Jump to content

Waypoint ID's

TheAprilFools

By TheAprilFools
in Website

Followers 2

Recommended Posts

TheAprilFools

+TheAprilFools

Posted
Posted (edited)

I was wondering about what the plan was for the waypoint ID's in the future. I was doing some thinking on the subject and this is what I discovered.

 

For caches up to and including GCFFFF the last four digits of the waypoint ID is a hex number and there can be 65536 caches.

 

For caches starting GCG000 and later the numbers are all numbers and letters except I L O S and U, giving 31 numbers, resulting in 446865 possible cache ID's.

 

The first "Q" cache (GCQ000) was placed on 2005/8/1, the first "P" cache (GCP000) was placed 2005/5/20. Given that rate, unless the "GC" at the beginning of the waypoint ID or change the length of the waypoint ID we will out if ID's in March of 2007.

 

Is there a plan for what comes after GCZZZZ?

 

I know its not a problem for a little while but I was wondering.

Edited by Blanston12
Link to comment
Jeremy

Jeremy

Posted
Posted
Is there a plan for what comes after GCZZZZ?

It's exactly what happened when the Y2K bug arrived. Total and absolute chaos, people screaming, and explosions.

 

Then the fireworks ended and everyone went home.

 

Seriously, the worse that will happen is a digit will be appended to the end. Yes, many older GPS units have a 6 character limit, but the GC is just prepended to the front of the number anyway so it can be stripped for older units and added back for lookups.

Link to comment
TheAprilFools

+TheAprilFools

Posted
  • Author
Posted
I thought zeros werent used in the codes to avoid confusion with O.

 

Im so c0nfused. :lol:

Actually I thought that it was the O's (the letter between N and P) that is not used.

 

Thanks for the info Jeremy, hopefully the software developers will have enough time to fix there programs and all those old GPSr's will break and have to be replaced.

 

On a related note, the Y2K problem will be nothing to 2038/1/19, thats the date that TIME ENDS! :lol:

Link to comment
fizzymagic

+fizzymagic

Posted
Posted

Thanks for bringing this up. As you may know, GeoCalc does conversions between cache number and GC waypoint. It had a bug, though, and didn't deal with waypoints beyond GCZZZZ correctly. Seeing this topic spurred me to fix it and release a new version.

 

GCZZZZ = 512,400

GCZZZZZ = 28,218,030

 

We're not likely to get to 8-letter waypoint names any time soon.

 

The new GeoCalc also does rhumb line distances and projections, FWIW. :lol:

Link to comment
The Leprechauns

+The Leprechauns

Posted
Posted (edited)

"GC number" is used as a convenient shorthand by geocachers to identify caches. In a group on a long day of finding caches, you'll hear people ask, "what's the GC number for the next one down the trail?" I think consistency with that is more important than the six-digit limitation. I strip out the letters "GC" before loading waypoints onto my GPS, so that I can fit other information like difficulty and terain. As was noted above, people with a six-digit waypoint name limit could just as easily remove the "GC" in a batch process, leaving them with a five-digit name.

 

Also, "GD" is used as a waypoint identifier by Geodashing, another GPS game, so it wouldn't be nice to step on any toes.

Edited by The Leprechauns
Link to comment
sTeamTraen

+sTeamTraen

Posted
Posted
This reminds me of the famous quote by Bill Gates: "Nobody will ever need more than 640 KB RAM." :anibad:

I believe the following is attributed to the mayor of Cleveland (?) on seeing a demonstration of the first telephone:

 

"I can see a time when every city will have one".

 

I think Bill Gates' quote was "640K ought to be enough for anybody", which if you're running MS-DOS isn't too far from the truth. Wow, did I just defend something Bill Gates once sent ? Bring on the penguins...

Link to comment
eigengott

+eigengott

Posted
Posted
This reminds me of the famous quote by Bill Gates: "Nobody will ever need more than 640 KB RAM." :)

Good point! That's why one should calculate the upper limit of such things...

 

OK, according to http://en.wikipedia.org/wiki/Earth we have about 148939063 km^2 of dry land. 1 km^2 is 2.566404 sqm, so we have a total of 58034145 sqm on this planet to place caches on.

 

Considering the 0.1 miles rule, we have 100 caches per square mile (OK, this is simplified, with a triangular pattern you could fit in some more caches). Makes a total of 5803414500 caches on this planet.

 

With a base 36 system this get's you log 5803414500 / log 36 ~ 6.27 digits at most. Rounded up and added the GC prefix, this leads to a maximum of 9 digits for the waypoint id - if we confine geocaching to this planet and within the 0.1 mile rule...

 

:)

Link to comment
GrizzlyJohn

+GrizzlyJohn

Posted
Posted (edited)

Why dig up such an old thread?

 

Because if he started a new thread he would have:

 

A) been markwelled back to this one

B) people telling him to use the search function

C) his new thread locked because this topic was already being discussed in another thread.

D) all of the above

Edited by GrizzlyJohn
Link to comment
Markwell

+Markwell

Posted
Posted

A little more detail: sure - some six digit codes are on caches that were never published and will never be. But since we're on GCTC** (345187-346147) right now, it's not too terribly pressing. According to the trends of growth that we've seen in the number of caches submitted since August 2005 (when cache 275000 entered the system), GCZZZZ (cache 512400) will be entered into the system sometime around late April or early May of 2007 (although that may be slowed by the demise of new virtuals and locationless caches).

Link to comment
Airmapper

+Airmapper

Posted
Posted (edited)

I'm not sure, but aren't some letter combinations eliminated. I mean you would have caches like GCCRAP, that being on the nice side of the possibilities. I assume there is something that prevents GC codes with interesting (or rather offensive) combinations of "four letters."

Edited by Airmapper
Link to comment
DavidMac

+DavidMac

Posted
Posted (edited)

I actually meant:

* archived

* deleted for any reason

* never used before

 

When managing a database, it is critical that all entries (the cache listings, in this case) be set apart with a unique identifier. In the case of GC.com, each listing has a number that's assigned to it in chronological order (AFAIK, not the GCxxxx waypoint, but an integer, as mentioned in Markwell's post above). You can lookup a cache by ID (number ID, not GCxxxx ID) using the following URL (replace the X's with an integer):

 

http://www.geocaching.com/seek/cache_details.aspx?id=xxxxxx

 

I think the GC waypoint ID is derived from this number using some complex formula that drops the letters I, O, U, L, and S. Originally, the generated GC waypoint ID was hex, but other letters were incorporated after GCFFFF. Therefore, ID=8 -> GC8, ID=30 -> GC1E, and ID=20000 -> GC4E20. You can try it with the URL above- it works up to GCFFFF, when the site added extra letters (they went away from a base 16 system and used a base 20-something system for waypoint IDs, if that makes sense).

 

This is why there aren't really any "unused" IDs. Each cache listing submitted, whether it is approved or not, is given an entry in a database. As of this morning, the site is up to more than 348,000 listings, and although many of those are archived, retracted, or never approved, they still exist in the database somewhere. It's considered bad practice to attempt to overwrite an old "unwanted" entry (say, an archived cache listing) with new data; instead, the unwanted entry is hidden (for example, a cache is archived or retracted) and new entries are tacked on to the end of the list. Even if you edit your cache to remove the description, hints, etc before archiving, you wouldn't be deleting the entry, you would just end up with an entry that contains no data and is hidden from view.

 

If that makes sense, it means that every cache listing from #1 to #340,000 contains some data, so every GCxxx waypoint up to the newest one has been "used". If you try to look up an unpublished or retracted cache, you'll probably get an error reading, "Sorry, you cannot view this cache listing until it has been published". Something is there, but it's just hidden from view unless you're an admin or you get a chance to peek at the actual database itself.

 

Of course, there are waypoint IDs that are not linked to a listing (GCCHJK, for example), but to go back and find them would be very difficult, as would implimenting a system that would assign them to an ID. There probably aren't enough of them to make it worth the effort, since they'd even run out of them after a short time.

Edited by DavidMac
Link to comment
Allen_L

+Allen_L

Posted
Posted

And of course it is always provided without a source.

 

The fact is that the memory limitation was hardware based: The intel 8086 processor imposed a toal 1MB memory space limit. the inflexible hardware architechture of the early IBM PC only allowed 640KB of that to be used by RAM

Link to comment
GizmoGuy411

+GizmoGuy411

Posted
Posted (edited)

Back to the topic.....

 

Here is an Excel formula you can use to convert to and from the waypoint name and the cache id number.

 

First add this 31 character sting to the cell at position A4:

0123456789ABCDEFGHJKMNPQRTVWXYZ

 

Add this text to cell A8: Input Cache ID Number

Add this text to cell B8: Input Waypoint Name Suffix

Add this text to cell C8: ID Result

Add this test to cell C8: WP Result

 

Add this formula to cell C10:

=IF($B10<"G000",(FIND(MID($B10,1,1),$A$4,1)-1)*16^3+(FIND(MID($B10,2,1),$A$4,1)-1)*16^2+(FIND(MID($B10,3,1),$A$4,1)-1)*16^1+(FIND(MID($B10,4,1),$A$4,1)-1),((FIND(MID($B10,1,1),$A$4,1)-1)-16)*31^3+(FIND(MID($B10,2,1),$A$4,1)-1)*31^2+(FIND(MID($B10,3,1),$A$4,1)-1)*31^1+(FIND(MID($B10,4,1),$A$4,1)-1)+65536)

 

Add this forumla to cell D10:

=IF($A10<65536,MID($A$4,MOD(INT($A10/16^3),16)+1,1)&MID($A$4,MOD(INT($A10/16^2),16)+1,1)&MID($A$4,MOD(INT($A10/16^1),16)+1,1)&MID($A$4,MOD(INT($A10/16^0),16)+1,1),MID(($A$4),MOD(INT(($A10-65536)/31^3),31)+16+1,1)&MID(($A$4),MOD(INT(($A10-65536)/31^2),31)+1,1)&MID(($A$4),MOD(INT(($A10-65536)/31^1),31)+1,1)&MID(($A$4),MOD(INT(($A10-65536)/31^0),31)+1,1))

 

This should convert correctly between the ID and the WP for the old Hexadecial WP naming system and the current Base 31 system.

 

You must input WP letters in capitals only, as I did not provide for case conversion.

 

 

Note that the orginal Hexidecimal naming system used the number symbols 0-9 and the letter symbols A-F.

The current system still uses the Hexidecimal system up to waypoint "GCFFFF", then starting with "GCG000" it cleverly switches to a Base 31 numbering system using the number symbols 0-9 and the letters symbols A-Z, with the letters "ILOSU" omitted to obviously avoid confusion with the allowed symbols, "1l05V".

Just think, "I Love OSU" to remember them. :laughing:

 

It was pretty clever of Jeremy to invent this way of expanding the naming range without disrupting the original naming convention! Since "GC0000" never seemed to exist, the range changed from a limit of 65535 to 512400.

 

Once I see how the "NEW" system will work I'll modify the Excel formulas.

 

I am curious if the old waypoints will still be represented as "GCxxxx" or as "GC0xxxx".

 

It may also be nice if the website added an option to strip the "GC" from the waypoint names for those downloading from the website to GPS units with only 6 waypoint character positions.

 

My quick calculations today, show about 483 waypoint names were used per day since the beginning of the year. If that rate remains steady, we should see "GC10000" at about mid January 2007! My hunch is it will happen a lot sooner.

 

Maybe someone will offer a prize to whoever snags that pivital Waypoint name!

 

(Please don't flame me here if my quick calculations are off, just be sure to let me know if they are!)

Edited by GizmoGuy411
Link to comment
Kryten

+Kryten

Posted
Posted

Waypoint IDs lengths need to be kept to 6 characters because there a lot of GPSr units out there that can't accept anything longer. Waypoint IDs also need to be unique because GPSr units have a nasty habbit of over overwriting duplicate waypoints without warning the user. If the unique component of a waypoint ID is now to be 5 characters in length then the new additional waypoints will need to contain the 5 unique characters belonging to their parent. This then leaves only a single character to differentiate between the additional waypoints belonging to any one cache. At present we ask the user to name additional waypoints by entering a 2 character code (which we use) and a 4 character lookup (which we ignore). It would be better, and less confusing, to stop asking for this information and instead allocate the single remaining character automatically.

Link to comment
SAWKS

SAWKS

Posted
Posted

Back to the topic.....

 

Here is an Excel formula you can use to convert to and from the waypoint name and the cache id number.

 

First add this 31 character sting to the cell at position A4:

0123456789ABCDEFGHJKMNPQRTVWXYZ

 

Add this text to cell A8: Input Cache ID Number

Add this text to cell B8: Input Waypoint Name Suffix

Add this text to cell C8: ID Result

Add this test to cell C8: WP Result

 

Add this formula to cell C10:

=IF($B10<"G000",(FIND(MID($B10,1,1),$A$4,1)-1)*16^3+(FIND(MID($B10,2,1),$A$4,1)-1)*16^2+(FIND(MID($B10,3,1),$A$4,1)-1)*16^1+(FIND(MID($B10,4,1),$A$4,1)-1),((FIND(MID($B10,1,1),$A$4,1)-1)-16)*31^3+(FIND(MID($B10,2,1),$A$4,1)-1)*31^2+(FIND(MID($B10,3,1),$A$4,1)-1)*31^1+(FIND(MID($B10,4,1),$A$4,1)-1)+65536)

 

Add this forumla to cell D10:

=IF($A10<65536,MID($A$4,MOD(INT($A10/16^3),16)+1,1)&MID($A$4,MOD(INT($A10/16^2),16)+1,1)&MID($A$4,MOD(INT($A10/16^1),16)+1,1)&MID($A$4,MOD(INT($A10/16^0),16)+1,1),MID(($A$4),MOD(INT(($A10-65536)/31^3),31)+16+1,1)&MID(($A$4),MOD(INT(($A10-65536)/31^2),31)+1,1)&MID(($A$4),MOD(INT(($A10-65536)/31^1),31)+1,1)&MID(($A$4),MOD(INT(($A10-65536)/31^0),31)+1,1))

 

This should convert correctly between the ID and the WP for the old Hexadecial WP naming system and the current Base 31 system.

 

You must input WP letters in capitals only, as I did not provide for case conversion.

 

 

Note that the orginal Hexidecimal naming system used the number symbols 0-9 and the letter symbols A-F.

The current system still uses the Hexidecimal system up to waypoint "GCFFFF", then starting with "GCG000" it cleverly switches to a Base 31 numbering system using the number symbols 0-9 and the letters symbols A-Z, with the letters "ILOSU" omitted to obviously avoid confusion with the allowed symbols, "1l05V".

Just think, "I Love OSU" to remember them. :laughing:

 

It was pretty clever of Jeremy to invent this way of expanding the naming range without disrupting the original naming convention! Since "GC0000" never seemed to exist, the range changed from a limit of 65535 to 512400.

 

Once I see how the "NEW" system will work I'll modify the Excel formulas.

 

I am curious if the old waypoints will still be represented as "GCxxxx" or as "GC0xxxx".

 

It may also be nice if the website added an option to strip the "GC" from the waypoint names for those downloading from the website to GPS units with only 6 waypoint character positions.

 

My quick calculations today, show about 483 waypoint names were used per day since the beginning of the year. If that rate remains steady, we should see "GC10000" at about mid January 2007! My hunch is it will happen a lot sooner.

 

Maybe someone will offer a prize to whoever snags that pivital Waypoint name!

 

(Please don't flame me here if my quick calculations are off, just be sure to let me know if they are!)

WOW! :laughing::laughing:

Link to comment
Guest
This topic is now closed to further replies.
Followers 2
Go to topic listing
×
×
  • Create New...