Pocket Query production times

[+Tre-da]

It appears to me that all scheduled PQs are produced at midnight server time. This may be fine in the US and further west but for those of us that live east of the US then this means we do not get the PQ until the morning. I live in the UK and the erver is currently 8 hours behind us. This means that the earliest I get a PQ is 8am and often hours later.

 

If all PQs were sent at midnight local time to the user then this would spread the load on the server and allow us all to get the PQs in the very early hours of the morning before loading up for the day off caching.

 

I suggest that the PQ sending time is changed to midnight local time to the user.

[+dprovan]

I'm not arguing with you, but I find it amusing that I have the exact opposite problem: living here on the west coast of the US, if I schedule a PQ in advance, it will run at midnight. So I always run them manually about 8am in order to pick up any caches published that morning...

 

The perfect answer would be for us to specify an exact time our PQs should be run rather than just a date. As imperfect solutions go, the current one seems OK, but your suggestion might be a little better without being any more complicated for us or them.

[+Markwell]

Ah - but I think they chose midnight server time because that's when the server usages is at its lowest. Doesn't make sense to start running PQs at 2:00 on a Friday afternoon in Seattle when all of California is gearing up for a nice long weekend of caching.

[+lodgebarn]

It appears to me that all scheduled PQs are produced at midnight server time. This may be fine in the US and further west but for those of us that live east of the US then this means we do not get the PQ until the morning. I live in the UK and the erver is currently 8 hours behind us. This means that the earliest I get a PQ is 8am and often hours later.

 

If all PQs were sent at midnight local time to the user then this would spread the load on the server and allow us all to get the PQs in the very early hours of the morning before loading up for the day off caching.

 

I suggest that the PQ sending time is changed to midnight local time to the user.

Oh no this would be a bad idea. Much more complicated scheduling would probably rule this out anyway. This is what I do, just get up at 7:00 am select your 5 PQs and they are with you in a trice

[+DonB]

If they only run PQs after midnight someone explain why I usually get mine within minutes no matter what time of the day I request them.

[+lodgebarn]

If they only run PQs after midnight someone explain why I usually get mine within minutes no matter what time of the day I request them.

My experience would agree with you. However what I understand is that the order they are run is not random and depends on when they were intitially created i.e lower numbers run first. Also just from observation those that have not been run for a while might also get some priority. Hence I keep all my old queries and re-purpose them when I need to.

[+Walts Hunting]

I thnk what is happening is he has set for say Tuesday so at midnight that morning it gets processed. If you submit one for the current day it queues it up and runs it based on the froggies hierarchy rules.

[+Beach_hut]

If they only run PQs after midnight someone explain why I usually get mine within minutes no matter what time of the day I request them.

My experience would agree with you. However what I understand is that the order they are run is not random and depends on when they were intitially created i.e lower numbers run first. Also just from observation those that have not been run for a while might also get some priority. Hence I keep all my old queries and re-purpose them when I need to.

 

It's in order of the last time it's run - hence why when you run a query for the first time you get it almost immediately (depending on time of day)

[+dprovan]

If they only run PQs after midnight someone explain why I usually get mine within minutes no matter what time of the day I request them.

We're talking about "scheduled" PQs, i.e., PQs that are requested on one day for a future day. Those are all started at the stroke of midnight, Seattle time. Well, by "all started", I mean the servers start to work through them. I have no idea what order they're done in or how long it takes to get through them all, although my experience is that mine are normally done within a couple hours. I assume if you were to manually schedule one during this period, it would be similarly delayed, but maybe they accommodate that.

[+DonB]

If they only run PQs after midnight someone explain why I usually get mine within minutes no matter what time of the day I request them.

My experience would agree with you. However what I understand is that the order they are run is not random and depends on when they were intitially created i.e lower numbers run first. Also just from observation those that have not been run for a while might also get some priority. Hence I keep all my old queries and re-purpose them when I need to.

 

It's in order of the last time it's run - hence why when you run a query for the first time you get it almost immediately (depending on time of day)

I question that theory because I have ran the same PQ twice within minutes of one another because I forgot to enter some caches and I still got the second one within a minute or two.

[+larryc43230]

I question that theory because I have ran the same PQ twice within minutes of one another because I forgot to enter some caches and I still got the second one within a minute or two.

How do you manage to run the same pocket query twice within minutes of each other? I've always had to wait 24 hours, and I thought everybody had to.

 

If you're referring to cloning a pocket query, then running the clone, that's not the same thing as far as the system is concerned.

 

--Larry

[+dprovan]

How do you manage to run the same pocket query twice within minutes of each other? I've always had to wait 24 hours, and I thought everybody had to.

Yeah, me, too. Drives me crazy when I didn't notice the pocket query ran at midnight and I try to run it again the next afternoon to pick up some newly published caches.

[+DonB]

I question that theory because I have ran the same PQ twice within minutes of one another because I forgot to enter some caches and I still got the second one within a minute or two.

How do you manage to run the same pocket query twice within minutes of each other? I've always had to wait 24 hours, and I thought everybody had to.

 

If you're referring to cloning a pocket query, then running the clone, that's not the same thing as far as the system is concerned.

 

--Larry

I run 99% of my PQs from several bookmark lists I keep edited, and what I'm saying is I have ran a PQ from one of my bookmarks and then a few minutes later realized I forgot to add a couple of caches. I add the caches and run the PQ again. I don't know what your definition of cloning a PQ is so call it whatever you want, all I know is it works and I'm happy.

[+chillypenguin]

How about a option a "Run PQ now" option in place of the scheduled system. To ensure that I have fresh PQ's of our local area I run a series every day, but often go weeks without caching so all those processor cycles go to waist.

 

I understand there is a issue of managing demand. But I reckon I only use 1 in 50 PQ's of the scheduled PQ's that I have set up.

Also a priority system could be implemented based on the number of PQ's you have run that day.

[+larryc43230]

I run 99% of my PQs from several bookmark lists I keep edited, and what I'm saying is I have ran a PQ from one of my bookmarks and then a few minutes later realized I forgot to add a couple of caches. I add the caches and run the PQ again. I don't know what your definition of cloning a PQ is so call it whatever you want, all I know is it works and I'm happy.

We must have different definitions of "run the PQ again."

 

Some of my pocket queries are also based on bookmark lists. One of them is named "Bookmarks - DNFs still haunting me". Here's its entry on my Pocket Queries page:

 

 

(By the way, the box just to the left of the Sunday box, with the column header that reads "Copy", is what I was referring to as "cloning". By checking that box, you can make a copy, or clone, of the existing pocket query and then, if you want, run the query. But that's a new pocket query as far as "the system" is concerned.)

 

As a test, I clicked the box for today (Sunday) and ran the pocket query.

 

 

I waited until I received the e-mail notifying me the pocket query was ready, then downloaded the pocket query file.

 

I returned to the Pocket Queries page. The "Last Generated (PST)" entry for that pocket query now read 10/20/2013 05:01:55, and the date and time were bolded.

 

 

I added a random geocache to the bookmark list that pocket query is based on.

 

I returned to the Pocket Queries page and refreshed the display. The "Last Generated (PST)" entry for that pocket query was still bolded.

 

Once again, I checked the box for today (Sunday), and waited. And waited... and waited...

 

About an hour later, that box is still checked, the date and time still says 10/20/2013 05:01:55, the date and time are still bolded, and I'm not getting a new pocket query. And I'll lay bets that that pocket query won't run until next Sunday, a week from now.

 

I'm guessing you're doing something different to get a second download for the "same" pocket query on the same day.

 

Here's one possibility:

 

Here's the entry for that bookmark list of mine on the Bookmark Lists page:

 

 

If I click the link that says "Create Pocket Query" for that bookmark list, I can create a new pocket query and run that pocket query immediately. But that's a new pocket query as far as the system is concerned, and that's not the same as running the same pocket query twice on the same day, and that's what I'm referring to.

 

If that's not how you're doing this, would you care to share your secret?

 

--Larry

[+larryc43230]

I figured I should mention that the only reason I'm getting all pedantic about this is to hopefully prevent newbies from reading about how someone is able to run the same pocket query more than once a day, then when they try the same thing it doesn't work for them.

 

Pocket queries aren't exactly the easiest feature of the Web site site to figure out; we don't need to add to any confusion they might have.

 

--Larry

[+dprovan]

I figured I should mention that the only reason I'm getting all pedantic about this is to hopefully prevent newbies from reading about how someone is able to run the same pocket query more than once a day, then when they try the same thing it doesn't work for them.

Gee, I was hoping that there is a way to rerun a PQ in the same day, and this guy was going to show us how. Unfortunately, I'm afraid your pessimistic and depressing analysis is probably correct.

[+larryc43230]

Gee, I was hoping that there is a way to rerun a PQ in the same day, and this guy was going to show us how. Unfortunately, I'm afraid your pessimistic and depressing analysis is probably correct.

I really don't think any of this limits what we can do with pocket queries very much. If DonB is generating a new pocket query based on a modified bookmark list, all that's "wasted" is one of the five pocket queries we can run every 24 hours. And there are always ways to get more cache data if you're willing to throw money at it. A second premium membership (which gives you five more pocket queries per day) and GSAK's Get Geocaches command and Get Caches on Route macro come to mind. And folks who use smart phone apps to geocache have additional, pretty much unlimited, ways to get cache information.

 

Actually, I'm still hoping DonB has come up with a way to run the exact same pocket query more often than once a day. I just can't figure out a way to do it myself.

 

--Larry

[+DonB]

Gee, I was hoping that there is a way to rerun a PQ in the same day, and this guy was going to show us how. Unfortunately, I'm afraid your pessimistic and depressing analysis is probably correct.

I really don't think any of this limits what we can do with pocket queries very much. If DonB is generating a new pocket query based on a modified bookmark list, all that's "wasted" is one of the five pocket queries we can run every 24 hours. And there are always ways to get more cache data if you're willing to throw money at it. A second premium membership (which gives you five more pocket queries per day) and GSAK's Get Geocaches command and Get Caches on Route macro come to mind. And folks who use smart phone apps to geocache have additional, pretty much unlimited, ways to get cache information.

 

Actually, I'm still hoping DonB has come up with a way to run the exact same pocket query more often than once a day. I just can't figure out a way to do it myself.

 

--Larry

I don't know what part of my explanation you guys don't understand. I get 5 PQs a day and have never used my quota that I can remember, so by rerunning a PQ I'm not exceeding my quota. If I'm wasting a PQ by doing that, so be it. As I said before, I'm happy.

[+larryc43230]

I don't know what part of my explanation you guys don't understand. I get 5 PQs a day and have never used my quota that I can remember, so by rerunning a PQ I'm not exceeding my quota. If I'm wasting a PQ by doing that, so be it. As I said before, I'm happy.

You haven't actually explained the steps you take. Personally, I'm still curious about what you mean by "rerunning a PQ". Since you mention wasting a PQ, I have to assume you're actually creating a new pocket query from the Bookmarks List page. Please correct me if I'm wrong.

 

--Larry

Edited October 21, 2013 by larryc43230