Jump to content

Demonising the DNF


Recommended Posts

2 minutes ago, coachstahly said:

The follow up question then becomes - Do you support Jeff's same use of previous logs and other criteria (like the algorithm) to make an assumption that a cache might still be there and that he just missed it?

 

I'm NOT asking about your thoughts about the dichotomy between his two points of view.  That's been made clear.

Jeff can do whatever he wants.

To come here though and offer one finger-in-the-wind argument as somehow reducing the validity of the other finger-in-the-wind argument just doesn't fly for me.

  • Upvote 1
Link to comment
2 minutes ago, Team Microdot said:

Jeff can do whatever he wants.

To come here though and offer one finger-in-the-wind argument as somehow reducing the validity of the other finger-in-the-wind argument just doesn't fly for me.

That's perfectly fine.  My point then becomes if the algorithm can be shown to be both right and wrong in its assessment of caches, how can it be used in such a manner to reward cachers with a virtual cache (or any future rewards as well)?  Did people who had false positives (and there were 7 in my area who received emails but had no issues with their caches) get booted from the selection process based solely on this fact?  If so, then perhaps something needs to be done (I have no idea what that could be).  I realize I'm making an assumption about the selection process for a virtual reward, but CHS has certainly been mentioned as one of the possible criteria for selection.

Link to comment
3 minutes ago, coachstahly said:

That's perfectly fine.  My point then becomes if the algorithm can be shown to be both right and wrong in its assessment of caches, how can it be used in such a manner to reward cachers with a virtual cache (or any future rewards as well)?  Did people who had false positives (and there were 7 in my area who received emails but had no issues with their caches) get booted from the selection process based solely on this fact?  If so, then perhaps something needs to be done (I have no idea what that could be).  I realize I'm making an assumption about the selection process for a virtual reward, but CHS has certainly been mentioned as one of the possible criteria for selection.

While I think that's a very valid and fair question I do think that that particular question deserves a thread of its own.

Link to comment
2 minutes ago, Team Microdot said:

While I think that's a very valid and fair question I do think that that particular question deserves a thread of its own.

The ONLY reason I posted it here is that DNFs are part of what appear to trigger a CHS notification.  For me, a DNF means that I couldn't find it.  It appears the CHS is taking it a step further by attributing multiple DNFs as a cache possibly having problems and therefore worth a CO visit.  The subsequent drop of the CHS score (assuming it drops beneath the threshold and generates an email) due to DNFs then becomes a negative stain on a COs record, which could, conceivably, eliminate them from consideration for recognition or reward for being a conscientious CO, thereby giving DNFs an added negative connotation.  I realize that the algorithm certainly takes into account factors that are controlled (D/T comes to mind immediately) but I can't imagine it takes into account the possible reasons for a DNF that don't warrant a CO check like muggles, not getting to GZ, felt unsafe going after the cache, or experience.  There's no way a computer program can be that subtle.

I don't want people to stop logging DNFs.  I don't want people to stop logging NM or NA logs.  In fact, I would prefer people to increase their use, when appropriate, to help clean up caches that need to be eliminated or tended to. I don't want the CHS to be discontinued.  I think it can certainly be a good thing but I don't necessarily think it should be relied upon beyond what it actually is, a tool to identify caches that might or might not need some attention.

Link to comment
2 hours ago, coachstahly said:

As to TeamDEMP's situations, I offer this one up as a counter to armchair logging of NM/NA logs.  It was a 1.5/1.5 with 10 previous DNFs over almost a year and a NM log suggesting that they think it's gone, as do many of the DNF logs.  Turns out I found the original container on the ground (it was a film can with a taped hook to hang on the fence) with signed logs dating to 2008.  It's my assumption the replacement went missing and then turned up again as it looks like a matchstick container in later logs.  My find is from 7/2/11.  It's my guess that TeamDEMP would have posted the NM/NA log if in the area based on the logs prior to my find.  Am I correct in that assumption?

https://coord.info/GCM69R

  • GCM69R  was archived in 2014. Your note after the 2014 reviewer disable says ..."I found this one back in 2011 and it was a film canister on the ground.  With the recent DNFs, it's either under some leaves along the fence or gone."
    Am I correct in understanding that you did not go back after the reviewer disable to determine if the film canister was in the leaf litter?
     
  • The cache owner has over 540 hides. He does not maintain them. Lets reviewers archive them. Also the cache owner does not live in Indiana, he moved to Florida.
    In this cache GC1ET72, which was archived by a reviewer in 2014, there's a 2013 NM that states:
    Needs Archived Needs Archived  12/23/2013
    Container is in pieces and I guess the owner has since moved to FL according to the local community.
     
  • There are 3 new caches in this cemetery posted in 2017 (apparently clever micros, based on the comments in the logs).
    Seems better then an old abandoned film canister that might be in the ground litter. 
    The new cache owner is active and does his own archiving--doesn't leave that task up to reviewers.
  • Upvote 2
Link to comment
1 hour ago, L0ne.R said:
  • GCM69R  was archived in 2014. Your note after the 2014 reviewer disable says ..."I found this one back in 2011 and it was a film canister on the ground.  With the recent DNFs, it's either under some leaves along the fence or gone."
    Am I correct in understanding that you did not go back after the reviewer disable to determine if the film canister was in the leaf litter?
     
  • The cache owner has over 540 hides. He does not maintain them. Lets reviewers archive them. Also the cache owner does not live in Indiana, he moved to Florida.
    In this cache GC1ET72, which was archived by a reviewer in 2014, there's a 2013 NM that states:
    Needs Archived Needs Archived  12/23/2013
    Container is in pieces and I guess the owner has since moved to FL according to the local community.
     
  • There are 3 new caches in this cemetery posted in 2017 (apparently clever micros, based on the comments in the logs).
    Seems better then an old abandoned film canister that might be in the ground litter. 
    The new cache owner is active and does his own archiving--doesn't leave that task up to reviewers.

You completely missed the point of my post, nor was it specifically addressed to you.  Allow me to re-quote the first sentence.  "As to TeamDEMP's situations, I offer this one up as a counter to armchair logging of NM/NA logs...."  TeamDEMP offered examples of caches they felt warranted an armchair NM/NA log.  I'm wanting to see if they would have filed such a log on this cache, which was actually there but covered under quite a few leaves, dead and otherwise.  I did NOT say the eventual archive wasn't warranted; I did NOT say the cache was a great cache worth saving; I did NOT say the CO was an amazing owner when it comes to maintenance.

Considering it is over 100 miles from my normal home (but close enough to my mom's lake house for summer weekends), no I didn't visit because it wasn't feasible.  How is it even valid in the context of whether or not TeamDEMP would have filed an armchair NM/NA log?  What is the point of bringing that up other than to point out I posted a note that let people know what I found and how I found it?  I didn't make any claims about it hopefully being saved or anything like that.  If I remember correctly, the only reason I posted was that I was downloading a new My Finds PQ into my GSAK database and every time I do that, I get a notification of a change in status in all caches in my finds.  I usually look at each one of interest to me to see what happened.  

As to the other caches, I know this CO doesn't maintain on a regular basis.  Again, irrelevant to the specific topic I was addressing.

3 new caches?  Again, I ask you, how is this SPECIFICALLY related to the question of an armchair NM/NA log?  It's not and is deflection away from the point I was trying to raise.  I don't care that the new cache owner is active and does his own archiving.  Good for them but it's completely irrelevant to the topic at hand.  We all know you have a thing about maintained caches and bring it up whenever you can.  However, this specific post was NOT about how well a CO maintains, how well the new CO maintains, or how much I may or may not have wanted this cache to stick around.  It was providing an example of what I thought might be an example of a hypothetical erroneous armchair NM/NA log when the cache wasn't actually MIA.  To top it off, this wasn't a great container (film can) but the log was dry as a bone for not having been signed since 2008.

Link to comment
11 hours ago, barefootjeff said:

Logging armchair NMs just seems a bit presumptuous to me, that's all, and if you're near enough to the cache for the DNFs to bother you, why not just take the time to actually go and have a look?

Yes, logging armchair NMs is presumptuous, which is why I always back it up with a careful explanation in the NM text saying why I think the armchair NM is justified. Yes, I prefer going to GZ "just because" even when I'm sure I won't learn anything, but sometimes it doesn't make sense to delay while waiting for me to get a chance or risk me forgetting about it entirely.

But that's NMs. For NAs, in most cases I don't consider a visit to GZ because I have no reason to doubt the judgement of the person filing the NM.

12 hours ago, barefootjeff said:

You never know, you might just be the one who spots the tricky camo and finds it.

No. When I file an armchair NM, there's no chance the problem is "tricky camo". That's not the kind of cache we're talking about. We're talking about caches that are missing and will almost certainly be archived by the time the smoke clears. We're talking about caches with DNFs saying "not under LPC" when the hint is "LPC" and the difficulty is 1.5 and the cache was found every other day for years but then suddenly has had no logs for months.

I think you're arguing against armchair caches because your caches will never deserve them. I have no doubt that's true. But other caches do deserve them, so I'd like to keep them in our toolbox even at the expense of us occasionally getting one that's in error.

  • Upvote 3
Link to comment
8 minutes ago, dprovan said:

No. When I file an armchair NM, there's no chance the problem is "tricky camo". That's not the kind of cache we're talking about. We're talking about caches that are missing and will almost certainly be archived by the time the smoke clears. We're talking about caches with DNFs saying "not under LPC" when the hint is "LPC" and the difficulty is 1.5 and the cache was found every other day for years but then suddenly has had no logs for months.

I think you're arguing against armchair caches because your caches will never deserve them. I have no doubt that's true. But other caches do deserve them, so I'd like to keep them in our toolbox even at the expense of us occasionally getting one that's in error.

Fair enough, it's probably a case of comparing apples and oranges. Our lamp posts don't have covers over their nuts so we're spared that scourge and most of the caches I do aren't anywhere near as clear cut as that. About the only DNF I've logged where I was really convinced the cache was missing was one where the CO provided a spoiler photo of the hidey-hole and I found the hole but it was empty.

Link to comment
On 10/10/2017 at 3:04 AM, barefootjeff said:

Because a DNF doesn't imply a cache needs maintenance, it only says someone didn't find it. It's up to the person logging the DNF as to whether they think the cache is likely missing or if it's more likely the hiding place has just defeated them on the day.

But a string of DNFs is an indicator that there might be a problem. Nowhere is it ever assumed that a string of DNFs means that there IS a problem. Even in the (problematic, imo) email wording. The suggestion is that the CO checks on it to ensure that it's still in good condition.  Is it an inconvenience doing that if it is still in good condition? Sure is. But that inconvenience can happen if someone logs a needless NM, or puts in a log that they think it's missing causing you to go check it and it's still there. CO's are quite often inconvenienced in this hobby. It's the bitter side of cache ownership responsibility :P

 

On 10/10/2017 at 5:45 AM, Team Microdot said:
On 10/10/2017 at 5:37 AM, anpefi said:

Obviously, you cannot be absolutely sure ;)

Quite.

All you have is a hunch devoid of any evidence to back it up.

---

Bayesian probability - statistical guessing.

I'm guessing the CHS relies significantly on the same sort of statistical guessing and, if this very thread is anything to go by - that's very far from reliable ;)

More like an educated guess by past related experiences. And I'd argue it's not "very far" from reliable. Except perhaps solely within the context of this thread. In the grand scheme, and based on feedback we have already received from GS and reviewers, I'm confident that false positives are a tiny fraction of caches identified to have a potential problem. Demand the statistics from GS.

 

On 10/10/2017 at 7:25 AM, Team Microdot said:

No - only to make an educated guess.

hah, I just didn't want to read the whole thred before replying =P

 

23 hours ago, Team Microdot said:

The icing on the cake I suppose was another poster suggesting that bayesian analysis together with gut feel was a sure-fire way to be certain that a cache one could not and had never found was probably still there.

:wacko: To "be certain" that a cache "was probably" still there... I see no problem with that. I think you're conflating certainty about a probability with certainty about a specific state. Nowhere was Jeff saying that he was ever certain that the cache was or was not there. So, you have a problem with his certainty that the cache may or may not be there? :P

Edited by thebruce0
Link to comment
23 hours ago, coachstahly said:

I realize the email doesn't MEAN anything and cachers are free to ignore it, but the CHS was used (apparently as one of many criteria) for selection of a virtual reward.  If it's going to have that type of importance/relevance in the future, then it needs to be adjusted so as not to generate as many false positives as it did recently, at least in my area and Red Sox Mark's area.  No, I do NOT know if the COs emailed Groundspeak to notify them of the fact that the email was sent erroneously but I know if I had received one in error, I would have emailed them.

This

 

23 hours ago, Team Microdot said:
23 hours ago, coachstahly said:

Do you support the algorithm's use with regard to the CHS and the subsequent email? 

Yes - I do.

Do I think it's perfect? No.

Do I think it needs to be perfect? No.

On balance, would I keep it the way it is rather than not have it at all? Yes.

Are there better solutions? Potentially - but they are so very unlikely to come to fruition unless there's a quantum shift in attitudes.

Wow, agreed on all points. *head asplode*

 

22 hours ago, coachstahly said:

For me, a DNF means that I couldn't find it.  It appears the CHS is taking it a step further by attributing multiple DNFs as a cache possibly having problems and therefore worth a CO visit.

And is there a problem with this?  There's definitely more probability that a cache has problems if it has a string (over days, not same-day) of DNF logs. Of course it's quite nuanced when taking context into consideration.  But, knowing the algorithm "learns" from feedback, is this fundamentally a legitimate problem? Especially if false positives are, at most, a minority, and that caches that do have problems needing a checkup are no longer left to rot, as it were, unawares or ignored.  Superficially (in most cases) incovenience the minority so that the majority which are problematic are (ideally) corrected and/or archived, to benefit the entire community?

 

22 hours ago, coachstahly said:

The subsequent drop of the CHS score (assuming it drops beneath the threshold and generates an email) due to DNFs then becomes a negative stain on a COs record, which could, conceivably, eliminate them from consideration for recognition or reward for being a conscientious CO, thereby giving DNFs an added negative connotation.

This is the part I agree with. The difference between using the CHS to identify potential problems (good), vs using the CHS to identify and reward "good cache owners".  Not "possibly" good cache owners. The actual reward renders the "possibly" meaningless. False positives - not-good owners - have already been rewarded by the unguaranteed CHS.  That, to me, is the biggest problem here.  Not whether the DNF should or shouldn't be counted in the algorithm.

 

22 hours ago, coachstahly said:

I don't want people to stop logging DNFs.  I don't want people to stop logging NM or NA logs.  In fact, I would prefer people to increase their use, when appropriate, to help clean up caches that need to be eliminated or tended to. I don't want the CHS to be discontinued.  I think it can certainly be a good thing but I don't necessarily think it should be relied upon beyond what it actually is, a tool to identify caches that might or might not need some attention.

This, agreed.

Link to comment

The comment you responded to, and your contesting of that comment: "a sure-fire way to be certain that a cache one could not and had never found was probably still there." As I quoted.   But I suppose it's easy to consider your own comments "that best I've heard" =P  Specifically, you seemed to be arguing against something I don't ever recall barefootjeff ever saying, that he was ever certain a cache was or was not there when logging a DNF.

Link to comment
12 minutes ago, thebruce0 said:

The comment you responded to, and your contesting of that comment: "a sure-fire way to be certain that a cache one could not and had never found was probably still there." As I quoted.   But I suppose it's easy to consider your own comments "that best I've heard" =P  Specifically, you seemed to be arguing against something I don't ever recall barefootjeff ever saying, that he was ever certain a cache was or was not there when logging a DNF.

Did you say earlier that you hadn't caught up on your reading?

Link to comment

I realize this cache has some previous DNFs, but looking through the cache logs, some didn't go inside, some didn't stay long (spiders) and some were new cachers probably reaching beyond their ability initially.  It's a 3/2.5, was found twice before (found by two cachers together) the singular DNF that generated the automated email.  It's my guess the history was responsible for the email but there have been NO replacement caches placed and quite a few of the DNFs returned to find it.  A couple found logs even mention that this one will most likely always be there.  The find rate is roughly 2/1 and without knowing what this cache is, it's my best guess that it's unlikely it will ever disappear due to muggles.  I would say the probability is high that the cache is still in place and doesn't need maintenance and would venture to guess that most posters on here would agree.  The CO has been told that they should post an OM from now on, instead of notes.

https://coord.info/GC1M3D3

Link to comment

And if the reviewer will essentially 'turn a blind eye' to couch-logged OM logs, then in this case that is the CO's solution (posting the OM log to restore the CHS, without a personal checkup)

Presuming the process implied by the reviewer is this:

* GC's CHS is low, triggering an automated email
* Reviewer on a routine scanning for low CHS caches is able to judge that the cache doesn't have a problem, takes no proactive action
* CO gets the email and ignores it knowing the cache is fine based on reviewing recent logs
* CO opts to post an OM log instead of a Note, solely to increase the CHS again
* Reviewer on a routine scanning for low CHS caches no longer sees the cache. Perhaps decides to check on recent OM logs - sees the coach-logged OM along with recent contextual logs, is able to judge that the cache didn't have a problem, doesn't immediately consider the owner a 'bad cache owner', takes no proactive action.

One "fairness" problem, imo, I see is:

* GC's CHS is low because of DNFs.
* Before the CO gets around to the afore-mentioned actions, GS decides to reward high-CHS cache owners solely by algorithmic results, without a human component in selection, and GC's owner is left out of possible consideration.

Link to comment
37 minutes ago, coachstahly said:

I realize this cache has some previous DNFs, but looking through the cache logs, some didn't go inside, some didn't stay long (spiders) and some were new cachers probably reaching beyond their ability initially.  It's a 3/2.5, was found twice before (found by two cachers together) the singular DNF that generated the automated email.  It's my guess the history was responsible for the email but there have been NO replacement caches placed and quite a few of the DNFs returned to find it.  A couple found logs even mention that this one will most likely always be there.  The find rate is roughly 2/1 and without knowing what this cache is, it's my best guess that it's unlikely it will ever disappear due to muggles.  I would say the probability is high that the cache is still in place and doesn't need maintenance and would venture to guess that most posters on here would agree.  The CO has been told that they should post an OM from now on, instead of notes.

https://coord.info/GC1M3D3

What's the context for posting this? Was this about how the CHS would react and did the CO receive some CHS email?

Or you asking if someone who hasn't visited the cache would possibly consider a NM/NA? Looking at this specific one, regardless of when I might would have visited this cache and the previous logs at that time, I'd not post a NM, so I certainly wouldn't post one without visiting the cache. No one has posted a NM log for this cache. 

I'm not sure there was any need for an OM to be posted by the owner and the note posted in August 2016 seems appropriate to me.  No NM needed to be cleared and it doesn't seem as if the owner visited the cache/performed maintenance based on the note text. 

Link to comment
10 minutes ago, Team DEMP said:

What's the context for posting this? Was this about how the CHS would react and did the CO receive some CHS email?

It's another cache I feel doesn't warrant an automated email based on what I see taking place within the logs.  They received the automated email based on some valid DNFs and some DNFs that weren't really valid.  Our reviewer said to post an OM to all caches so as to make sure it bumps up/resets the CHS so notes, while still verifying the cache is in play, won't do anything to the CHS and it will remain on the low side of the threshold until an OM is posted.

I just wondered if any of the cachers in this thread (or the forum) would have felt this warranted a check by the CO based on the logs at hand.  Does the human viewpoint feel this cache should have received an email?

Link to comment
22 minutes ago, coachstahly said:

It's another cache I feel doesn't warrant an automated email based on what I see taking place within the logs.  They received the automated email based on some valid DNFs and some DNFs that weren't really valid.  Our reviewer said to post an OM to all caches so as to make sure it bumps up/resets the CHS so notes, while still verifying the cache is in play, won't do anything to the CHS and it will remain on the low side of the threshold until an OM is posted.

I just wondered if any of the cachers in this thread (or the forum) would have felt this warranted a check by the CO based on the logs at hand.  Does the human viewpoint feel this cache should have received an email?

Having read through the logs it does seem that this one has a fairly routine cycle of finds interspersed with DNF's - for various good reasons.

It makes me wonder if the CHS is smart enough to establish a 'baseline' for an individual cache and then 'look for' deviations from that baseline i.e. look for exceptions to the norm...

Link to comment
3 minutes ago, Team Microdot said:

It makes me wonder if the CHS is smart enough to establish a 'baseline' for an individual cache and then 'look for' deviations from that baseline i.e. look for exceptions to the norm...

I don't think programs are that sophisticated yet, as much as we might like.  I'm reminded that it's only an email, but the CHS has fallen beneath a threshold pre-determined by Groundspeak, which goes to the "record" of the CO.  That's not a bad thing, in and of itself, but as I posited earlier, if the CHS is meant to be used as more than just a tool....

I know thebruce0 posted in the other thread a reviewer NM based on the CHS, which is different than the situation here. 

Link to comment
31 minutes ago, coachstahly said:
43 minutes ago, Team DEMP said:

What's the context for posting this? Was this about how the CHS would react and did the CO receive some CHS email?

It's another cache I feel doesn't warrant an automated email based on what I see taking place within the logs.  They received the automated email based on some valid DNFs and some DNFs that weren't really valid.  Our reviewer said to post an OM to all caches so as to make sure it bumps up/resets the CHS so notes, while still verifying the cache is in play, won't do anything to the CHS and it will remain on the low side of the threshold until an OM is posted.

I just wondered if any of the cachers in this thread (or the forum) would have felt this warranted a check by the CO based on the logs at hand.  Does the human viewpoint feel this cache should have received an email?

Ah - completely agree that https://www.geocaching.com/geocache/GC1M3D3_walk-down-an-old-railway-in-kettering-3 shouldn't have triggered a message to the cache owner to investigate the health of their cache.  As I noted in my earlier reply, I wouldn't NM it if I DNF'ed it.  I don't know if anyone reached back out to Groundspeak and suggested they review the algorithm to see what logic triggered the alert and that it be tweaked to account for whatever it detected here. 

Link to comment
4 hours ago, Team DEMP said:

I don't know if anyone reached back out to Groundspeak and suggested they review the algorithm to see what logic triggered the alert and that it be tweaked to account for whatever it detected here.

As I mentioned earlier in the thread, Groundspeak's response, nine months ago at least, to reports of false positives was a brush-off and reprimand. I'd like to hope that's changed but I haven't seen anyone say that it has.

Link to comment
17 minutes ago, barefootjeff said:

As I mentioned earlier in the thread, Groundspeak's response, nine months ago at least, to reports of false positives was a brush-off and reprimand. I'd like to hope that's changed but I haven't seen anyone say that it has.

This needs context and links.  And since the 'threshold' is subjective, even the definition of a 'false positive' is itself up for the judgement of HQ.  On the contrary, I'm confident I've heard at least one or two lakeys over the months either respond appreciatively for constructive feedback or indicate that people are effectively watching the discussions and taking things into consideration to varying degrees. And no I don't want to go scraping over months of threads to try to find the occasional post :P.  But those observations are what brought me to my current standing on the issue.

Edited by thebruce0
  • Upvote 1
Link to comment
8 minutes ago, thebruce0 said:
23 minutes ago, barefootjeff said:

As I mentioned earlier in the thread, Groundspeak's response, nine months ago at least, to reports of false positives was a brush-off and reprimand. I'd like to hope that's changed but I haven't seen anyone say that it has.

This needs context and links.  ...And no I don't want to go scraping over months of threads to try to find the occasional post :P

Demands evidence, can't be bothered to reciprocate :rolleyes:

Link to comment
23 minutes ago, thebruce0 said:

This needs context and links.  And since the 'threshold' is subjective, even the definition of a 'false positive' is itself up for the judgement of HQ.  On the contrary, I'm confident I've heard at least one or two lakeys over the months either respond appreciatively for constructive feedback or indicate that people are effectively watching the discussions and taking things into consideration to varying degrees. And no I don't want to go scraping over months of threads to try to find the occasional post :P.  But those observations are what brought me to my current standing on the issue.

I'm referring specifically to the email I got back from HQ in response to my attempt to report a blatant false positive (1 DNF on a 7-week-old D2/T5 cache), which explained how the CHS worked, said that it has been well-received by the community and finished off with a stern reminder that, when hiding the cache, I'd agreed to the guidelines including "Owner is responsible for visits to the physical location." I suspect this is just a boilerplate response that goes out to anyone contacting them about CHS issues, maybe something does get passed down to the developers but there was nothing in the reponse to indicate that or explain why the cache had been targeted.

Edited by barefootjeff
Link to comment
5 minutes ago, NanCycle said:
31 minutes ago, barefootjeff said:

Just curious what an "invalid" DNF is. Did they really find it and are lying? The opposite of "found it = didn't find it"?

Probably a DNS.  Didn't really look for it for whatever reason, weather, crabby kids, had to find a bathroom, etc.

On at least four occasions I've logged DNF when I've been able to see the cache but wasn't game to climb up or out to reach it because of my impaired sense of balance. Were they invalid DNFs? I've also had quite a few DNFs on my hides citing various reasons like approaching storms, failing light, not agile enough, too many mosquitoes, no internet access, dead batteries, etc. - are they all invalid?

Link to comment
1 hour ago, Team Microdot said:

can't be bothered to reciprocate :rolleyes:

I could say so many things, but I won't.

 

1 hour ago, barefootjeff said:

I'm referring specifically to the email I got back from HQ in response to my attempt to report a blatant false positive (1 DNF on a 7-week-old D2/T5 cache), which explained how the CHS worked, said that it has been well-received by the community and finished off with a stern reminder that, when hiding the cache, I'd agreed to the guidelines including "Owner is responsible for visits to the physical location." I suspect this is just a boilerplate response that goes out to anyone contacting them about CHS issues, maybe something does get passed down to the developers but there was nothing in the reponse to indicate that or explain why the cache had been targeted.

Ah ok, tho I fail to see where there's a "reprimand" there.  If it was indeed boilerplate, then they merely echoed what's already known, not made a personal chiding against you for asking. Rather, it may have seemed stand-offish if they appeared to dance around the question.

Also, quite often to avoid complex interactions a canned response that supports existing statements could be sent, while behind the lines the information is passed on to people in a better position to make productive decisions.  Is that actually what happened here? No idea. But your situation doesn't convince me that from the horse's mouth they admitted they don't consider feedback to adjust the algorithm.  Additionally, if they admit to making use of specific feedback, that part of the algorithm becomes 'known', and could thus increase concern for 'abuse'.  So as long as the email is still true, they can send it out.  And just as many reviewers are more often using as objective and factual a boilerplate reviewer note as possible to remove any semblance of personal input, opinion, or judgement that could be deemed flawed, due to reciprocated anger and debate in past situations, it's likely that emails from support like this have also grown to be 'stern' and direct, rather than friendly and understanding.

Edited by thebruce0
  • Upvote 1
Link to comment
11 minutes ago, thebruce0 said:
1 hour ago, barefootjeff said:

I'm referring specifically to the email I got back from HQ in response to my attempt to report a blatant false positive (1 DNF on a 7-week-old D2/T5 cache), which explained how the CHS worked, said that it has been well-received by the community and finished off with a stern reminder that, when hiding the cache, I'd agreed to the guidelines including "Owner is responsible for visits to the physical location." I suspect this is just a boilerplate response that goes out to anyone contacting them about CHS issues, maybe something does get passed down to the developers but there was nothing in the reponse to indicate that or explain why the cache had been targeted.

Ah ok, tho I fail to see where there's a "reprimand" there.  If it was indeed boilerplate, then they merely echoed what's already known, not made a personal chiding against you for asking. Rather, it may have seemed stand-offish if they appeared to dance around the question.

Also, quite often to avoid complex interactions a canned response that supports existing statements could be sent, while behind the lines the information is passed on to people in a better position to make productive decisions.  Is that actually what happened here? No idea. But your situation doesn't convince me that from the horse's mouth they admitted they don't consider feedback to adjust the algorithm.  Additionally, if they admit to making use of specific feedback, that part of the algorithm becomes 'known', and could thus increase concern for 'abuse'.  So as long as the email is still true, they can send it out.  And just as many reviewers are more often using as objective and factual a boilerplate reviewer note as possible to remove any semblance of personal input, opinion, or judgement that could be deemed flawed, due to reciprocated anger and debate in past situations, it's likely that emails from support like this have also grown to be 'stern' and direct, rather than friendly and understanding.

The stern reminder at the end did come across to me as a reprimand, effectively saying that if I'd kept up my side of the bargain and "visited the physical location" more often, this wouldn't have happened. Sheeze, the cache was only 7 weeks old and obviously I'd visited the location then to hide it, and the CHS email went out less than 4 days after the DNF was logged so they don't give you much time, particularly on a T5 cache that requires a fair bit of effort and planning with tides and winds to visit.

Sure, they're not saying that they don't consider feedback, but they're certainly not welcoming it either, or at least they weren't then. Maybe things have changed. I'm by nature an optimist, so yeah, maybe the next time I get a CHS email it'll go better.

Link to comment
1 hour ago, barefootjeff said:

On at least four occasions I've logged DNF when I've been able to see the cache but wasn't game to climb up or out to reach it because of my impaired sense of balance. Were they invalid DNFs? I've also had quite a few DNFs on my hides citing various reasons like approaching storms, failing light, not agile enough, too many mosquitoes, no internet access, dead batteries, etc. - are they all invalid?

I suppose that depends on whether the meaning of DNF is "Did Not Find", or the meaning of DNF is "1/2 a Needs Maintenance" (or some other fraction of a Needs Maintenance).

Link to comment
3 hours ago, barefootjeff said:

I'm referring specifically to the email I got back from HQ in response to my attempt to report a blatant false positive (1 DNF on a 7-week-old D2/T5 cache), which explained how the CHS worked, said that it has been well-received by the community and finished off with a stern reminder that, when hiding the cache, I'd agreed to the guidelines including "Owner is responsible for visits to the physical location." I suspect this is just a boilerplate response that goes out to anyone contacting them about CHS issues, maybe something does get passed down to the developers but there was nothing in the reponse to indicate that or explain why the cache had been targeted.

 

On 10/5/2017 at 6:27 PM, Keystone said:

On behalf of Signal the Frog, and all other amphibian mascots for geolocational gaming activities, I extend profuse apologies for the one email sent to Barefoot Jeff as a result of the Cache Health algorithm that was in effect in late 2016.

 

  • Upvote 3
Link to comment
4 hours ago, barefootjeff said:

As I mentioned earlier in the thread, Groundspeak's response, nine months ago at least, to reports of false positives was a brush-off and reprimand. I'd like to hope that's changed but I haven't seen anyone say that it has.

On 9/19/2017 at 1:55 PM, Keystone said:

The cache health algorithm is pretty sophisticated, and it's tweaked from time to time in response to feedback.  One such tweak is in testing right now, for deployment in the near future.  Some of that feedback comes from reviewers, who can see each cache's health score.  We can, and do, question why a cache triggered a reminder email to the owner under certain circumstances that, as geocachers, we'd call premature, and where, as reviewers, we would not take action.  The opposite is also true:  "why didn't this 1/1 cache with eight straight DNF's over the past year not trigger an email reminder to the owner?"

 

  • Upvote 3
Link to comment
3 hours ago, Team Microdot said:

I'm sure it'll be very wordy and strive to cover every possible eventuality. Or if not, a subsequent edit will plug any gaps.

It's only wordy (often to the same people) because I've learned that some people enjoy picking through a comment they're debating to find any possible satellite case or exception and stating it as if it nullifies the entire argument. It's quite annoying. As I fully expect you will reciprocate. So. Let's move along, yes?

 

Also, thanks Keystone for those quotes. Someone should bookmark the comments forperpetuity :P

  • Upvote 1
Link to comment
5 hours ago, thebruce0 said:
9 hours ago, Team Microdot said:

I'm sure it'll be very wordy and strive to cover every possible eventuality. Or if not, a subsequent edit will plug any gaps.

It's only wordy (often to the same people) because I've learned that some people enjoy picking through a comment they're debating to find any possible satellite case or exception and stating it as if it nullifies the entire argument. It's quite annoying.

It is not unheard of for even the most carefully crafted, essay-length, base-covering post, even after subsequent gap-plugging edits, to leave some opportunity for someone to hold and voice an alternative opinion - or even to identify one small but key element contrary to the overall thrust of the argument presented.

There are days, admittedly rare, where I miss Cezanne.

Link to comment
10 hours ago, niraD said:

I suppose that depends on whether the meaning of DNF is "Did Not Find", or the meaning of DNF is "1/2 a Needs Maintenance" (or some other fraction of a Needs Ma

In the context of this and other threads about the CHS,  I don't think it's so much the meaning of a DNF, but what a DNF apparently is supposed to indicate.  When a DNF or several DNFs is used to determine a CHS score it's based on the premise that the DNF indicates that some sort of maintenance is required (basically that the cache is missing).  I can think of many other reasons why a DNF may have been posted that would not indicate that there is anything wrong with the cache.

The cache just may be very difficult to find.  In that case, the relative experience of those trying to find it may be a factor.  Is the CHS algorithm considering the experience of those trying to find a cache when determining if a DNF is "valid"?   Often multiple DNFs are a result of a group of geocacher (which may be made of geocachers with little experience)  searching at the same time?  Can the CHS algorith determine when multiple DNFs were all made by the same group?

What was the weather like?  Trying to find a cache hidden near the ground after a foot of snow has fallen can be very difficult.   Although there may be times when a DNF was really a DNS I know that when if there is torrential rain I might still search for the cache but probably wouldn't search as thoroughly as I might if it were a clear sunny day.  Similarly, if there are a lot of muggles about that may also determine how thoroughly I search for a cache.  Can the CHS algorithm consider environmental factors which may impact how thoroughly a geocacher searches for a cache?  

I know that these factors can be considered by a reviewer before they take additional action, but that doesn't change the CHS score.  As far as I know, the only way to improve a CHS on a cache is to perform a maintenance visit and post a OM log. 

Link to comment
14 hours ago, barefootjeff said:

Just curious what an "invalid" DNF is. Did they really find it and are lying? The opposite of "found it = didn't find it"?

Invalid wasn't the best word choice there.  

"Got to GZ. Looked around. Hmm. Should I try this by myself? Answer. Nope. I'll bring someone with me next time."  I would have posted a note instead of a DNF.

"Plan to go back soon with my sneakers on instead of sandals! I will try again!"  My guess that they didn't even go in as there was water and they didn't want to get their feet wet.  Note for me.

I have no problem with these two cachers posting DNFs.  I would have gone in a different direction because it appears neither one actively began searching for the cache.  I also realize the reasons for DNF logs has been repeatedly raised so not looking to reopen that can of worms.

Link to comment

(moving along...)

2 hours ago, NYPaddleCacher said:

The cache just may be very difficult to find.  In that case, the relative experience of those trying to find it may be a factor.  Is the CHS algorithm considering the experience of those trying to find a cache when determining if a DNF is "valid"?   Often multiple DNFs are a result of a group of geocacher (which may be made of geocachers with little experience)  searching at the same time?  Can the CHS algorith determine when multiple DNFs were all made by the same group?

What was the weather like?  Trying to find a cache hidden near the ground after a foot of snow has fallen can be very difficult.   Although there may be times when a DNF was really a DNS I know that when if there is torrential rain I might still search for the cache but probably wouldn't search as thoroughly as I might if it were a clear sunny day.  Similarly, if there are a lot of muggles about that may also determine how thoroughly I search for a cache.  Can the CHS algorithm consider environmental factors which may impact how thoroughly a geocacher searches for a cache?  

Exactly all this, and why I strongly hold the position that debating or reverse engineering the CHS is pointless and a waste of time. There are far too many [potential] factors involved, and it's just appeasing the desire and demand to find out why something happened that we don't necessarily like. Jumping to blame-the-CHS-because-blah-blah-blah is problematic, and so "demonising the DNF" is just as much. Many things we do regularly can play a role in affected a CHS, and the DNF is just one of them. And we simply don't actually know how complex (or simple) the algorithm is when it decides how much to weigh the existence of one more DNFs on a cache.

If we find a 'false positive', we don't actually know if it's a false positive, or if there is indeed some other element of the algorithm that caused it.  All we can do it raise the instance to the attention of TPTB and hope they too see it as a false positive (by effectively defending why it should be considered one) and adjust the algorithm.

Edited by thebruce0
  • Upvote 1
Link to comment

As far back as I can remember multiple dnf's have been an indicator that something could be wrong with a cache.   It doesn't mean there is a problem,  only there could be.   As a cache owner, "could be"  is enough for me to want to take a look.  

I agree a dnf can mean many things,  one of them being the cache is missing.   

If GS is intent on including multiple dnf's as part of the CHS than it seems prudent to define what a dnf is and when it should be used.   

Link to comment
3 hours ago, justintim1999 said:

I agree a dnf can mean many things,  one of them being the cache is missing.  

Another is that the cache is not missing, but was simply not found.  As a cache owner, one can be very diligent about maintaining their caches, going out almost daily to confirm the cache is in place and does not need maintenance but that's will not prevent DNFs on a cache simply because someone couldn't find it.  Here are a few things that I CO could do to prevent or at least keep DNF logs at a minimum:

Only place caches that are very easy to find.

Only place caches close to home where one can spend as little time as possible walking from a vehicle to the cache.

Give explicit spoiler hints regarding the location (with pictures and arrows) of the cache, which makes it very easy to find.

Allow throw downs on the cache so nobody every has to post a DNF.

Allow people the log the cache as found even when they didn't find the cache.

None of those things are particularly good for the game (though providing useful hints doesn't really hurt), yet I'm concerned that's the direction some will take rather than going out  on almost a daily basis to check on their geocaches.

Link to comment

Or, considering upping the difficulty rating if a cache regularly attracts DNF's.  If only half the searchers succeed in their hunt, and the DNF's are not from brand new geocachers, then perhaps the cache shouldn't be rated at 1.5 difficulty stars. 

(This assumes that the cache health score takes the cache's difficulty rating into account, which I can neither confirm nor deny.)

  • Upvote 4
Link to comment
30 minutes ago, Keystone said:

Or, considering upping the difficulty rating if a cache regularly attracts DNF's.  If only half the searchers succeed in their hunt, and the DNF's are not from brand new geocachers, then perhaps the cache shouldn't be rated at 1.5 difficulty stars. 

(This assumes that the cache health score takes the cache's difficulty rating into account, which I can neither confirm nor deny.)

Correcting a difficulty rating to accurately reflect the difficulty of finding the cache is good.  Inflating the difficulty rating to a 3 or more on a cache that should be a 1.5 (assuming that the difficulty rating has an impact on the CHS), not so good.

Link to comment

Well that's the problem - it's what the CO believes it should be vs how it turns out in the real world. The CO could insist that it really is a 1.5 difficulty - but the CO isn't the one finding it. And if the find stats (and cacher finder stats) imply that it's really not that easy, maybe the CO should bite the bullet and start adjusting the D.

Ultimately it is up to the CO, of course. And, adjust the D won't automatically reduce DNFs. Adjusting the D would only (theoretically) affect whether the algorithm thinks the DNFs indicate a possibly missing cache. So... *shrug*

Link to comment
2 hours ago, Keystone said:

(This assumes that the cache health score takes the cache's difficulty rating into account, which I can neither confirm nor deny.)

I wonder whether the CHS takes a cache's terrain into account, because some of the examples of DNF logs that say nothing about whether the cache is really there seem to indicate that the seeker was foiled by being unprepared for the terrain. The cache could be a genuine 1-star or 1.5-star difficulty, but if the terrain keeps people from reaching GZ, and if a significant portion of those people post DNF logs, then the CHS could falsely flag that cache for a "reminder" email.

Hypothetically speaking, of course.

Edited by niraD
Link to comment
19 minutes ago, niraD said:

I wonder whether the CHS takes a cache's terrain into account, because some of the examples of DNF logs that say nothing about whether the cache is really there seem to indicate that the seeker was foiled by being unprepared for the terrain. The cache could be a genuine 1-star or 1.5-star difficulty, but if the terrain keeps people from reaching GZ, and if a significant portion of those people post DNF logs, then the CHS could falsely flag that cache for a "reminder" email.

Hypothetically speaking, of course.

Like what bruce0 said above, the CO might think a cache that requires a ladder is a terrain 2, but to the average cacher that doesn't carry a small ladder with them at all times, it may be more accurately rated a T4 (at the very least a T3.5).

(I've had this happen to me, in my case a 6 foot ladder would be required, the CO felt high terrain ratings are "rewards" and since only a short ladder (for tall people) was required it didn't warrant more then a T2).

Link to comment
37 minutes ago, niraD said:

I wonder whether the CHS takes a cache's terrain into account, because some of the examples of DNF logs that say nothing about whether the cache is really there seem to indicate that the seeker was foiled by being unprepared for the terrain. The cache could be a genuine 1-star or 1.5-star difficulty, but if the terrain keeps people from reaching GZ, and if a significant portion of those people post DNF logs, then the CHS could falsely flag that cache for a "reminder" email.

Hypothetically speaking, of course.

In that case those people shouldn't be posting a dnf .

That may help solve some of the problem.

Link to comment

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
×
×
  • Create New...