Geocache Health Score

[cezanne]

 

Deleting DNFs could be something some cachers will resort to and note that unlike finds there is no chance to have DNFs reinstalled by GS.

 

Cache logs are never "deleted" they're only removed from view but remain in the database. The only logical way for this algorithm to work would be for it to process the logs in the database, and not just the logs on the cache page - therefore COs deleting DNF/NM/NA logs should have no effect on the way the algorithm works. Of course many COs won't realise this and may still delete DNFs, which would be unfortunate, but many COs have been doing that for years anyway.

 

Do you really think that the algorithm takes into account archived logs? That would include logs that are deleted by the author out of mistake (wrong log type or wrong cache) too and that would not make sense at all.

 

I see a big difference between NM/NA and DNF logs.

 

While NM and NA have been introduced by GS with a specific purpose in mind, DNF logs have been handled quite individually since geocaching started off.

There are cachers who log DNF whenever they do not end up with a success after having pressed go to on their gps and some that only log DNF if they explored every possible place, asked previous finders and still could not find the cache to name only some extreme ends and not all the spectrum in between.

 

While of course it happens that some cachers use NM and NA in the wrong way, that's not a really widespread thing while the extremely different interpretation of DNF is an issue in my eyes which needs to be taken into account.

 

I still wonder whether those who work on the health score algorithm have taken into account that many DNF logs say absolutely nothing about the cache health.

Edited March 2, 2017 by cezanne

[+justintim1999]

Let me reiterate and focus bluntly on my point for you: Knowledge that "the algorithm is not a consequence, and that the email can be ignored if the CO knows that there is no problem with the cache".

That is all. That's it. Not an insult. Not one CO being in any way "better" than another. Simply, and strictly, knowledge that as a CO the email can be ignored. A CO who does not know this knows less than one who does.

The Health Score is a standard. It's not that important to me if the standard isn't rigidly enforced. I'm more worried about cezanne and fizzymagic deciding the standard isn't for them even if they aren't worried about whether they'll actually be held to it.

 

Furthermore, the more you guys talk about how the e-mail's just an e-mail, the more I can't help but hear a whisper "although I wish it were enforced" between the lines.

 

ETA: I suppose this question can be asked for those here who despise this Geocache Health Score function: If the email were worded differently, and the DNF component of the algorithm were adjusted to an agreeable result, would you no longer have a problem with the Geocache Health Score system?

No, it wouldn't make any difference to me. I don't consider the e-mail anything more than a symptom.

 

Deal with the problem, as a responsible CO. Or wait until the problem is raised with a reviewer who will then deal with your situation appropriately (giving them the benefiti of the doubt as responsible fair reviewers).

Why is that better than, "Deal with the problem as a responsible seeker. Or wait until the problem is raised by someone else with a reviewer who will then deal with the situation appropriately"? Since the time the cache remains listed is identical, it makes more sense to me to stress whose inaction is actually causing the delay. It's just a happy coincidence that that also stresses that the person being impacted by the bad cache should feel responsible for dealing with the problem.

 

That's why the human element (reviewer) is still a very important part of this whole thing.

This is the crux. I think the seeker is an even more important part of this whole thing. The reviewer should just be an ultra-reliable last resort.

 

None of this should happen without the human element being involved. I agree that a NM needs to start the ball rolling. From there it's up to the cache owner to stop it. Every cache (in good standing) should start at 100%. Every NM should deduct points. After fixing the problem the OML should bring the cache back to 100% You can fiddle with dnfs and possibly time between NMs and OMLs to try to get a more accurate picture of maintenance habits but in the end it's just a tool.

[+justintim1999]

 

Deleting DNFs could be something some cachers will resort to and note that unlike finds there is no chance to have DNFs reinstalled by GS.

 

Cache logs are never "deleted" they're only removed from view but remain in the database. The only logical way for this algorithm to work would be for it to process the logs in the database, and not just the logs on the cache page - therefore COs deleting DNF/NM/NA logs should have no effect on the way the algorithm works. Of course many COs won't realise this and may still delete DNFs, which would be unfortunate, but many COs have been doing that for years anyway.

 

Do you really think that the algorithm takes into account archived logs? That would include logs that are deleted by the author out of mistake (wrong log type or wrong cache) too and that would not make sense at all.

 

I see a big difference between NM/NA and DNF logs.

 

While NM and NA have been introduced by GS with a specific purpose in mind, DNF logs have been handled quite individually since geocaching started off.

There are cachers who log DNF whenever they do not end up with a success after having pressed go to on their gps and some that only log DNF if they explored every possible place, asked previous finders and still could not find the cache to name only some extreme ends and not all the spectrum in between.

 

While of course it happens that some cachers use NM and NA in the wrong way, that's not a really widespread thing while the extremely different interpretation of DNF is an issue in my eyes which needs to be taken into account.

 

I still wonder whether those who work on the health score algorithm have taken into account that many DNF logs say absolutely nothing about the cache health.

 

I've been waiting a while to say "Your absolutely right". Dnfs shouldn't hold as much weight as a NM but..... what good is the health of a cache if it's missing?

[+Team Microdot]

Just wondering, if a cache gets a low health score because of one or more DNFs, does someone subsequently logging a find restore the score to pristine or are those DNFs an indelible mark against it for all time to come?

 

I see this as more of a sliding scale than simple on/off switches.

 

Some things will add a benenficial score and the health rating will increase. Other things will add a detrimental score and the health rating will decrease.

 

The health rating can wobble up and down constantly without triggering any response - so long as it stays above any pre-determined thresholds which have been selected as points at which some action must take place.

 

We dont know the algorithm. I would think it is a bit complex, and it can be tweaked.

 

Generally, I would think that an OM would reset it to a 'neutral to good' score at that point. Then we can see what happens after that, finds, or more DNFs, or another NM/NA.

 

Again, I don't see the idea of a single log of any type resetting anything.

 

Positive actions yield beneficial scores = health score increases

 

Negative actions yield detrimental scores = health score decreases.

 

OM could be treated as a positive score but then again it might be an armchair OM so it might yield a beneficial score but not necessarily a huge one.

 

A find AFTER the OM might validate that OM and bump the score up a little bit - unless there's also an NM from the finder on the same day in which case the OM becomes questionable and any beneficial score it originally contributed might need to be deducted again.

 

Of course a reviewer or lackey could have the power to intervene, override the automated scoring and reset things as they see fit, thus providing a 'sanity check' for the automated systems.

[+MartyBartfast]

Do you really think that the algorithm takes into account archived logs? That would include logs that are deleted by the author out of mistake (wrong log type or wrong cache) too and that would not make sense at all.

 

Yes. And if it doesn't then it should, otherwise it the algorithm could be effectively disabled by the CO simply deleting any logs they don't like. The number of incorrect logs subsequently deleted by authors for whatever reason is going to be tiny, and if the algorithm is correctly balanced it will be statistically insignificant.

[+TheOldfields]

I've not been through the entire thread so not sure if this has been mentioned before, but wouldn't it make sense for HQ to simply make public the algorithm they're using?

[+MartyBartfast]

 

Again, I don't see the idea of a single log of any type resetting anything.

 

Positive actions yield beneficial scores = health score increases

 

Negative actions yield detrimental scores = health score decreases.

 

OM could be treated as a positive score but then again it might be an armchair OM so it might yield a beneficial score but not necessarily a huge one.

 

A find AFTER the OM might validate that OM and bump the score up a little bit - unless there's also an NM from the finder on the same day in which case the OM becomes questionable and any beneficial score it originally contributed might need to be deducted again.

 

Of course a reviewer or lackey could have the power to intervene, override the automated scoring and reset things as they see fit, thus providing a 'sanity check' for the automated systems.

 

That's pretty much how I would approach it if I were writing it.

[+Team Microdot]

I've not been through the entire thread so not sure if this has been mentioned before, but wouldn't it make sense for HQ to simply make public the algorithm they're using?

 

No.

 

For some of same reasons Google doesn't make its algorithms public.

[cezanne]

Do you really think that the algorithm takes into account archived logs? That would include logs that are deleted by the author out of mistake (wrong log type or wrong cache) too and that would not make sense at all.

 

Yes. And if it doesn't then it should, otherwise it the algorithm could be effectively disabled by the CO simply deleting any logs they don't like. The number of incorrect logs subsequently deleted by authors for whatever reason is going to be tiny, and if the algorithm is correctly balanced it will be statistically insignificant.

 

Is it really tiny?

[+Team Microdot]

Do you really think that the algorithm takes into account archived logs? That would include logs that are deleted by the author out of mistake (wrong log type or wrong cache) too and that would not make sense at all.

 

Yes. And if it doesn't then it should, otherwise it the algorithm could be effectively disabled by the CO simply deleting any logs they don't like. The number of incorrect logs subsequently deleted by authors for whatever reason is going to be tiny, and if the algorithm is correctly balanced it will be statistically insignificant.

 

Is it really tiny?

 

Is it not?

[cezanne]

 

Negative actions yield detrimental scores = health score decreases.

 

So how would your perceived system cope with 12 DNFs coming from the same group who cached together and visited the wrong final location or who had a car accident and then went back home vs 12 DNFs of individual cachers who all could not find the cache on that day?

 

I guess it would consider this as something very negative, wouldn't it?

 

The spam example seems less complex to me (though it is complex enough) as this widespread spectrum of different uses of DNFs is something that really makes things complicated in my opinion.

[+TheOldfields]

I've not been through the entire thread so not sure if this has been mentioned before, but wouldn't it make sense for HQ to simply make public the algorithm they're using?

 

No.

 

For some of same reasons Google doesn't make its algorithms public.

 

Google's algorithms are extreme high tech worked on by leading members of the artificial intelligence community. Comparing that to a rating on the accessibility of a piece of tupperware is a little bit outrageous.

[cezanne]

Do you really think that the algorithm takes into account archived logs? That would include logs that are deleted by the author out of mistake (wrong log type or wrong cache) too and that would not make sense at all.

 

Yes. And if it doesn't then it should, otherwise it the algorithm could be effectively disabled by the CO simply deleting any logs they don't like. The number of incorrect logs subsequently deleted by authors for whatever reason is going to be tiny, and if the algorithm is correctly balanced it will be statistically insignificant.

 

Is it really tiny?

 

Is it not?

 

Maybe, maybe not. I do not have data. I only know that it is not too uncommon that I choose the wrong log types or make mistakes when splitting my logs into parts or log caches in the wrong order and then delete logs to get the order right and repost logs.

[+justintim1999]

Just wondering, if a cache gets a low health score because of one or more DNFs, does someone subsequently logging a find restore the score to pristine or are those DNFs an indelible mark against it for all time to come?

 

I see this as more of a sliding scale than simple on/off switches.

 

Some things will add a benenficial score and the health rating will increase. Other things will add a detrimental score and the health rating will decrease.

 

The health rating can wobble up and down constantly without triggering any response - so long as it stays above any pre-determined thresholds which have been selected as points at which some action must take place.

 

We dont know the algorithm. I would think it is a bit complex, and it can be tweaked.

 

Generally, I would think that an OM would reset it to a 'neutral to good' score at that point. Then we can see what happens after that, finds, or more DNFs, or another NM/NA.

 

Again, I don't see the idea of a single log of any type resetting anything.

 

Positive actions yield beneficial scores = health score increases

 

Negative actions yield detrimental scores = health score decreases.

 

OM could be treated as a positive score but then again it might be an armchair OM so it might yield a beneficial score but not necessarily a huge one.

 

A find AFTER the OM might validate that OM and bump the score up a little bit - unless there's also an NM from the finder on the same day in which case the OM becomes questionable and any beneficial score it originally contributed might need to be deducted again.

 

Of course a reviewer or lackey could have the power to intervene, override the automated scoring and reset things as they see fit, thus providing a 'sanity check' for the automated systems.

 

This is all good. I just think when you start trying to determine the validity of OMLs your getting into a guessing game that most reviewers would probably avoid.

[+barefootjeff]

Just wondering, if a cache gets a low health score because of one or more DNFs, does someone subsequently logging a find restore the score to pristine or are those DNFs an indelible mark against it for all time to come?

 

I see this as more of a sliding scale than simple on/off switches.

 

Some things will add a benenficial score and the health rating will increase. Other things will add a detrimental score and the health rating will decrease.

 

The health rating can wobble up and down constantly without triggering any response - so long as it stays above any pre-determined thresholds which have been selected as points at which some action must take place.

 

We dont know the algorithm. I would think it is a bit complex, and it can be tweaked.

 

Generally, I would think that an OM would reset it to a 'neutral to good' score at that point. Then we can see what happens after that, finds, or more DNFs, or another NM/NA.

 

Again, I don't see the idea of a single log of any type resetting anything.

 

Positive actions yield beneficial scores = health score increases

 

Negative actions yield detrimental scores = health score decreases.

 

OM could be treated as a positive score but then again it might be an armchair OM so it might yield a beneficial score but not necessarily a huge one.

 

A find AFTER the OM might validate that OM and bump the score up a little bit - unless there's also an NM from the finder on the same day in which case the OM becomes questionable and any beneficial score it originally contributed might need to be deducted again.

 

Of course a reviewer or lackey could have the power to intervene, override the automated scoring and reset things as they see fit, thus providing a 'sanity check' for the automated systems.

People don't log DNF if they find a cache with a cracked lid or damp log. The only cache health issue a DNF can reasonably infer is a missing cache. It doesn't matter whether there's been one, ten or a hundred DNFs, a subsequent "Found It!" log proves the cache isn't missing, so why should those DNFs continue to carry any weight?

 

Of course there's always the possibility of a throwdown in that scenario, but heaven help us if the algorithm is trying to predict those! I would've thought the presumption of innocence would apply here.

[+niraD]

A find AFTER the OM might validate that OM and bump the score up a little bit - unless there's also an NM from the finder on the same day in which case the OM becomes questionable and any beneficial score it originally contributed might need to be deducted again.

Weighting certain combinations of logs together could make sense.

 

If I log a DNF today, and log a Find tomorrow, then that could increase the score for the cache being hard to find. The more DNFs I log before finally logging a Find, the higher the "hard to find" score would be.

 

In turn, the weight of DNFs logged against a "hard to find" cache could be reduced.

[+Team Microdot]

Just wondering, if a cache gets a low health score because of one or more DNFs, does someone subsequently logging a find restore the score to pristine or are those DNFs an indelible mark against it for all time to come?

 

I see this as more of a sliding scale than simple on/off switches.

 

Some things will add a benenficial score and the health rating will increase. Other things will add a detrimental score and the health rating will decrease.

 

The health rating can wobble up and down constantly without triggering any response - so long as it stays above any pre-determined thresholds which have been selected as points at which some action must take place.

 

We dont know the algorithm. I would think it is a bit complex, and it can be tweaked.

 

Generally, I would think that an OM would reset it to a 'neutral to good' score at that point. Then we can see what happens after that, finds, or more DNFs, or another NM/NA.

 

Again, I don't see the idea of a single log of any type resetting anything.

 

Positive actions yield beneficial scores = health score increases

 

Negative actions yield detrimental scores = health score decreases.

 

OM could be treated as a positive score but then again it might be an armchair OM so it might yield a beneficial score but not necessarily a huge one.

 

A find AFTER the OM might validate that OM and bump the score up a little bit - unless there's also an NM from the finder on the same day in which case the OM becomes questionable and any beneficial score it originally contributed might need to be deducted again.

 

Of course a reviewer or lackey could have the power to intervene, override the automated scoring and reset things as they see fit, thus providing a 'sanity check' for the automated systems.

 

This is all good. I just think when you start trying to determine the validity of OMLs your getting into a guessing game that most reviewers would probably avoid.

 

That would depend on the detail in the logs.

 

Let's say the first nm says that the box has a broken lid and is soaking wet and filthy.

 

The CO posts an armchair OM log.

 

The next finder logs NM and says that the box has a broken lid and is soaking wet and filthy.

 

This would be complex for an algorithm to pick up but a human reviewer would spot the armchair OM easily.

[+barefootjeff]

A find AFTER the OM might validate that OM and bump the score up a little bit - unless there's also an NM from the finder on the same day in which case the OM becomes questionable and any beneficial score it originally contributed might need to be deducted again.

Weighting certain combinations of logs together could make sense.

 

If I log a DNF today, and log a Find tomorrow, then that could increase the score for the cache being hard to find. The more DNFs I log before finally logging a Find, the higher the "hard to find" score would be.

 

In turn, the weight of DNFs logged against a "hard to find" cache could be reduced.

Only if the DNFs were due to the cache being well concealed. There are so many other things that can result in a DNF, like sudden rain, nightfall, an influx of muggles, a fear of heights or whatever. Trying to infer anything from DNF logs without reading their content is pure guesswork.

[cezanne]

The only cache health issue a DNF can reasonably infer is a missing cache. It doesn't matter whether there's been one, ten or a hundred DNFs, a subsequent "Found It!" log proves the cache isn't missing, so why should those DNFs continue to carry any weight?

 

Of course there's always the possibility of a throwdown in that scenario, but heaven help us if the algorithm is trying to predict those! I would've thought the presumption of innocence would apply here.

 

The more I think about the issue the more I come to the conclusion that it is a different thing whether an algorithm is developped that should provide reviewers with a list of caches they might want to have a look at.

If a reviewer is willing to do the work, it could not create a damage if they get listed a cache with say 1 or 2 DNFs and if they then looked at the logs, they would quickly see that in such cases typically no action is necessary. It should also be up to the reviewers to provide their kind of feedback to GS when it comes to which type of system is useful for them.

 

The problematic area starts when the same algorithm and produced cache list is used to trigger the mentioned e-mails which are sent to cache owners. When cache owners are effected they should be able to provide feedback too and more importantly, it would be reasonable to use a different parameter setting or a different algorithm to produce a different sort of list (DNFs e.g. should definitely play a smaller role for this case).

[+justintim1999]

Just wondering, if a cache gets a low health score because of one or more DNFs, does someone subsequently logging a find restore the score to pristine or are those DNFs an indelible mark against it for all time to come?

 

I see this as more of a sliding scale than simple on/off switches.

 

Some things will add a benenficial score and the health rating will increase. Other things will add a detrimental score and the health rating will decrease.

 

The health rating can wobble up and down constantly without triggering any response - so long as it stays above any pre-determined thresholds which have been selected as points at which some action must take place.

 

We dont know the algorithm. I would think it is a bit complex, and it can be tweaked.

 

Generally, I would think that an OM would reset it to a 'neutral to good' score at that point. Then we can see what happens after that, finds, or more DNFs, or another NM/NA.

 

Again, I don't see the idea of a single log of any type resetting anything.

 

Positive actions yield beneficial scores = health score increases

 

Negative actions yield detrimental scores = health score decreases.

 

OM could be treated as a positive score but then again it might be an armchair OM so it might yield a beneficial score but not necessarily a huge one.

 

A find AFTER the OM might validate that OM and bump the score up a little bit - unless there's also an NM from the finder on the same day in which case the OM becomes questionable and any beneficial score it originally contributed might need to be deducted again.

 

Of course a reviewer or lackey could have the power to intervene, override the automated scoring and reset things as they see fit, thus providing a 'sanity check' for the automated systems.

 

This is all good. I just think when you start trying to determine the validity of OMLs your getting into a guessing game that most reviewers would probably avoid.

 

That would depend on the detail in the logs.

 

Let's say the first nm says that the box has a broken lid and is soaking wet and filthy.

 

The CO posts an armchair OM log.

 

The next finder logs NM and says that the box has a broken lid and is soaking wet and filthy.

 

This would be complex for an algorithm to pick up but a human reviewer would spot the armchair OM easily.

 

For that purposes of the automated health score you'd have to treat every OML the same and as you say let the reviewer dig deeper and adjust as they see fit.

[+barefootjeff]

If you want to get pedantic, a "Found It!" log should carry a small negative score, since for each one of those, there's one less space left in the logbook and one extra chance for water to get into the container.

[+Team Microdot]

Just wondering, if a cache gets a low health score because of one or more DNFs, does someone subsequently logging a find restore the score to pristine or are those DNFs an indelible mark against it for all time to come?

 

I see this as more of a sliding scale than simple on/off switches.

 

Some things will add a benenficial score and the health rating will increase. Other things will add a detrimental score and the health rating will decrease.

 

The health rating can wobble up and down constantly without triggering any response - so long as it stays above any pre-determined thresholds which have been selected as points at which some action must take place.

 

We dont know the algorithm. I would think it is a bit complex, and it can be tweaked.

 

Generally, I would think that an OM would reset it to a 'neutral to good' score at that point. Then we can see what happens after that, finds, or more DNFs, or another NM/NA.

 

Again, I don't see the idea of a single log of any type resetting anything.

 

Positive actions yield beneficial scores = health score increases

 

Negative actions yield detrimental scores = health score decreases.

 

OM could be treated as a positive score but then again it might be an armchair OM so it might yield a beneficial score but not necessarily a huge one.

 

A find AFTER the OM might validate that OM and bump the score up a little bit - unless there's also an NM from the finder on the same day in which case the OM becomes questionable and any beneficial score it originally contributed might need to be deducted again.

 

Of course a reviewer or lackey could have the power to intervene, override the automated scoring and reset things as they see fit, thus providing a 'sanity check' for the automated systems.

People don't log DNF if they find a cache with a cracked lid or damp log. The only cache health issue a DNF can reasonably infer is a missing cache. It doesn't matter whether there's been one, ten or a hundred DNFs, a subsequent "Found It!" log proves the cache isn't missing, so why should those DNFs continue to carry any weight?

 

Of course there's always the possibility of a throwdown in that scenario, but heaven help us if the algorithm is trying to predict those! I would've thought the presumption of innocence would apply here.

 

Good points. Needs some thought.

 

I might be in danger of making things overly complex but let's at least explore.

 

Let's give the first DNF a negative score of -1. No massive impact on the health rating.

 

The second DNF by a DIFFERENT cacher on DIFFERENT day scores -2.

 

Third scores -3 and so on. Each further DNF makes it more likely the cache is missing. For normal everyday caches this fits reasonably well with human thought patterns I think.

 

Subsequent FOUND logs reverse this (sort of). First find increases health by +1, second by +2, third by +3.

 

I might describe it as an algorithmic trust/confidence scale.

 

Multiple confirmations from multiple sources increases confidence and trust.

[Keystone]

If the cache has no problem, then it can be ignored.

Then the email should present that as an option.

 

To quote someone who has received the email (when the T5 cache in question had received only a single DNF): "The options it gave me were to immediately visit the cache, "repair" it and log an OM, disable it until I could do so, or archive it."

 

Do you really want to advertise "do nothing" as an option?

This was likely the line of thinking used, though I didn't author the standard email text. Rather than watering down the message to give an excuse to a lazy CO who ought to be checking on a legitimate problem, the more direct list of options is used in the interest of brevity and on the assumption that a smart and conscientious CO would figure out there is no need to check on their cache. Again, just my guess.

[+justintim1999]

Just wondering, if a cache gets a low health score because of one or more DNFs, does someone subsequently logging a find restore the score to pristine or are those DNFs an indelible mark against it for all time to come?

 

I see this as more of a sliding scale than simple on/off switches.

 

Some things will add a benenficial score and the health rating will increase. Other things will add a detrimental score and the health rating will decrease.

 

The health rating can wobble up and down constantly without triggering any response - so long as it stays above any pre-determined thresholds which have been selected as points at which some action must take place.

 

We dont know the algorithm. I would think it is a bit complex, and it can be tweaked.

 

Generally, I would think that an OM would reset it to a 'neutral to good' score at that point. Then we can see what happens after that, finds, or more DNFs, or another NM/NA.

 

Again, I don't see the idea of a single log of any type resetting anything.

 

Positive actions yield beneficial scores = health score increases

 

Negative actions yield detrimental scores = health score decreases.

 

OM could be treated as a positive score but then again it might be an armchair OM so it might yield a beneficial score but not necessarily a huge one.

 

A find AFTER the OM might validate that OM and bump the score up a little bit - unless there's also an NM from the finder on the same day in which case the OM becomes questionable and any beneficial score it originally contributed might need to be deducted again.

 

Of course a reviewer or lackey could have the power to intervene, override the automated scoring and reset things as they see fit, thus providing a 'sanity check' for the automated systems.

People don't log DNF if they find a cache with a cracked lid or damp log. The only cache health issue a DNF can reasonably infer is a missing cache. It doesn't matter whether there's been one, ten or a hundred DNFs, a subsequent "Found It!" log proves the cache isn't missing, so why should those DNFs continue to carry any weight?

 

Of course there's always the possibility of a throwdown in that scenario, but heaven help us if the algorithm is trying to predict those! I would've thought the presumption of innocence would apply here.

 

Good points. Needs some thought.

 

I might be in danger of making things overly complex but let's at least explore.

 

Let's give the first DNF a negative score of -1. No massive impact on the health rating.

 

The second DNF by a DIFFERENT cacher on DIFFERENT day scores -2.

 

Third scores -3 and so on. Each further DNF makes it more likely the cache is missing. For normal everyday caches this fits reasonably well with human thought patterns I think.

 

Subsequent FOUND logs reverse this (sort of). First find increases health by +1, second by +2, third by +3.

 

I might describe it as an algorithmic trust/confidence scale.

 

Multiple confirmations from multiple sources increases confidence and trust.

 

I wouldn't like to see the health score rebound faster after the first find. I think the finds should be worth more than the dnfs. I do like the increasing multiple dnf factor. Fair but doesn't attack the health score too much.

[+barefootjeff]

If the cache has no problem, then it can be ignored.

Then the email should present that as an option.

 

To quote someone who has received the email (when the T5 cache in question had received only a single DNF): "The options it gave me were to immediately visit the cache, "repair" it and log an OM, disable it until I could do so, or archive it."

 

Do you really want to advertise "do nothing" as an option?

This was likely the line of thinking used, though I didn't author the standard email text. Rather than watering down the message to give an excuse to a lazy CO who ought to be checking on a legitimate problem, the more direct list of options is used in the interest of brevity and on the assumption that a smart and conscientious CO would figure out there is no need to check on their cache. Again, just my guess.

I must be stupid or unconscientious then. When I got the email, my immediate reaction was to try to figure out what I needed to do to make things right. I'd already ascertained from the DNFer that she'd been looking in the wrong place so a cache visit was pointless, my query to HQ's support centre only got me a boilerplate response telling me about the virtues of the Cache Health system and referring me to the Help Centre article, and it wasn't until the DNFer went back and found it and I reported that to the support person, that I was told the black mark against my cache had been removed. Whether that happened automatically as a result of the "Found It!" log or whether it was done manually, I don't know.

 

Even if a "do nothing" option is unwise, an option to report it as a false positive with a reason attached would have been greatly appreciated.

[+thebruce0]

Basing the algorithm on consecutive logs is problematic, for obvious reasons. You could base it on consecutive days, but on active caches any semblance actual log order or meaningful adjacent experiences is completely lost.

[cezanne]

 

I must be stupid or unconscientious then.

 

Me too.

[+thebruce0]

Even if a "do nothing" option is unwise, an option to report it as a false positive with a reason attached would have been greatly appreciated.

This I'd support, as described. It's a suggested improvement to the system, rather than just complaints about the system as a whole based on very specific hypotheticals and *feelings*

[+niraD]

If I log a DNF today, and log a Find tomorrow, then that could increase the score for the cache being hard to find. The more DNFs I log before finally logging a Find, the higher the "hard to find" score would be.

 

In turn, the weight of DNFs logged against a "hard to find" cache could be reduced.

Only if the DNFs were due to the cache being well concealed. There are so many other things that can result in a DNF, like sudden rain, nightfall, an influx of muggles, a fear of heights or whatever. Trying to infer anything from DNF logs without reading their content is pure guesswork.

If a cache shows a pattern of DNF logs followed by Find logs by the same users, does it really matter why that pattern exists? Whether it's well camouflaged, or subject to environmental obstacles, or anything else, the pattern exists that people often log a Find after logging a prior DNF.

 

As such, the weight of a single DNF should be less, as far as indicating potential "health concerns" about the cache.

[+barefootjeff]

Just wondering, if a cache gets a low health score because of one or more DNFs, does someone subsequently logging a find restore the score to pristine or are those DNFs an indelible mark against it for all time to come?

 

I see this as more of a sliding scale than simple on/off switches.

 

Some things will add a benenficial score and the health rating will increase. Other things will add a detrimental score and the health rating will decrease.

 

The health rating can wobble up and down constantly without triggering any response - so long as it stays above any pre-determined thresholds which have been selected as points at which some action must take place.

 

We dont know the algorithm. I would think it is a bit complex, and it can be tweaked.

 

Generally, I would think that an OM would reset it to a 'neutral to good' score at that point. Then we can see what happens after that, finds, or more DNFs, or another NM/NA.

 

Again, I don't see the idea of a single log of any type resetting anything.

 

Positive actions yield beneficial scores = health score increases

 

Negative actions yield detrimental scores = health score decreases.

 

OM could be treated as a positive score but then again it might be an armchair OM so it might yield a beneficial score but not necessarily a huge one.

 

A find AFTER the OM might validate that OM and bump the score up a little bit - unless there's also an NM from the finder on the same day in which case the OM becomes questionable and any beneficial score it originally contributed might need to be deducted again.

 

Of course a reviewer or lackey could have the power to intervene, override the automated scoring and reset things as they see fit, thus providing a 'sanity check' for the automated systems.

People don't log DNF if they find a cache with a cracked lid or damp log. The only cache health issue a DNF can reasonably infer is a missing cache. It doesn't matter whether there's been one, ten or a hundred DNFs, a subsequent "Found It!" log proves the cache isn't missing, so why should those DNFs continue to carry any weight?

 

Of course there's always the possibility of a throwdown in that scenario, but heaven help us if the algorithm is trying to predict those! I would've thought the presumption of innocence would apply here.

 

Good points. Needs some thought.

 

I might be in danger of making things overly complex but let's at least explore.

 

Let's give the first DNF a negative score of -1. No massive impact on the health rating.

 

The second DNF by a DIFFERENT cacher on DIFFERENT day scores -2.

 

Third scores -3 and so on. Each further DNF makes it more likely the cache is missing. For normal everyday caches this fits reasonably well with human thought patterns I think.

 

Subsequent FOUND logs reverse this (sort of). First find increases health by +1, second by +2, third by +3.

 

I might describe it as an algorithmic trust/confidence scale.

 

Multiple confirmations from multiple sources increases confidence and trust.

 

I wouldn't like to see the health score rebound faster after the first find. I think the finds should be worth more than the dnfs. I do like the increasing multiple dnf factor. Fair but doesn't attack the health score too much.

Sorry, but I still don't see why DNFs should continue to carry weight after a subsequent find. What health issue can a cache that's had a few DNFs but has then been found still have? Correct me if I'm wrong, but I can't think of any health issue other than "cache is missing" that can be reasonably inferred from a DNF, and surely a find, any find, proves that this inference from all those DNFs was wrong.

 

If DNFs continue to carry weight after the cache is subsequently found, I'll be even more reluctant to log a DNF on anything I can't put a signature in the log because of my own shortcomings or ineptitude.

[+justintim1999]

Just wondering, if a cache gets a low health score because of one or more DNFs, does someone subsequently logging a find restore the score to pristine or are those DNFs an indelible mark against it for all time to come?

 

I see this as more of a sliding scale than simple on/off switches.

 

Some things will add a benenficial score and the health rating will increase. Other things will add a detrimental score and the health rating will decrease.

 

The health rating can wobble up and down constantly without triggering any response - so long as it stays above any pre-determined thresholds which have been selected as points at which some action must take place.

 

We dont know the algorithm. I would think it is a bit complex, and it can be tweaked.

 

Generally, I would think that an OM would reset it to a 'neutral to good' score at that point. Then we can see what happens after that, finds, or more DNFs, or another NM/NA.

 

Again, I don't see the idea of a single log of any type resetting anything.

 

Positive actions yield beneficial scores = health score increases

 

Negative actions yield detrimental scores = health score decreases.

 

OM could be treated as a positive score but then again it might be an armchair OM so it might yield a beneficial score but not necessarily a huge one.

 

A find AFTER the OM might validate that OM and bump the score up a little bit - unless there's also an NM from the finder on the same day in which case the OM becomes questionable and any beneficial score it originally contributed might need to be deducted again.

 

Of course a reviewer or lackey could have the power to intervene, override the automated scoring and reset things as they see fit, thus providing a 'sanity check' for the automated systems.

People don't log DNF if they find a cache with a cracked lid or damp log. The only cache health issue a DNF can reasonably infer is a missing cache. It doesn't matter whether there's been one, ten or a hundred DNFs, a subsequent "Found It!" log proves the cache isn't missing, so why should those DNFs continue to carry any weight?

 

Of course there's always the possibility of a throwdown in that scenario, but heaven help us if the algorithm is trying to predict those! I would've thought the presumption of innocence would apply here.

 

Good points. Needs some thought.

 

I might be in danger of making things overly complex but let's at least explore.

 

Let's give the first DNF a negative score of -1. No massive impact on the health rating.

 

The second DNF by a DIFFERENT cacher on DIFFERENT day scores -2.

 

Third scores -3 and so on. Each further DNF makes it more likely the cache is missing. For normal everyday caches this fits reasonably well with human thought patterns I think.

 

Subsequent FOUND logs reverse this (sort of). First find increases health by +1, second by +2, third by +3.

 

I might describe it as an algorithmic trust/confidence scale.

 

Multiple confirmations from multiple sources increases confidence and trust.

 

I wouldn't like to see the health score rebound faster after the first find. I think the finds should be worth more than the dnfs. I do like the increasing multiple dnf factor. Fair but doesn't attack the health score too much.

Sorry, but I still don't see why DNFs should continue to carry weight after a subsequent find. What health issue can a cache that's had a few DNFs but has then been found still have? Correct me if I'm wrong, but I can't think of any health issue other than "cache is missing" that can be reasonably inferred from a DNF, and surely a find, any find, proves that this inference from all those DNFs was wrong.

 

If DNFs continue to carry weight after the cache is subsequently found, I'll be even more reluctant to log a DNF on anything I can't put a signature in the log because of my own shortcomings or ineptitude.

 

I would think to protect against finds that could potentially be bogus. Two finds in a row would be a better indicator that the cache is indeed there and in tact.

[+barefootjeff]

Just wondering, if a cache gets a low health score because of one or more DNFs, does someone subsequently logging a find restore the score to pristine or are those DNFs an indelible mark against it for all time to come?

 

I see this as more of a sliding scale than simple on/off switches.

 

Some things will add a benenficial score and the health rating will increase. Other things will add a detrimental score and the health rating will decrease.

 

The health rating can wobble up and down constantly without triggering any response - so long as it stays above any pre-determined thresholds which have been selected as points at which some action must take place.

 

We dont know the algorithm. I would think it is a bit complex, and it can be tweaked.

 

Generally, I would think that an OM would reset it to a 'neutral to good' score at that point. Then we can see what happens after that, finds, or more DNFs, or another NM/NA.

 

Again, I don't see the idea of a single log of any type resetting anything.

 

Positive actions yield beneficial scores = health score increases

 

Negative actions yield detrimental scores = health score decreases.

 

OM could be treated as a positive score but then again it might be an armchair OM so it might yield a beneficial score but not necessarily a huge one.

 

A find AFTER the OM might validate that OM and bump the score up a little bit - unless there's also an NM from the finder on the same day in which case the OM becomes questionable and any beneficial score it originally contributed might need to be deducted again.

 

Of course a reviewer or lackey could have the power to intervene, override the automated scoring and reset things as they see fit, thus providing a 'sanity check' for the automated systems.

People don't log DNF if they find a cache with a cracked lid or damp log. The only cache health issue a DNF can reasonably infer is a missing cache. It doesn't matter whether there's been one, ten or a hundred DNFs, a subsequent "Found It!" log proves the cache isn't missing, so why should those DNFs continue to carry any weight?

 

Of course there's always the possibility of a throwdown in that scenario, but heaven help us if the algorithm is trying to predict those! I would've thought the presumption of innocence would apply here.

 

Good points. Needs some thought.

 

I might be in danger of making things overly complex but let's at least explore.

 

Let's give the first DNF a negative score of -1. No massive impact on the health rating.

 

The second DNF by a DIFFERENT cacher on DIFFERENT day scores -2.

 

Third scores -3 and so on. Each further DNF makes it more likely the cache is missing. For normal everyday caches this fits reasonably well with human thought patterns I think.

 

Subsequent FOUND logs reverse this (sort of). First find increases health by +1, second by +2, third by +3.

 

I might describe it as an algorithmic trust/confidence scale.

 

Multiple confirmations from multiple sources increases confidence and trust.

 

I wouldn't like to see the health score rebound faster after the first find. I think the finds should be worth more than the dnfs. I do like the increasing multiple dnf factor. Fair but doesn't attack the health score too much.

Sorry, but I still don't see why DNFs should continue to carry weight after a subsequent find. What health issue can a cache that's had a few DNFs but has then been found still have? Correct me if I'm wrong, but I can't think of any health issue other than "cache is missing" that can be reasonably inferred from a DNF, and surely a find, any find, proves that this inference from all those DNFs was wrong.

 

If DNFs continue to carry weight after the cache is subsequently found, I'll be even more reluctant to log a DNF on anything I can't put a signature in the log because of my own shortcomings or ineptitude.

 

I would think to protect against finds that could potentially be bogus. Two finds in a row would be a better indicator that the cache is indeed there and in tact.

Unless bogus finds are rampant, the chances of that are far far less than the likelihood that the DNF was for reasons other than a missing cache. A found log should be taken in good faith unless there's strong evidence to prove otherwise, whereas a DNF could be due to a whole manner of things unrelated to cache health.

[+justintim1999]

Just wondering, if a cache gets a low health score because of one or more DNFs, does someone subsequently logging a find restore the score to pristine or are those DNFs an indelible mark against it for all time to come?

 

I see this as more of a sliding scale than simple on/off switches.

 

Some things will add a benenficial score and the health rating will increase. Other things will add a detrimental score and the health rating will decrease.

 

The health rating can wobble up and down constantly without triggering any response - so long as it stays above any pre-determined thresholds which have been selected as points at which some action must take place.

 

We dont know the algorithm. I would think it is a bit complex, and it can be tweaked.

 

Generally, I would think that an OM would reset it to a 'neutral to good' score at that point. Then we can see what happens after that, finds, or more DNFs, or another NM/NA.

 

Again, I don't see the idea of a single log of any type resetting anything.

 

Positive actions yield beneficial scores = health score increases

 

Negative actions yield detrimental scores = health score decreases.

 

OM could be treated as a positive score but then again it might be an armchair OM so it might yield a beneficial score but not necessarily a huge one.

 

A find AFTER the OM might validate that OM and bump the score up a little bit - unless there's also an NM from the finder on the same day in which case the OM becomes questionable and any beneficial score it originally contributed might need to be deducted again.

 

Of course a reviewer or lackey could have the power to intervene, override the automated scoring and reset things as they see fit, thus providing a 'sanity check' for the automated systems.

People don't log DNF if they find a cache with a cracked lid or damp log. The only cache health issue a DNF can reasonably infer is a missing cache. It doesn't matter whether there's been one, ten or a hundred DNFs, a subsequent "Found It!" log proves the cache isn't missing, so why should those DNFs continue to carry any weight?

 

Of course there's always the possibility of a throwdown in that scenario, but heaven help us if the algorithm is trying to predict those! I would've thought the presumption of innocence would apply here.

 

Good points. Needs some thought.

 

I might be in danger of making things overly complex but let's at least explore.

 

Let's give the first DNF a negative score of -1. No massive impact on the health rating.

 

The second DNF by a DIFFERENT cacher on DIFFERENT day scores -2.

 

Third scores -3 and so on. Each further DNF makes it more likely the cache is missing. For normal everyday caches this fits reasonably well with human thought patterns I think.

 

Subsequent FOUND logs reverse this (sort of). First find increases health by +1, second by +2, third by +3.

 

I might describe it as an algorithmic trust/confidence scale.

 

Multiple confirmations from multiple sources increases confidence and trust.

 

I wouldn't like to see the health score rebound faster after the first find. I think the finds should be worth more than the dnfs. I do like the increasing multiple dnf factor. Fair but doesn't attack the health score too much.

Sorry, but I still don't see why DNFs should continue to carry weight after a subsequent find. What health issue can a cache that's had a few DNFs but has then been found still have? Correct me if I'm wrong, but I can't think of any health issue other than "cache is missing" that can be reasonably inferred from a DNF, and surely a find, any find, proves that this inference from all those DNFs was wrong.

 

If DNFs continue to carry weight after the cache is subsequently found, I'll be even more reluctant to log a DNF on anything I can't put a signature in the log because of my own shortcomings or ineptitude.

 

I would think to protect against finds that could potentially be bogus. Two finds in a row would be a better indicator that the cache is indeed there and in tact.

Unless bogus finds are rampant, the chances of that are far far less than the likelihood that the DNF was for reasons other than a missing cache. A found log should be taken in good faith unless there's strong evidence to prove otherwise, whereas a DNF could be due to a whole manner of things unrelated to cache health.

 

I don't know if I would dismiss them all together but if it was a deal breaker I'd be willing to not include them as part of the health score.

[Keystone]

Even if a "do nothing" option is unwise, an option to report it as a false positive with a reason attached would have been greatly appreciated.

This I'd support, as described. It's a suggested improvement to the system, rather than just complaints about the system as a whole based on very specific hypotheticals and *feelings*

If I were the King of the Geocaching HQ Customer Support desk, I'd have a procedure in place to forward reports of algorithm oddities to a developer responsible for the maintenance of the Cache Health toolset. This procedure would include a helpful reply script for the inquirer, like "Thanks for your feedback! We're always striving to improve the Cache Health system. I've passed your specific example along to the developer."

 

Failing that, you could collect some examples and share them with your local Community Volunteer Reviewer. He or she may be motivated (but isn't required) to share them in our ongoing discussion thread in the Reviewers' Forum. I know that, from that feedback channel, the algorithm parameters have been adjusted a few times to produce better results.

Edited March 3, 2017 by Keystone

[+dprovan]

None of this should happen without the human element being involved. I agree that a NM needs to start the ball rolling. From there it's up to the cache owner to stop it.

Why should an NM start the ball rolling? Without the Health Score, an NA starts the ball rolling. Why is that not good enough?

 

Pages and pages of discussion about the pluses and minus of the Health Score, pages and pages discussing various tweaks to the Health Score to fix the myriad of problems with it, and then it turns out the only problem with the existing system is that although you're happy to depend on people filing NMs, you refuse to depend on people filing NAs.

[+justintim1999]

None of this should happen without the human element being involved. I agree that a NM needs to start the ball rolling. From there it's up to the cache owner to stop it.

Why should an NM start the ball rolling? Without the Health Score, an NA starts the ball rolling. Why is that not good enough?

 

Pages and pages of discussion about the pluses and minus of the Health Score, pages and pages discussing various tweaks to the Health Score to fix the myriad of problems with it, and then it turns out the only problem with the existing system is that although you're happy to depend on people filing NMs, you refuse to depend on people filing NAs.

 

Why wait until a NA is necessary to start the process of weeding out bad caches? What's wrong with identifying them early and getting them off the map?

[+niraD]

Why wait until a NA is necessary to start the process of weeding out bad caches? What's wrong with identifying them early and getting them off the map?

If an NA is not necessary, then why do we need to get them off the map?

[+justintim1999]

Why wait until a NA is necessary to start the process of weeding out bad caches? What's wrong with identifying them early and getting them off the map?

If an NA is not necessary, then why do we need to get them off the map?

 

Because in many of these cases an NA is inevitable. Sooner or later these caches will wind up being archived. NAs are still necessary and the current system works to a point but I don't think it's efficient enough and could be streamlined.

 

So your also against any system that will identify bad cache maintenance more effectively? It's already been said, by the people who use the information, that the cache health score helps.

 

What more do you need to know.

[+Touchstone]

.....but I don't think it's efficient enough and could be streamlined.

 

So tell me again what the mad rush is all about?

[+justintim1999]

.....but I don't think it's efficient enough and could be streamlined.

 

So tell me again what the mad rush is all about?

 

I guess I just don't like finding trash. Could be I'd like to see spaces open up so responsible cachers can hide something new and nice I can go find. It's more likely I'm jealous I have to spend time maintaining my caches while other owners are busy organizing their sock drawer.

 

Why put off till tomorrow what you can do today, right?

[+barefootjeff]

.....but I don't think it's efficient enough and could be streamlined.

 

So tell me again what the mad rush is all about?

 

I guess I just don't like finding trash. Could be I'd like to see spaces open up so responsible cachers can hide something new and nice I can go find. It's more likely I'm jealous I have to spend time maintaining my caches while other owners are busy organizing their sock drawer.

 

Why put off till tomorrow what you can do today, right?

Trashy caches without an outstanding NM log won't be picked up by the Cache Health algorithm since it's then only looking at DNFs and people DNF because they can't find a cache at all, not because it's trashy.

 

If there are trashy caches in your area with outstanding NMs, you should feel free to log NAs on them as long as you think you've given the CO enough time to respond to the NM.

 

So what trashy caches will the Cache Health system get off the map that you can't already clear just as quickly with an NA?

[+Touchstone]

.....but I don't think it's efficient enough and could be streamlined.

 

So tell me again what the mad rush is all about?

 

I guess I just don't like finding trash. Could be I'd like to see spaces open up so responsible cachers can hide something new and nice I can go find. It's more likely I'm jealous I have to spend time maintaining my caches while other owners are busy organizing their sock drawer.

 

Why put off till tomorrow what you can do today, right?

Y'know, I looked briefly around the neighborhood of one of the caches you own, and was actually struck by how few of the nearby Listings were Disabled, or even had the dreaded red wrench for that matter. Looks like you might have to travel a bit further to get out of this apparent land of milk and honey you live in, to find this so called *trash* you speak of.

[+L0ne.R]

 

So what trashy caches will the Cache Health system get off the map that you can't already clear just as quickly with an NA?

In my area within 25 miles there are 489 red-wrench caches. Shall I start going through the logs and place NAs on those caches deserving of an NA?

I think reviewers might prefer the algorithm help with that work. And I expect it might tick off fewer cache owners and people who think NAers are self-appointed cache cops, if the algorithm help draw out the problem caches.

[+L0ne.R]

.....but I don't think it's efficient enough and could be streamlined.

 

So tell me again what the mad rush is all about?

 

I guess I just don't like finding trash. Could be I'd like to see spaces open up so responsible cachers can hide something new and nice I can go find. It's more likely I'm jealous I have to spend time maintaining my caches while other owners are busy organizing their sock drawer.

 

Why put off till tomorrow what you can do today, right?

Y'know, I looked briefly around the neighborhood of one of the caches you own, and was actually struck by how few of the nearby Listings were Disabled, or even had the dreaded red wrench for that matter. Looks like you might have to travel a bit further to get out of this apparent land of milk and honey you live in, to find this so called *trash* you speak of.

 

I did a PQ for caches with a NM attribute within 25 miles of one of his caches and get 467 caches with a red wrench.

[+niraD]

Why wait until a NA is necessary to start the process of weeding out bad caches? What's wrong with identifying them early and getting them off the map?

If an NA is not necessary, then why do we need to get them off the map?

Because in many of these cases an NA is inevitable. Sooner or later these caches will wind up being archived. NAs are still necessary and the current system works to a point but I don't think it's efficient enough and could be streamlined.

To me, NA means we need to get the cache off the map (unless the owner fixes the problem promptly). And caches we need to get off the map are caches where an NA is necessary.

 

I'm confused by the distinction you seem to be making between caches where an NA is necessary, and caches that we need to get off the map because an NA is "inevitable" (even though it is not yet "necessary").

[+barefootjeff]

 

So what trashy caches will the Cache Health system get off the map that you can't already clear just as quickly with an NA?

In my area within 25 miles there are 489 red-wrench caches. Shall I start going through the logs and place NAs on those caches deserving of an NA?

I think reviewers might prefer the algorithm help with that work. And I expect it might tick off fewer cache owners and people who think NAers are self-appointed cache cops, if the algorithm help draw out the problem caches.

Okay, fair enough. I've just had a quick look at red wrench caches around here - most are micros with full logs, in most cases the log's been replaced by a subsequent finder but the CO hasn't cleared the NM. As I said earlier, I don't have any qualms about the Cache Health thing sending emails to COs with outstanding NMs, that's pretty clear cut. It's its reliance on DNFs that bothers me, but I guess I'm going to have to admit defeat on that.

[+dprovan]

Why wait until a NA is necessary to start the process of weeding out bad caches? What's wrong with identifying them early and getting them off the map?

This is caching 101. The NM tells the CO there's a problem and gives him time to react. The NA tells the reviewer that the CO didn't react. I think the amount of time spent is both reasonable and fair.

[cezanne]

In my area within 25 miles there are 489 red-wrench caches. Shall I start going through the logs and place NAs on those caches deserving of an NA?

I think reviewers might prefer the algorithm help with that work. And I expect it might tick off fewer cache owners and people who think NAers are self-appointed cache cops, if the algorithm help draw out the problem caches.

 

Actually I expect any cacher who complains about having no enjoyable caches to visit to look at all caches in the area that could be potentially enjoyable. So it would be a natural consequence to look at logs and one could place NAs wherever appropriate without having a look at all caches with a red wrench and going through that list.

A red wrench is certainly no obstacle for a cache to be enjoyable, not even for you. It should be about the real condition of the cache and not about formalities.

 

For me it's like that: Either I want to visit a cache or I don't. In the latter case it's pretty much irrelevant for me in which condition the cache is, but as cachers are different there will be other cachers that care and if none cares, then I see no problem.

 

I never understood why the main interest of some people seems to get caches off the map. I'm selecting all my caches manually. I could not even tell how many caches are there in area with a certain property and I never understood the people who e.g. create a PQ of all caches which are disabled since say x months, or all with a red wrench and are not reviewers. Why should I care? I look at caches that I have not found and at some I found for certain reasons. But why should I look at caches with a certain property without being a reviewer and somehow only with the goal to argue about clearing the map?

Edited March 3, 2017 by cezanne

[+Team Microdot]

Sorry, but I still don't see why DNFs should continue to carry weight after a subsequent find. What health issue can a cache that's had a few DNFs but has then been found still have? Correct me if I'm wrong, but I can't think of any health issue other than "cache is missing" that can be reasonably inferred from a DNF, and surely a find, any find, proves that this inference from all those DNFs was wrong.

 

If DNFs continue to carry weight after the cache is subsequently found, I'll be even more reluctant to log a DNF on anything I can't put a signature in the log because of my own shortcomings or ineptitude.

 

With respect, allow me to present the Found It = Didn't Find It thread

[+Team Microdot]

 

Negative actions yield detrimental scores = health score decreases.

 

So how would your perceived system cope with 12 DNFs coming from the same group who cached together and visited the wrong final location or who had a car accident and then went back home vs 12 DNFs of individual cachers who all could not find the cache on that day?

 

I guess it would consider this as something very negative, wouldn't it?

 

The spam example seems less complex to me (though it is complex enough) as this widespread spectrum of different uses of DNFs is something that really makes things complicated in my opinion.

 

I missed this yesterday.

 

I've noticed that your strategy seems to be to come up with extreme (I would call them stupid) examples in a bid to prove someone else wrong.

 

To answer your question - it wouldn't. It's an automated system. It's not perfect and it never will be. There - you win - congratulations. I hope that gives you a warm, fuzzy feeling.

 

Now - a challenge - find me some logs that fit the precise scenario you describe - take as long as you need.

 

And another challenge - if you believe someone else's suggestion to be poor or flawed - come up with a BETTER suggestion. Just try it - just once - see how it feels. Or is sitting and sniping from the sidelines the best you can do?

 

#RantOver.