Talk:Gotthard Tunnel

How many tunnels?

"It is built as one double-track, standard gauge tunnel."

• 

Then why does this picture clearly show two double-track tunnels?

The two tubes merge after a few hundred meters, its just some switches of Göschenen railway station that were put in the mountain. The single double-track tube can be clearly seen at the other end, near Airolo. --Kabelleger 11:37, 29 April 2007 (UTC)

Design

I have heard the Gotthard Rail Tunnel uses a design that requires it to circle while changing altitude. If any experts can verify this, I believe it would be interesting in the article.Voiceperson (talk) 18:46, 19 August 2008 (UTC)

The Gotthard rail tunnel doesn't circle nor change altitude (well, a bit, but it's not really relevant). Maybe you're confusing it with other spiral tunnels of the Gotthardbahn? --Kabelleger (talk) 19:57, 19 August 2008 (UTC)

1921-05-29 07:50-08:30

This is hard to read. Is it to some standard? —Preceding unsigned comment added by 80.189.103.145 (talk) 18:02, 30 March 2010 (UTC)

Steam in the first years?

The rail was electrified in 1920. But how were the trains propelled until then? If it was steam, then were the boilers able to build enough pressure before entering the tunnel to drive it through? Did they stop firing when or before they entered the tunnel?

88.233.120.200 (talk) 19:21, 20 October 2011 (UTC)

Precision of length.

The length of the tunnel is given as 15.003 kilometres (9.322 mi). That degree of precision is both unnecessary and impossible. Such accuracy of measurement is impossible to achieve as it requires the length to be measured with an uncertainty of measurement of just 0.02%. A laser rangefinder could only be used if the tunnel were gun barrel straight (no curves in any plane which is not the case), but such a measurement would be severely degraded by changes in air density caused by humidity and temperature changes throughout the tunnel.

GPS co-ordinates are similarly inaccurate because although GPS can have an accuracy of around 2.5 metres (8.2 ft) (if using a system that uses both NAVSTAR and GLONASS), the distance between a pair of co-ordinates is calculated following a curved path representing the circumference of a theoretical geoid connecting the two points. That is: if you want the distance between (say) London and San Francisco, GPS gives you the great circle distance over the surface of the globe not the direct distance through the planet. The error in any case is similar to the precision claimed so the length could be equally be 14.995 kilometres (9.317 mi). RFenergy (talk) 12:09, 13 February 2020 (UTC)

We live in the 21st century! https://shop.swisstopo.admin.ch/en/products/geoservice/swipos -- ZH8000 (talk) 21:05, 28 February 2020 (UTC)

Even in 2020, you cannot get centimetre accuracy over a distance of 15km using a carrier-phase GPS (used for surveying). GPS cannot give you the length of the tunnel, only the distance between the ends of the tunnel, which is less than the length due to the fact that it ain't straight. But even so, 0.02% precision is totally unnecessary for an encyclopaedic article.

Interesting how your ref. claims centimetre accuracy and then goes on to claim an accuracy of half a metre (for conventional GPS receivers). Half a metre would only be possible if the receiver was able to receive the military L2 signals (and I doubt that the Americans would release the crypto codes). Two metres is the gold standard for a differentially corrected GPS system, improved from 10 metres for uncorrected. Even then, that accuracy is only valid if the correction data applies to the receiver's precise location which it will rarely do so as the differential nodes are around a kilometre apart. Carrier phase systems are capable of sub-centimetre accuracy but only under the most ideal of conditions where the available satellites and position does not significantly change between measurements. This could never apply to two positions 15 km apart. Carrier-phase was not considered as part of the original GPS system, but was subsequently developed once it was realised that all the NAVSTAR GPS satellites transmit on exactly the same pair of frequencies, it being unnecessary to decrypt the L2 signal as only its carrier signal is required. RFenergy (talk) 15:24, 29 February 2020 (UTC)

User:ZH8000 has, once agian, decided to revert the claim that the tunnel is 15.003 km long claiming that this is the official length. However; he has provided no reference supporting it. Looking for any proper modern reliable references only turns up claims that the tunnel is 15 km long (without the extra three metres) or more usually just "approximately 15 metres".

I am still fully aware that the extra three metres was an impossible degree of accuracy (even in 2020). ZH8000 defended the figure by claiming that GPS had an accuracy in Switzerland that cannot be matched anywhere else on the planet (conveniently ignoring the fact that GPS can only give you the 'as the crow flies' distance between the tunnel portals and not the length of a non straight tunnel).

I started to wonder if that three metres had come from a conversion error somewhere, but I couldn't, initially, find a reasonable measurement (i.e. not overly precise, but a reasonable figure that reflected a sensible probable error of measurement) that converted to 15.003 km.

A search of newspaper and magazine articles turned up an article from the Illustrated London News published in 1882 which gave the length of the tunnel as "15.003 kilometres" (and helpfully converted it to UK distances (a little inacurately) as "9+1⁄3" miles) because few, if any, of their readers of that era would recognise kilometres. But there was that three metres. However, in 1882, such accuracy was completely unheard of (and completely impossible - and ZH8000's claimed GPS didn't exist). The official accuracy of any survey using triangulation (the only available way of arriving at the tunnel length in 1882) was a little under 50 metres.

Further searching turned up a press release from the same year issued by the "Gotthardbahn", apparently the railway company that first operated services through the tunnel. The press release gave the length of the tunnel as approximately 3+1⁄8 meiles (en: miles) (a believably imprecise statement). Problem was: that is nowhere near 15 kilometres, or so I initially thought.

But then I remembered that the distance unit of the mile was not standardised across the world until 1959 with the mile in various countries varying wildly in size and often within areas of the same country (Switzerland being no exception on this latter point). The Canton of Ticino was in an area that had an unusually large mile. Various lengths have been quoted for the Tacinoan (is that a word?) mile varying from nominally 4800 to 4810 metres (4808 metres being inexpicably popular). But these figures are too conveniently exact. An interogation of a historical units database turned up that the mile in use in that part of Switzerland was actually 4800.96072 metres. Multiply that by 3.125 (3+1⁄8) and you get 15.00300225. So that is where the errant three metres comes from (ignoring the extra 2.25 millimetres as including them would be just plain silly). I had hoped to find a statement of the length from the construction company but came up empty. There is no reason to suspect that the Gotthardbahn used anything other than the officially quoted length.

It should be noted: that the use of the metric system of units was by no means universal anywhere in the world in 1882 and generally more the exception than the rule. I would conjecture that the length was originally converted to kilometres so that the length of the tunnel could be publicised in Italy (the tunnel is close enough to Italy that people travelling between Switzerland and Italy in that region would pass through the tunnel).

Italy had inherited a number of different miles following its unification but Italians were, by this time, familiar with the kilometre as it was (slowly) adopted across the old kingdoms over about 25 years, it becomming the official measurement system in Italy in 1861. However, in true Italian style, there were a few pockets of resistance that held out until 1870. RFenergy (talk) 14:14, 24 March 2020‎

You are fully mislead.

Firstly, lear hot to use a talk page correctly.

Secondly, your whole text is totally futile: WP:NOR, WP:NOTFORUM, WP:NOTPAPERS, WP:NOTBATTLEGROUND. All of them you are violating. It is fully useless to discuss wether something is possible or not. As a cotributer to WP you are "only" expected to document, not questioninghwoever boring it might be for you.

And thirdly, you are repeatingly violating the WP:BRD procedure: Your initial change has been challanged and since then you were not able to provide a source supporting your initial change, therefore you fail in WP:VER as well.

Therefore I finally reset the original state of the article since your first change.

-- ZH8000 (talk) 20:30, 29 March 2020 (UTC)

@ZH8000:Please learn how to use a talk page correctly. DO NOT reformat other people's posts. It was indented exactly as I had intended. It was not directly answering any of the points above but introducing a new argument.

After translating your post into intelligible English, regardless: neither WP:BRD (which is not an enforceable policy - only an essay) nor WP:3RR (which is) applies in this case. The claimed length of 15.003 metres is entirely without a reliable reference and, as such, may be freely removed by anyone. It is not up to the challenger to source a refutation of a specific claim but the person making the claim, or it having been challenged, the person subsequently restoring it (WP:BURDEN). If you wish to restore your claimed length, then in accordance with said WP:BURDEN, you must provide a reliable reference for this impossible precision. A failure to do so is edit warring at best but at least disruptive editing. Although I did find a reference supporting the figure, it is far from reliable given its age.

"Any material lacking a reliable source directly supporting it may be removed and should not be restored without an inline citation to a reliable source." (WP:BURDEN - an enforceable policy)

I have not violated anything by attempting to discuss the false claim along with your disruptive editing or your edit warring. This is valid talk page usage. I would have been entirely within my rights to delete your post as a personal attack and making no attempt to discuss the issue. I chose instead to respond as is my right. -RFenergy (talk) 13:04, 30 March 2020 (UTC)

References

Precision of length Once More.

It's now given as 15,002.64-metre-long (49,221 ft 3 in). I'd like to change it to "9 miles (15 km)" but found the text was machine-generated, and I don't know how that works.

Could someone more expert make the change, and save us from being a laughing stock. John Wheater (talk) 16:27, 12 August 2022 (UTC)

@ZH8000: John, and all the others in the previous section above, say it better than me. Your insistence that for some reason we need to include measurements which are literally inch-perfect is at best overprecise and at worst just deliberately wasting people's time. I challenge you to find me a single example of a multi-kilometre long object for which the exact length down to the centimetre is necessary or frequently quoted. Wikipedia is written for a general audience, not for specialists or enthusiasts, and that means that a selection and a summary (and, yes, rounding, especially in cases like this were the measurement is clearly overprecise) of the most pertinent information needs to be made. 173.179.105.16 (talk) 21:37, 10 November 2022 (UTC)

It is not about being overprecise, but simply a quotation of a serious publication. On the other hand, you, as a WP editor, are not in the position to interpret the reason why the publisher used the form they decided to use (WP:NOR). As a compromise: Let's have the adjusted figure in the text, but the quoted figure in the info box. -- ZH8000 (talk) 15:49, 14 November 2022 (UTC)

You don't get it, do you? The figure of exactly 15002.64 metres is not acceptable, because it is indeed overprecise and raises more questions than it answers:

-How exactly was it measured? (did someone walk through the tunnel with a measuring tape? is this the distance as measured by a train running on one of the tracks? ...)

-What exactly is being measured? (is this from one of the portals to the other? from which point exactly? if measured along a track, which one (if there is a curve at some point, then the distance will likely differ by several metres depending on whether one measures along the inner or outer rail - or the inner or outer edge of the tunnel, for that matter)

In short, the measurement uncertainty is certainly larger than the 0.5 cm / 1500264 cm (a staggering 0.33 ppm) that is implied, and there is only one source which gives this exact number (and no, we are not bound to accept on faith that the more "precise" source is correct). As the previous section shows, there are sources which do not bother with such fastidious "precision". I also see that I'm not the first one to notice how ridiculous it is to claim we have a centimetre-precise measurement of this. Since you seem to like lecturing others about what they ought to do, you might want to take a look at the dictionary definition of the word "consensus", which is not "the one that sticks around the longest and reverts whoever disagrees and accuses them of edit-warring gets to say how the article should be". You also haven't addressed the consistency issue with other similar articles about railway tunnels (for ex. the Channel Tunnel or the Severn Tunnel) or bridges (ex. Confederation Bridge, or pick any other example), none of which seem to want to bother with such artificial precision. 173.179.105.16 (talk) 16:49, 14 November 2022 (UTC)