Return to Unfiction unforum
 a.r.g.b.b 
FAQ FAQ   Search Search 
 
Welcome!
New users, PLEASE read these forum guidelines. New posters, SEARCH before posting and read these rules before posting your killer new campaign. New players may also wish to peruse the ARG Player Tutorial.

All users must abide by the Terms of Service.
Website Restoration Project
This archiving project is a collaboration between Unfiction and Sean Stacey (SpaceBass), Brian Enigma (BrianEnigma), and Laura E. Hall (lehall) with
the Center for Immersive Arts.
Announcements
This is a static snapshot of the
Unfiction forums, as of
July 23, 2017.
This site is intended as an archive to chronicle the history of Alternate Reality Games.
  

The time now is Sun Nov 17, 2024 7:32 am
All times are UTC - 4 (DST in action)		 View posts in this forum since last visit
View unanswered posts in this forum
Calendar
 Forum index » Archive » Archive: The Haunted Apiary (Let Op!) » The Haunted Apiary (Let Op!): Puzzles
[INFO] Problem with mapping with aerial and street maps
View previous topicView next topic
Page 1 of 2 [17 Posts]   Goto page: 1, 2 Next
Author Message
hamatoyoshi
Veteran

Joined: 02 Aug 2004
Posts: 127

 		[INFO] Problem with mapping with aerial and street maps
It seems to me that there's a lot speculation on the exact point of the coordinates based on the accuracy of the decimal degree data.

The major problem we have with these coordinates is that we don't know what datum they are in.

Info for you all:

What's a datum?

Why is there more than one datum?

And super-technical Converting between datums.

So what does this mean? Most GPS units can function in severals datums, generally one known as WGS84. Most USGS maps are in NAD27. Mapquest does not specify, but it is speculated that they mix WGS84 and NAD27.

Depending on where you are on Earth, converting coordinates between datums can result in over a kilometer of difference of location depending on which datum you are in. Most GPS units have an internal calculator that allows you to switch between datums, try switching yours and see where you end up with the same coordinates.

Here's a coordinate translation site.

Honestly, I'm guessing that WGS84 is the datum we are operating in based on the experience of people visiting points in Chicago with their GPS units. My asssumptions on that last comment: GPS tends to be by default WGS84, and the Sears Tower, if any specific locations are in fact important, seems to be a specific location.

We can't know if the PMs got this data from a map, or from walking around with GPS units, and to analyze those locations, we cannot tell what datum our street maps are in unless they are explicitly named.

PostPosted: Mon Aug 16, 2004 7:42 pm
 View user's profile
 Back to top 
usernameguy
Boot

Joined: 15 Aug 2004
Posts: 68

 		Here's the thing, though
Here's the thing, though: Do the PM's realize this? This is pretty esoteric. Not many people get that.

If they didn't, and they intended us to know what was exactly at those spots, we are pretty SOL. There's is nearly zero way to tell what's there.

Personally I think they screwed up, and assumed GPS coordinates not only uniquely identified buildings, but they also assumed we could just use a reverse geocoder to check stuff up.

I also think the problem is somewhat surmountable. If the there are a limited number of sources, and one source always returns sensisical results (eg always exactly on buildings, not in the street) then we know which source they used.

PostPosted: Mon Aug 16, 2004 9:27 pm
 View user's profile
 Back to top 
Extrasonic
Decorated


Joined: 03 Aug 2004
Posts: 233
Location: Suburban Chicago

 		Re: Here's the thing, though
usernameguy wrote:
Here's the thing, though: Do the PM's realize this? This is pretty esoteric. Not many people get that.

If they didn't, and they intended us to know what was exactly at those spots, we are pretty SOL. There's is nearly zero way to tell what's there.


Just to follow through on your thought - if the PMs did understand that, then that lends some credibility to all of the non-"real space" theories floating around here - e.g. that the coordinates aren't really coordinates but some sort of math/word puzzle. As many in the "hey, these aren't coordinates" camp frequently point out, Dana's latest blog entry says, "Lesson learned: Looks can be deceiving here, even when dealing with something as simple as the donut", which they suggest is a not-so-subtle hint that we're not looking at what we think we're looking at.

PostPosted: Mon Aug 16, 2004 9:49 pm
 View user's profile
 Back to top 
hamatoyoshi
Veteran

Joined: 02 Aug 2004
Posts: 127

 		Re: Here's the thing, though
usernameguy wrote:
Here's the thing, though: Do the PM's realize this? This is pretty esoteric. Not many people get that.

If they didn't, and they intended us to know what was exactly at those spots, we are pretty SOL. There's is nearly zero way to tell what's there.

Personally I think they screwed up, and assumed GPS coordinates not only uniquely identified buildings, but they also assumed we could just use a reverse geocoder to check stuff up.

I also think the problem is somewhat surmountable. If the there are a limited number of sources, and one source always returns sensisical results (eg always exactly on buildings, not in the street) then we know which source they used.


Well, I think it all depends on their purposes and how they collected their data.

As far as their purposes go, individual spots may not be especially significant at all, but we simply don't know yet.

Supposing they sent someone to every spot with a GPS unit, we're pretty safe with our GPS units and WGS84. Also, the company that contributes the majority of Mapquest, Navteq, data uses WGS84 I understand.

However, if they picked their spots off of detailed maps, then those maps would most likely be WGS84, NAD27, or NAD83. We can't be sure unless we know their sources. Potentially they could have a mix of sources.

The coordinates can be converted between the formats however. Here's an interesting looking tool to do it as well.

As it is though, I think based on some basic location reports with individuals with GPS, that they're probably operating off of WGS84, but depending on how they picked their spots, it could be a mix and the Chicago area just happens to come up correctly in our books.

Another thing I always questioned was why they used Decimal-degrees as the unit when they could have used Degrees decimal-minutes as used by many GPS units and is used at The International GPS Waypoint Registry. The thing I found is that of the handful of coordinates I converted and tried to look up, none are listed in the registry. Many of the registry spots are "significant" spots and landmarks.

And the last thing I thought of is that they would have had to have clear access to 12 GPS satellites to get the < 15 meter accuracy that they are specifying on the coordinates. Most GPS units these days can receive 12 satellite signals, but I'm not really sure how feasible it would be to get reception to 12 satellites for 220 points. Losing one significant figure of accuracy widens the accuracy radius from around 13 meters to over 100 meters.

PostPosted: Mon Aug 16, 2004 10:15 pm
 View user's profile
 Back to top 
usernameguy
Boot

Joined: 15 Aug 2004
Posts: 68

 		I personally think they:

a) found buildings/organizations/spots that fit their criteria
b) found their addresses, or itheir locations on TerraServer or something similiar, and
c) used either a geocoder or TerraServer to convert to GPS coordinates.

and, of course,
d) didn't realize there's this inherent ambiguity in GPS.

oh, and
e) didn't realize that geocoders don't pinpoint, they interpolate.

I think there was meant to be meaning in each location, kinda like the links puzzle, but that meaning is lost when the GPS gets screwed up. Which is why we're having such problems with this puzzle. Maybe we should email Dana in some vaguely out-of-game way asking her about this.

PostPosted: Mon Aug 16, 2004 10:38 pm
 View user's profile
 Back to top 
SpghEddy
Veteran


Joined: 24 Jul 2004
Posts: 88

 		I think it's possible that they did not physically visit these locations, but did an address -> GPS lookup using some service.

Wishi-san did a reverse address lookup which has been used by many ARGers as the definitive guide to the original addresses. However, numerous errors have been found in that data (where two roads intersect, for instance, or when the physical addresses are different, since the addresses are just interpolations)

The only way we can really know what range of addresses the PMs started with would be to do a forward (address -> GPS) lookup which matches the GPS coordinates.

So far, I have tried terraserver.com, terraserver.microsoft.com, and maporama.com.
The first two have over four decimal places of precision, but maporama has exactly four, which leads me to believe that it might use the same database that the PMs used.

Once we find the right street and range of addresses, we still need to discover whether it's features of the actual place that matter, or just a code that uses the address itself.

PostPosted: Mon Aug 16, 2004 11:41 pm
 View user's profile
 Back to top 
usernameguy
Boot

Joined: 15 Aug 2004
Posts: 68

 		Wow, SpghEddy, I think you may have it. I just put in my hand-tuned resolution for 40.7699 -73.9847, 113 W 60th St, New York, NY into maporama.com, and it returned: 40.7699 , -73.9847. Exactly right.

Good job!

I think this is it. The PM's checked up "Fordham University Lincoln Center", for some reason, on Google, then used maporama.com to translate that address to GPS.

PostPosted: Tue Aug 17, 2004 12:29 am
 View user's profile
 Back to top 
chaotic_mind
Unfettered


Joined: 03 Aug 2004
Posts: 325
Location: Inside my head, behind the eyes

 		What about those places that don't have address? The reverse address look up on the Wiki for my coord. is very wrong, placing it in a city a few miles away.

Gah, my coordinate is such a misbehaving black sheep.

(44.9784 -99.2239)

Luke P.

edit- I'm not even sure where to look up an address, short of going all the way out there, and I don't know if that'd help. It's not in a town, the coord. is part of a township.

PostPosted: Tue Aug 17, 2004 12:35 am
Last edited by chaotic_mind on Tue Aug 17, 2004 12:44 am; edited 2 times in total
 View user's profile MSN Messenger
 Back to top 
usernameguy
Boot

Joined: 15 Aug 2004
Posts: 68

 		I think they either moved the center of the map from maporama onto the point they wanted, or they used another service (unlikely).

How would Melissa, The Flea, The Princess, or the SPDR get a set of GPS addresses? Not by going out in the field.

PostPosted: Tue Aug 17, 2004 12:40 am
 View user's profile
 Back to top 
jbd
Decorated

Joined: 27 Jul 2004
Posts: 295

 		I think it's possible there was a mixed use of methods (both reverse lookup and hand-checking).

I think (at least based on the logic of what the AI would be capable of) anything not referenced by the Internet would at least have to be viewable by some satellite system (including, say, a secret one that requires hacking to get into). Do we have any points that are definitely indoors? I heard of a couple that were possible but nothing certain.

(Of course, in that case the indoors ones might be simply referencing the structure, so I guess that still doesn't kill the theory altogether.)

PostPosted: Tue Aug 17, 2004 1:51 am
 View user's profile
 Back to top 
xnbomb
Unfettered


Joined: 13 Oct 2003
Posts: 660
Location: J302B S8JDC

 		Re: Here's the thing, though
hamatoyoshi wrote:
The major problem we have with these coordinates is that we don't know what datum they are in.

If for no other reason, I do like this coordinate conundrum because it is contributing to the geographic literacy of so many Smile . Hamatoyoshi is right, when we're talking about points that specified as precisely as these are, the datum used is a significant issue.

I agree with hamatoyoshi's speculation that if these coordinates are meant to be used in a GPS context, WGS84 is most likely the datum that has been used.

hamatoyoshi wrote:
As it is though, I think based on some basic location reports with individuals with GPS, that they're probably operating off of WGS84, but depending on how they picked their spots, it could be a mix and the Chicago area just happens to come up correctly in our books.

It's worth noting, however, that just because sets of coordinates have been specified, that doesn't necessarily mean that they were selected/located using GPS. These locations could just as easily have been selected/located using high resolution imagery and/or street coverages. Our coordinate problem is not necessarily a GPS problem, although GPS is a convenient way to find these locations in the field.

I think it's safe to assume that by giving us these sets of coordinates, some players would feel compelled to visit them to see what's there. It only makes sense that the coordinates be provided in a format and datum that would allow us to find the locations intended (if that was the intention).

hamatoyoshi wrote:
Another thing I always questioned was why they used Decimal-degrees as the unit when they could have used Degrees decimal-minutes as used by many GPS units and is used at The International GPS Waypoint Registry

That's got me wondering too. The one nice thing about DD is you can store the coordinates in standard numeric database formats, whereas DM or DMS you either need some sort of more specialized data structure, or you need to store them as strings. If you have a lat/long coverage and placed a bunch of points on it, and then extracted coordinates for those points into some kind of table, DD is ideal. It also opens the door to all sorts of numeric interpretations of what to do with the coordinates ...

It's important that our field investigators understand these subtleties, though. I suspect that some of the confusion we've seen when two people come up with different things at the same coordinates is a results of someone not knowing how to read decimal degrees coordinates and assuming that they can take the digits and read them as DMS.

hamatoyoshi wrote:
And the last thing I thought of is that they would have had to have clear access to 12 GPS satellites to get the < 15 meter accuracy that they are specifying on the coordinates.

You're always going to have trouble getting a reading good enough to reproduce the 4 decimal places specified in these coordinates with consumer equipment, especially in urban environments with tall buildings where lots of GPS satellites are going to be obscured. Yes, more satellites in your constellation helps you get better accuracy sooner, but so does sitting at the same location for a long time and taking lots of readings and averaging them out. I take your point, though, and do wonder how they expected us to find these locations (through GPS in the field, or through the use of online and other electronic mapping resources beforehand to find landmarks, like street addresses, that would allow us to easily locate these points in the field).
_________________
My location is a little tricky, but sooner or later, you'll get the knack.

{J302B S8JDC, 8996N M8L4W, 92D40 Q1JX5, 4PPRN R2B97, 8DC7C NZJNV, 8CH7V Q891H}

PostPosted: Tue Aug 17, 2004 3:03 am
 View user's profile
 Back to top 
SpghEddy
Veteran


Joined: 24 Jul 2004
Posts: 88

 		If the PMs were just doing Address -> GPS lookups on some site, it doesn't really matter what datum they're using, as long as we can figure out which site it is.

Leaving aside the coordinates in Sedalia, SD for now, I'm really starting to think that maporama.com might be the answer. So far, Usernameguy found out that one of the NY coordinates is *exactly* matched by typing in the address for the Fordham University Lincoln Center.

I picked two coordinates at random, and was also able to come up with better addresses, that also seem to point to something.

For the KINGMAN, AZ coordinate, I found a different address, 3320, which matches exactly, and also is the location of a Walmart.

Interestingly, for the Berkeley, CA row 88, col 1 coordinate, while Wishi-san originally found a Shattuck st. address, Maporama finds a perfectly matching address on Kittredge, which just so happens to be the Berkeley Central Public Library!

This seems to be progress, though I have to admit that it is annoying to do these lookups... you basically have to tweak the address until you get the resulting GPS just right. As soon as the Wiki is back up, if anyone wants to help, I'll set up a page for Maporama-derived addresses.

PostPosted: Tue Aug 17, 2004 4:29 am
 View user's profile
 Back to top 
kalamazoo
Veteran

Joined: 24 Jul 2004
Posts: 83

 		And interestingly, some of the coordinates DO NOT change when you switch from mapquest to maporama. An exhaustive search on (40.7520, -73.9819) using Maporama found it at W 40th St by 5th Ave, NYC: the New York Public Library, just 2 blocks off the location given by Mapquest.

Still, this is interesting... I had noticed that famous landmarks were concentrated in the East and were not well represented in the West. Let's look at the Seattle, San Fran, and LA points to see if they hit known landmarks.

EDIT: Spot check on (47.612, -122.3353), listed as 1577 6th Ave., Seattle; Maporama pinpoints at 6th Ave. and Pine Street (600 Pine; 1600 6th Ave.), exactly on target.

EDIT: Spot Check on (37.7928, -122.3968), listed as 224 Market St., San Fran; Maporama pinpoints on north side of Market Street, a bit west of the corner of Main and Market: approximately 260 Market Street.

EDIT: Spot Check on (34.1015, -118.3405), listed as 6927 Hollywood Blvd (just east of La Brea Ave); Maporama pinpoints precisely as listed.

EDIT: Spot Check on (37.8684, -122.2679), listed as Shattuck St., Berkeley; Maporama pinpoints at Shattuck & Kittredge, NW corner, near 2098 Kittredge. 2090 Kittredge is indeed the Berkeley Public Library.

CONCLUSION: Maporama shows deviations of not more than one block from Mapquest results.

PostPosted: Tue Aug 17, 2004 8:19 am
 View user's profile
 Back to top 
Dorkmaster
Unfictologist


Joined: 27 Jul 2004
Posts: 1328
Location: The People's Republic of Dork

 		I agree with xnbomb mostly here... although everyone has had some great valid tangential points here! I'm in agreement that everything has been potentially off because of the potential datum switching that mapping programs can have. And nothing against Wishi-san's AMAZING work getting all those addresses, but I've been skeptical of the "addresses" from the beginning, since we weren't given addresses, we were given decimal-degree geo-points...

My only query with all this, (and it's been brought up before to a certain degree,) is this: There seems to be a double standard in terms of what we should be expecting the PMs to expect of us... We are either to all get GPS devices and track these locations as precisely as possible, or the PMs understand that maybe 1/10 (if even that ratio) players have access to GPS devices. My point is this: When we've got the datum right, of course, we then need to decide "how close is close enough"... and I don't know if we've been given enough info yet to accurately determine that... I think we should still be going to these points and logging our finds, but I still think the update today will be QUITE CLEAR in telling us what we will need to do with these "axon points"... That's my spec at least...
_________________
"The future is here. It's just not widely distributed yet." -William Gibson
"Always read stuff that will make you look good if you die in the middle of it." - PJ O'Rourke
"ACADEMY, n. A modern school where football is taught." - Ambrose Bierce

PostPosted: Tue Aug 17, 2004 10:32 am
 View user's profile Visit poster's website AIM Address
 Back to top 
devolver
Guest
kalamazoo wrote:

EDIT: Spot check on (47.612, -122.3353), listed as 1577 6th Ave., Seattle; Maporama pinpoints at 6th Ave. and Pine Street (600 Pine; 1600 6th Ave.), exactly on target.


However, the address makes a big difference, I believe, esp. for those of us (like me) who don't have any GPS equipment. For example, 1577 6th ave is a non-existant address; 600 6th Avenue is Pacific Place, a large shopping center in downtown Seattle. (For the record, that corner has Old Navy, ColdWater Creek, Nordstrom, and Pacific Place.) And the library talk up above got me thinking, because there is a Barnes and Noble in Pacific Place.

In addition, another of the Seattle addresses on the Wiki was 1021 E. John (or something--can't get to the Wiki right now). Anyway, it was also a non-existant address. However, at 905 E. John there is a bookstore, so I used maporama and it is much, much closer GPS-wise than the 1021 address. Maporama gives the coordinates as 47° 37' 11" , -122° 19' 14". On ILB, the coordinates for that location are 47.3722 -122.1914. Which seem pretty close to me, although I've just realized that the decimals probably mean a different way of measuring? Embarassed (This is what happens when you rely on street signs instead of GPS to get around.) Anyway, could be something to the book theory--I can think of a few other Seattle locations that, with a little fudging one way or the other, would be near or on a bookstore.

PostPosted: Tue Aug 17, 2004 11:43 am
 Back to top 
Display posts from previous:   Sort by:   
Page 1 of 2 [17 Posts]   Goto page: 1, 2 Next
View previous topicView next topic
 Forum index » Archive » Archive: The Haunted Apiary (Let Op!) » The Haunted Apiary (Let Op!): Puzzles
Jump to:  


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
You cannot attach files in this forum
You can download files in this forum
You cannot post calendar events in this forum



Powered by phpBB © 2001, 2005 phpBB Group