If you discount random experiments hosted on someone's laptop or something, there's probably something in La Paz, Bolivia which would get us to around 4000m elevation. Chile has a data center at 5000m elevation but it's for a supercomputer. Some Googling indicates China thought they were on to something with a data centre in Lhasa at around 3600m elevation: https://www.passionateinmarketing.com/china-to-build-the-wor...
Perhaps an interesting way to further qualify "highest web site" might be to demand it's available directly from the broader Internet. While even a permanent TCP/IP based device in space would be pretty cool (and, I imagine, actually exists) the cost of the connection is probably onerous to make it open to public use(?)
So perhaps if someone could rig up a microwave link to a solar powered Raspberry Pi Zero hanging off of a weather balloon or something.. we could have a truly publicly accessible Web server at above jet-liner height? :-D This feels like something Tom Scott might try..
We could definitely do better, but could something in orbit truly be considered "highest in the world"? I'd probably not credit a website as highest in the world unless it were actually on the world, say, on a mountain, or maybe arguably inside the atmosphere.
I had an interesting discussion with my 8yo recently about the highest mountain. It had never occurred to me, but "highest" does not have an agreed upon definition. There are actually a few competing definitions[1]:
There's Muana Kea, which is the _tallest_, meaning the furthest distance from the mountain's base to its peak.
There's Mount Everest (which I had assumed was the agreed upon "highest"), which has its peak at the highest altitude.
Chimborazo has the most unexpected definition of "highest", which is a peak that is furthest from the center of the Earth! We don't live on a perfect sphere, so some mountains can be "shorter" but still further from the core!
Would be interesting to see if the "highest websites" differ for each of these definitions!
I share a propensity for getting interested in things like this more than I perhaps should(!) In most cases, Everest doesn't really look particularly impressive given its low prominence in relation to its neighbors, although Wikipedia seems to treat sea level as being its true "base" as a sort of special case so it still comes in at #1 on https://en.wikipedia.org/wiki/List_of_mountain_peaks_by_prom... !
Chimborazo is a fun one and I would accept a strident enough argument for it to be #1. It truly looks like a magnificent mountain too, indeed perhaps the archetypal 'mountain', if we can take aesthetics into account ;-)
> Wikipedia seems to treat sea level as being its true "base" as a sort of special case so it still comes in at #1
This is due to the definition of prominence, which is roughly "how far down do you have to go from the summit before you can go up to a higher point." This is undefined for the highest mountain, since there is no higher mountain to go up to, so height above sea level is used instead.
You make it seem that something funny is going on with Everest's prominence. That's just the definition of prominence:
> prominence measures the height of a mountain or hill's summit relative to the lowest contour line encircling it but containing no higher summit within it
By definition, this means that the highest mountain's height is its prominence.
I don't think there's anything "funny" logically ;-) It's hard to explain, but I just don't find it very gratifying as a measure given Everest is surrounded by almost equally tall mountains with small prominences (Lhotse's is about 600m) and the way Wikipedia refers to a "special definition for Everest" just illogically raises my hackles much in the same way as 1 not being considered prime does and at least that's useful ;-)
Prominence is a weird property. Consider Grays and Torreys in Colorado, two 14,000ft mountains that are very close to each other. Both are some of the tallest mountains in the country and almost the same height, but because Grays is about 3ft taller than Torreys it ends up having 5 times the prominence (2750 vs 560).
I live in Teton Valley, Idaho (6,500 feet above sea level) and I see the Grand Teton (13,775 feet) everyday (looking at it now actually). That's a prominence of 7,275 feet. I have also been to the base camp (13,550 feet) of Annapurna (26,545 feet). That's a prominence of 13,000 feet or almost double what I'm used to seeing - and that difference is stunningly obvious. I've mentioned this to people often - that just standing in the Himalaya creates a feeling of majesty that is (literally) not possible to experience anywhere else unless something like Mt. Rainer were a couple thousand feet from the Pacific coast (instead of 112 miles) or unless the oceans were drained.
The joke being that the Toblerone bar has a mountain based on the Matterhorn as it’s logo. Either that, or someone very dedicated to Toblerone chocolate created the world’s largest advertisement in the Alps. That would be quite the story. If so, they forgot the bear.
> There's Muana Kea, which is the _tallest_, meaning the furthest distance from the mountain's base to its peak.
I had thought Denali in Alaska held this honor, but apparently the base of Mauna Kea is considered to be the ocean floor which makes it taller. Interesting.
You're thinking about a peak's prominence. A lot of people put a lot of thought into that, which is summed up by Wikipedia [1], including a list of talles peaks by prominence. By definition, Mount Everest wins again, as there is no col to be considered it's parent.
The Muana Kea claim is highly questionable. The only complete and sound definition of the "base" of a mountain would make the height the same as topographic prominence, and by definition the highest peak must also have the greatest prominence (whereas other peaks will vary wildly in ordering).
I can put my phone in my plane later today and host a website from 6 km if it'll work, though mobile antennas don't really aim high so they don't work well at altitude.
(If it was, would a website hosted on Olympus Mons (Mars) be higher or lower than one hosted on Everest? Closer to its plant's core, further from the sun, not sure about total relative gravitational potential…)
At best you could extend that argument to the karman line. Surely the atmosphere is considered part of earth (assuming the narrow definition of world to be planet)
Surely I am not departing the world by taking a flight to orlando. Otherwise, only subterranean servers would count.
I'm sure the ISS's public facing comms (I believe they host a public HAM repeater as well) are isolated. Also, there's no reason that something added well after Twitter (the first ISS direct communication was a tweet - before that tweets were related through a ground station) would be done in an obsolete way.
But lastly, unless they're hosting a forum, XSS is kinda irrelevant and 1995 predates XSS issues (and for most of the year JS)
I mean, talking to a server in Australia as the ground station, and having it launch a signal to and from the ISS (the "a bit") works. But it's connecting via some listening post like in Alice Springs, not a data center in Sydney, so you have to wonder about Australias internal internet service.
> Gecko-based browsers like Firefox show an interesting behavior when trying to make the site any higher than 18.939583 kilometers, “shrinking” or “collapsing” the main container.
I posted this on HN hoping to find an explanation. I tried all sorts of calculations, including measuring my screen size to find how many centimetres 1 pixel is but couldn't make sense of any it.
Small nitpick: an inch is 2.54 cm, not 2.56 cm. That makes the html element 284035.5 cm high, which I guess is the height of that "whws" div plus the height of text and margins above and below.
In any case about 28.4 m is a long way of from 18.94 km.
Although for css cm, browsers will use 37.8 px/cm instead of 96/2.54=37.7952 when making the conversion which gives the even 284000cm
The only way I can think of to get 18.94 km from the browser's interpreted 2.84 km is if the website is physically projected/printed on a 14.4 dpi display like a billboard
2. I can increase 2840em to somewhere below 5000em no problem on both Chrome and Firefox. Beyond that Firefox hides the element and Chrome cuts off at some point. Not sure if it's dependent on vram or something? Another interesting thing is Firefox's Layout tab shows a wrong, capped height in the Box Model diagram; currently says 8947820px for me, but it even changes from time to time.
Or you could use the browser/windows(?) shortcut Ctrl+Home/Ctrl+End, or Ctrl+Up/Ctrl+Down. OR just use the 'take the elevator' link on the page itself.
This site is very, very old. I believe it predates Google Chrome.
It's based on a rendering limitation found in Firefox back then. It never really worked in Internet Explorer - and Firefox (as well as Chrome) can render much higher websites these days.
For those on mobile, here's the end of the webpage after all the scrolling:
Awesome. Restart? Scroll up (we’ll wait for you), use the elevator, or press Ctrl and Pos1.
For everyone else there’s some fine print.
This website is a CSS experiment. Due to its experimental nature, there are some accessibility limitations.
The technical and design principles of this site are simple:
one HTML container;
pure CSS styling;
no workarounds, filters, or hacks.
Known issues
Gecko-based browsers like Firefox show an interesting behavior when trying to make the site any higher than 18.939583 kilometers, “shrinking” or “collapsing” the main container.
Internet Explorer has problems following internal links, and it’s unclear whether the container actually is said 18.94 kilometers high (or long, or tall).
If you do come up with fixes or an even higher element whose formatting is broadly supported, email info@worlds-highest-website.com. There may be a reward.
WHWS is presented by web artist Jens Oliver Meiert and the secret “Flying Standardistas Club.”
Design: Alessandro Lettieri.
Contact: Jens Oliver Meiert (Hamburg, Germany) · info@worlds-highest-website.com · +1-754-400-0999.
The font rendering in Firefox on macOS (2x dpi scaling) is all "wobbly" - the glyphs are on inconsisent baselines, varying by a few pixels (but only the text at the bottom, not at the top). I suppose this is due to floating point precision/rounding errors, which is interesting to see.
Also, the "elevator" back to the top only takes me about 80% of the way there, which is probably due to something similar.
It took me less than 2 minutes of scrolling to reach the bottom. This implies I'm scrolling at 568 kph or 353 mph ! This seems impossible. Either this site is not really 18.94 km long, or Chrome on Android implements scrolling acceleration to such speeds which I don't believe it does.
100 comments
[ 3.6 ms ] story [ 167 ms ] threadNow it's the world's widest website!
So perhaps if someone could rig up a microwave link to a solar powered Raspberry Pi Zero hanging off of a weather balloon or something.. we could have a truly publicly accessible Web server at above jet-liner height? :-D This feels like something Tom Scott might try..
There's Muana Kea, which is the _tallest_, meaning the furthest distance from the mountain's base to its peak.
There's Mount Everest (which I had assumed was the agreed upon "highest"), which has its peak at the highest altitude.
Chimborazo has the most unexpected definition of "highest", which is a peak that is furthest from the center of the Earth! We don't live on a perfect sphere, so some mountains can be "shorter" but still further from the core!
Would be interesting to see if the "highest websites" differ for each of these definitions!
[1]: https://geology.com/records/highest-mountain-in-the-world.sh...
Chimborazo is a fun one and I would accept a strident enough argument for it to be #1. It truly looks like a magnificent mountain too, indeed perhaps the archetypal 'mountain', if we can take aesthetics into account ;-)
This is due to the definition of prominence, which is roughly "how far down do you have to go from the summit before you can go up to a higher point." This is undefined for the highest mountain, since there is no higher mountain to go up to, so height above sea level is used instead.
> prominence measures the height of a mountain or hill's summit relative to the lowest contour line encircling it but containing no higher summit within it
By definition, this means that the highest mountain's height is its prominence.
I had thought Denali in Alaska held this honor, but apparently the base of Mauna Kea is considered to be the ocean floor which makes it taller. Interesting.
[1] https://en.wikipedia.org/wiki/Topographic_prominence
(If it was, would a website hosted on Olympus Mons (Mars) be higher or lower than one hosted on Everest? Closer to its plant's core, further from the sun, not sure about total relative gravitational potential…)
Surely I am not departing the world by taking a flight to orlando. Otherwise, only subterranean servers would count.
But lastly, unless they're hosting a forum, XSS is kinda irrelevant and 1995 predates XSS issues (and for most of the year JS)
Of course, I am in Australia.
> Gecko-based browsers like Firefox show an interesting behavior when trying to make the site any higher than 18.939583 kilometers, “shrinking” or “collapsing” the main container.
Not necessarily related to what you're seeing.
I don't get this. Could someone not missing something explain?
Edit, to make the question clearer: The CSS of the 'spacer' div has:
Font size: 100cm;
Height: 2840em;
How does this result in 18.94km?
In any case about 28.4 m is a long way of from 18.94 km.
Although for css cm, browsers will use 37.8 px/cm instead of 96/2.54=37.7952 when making the conversion which gives the even 284000cm
The only way I can think of to get 18.94 km from the browser's interpreted 2.84 km is if the website is physically projected/printed on a 14.4 dpi display like a billboard
2. I can increase 2840em to somewhere below 5000em no problem on both Chrome and Firefox. Beyond that Firefox hides the element and Chrome cuts off at some point. Not sure if it's dependent on vram or something? Another interesting thing is Firefox's Layout tab shows a wrong, capped height in the Box Model diagram; currently says 8947820px for me, but it even changes from time to time.
https://www.youtube.com/watch?v=4J9MRYJz9-4
It may share the idea, but not the spirit.
It's based on a rendering limitation found in Firefox back then. It never really worked in Internet Explorer - and Firefox (as well as Chrome) can render much higher websites these days.
Actually you don't even need js
For instance:
git show 47d8c02bdbec91f8e0ce4b4453010ea9b7b241ef
commit 47d8c02bdbec91f8e0ce4b4453010ea9b7b241ef
Author: txxxxxxx <txxxxxxx@users.noreply.github.com>
Date: Sat Nov 30 17:25:32 2019 +0100
Awesome. Restart? Scroll up (we’ll wait for you), use the elevator, or press Ctrl and Pos1. For everyone else there’s some fine print.
This website is a CSS experiment. Due to its experimental nature, there are some accessibility limitations.
The technical and design principles of this site are simple:
one HTML container;
pure CSS styling;
no workarounds, filters, or hacks.
Known issues
Gecko-based browsers like Firefox show an interesting behavior when trying to make the site any higher than 18.939583 kilometers, “shrinking” or “collapsing” the main container.
Internet Explorer has problems following internal links, and it’s unclear whether the container actually is said 18.94 kilometers high (or long, or tall).
If you do come up with fixes or an even higher element whose formatting is broadly supported, email info@worlds-highest-website.com. There may be a reward.
WHWS is presented by web artist Jens Oliver Meiert and the secret “Flying Standardistas Club.”
Design: Alessandro Lettieri.
Contact: Jens Oliver Meiert (Hamburg, Germany) · info@worlds-highest-website.com · +1-754-400-0999.
Also, the "elevator" back to the top only takes me about 80% of the way there, which is probably due to something similar.