6 ms·

Australia's Entire GPS Navigation Is Off by 5 Feet

24 points by ghosh 10y ago

cocotino 10y ago
What is a 5 Feet?
- mavhc 10y ago
  1.524 metres
- jwn 10y ago
  It's approximately 5.6 Subway Footlong Subs.
- tfm 10y ago
  It's a bug that's lost its foot in the door. Previous discussion: https://news.ycombinator.com/item?id=12188144 https://news.ycombinator.com/item?id=12188144
- Intermernet 10y ago
  Simple. One and a quarter Wombats, or five eighths of a Funnel Web Spider.
avs733 10y ago
and shifting about 7cm per year. I did the math last night and realized that Australia has moved further than I am tall since my birth. For some reason that messed with my head.
- Klathmon 10y ago
  The thought that the ground we stand on grows more than we do is weird...
- jacquesm 10y ago
  Wait until it shifts underneath you while you're standing on it and see how that messes with your head. The first time I experienced a very minor earthquake the feeling did not leave me for weeks.
  - toomanybeersies 10y ago
    There's nothing as scary as being woken up by an earthquake in the middle of the night, and it feels and sounds like the apocalypse is happening. Daytime earthquakes aren't too bad, they're actually really cool if you're in the wilderness, you can see and hear it coming towards you.
  - avs733 10y ago
    I grew up in California...it has to be a big one to even wake me up nowadays.
    - jacquesm 10y ago
      Here's to hoping you sleep soundly then!
ungzd 10y ago
But GPS is using WGS84 datum (usually). And they're changing their own local datum. Nothing will be changed for GPS. It will still use WGS84. Or there are weird Australian upside-down phones and navigators where GPS calculates position in their local datum?
- tfm 10y ago
  Official local datum is GDA94, which is usually within about a metre of WGS84. /me pats pocket protector I think the idea here though is that the underlying terrain has scooted around a bit in the last twenty years, so a new datum would save on the need to print new maps and update the databases containing trig points and such. New coordinates can be labelled as explicitly aligned to the new datum, get some versioning happening. Unfortuately it seems impossible to confirm this, because the Geoscience Australia website does not appear to make any reference to this purported new effort that's been making the rounds of late. New website layout, maybe their CMS is broken.
  - ellisv 10y ago
    This is similar to the difference between WGS84 and NAD83 (North American Datum 1983). WGS84 is defined by the average of stations world wide while NAD83 is defined in a way that it is relatively constant across North America. NAD83 initially matched WGS84 but their difference increases each year by a small amount (negligible for most consumer use). Of course there are other differences between NAD83 and WGS84 as well (reference ellipsoids) and often they are treated as the same by non-cartographers.
abhikandoi2000 10y ago
"“We’re fast approaching the day when people will expect accuracies of centimeters in real time out of their handheld devices and then we’ll see a lot of head scratching as things no longer line up,” Smith told Scientific American three years ago. It looks like that day has arrived" Has that day really arrived? I doubt, given the fact that most consumer navigation systems are reliable (directions on Google Maps, pick up on Uber and alike) as long as they are not being blocked by major hurdles.
- rburhum 10y ago
  Think AR applications. Anything that requires accuracy that is not a "Pokemon Go" game would need that.
anotheryou 10y ago
I wonder if google maps and OSM will be more accurate before or after the change. (I guess after, be because the maps are probably based on satellite images)
- maxerickson 10y ago
  It depends on how that imagery has been rectified. If they were aligned to match up with the previous system, then any info derived from them would show the offset.
sctb 10y ago
Previously: https://news.ycombinator.com/item?id=12188144 https://news.ycombinator.com/item?id=12188144
ChrisFoster 10y ago
Ugh, all the recent articles I've seen on this are subtly or entirely wrong about the reasons for a datum change (though this article is a lot better than most). It's possible to account and correct for most of the 1.5 meters when transforming between GDA94 (the current Australian national datum) and ITRF2008 (effectively what's used by GPS), and the transformation published by Geoscience Australia is particularly simple, having only 14 parameters. If you want to know about this from a reliable source, the ICSM site is the place to go. They actually know what they're talking about, and their faq in particular explains a lot of points of jargon and convention rather clearly: http://www.icsm.gov.au/gda2020/faqs-2.html http://www.icsm.gov.au/gda2020/faqs-2.html

24 points by ghosh 10y ago

cocotino 10y ago

What is a 5 Feet?

mavhc 10y ago
1.524 metres
jwn 10y ago
It's approximately 5.6 Subway Footlong Subs.
tfm 10y ago
It's a bug that's lost its foot in the door. Previous discussion: https://news.ycombinator.com/item?id=12188144 https://news.ycombinator.com/item?id=12188144
Intermernet 10y ago
Simple. One and a quarter Wombats, or five eighths of a Funnel Web Spider.

avs733 10y ago

and shifting about 7cm per year. I did the math last night and realized that Australia has moved further than I am tall since my birth. For some reason that messed with my head.

Klathmon 10y ago
The thought that the ground we stand on grows more than we do is weird...
jacquesm 10y ago
Wait until it shifts underneath you while you're standing on it and see how that messes with your head. The first time I experienced a very minor earthquake the feeling did not leave me for weeks.
- toomanybeersies 10y ago
  There's nothing as scary as being woken up by an earthquake in the middle of the night, and it feels and sounds like the apocalypse is happening. Daytime earthquakes aren't too bad, they're actually really cool if you're in the wilderness, you can see and hear it coming towards you.
- avs733 10y ago
  I grew up in California...it has to be a big one to even wake me up nowadays.
  - jacquesm 10y ago
    Here's to hoping you sleep soundly then!

ungzd 10y ago

But GPS is using WGS84 datum (usually). And they're changing their own local datum. Nothing will be changed for GPS. It will still use WGS84. Or there are weird Australian upside-down phones and navigators where GPS calculates position in their local datum?

tfm 10y ago
Official local datum is GDA94, which is usually within about a metre of WGS84. /me pats pocket protector I think the idea here though is that the underlying terrain has scooted around a bit in the last twenty years, so a new datum would save on the need to print new maps and update the databases containing trig points and such. New coordinates can be labelled as explicitly aligned to the new datum, get some versioning happening. Unfortuately it seems impossible to confirm this, because the Geoscience Australia website does not appear to make any reference to this purported new effort that's been making the rounds of late. New website layout, maybe their CMS is broken.
- ellisv 10y ago
  This is similar to the difference between WGS84 and NAD83 (North American Datum 1983). WGS84 is defined by the average of stations world wide while NAD83 is defined in a way that it is relatively constant across North America. NAD83 initially matched WGS84 but their difference increases each year by a small amount (negligible for most consumer use). Of course there are other differences between NAD83 and WGS84 as well (reference ellipsoids) and often they are treated as the same by non-cartographers.

abhikandoi2000 10y ago

"“We’re fast approaching the day when people will expect accuracies of centimeters in real time out of their handheld devices and then we’ll see a lot of head scratching as things no longer line up,” Smith told Scientific American three years ago. It looks like that day has arrived" Has that day really arrived? I doubt, given the fact that most consumer navigation systems are reliable (directions on Google Maps, pick up on Uber and alike) as long as they are not being blocked by major hurdles.

rburhum 10y ago
Think AR applications. Anything that requires accuracy that is not a "Pokemon Go" game would need that.

anotheryou 10y ago

I wonder if google maps and OSM will be more accurate before or after the change. (I guess after, be because the maps are probably based on satellite images)

maxerickson 10y ago
It depends on how that imagery has been rectified. If they were aligned to match up with the previous system, then any info derived from them would show the offset.

sctb 10y ago

Previously: https://news.ycombinator.com/item?id=12188144 https://news.ycombinator.com/item?id=12188144

ChrisFoster 10y ago

Ugh, all the recent articles I've seen on this are subtly or entirely wrong about the reasons for a datum change (though this article is a lot better than most). It's possible to account and correct for most of the 1.5 meters when transforming between GDA94 (the current Australian national datum) and ITRF2008 (effectively what's used by GPS), and the transformation published by Geoscience Australia is particularly simple, having only 14 parameters. If you want to know about this from a reliable source, the ICSM site is the place to go. They actually know what they're talking about, and their faq in particular explains a lot of points of jargon and convention rather clearly: http://www.icsm.gov.au/gda2020/faqs-2.html http://www.icsm.gov.au/gda2020/faqs-2.html