --Redrose64 🌹 (talk) 20:38, 28 July 2018 (UTC)

Is it me, or does that closure request entirely misrepresent the apparent consensus reached in the discussion above? Mjroots (talk) 20:48, 28 July 2018 (UTC)

Mjroots I'd agree with you on that. I took the conversation's petering out to indicate that people were all mildly unhappy about the solution I stated previously, but only mildly so rather than the vehement opposition to the numbered proposals. -mattbuck (Talk) 21:40, 28 July 2018 (UTC)

Let's look at a real example, which reads: "It required a 1 in 50 climb and a 15 chains (990 ft; 300 m) turn from the LSWR main line to reach the bridge". It does not matter where this appears in an article. There are two problems with it:

it presents a quantity in an unfamiliar unit (chains) before converting to familiar units;
it doesn't tell us what physical quantity is being reported in chains.

I just find it unnecessarily confusing. Dondervogel 2 (talk) 21:49, 28 July 2018 (UTC)

It's only confusing because the article should clarify whether it's a curve radius or a distance. That would be the case whether measuring in chains, metres or cubits. Wording such as a turn of radius 15 chains (990 ft; 300 m) would be quite clear. Certes (talk) 22:04, 28 July 2018 (UTC)

Specifying the nature of the quantity resolves the ambiguity but not the unfamiliarity. I strongly prefer starting with a familiar unit, e.g., "300 m (990 ft, or 15 chains)". Readers understand metres and feet. Few will have encountered the chain. Dondervogel 2 (talk) 22:24, 28 July 2018 (UTC)

The text in question is sourced, and it was added by Ritchie333 (talk · contribs). Have you checked the source (or even asked Ritchie333) to see what units it uses? Chains, I'll warrant. Have you looked for an alternative source that uses metres or yards? Also, since that edit was made, the article passed a GA review, in which the reviewer (Tim riley) did not call out the use of chains as a unit. --Redrose64 🌹 (talk) 22:42, 28 July 2018 (UTC)

And I thought it would be easy to summarise this discussion! But let me try again

@Dondervogel 2, it is clear that there is a strong consensus against your proposal, further argument is pointless. I for one agreed with you at the outset; I have been persuaded by the debate and no longer do so.
there is a consensus for a change to show familiar units in the lead, 'as engineered' units in the body.
there is no consensus to close the RFC to status quo as though this discussion was a complete waste of time. Most contributors have moved towards the compromise at 2 just stated.

is this reasonable? --John Maynard Friedman (talk) 12:24, 29 July 2018 (UTC)

Quite frankly, this subsection is a waste of time, because the discussion should be closed by an uninvolved editor. Jc86035 (talk) 12:37, 29 July 2018 (UTC)

I have made a request at WP:AN/RFC#Wikipedia talk:WikiProject UK Railways#Chains RFC for a netral party to review. --John Maynard Friedman (talk) 18:48, 29 July 2018 (UTC)

OK Jc86035, as you've objected to my closing the RFC, I'm happy to step back and allow an uninvolved admin close. Mjroots (talk) 19:38, 29 July 2018 (UTC)

FYI (and hopefully to assist the reviewer), I have updated Bletchley as per discussion. Please use the talk page of that article for initial comments so that only amy new points of principle come back here. --John Maynard Friedman (talk) 11:04, 30 July 2018 (UTC)

What about precision?

[The following discussion was originally moved from Clean up of Chain (unit). Dondervogel 2 (talk) 20:19, 31 July 2018 (UTC)]

In regard to units, in WT:MOS I wrote:

Do we have (copies of) the original engineering drawings for the terra-cotta army? Or documents close to the appropriate time period? If so, it is probably best to use those units. But for measurements made (much) later, in a more current unit system, I don't see a reason to back convert. One complication with unit conversion is significant figures and uncertainty. Many values are written with an appropriate number of figures, given the uncertainty in the measurement. Conversions will often either lose precision, or imply greater precision than the original. (This happens naturally on rounding decimal values.) As to railway gauge, I suspect that it was converted, with appropriate rounding and within uncertainty, years ago. This is even done in the definition of many units, when new methods allow for increased accuracy in measurement. If the primary source uses a certain unit, and even if the data comes from a reliable secondary source, it seems reasonably to use the primary source unit, with conversions to modern units. If there is no primary source, with reported measurements being made much later, I don't see any reason to use units that might have been used at the time.

(This followed a suggestion that the terra-cotta army measurement be in Qin dynasty units.)

It does happen that 1 chain is close enough to 0.01 mile, that implied uncertainty is close enough, though. Gah4 (talk) 23:26, 29 July 2018 (UTC)

The comment was originally posted at one of those other locations and I suggested that it would be best to keep arguments for and against the use of the chain in one place, namely here, where it started. Gah4 is simply responding to that request. I consider his point about precision to be a potentially valid reason to use chains that partly counters my concern about unfamiliarity. I therefore think it is worth discussing (on this page). Dondervogel 2 (talk) 08:54, 30 July 2018 (UTC)

Thanks. I believe that this could have general use, so the one in WT:MOS should stay. The additional complication is that units sometimes have more meaning than seems obvious. Torque is measured in units of N m (Newton meters), even though J (Joule) is dimensionally equivalent. Similarly, the Volt Ampere (V A) is dimensionally the same a Watt (W), but implies that a power factor is not included. I don't see that is important here, but one should be careful when changing units. Gah4 (talk) 10:39, 30 July 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.

[1] According to this chart the average orange has a surface area of around 140cm²

[2] 3.5 µacres

[1]

[2]