Feature request: all caches in a state PQ

[+steve0mania]

I haven't keep up with all of the recent discussions about users needing to update their PQ requests every month or else gc.com will disable them, but I am guessing this is to try to decrease some of the load on the servers.

 

An alternate way to decrease the load would be to have a few downloadable PQs available on the gc.com website.

 

For example, in Utah (and I would guess in many other states), users are frequently running multiple PQs per week to keep the entire state's worth of caches updated in their GSAK databases. I'm sure this has been brought up before, but I would suggest that gc.com make downloadable files with an entire state's (or country's, for smaller countries) list of caches. If this were updated frequently (weekly? daily?), many of us wouldn't need to run multiple PQs, but would simply download the appropriate file for our state. This would also make it easier if one were going on a trip to a new location, then one could just download the appropriate file for the travel destination.

[+alexrudd]

I agree - it would be very useful and would eliminate almost all my pocket queries except for going on trips.

 

Unfortunately, it's been suggested repeatedly and hasn't been implemented yet, nor does it look like it will be ever implemented.

[+LeGodFather]

It would eliminate most of my queries too. And, I would only need it twice a week - instead of my 9 pqs running on 2-3 days.. And, if the files is already built, no need for a PQ - just a download.

[+StarBrand]

Why must you have a whole state? Do you plan to cache the whole thing today? Just curious to know.

 

 

I usually can get a query of the route/area I plan to be at within 5 minutes and can be on my way. When you consider how often changes are made to the site and archived caches (not in PQs) - it just doesn't make sense to "need" that many caches in a download. Sure it might be nice but not really a "need".

 

Also think California...way too many in any one file. The 5 per day - 2500 cache limit exists for a reason.

[+darus67]

If there are 100 people in an area that run 5 maxed out PQs every day

that adds up to 250000 database hits.

 

If Groundspeak pre-generated a PQ containing 2500 waypoints every day and just shipped it out to those same 100 people, that would save 247500 database hits.

 

It seems to me that if they planned it out right, they could have a huge net win by making

canned PQs available, at least in some populated areas.

 

The problem is, I get the impression that Groundspeak has an aversion to having people maintain offline databases of caches like that. Whether it makes sense to us, or not, policy is policy.

[+LeGodFather]

Why must you have a whole state? Do you plan to cache the whole thing today? Just curious to know.

 

 

I usually can get a query of the route/area I plan to be at within 5 minutes and can be on my way. When you consider how often changes are made to the site and archived caches (not in PQs) - it just doesn't make sense to "need" that many caches in a download. Sure it might be nice but not really a "need".

 

Also think California...way too many in any one file. The 5 per day - 2500 cache limit exists for a reason.

You're right.. I don't need it. But, I love to have it and see how many caches are being published everyday/week.. and make all sorts of stats about it.

 

And.. when I get my 60csx, I will be able to convert them all into POIs and load'em all in my GPS!

[+welch]

Why must you have a whole state? Do you plan to cache the whole thing today? Just curious to know.

 

I usually can get a query of the route/area I plan to be at within 5 minutes and can be on my way. When you consider how often changes are made to the site and archived caches (not in PQs) - it just doesn't make sense to "need" that many caches in a download. Sure it might be nice but not really a "need".

 

Also think California...way too many in any one file. The 5 per day - 2500 cache limit exists for a reason.

 

Who cares why they want it?

If they want the whole state, they'll get it anyways. Only it will be broken up over days (or weeks) and involve many PQs. It would seem like there could be a benefit to figuring out what people (multiple people) want and giving it to them instead of having them all repeatly hitting the PQ system and grabbing little chunks each time. By which time they probably start over again to see whats changed since the first chunk

Just give them 5,000 or 10,000 waypoints and cut off the rest of their PQ for two days, or four, or whatever an equal amount of time would be. They get want they want, and the system doesn't have to build 9 or 19 more PQs. Yes some states have more than 5 or 10k cache, so maybe doing whole states won't work, maybe it states should be broken into large sections, or maybe just certain areas (big metros) should be the target. Whatever it is, there is bound to be redundent PQs that could be reduced with planning.

[+alexrudd]

Why must you have a whole state? Do you plan to cache the whole thing today? Just curious to know.

I don't really want a whole state - just Chicago would be fine in my case. It's just that there aren't many other ways to geographically sort, and I can just delete the parts of the state I don't need. It could feasibly be broken down farther, it's just that states are rather convenient and all caches already have state data.

[+The Leprechauns]

Why must you have a whole state? Do you plan to cache the whole thing today? Just curious to know.

I don't really want a whole state - just Chicago would be fine in my case. It's just that there aren't many other ways to geographically sort, and I can just delete the parts of the state I don't need. It could feasibly be broken down farther, it's just that states are rather convenient and all caches already have state data.

I'm sorry, but downloading all Illinois caches strikes me as an odd way to obtain all the caches near Chicago. Do you have an aversion to caches in Indiana and Wisconsin? It seems more likely that you'd wander across those nearby state line boundaries before you would find a cache in Peoria or Springfield or East St. Louis. Unless you live near the middle of a regularly-shaped state, political subdivision boundaries are rather irrelevant for a game based on latitude and longitude coordinates.

 

I live in Pittsburgh, meaning I am less than 50 miles from West Virginia and Ohio -- not too different than your situation in Chicago. I order my pocket queries in a radius around my home coordinates, regardless of what state the caches are located in. To get around the problem of maxing out, there are a number of ways to divide up the queries. The most reliable is to split them up by the date of placement, which is a fairly constant number over time. Another way is "easy" caches (for cache run days) vs. "hard" caches (when I'm in the mood for just one or two challenging caches). Both methods avoid the inefficient "overlapping circles" method.

 

I have 300 cache finds in Eastern Ohio, but zero in the Scranton-Wilkes Barre area of Northeast Pennsylvania. I cannot fathom the need for downloading the many thousands of caches in areas of Pennsylvania that I am unlikely to visit. When I get around to making a roadtrip to Scranton, I will run a pocket query a week or so prior to my departure, then freshen it up at the last minute.

 

But that's just me.

[+budd-rdc]

Once people obtain the PQ for the entire state, they'll want to keep their databases "current" so they'll go back to Groundspeak and run some more PQs, especially since caches will become updated, archived, etc. Not sure how much hit the database will get when that happens, but I suspect it's not negligible. Never underestimate the pack rat mentality.

 

For states and countries that are cache sparse, this can already be done with very small number of PQs. For example, entire Japan (about the size of California) can be covered with 2 PQs or less.

[+Glenn]

I like the idea of pre-generated PQs. But I think they would be more useful if they were based on popular queries.

[+alexrudd]

I'm sorry, but downloading all Illinois caches strikes me as an odd way to obtain all the caches near Chicago. Do you have an aversion to caches in Indiana and Wisconsin?

<snip>

Currently I do exactly as you suggest, and I have a weekly set of PQ's that pulls in caches from my home coordinates, filtered by date placed. However, I could also get all of IL, WI, and IN and just delete the caches I don't want with very little additional effort with pregenerated queries.

 

Obviously, it's not a benefit if I'm the only person ever generating Pocket Queries. However, since I'm sure there are plently of other people in IL maintaining offline databases, it would be much more efficient to generate one query (even with excess information) than a separate set of queries for each person. Heck, it even pays off if I were to download WI, IL, IN, MI, IA, and MO. Even though that's 6 queries instead of five, the six are only created once and reused many times. Although states aren't the optimal solution, it'd be easy to implement and use.

 

Once people obtain the PQ for the entire state, they'll want to keep their databases "current" so they'll go back to Groundspeak and run some more PQs, especially since caches will become updated, archived, etc.

This wouldn't happen if the state PQ were generated daily, or even weekly if guaranteed to contain all the logs since the last week.

[+gcslim]

If there are 100 people in an area that run 5 maxed out PQs every day

that adds up to 250000 database hits.

 

If Groundspeak pre-generated a PQ containing 2500 waypoints every day and just shipped it out to those same 100 people, that would save 247500 database hits.

 

Actually, databases don't work like web servers, The database queries get all the info in one shot. 100 people with 5 PQs is only 500 database hits.

 

Assuming that Groundspeak has a decent database engine (Oracle, SQL Server, etc) the only strain that a 5000 cache query would have over a 5 cache query is the time the program would take formatting the database results into a GPX file.

 

I used to download statewide caches, but not so much anymore because I found that I regularly used only my 50 mile-or-so radius. It also became a problem when I would go somewhere else, I'd download a fresh PQ for the area and then I'd have bunches of archived caches mixed in with the fresh ones. Now I keep it nice and clean and I'm sure to clear our an area completely before merging in fresh data that I don't get archive notifications for.

 

The thing that I would like is the option of having maybe one 500+ PQ. perhaps the limit could be that it would only be able to run once a week and have a limit of 5000 caches or something. I went on vacation to western NC earlier this year and wasn't going to have internet access the whole time . I had a heck of a time trying to get caches for all the areas I was considering visiting since it spanned NC, SC, GA and TN. I didn't want the full state for all of them (so dates didn't work) and one PQ would only get me about 25 miles (thanks to ozguff! Great caches!). One 5000 cache query would have been perfect for the vacation.

[+darus67]

 

Actually, databases don't work like web servers, The database queries get all the info in one shot. 100 people with 5

PQs is only 500 database hits.

 

Assuming that Groundspeak has a decent database engine (Oracle, SQL Server, etc) the only strain that a 5000 cache query would have over a 5 cache query is the time the program would take formatting the database results into a GPX file.

 

 

So my numbers are off. It would still reduce the number of machine cycles by 99% if they could cache the GPX file when the first person generates it and then for the other 99 just ship it out.

 

Of course this only works if you have 100 people who are all happy using the same PQ. Maybe what we need are public PQs, in the same manner as public routes.

 

On the subject of routes, I can say that the advent of "Caches Along a Route" has eliminated my desire to download PQs that cover

huge areas.

[+budd-rdc]

Of course this only works if you have 100 people who are all happy using the same PQ. Maybe what we need are public PQs, in the same manner as public routes.

 

On the subject of routes, I can say that the advent of "Caches Along a Route" has eliminated my desire to download PQs that cover

huge areas.

I think the "Public PQ" idea might be the solution we are looking for. I like the idea.

[+sbell111]

Of course this only works if you have 100 people who are all happy using the same PQ. Maybe what we need are public PQs, in the same manner as public routes.

 

On the subject of routes, I can say that the advent of "Caches Along a Route" has eliminated my desire to download PQs that cover

huge areas.

I think the "Public PQ" idea might be the solution we are looking for. I like the idea.

Jeremy has been pretty clear on the statewide PQ idea in the past. I don't see it happening.

[+The Leprechauns]

I would prefer a customizable PQ over a generic one. I like having ALL my logs on a cache get returned with the PQ, which would not happen with a generic PQ. I like being able to exclude caches of a type that I don't like. And so forth. I have a nice database for the area where I am most likely to hunt for caches -- i.e., within 100 miles of my home. That area covers parts of four states.

 

But that's just me. I don't keep a vast offline database with dozens of previous logs. Somehow I've managed to find a few caches without having to read 30 logs deep.

[+darus67]

The public PQ idea is only a win for users if its cheaper than a regular PQ.

Something like, you can download 10 pre-generated PQs per day or the 5

regular PQs you get. The limitation is, you're stuck with whatever someone else has already run.