Can challenge checker send e-mail on logged find if not qualified?

Question

Project-gc recently made it possible for CO's to get emails when people post photos that betray the location of the cache. This seems a cool feature that I have not encountered yet.

It would be nice if challenges, it notified folks if someone has issued a found log and according to the checker the cache has not been completed. The reason for using the word folks, is there seems to be several people who might be interested in getting this information.  First the "faux finder" might want to get the information to correct their log. Secondly the CO might want it to either delete the log or see that the checker gets fixed.  Additionally whoever tagged it might want to improve the tag.  Finally the script author might feel the same way as the tagger. This seems a good tool to improve the quality of the checkers and the finds.

Thanks.

Answer 1

Our opinion is the following. Our challenge checkers are contributed by the community. In other words, we are not creating the code ourselves. Therefore, we can not guarantee its correctness, if the rules of the challenge has been correctly interpreted, if the rules has changed, or if the code is bug free.

It would be wrong for us to send emails to people that might not have fulfilled it. In my opinion, it would be fairly easy for a cache owner to run the checker on every new found log he/she gets on challenges, especially when it's doubtful. If the checker says no, then he/she should look closer at the users proof, or contact the user if there is no proof attached.

I am also a bit unsure about your idea on one point. Whom do you suggest we contact? We never send emails to anyone who hasn't actively asked for them. So we can not send to users who logged a challenge and shouldn't. In fact, we don't have any email addresses if they have not authenticated at Project-GC and added the email address, and verified it.

It could be another notifier service at Project-GC, for challenge cache owners. This we could consider if there was demand for it. But personally, I think it's just as easy to check them one by one. At least if you don't own a lot of them.

Please correct me if I misunderstood your question/suggestion.

Comments

An easy way to implement the possibility for a CO is to a make it possible to use the validate function for more then the last 10 logs. All and a specified number or after a date.To enable it only for the CO is a possibility

If that used the known ok result and only test unknown and not ok it would also be quite fast. Perhaps not the first time but after that depending on how long the results are stored.
I suspect that will result in few false positive because logs are not that often removed and other changes means that you did qualified and that is ok according to at leas my interpretation of the guidelines

Another problem with mail is that some challenges will result in not ok even if it was ok when you logged the challenge even if nothing is change. Only new logs, fp, caches and time can do that.
The checker can only check if you qualified now and not the condition when you logged the challenge witch is the important point.
The Jasmer Challenges(Logged all hidden month) will fail the each new month when another hidden month is require is the most obvious one.
There is other that uses top FP in an area, empty area, % cachetype, Archived not allowed will.
An of course DT, attribute, name etc on caches can change. It is impossible to check for the past because not all data is tracked and if was the rewrite and extra complexity would not be worth it.

---

I think you get the gist of the question. I do disagree with you on one point and I think there is some confusion.  

As someone who has only 1 challenge cache out there with a checker I tagged myself, getting an automatic response would be useful. Life gets busy and automation would allow quicker feedback to the not finders. The biggest feature of a notifier is that I can check offline as match the found log and the notifier e-mail. I cannot imagine what it is like for some of the challenge power trails out there. Some friends have a 100+ trail and it seems a pain quite painful.

As for who we contact, there were two or three options for folks to contact. The first two are developers who might want to fix their/debug code.  That would be the script and tag developer. Many notifications and they might realize that the checker is not working as intended. The other person is the CO. For the reasons listed above it make it convenient.  

Targets comments about Jasmer and other time varying challenges are correct. However processing at the time could just be the arrival of the found log.

---

A a developer i suspect that most incorrect logs is because of cacher that have not checked if thy are qualified. or they have misunderstand the requirements
That is after looking at failing finders in the validate function.
Newbies is a large group where they log the challenges as it is a trad.

I have 6219 tags (48% of all tags) on my script and suspect that i would get alot of mail each day from incorrect logs
The best way to detect incorrect scripts or more likely incorrect tags if for the CO to contact that tages/writer if if that is the case. I hada a lot of empty tags before i changed the code to require a config

I get  a few (<5 i guess) mail each month about incorrect tags. The tager might get more but i am not sure

I hope that you have seen the verify tag function that test the last 10 finds I think you will se it on the normal checker page if you taged it och is the script owner but if you use the tag script button and look for a that checker it is there for all tags.

I dont know i you get is on the normal page as a challenge owner but if you dont i would be an ide to att it there.( Cant check when i have taged all my challenges myself)

---

Just to be clear I am not expecting anyone to automatically get emails. I expect them to be opt in.

I don't know what the percentages of errors are but I can point to a couple of examples.

Certainly misinterpretation by finders is a condition to trap. I expect this is the cos problem.

I have seen tagged scripts that do not match the requirements. In this case it seems to be the taggers role to resolve. My expectation is after a nudge from the co. They might turn on notifications for a short time to fix it.

There is a third case where the documentation of the script does not align with the code. The example here documented something as cache owner but used placedby. Here maybe the scripted should get involved. There is low likely hood of the scripter needing to turn notifications but they might for that one tagging.

As for the last 10, 10 is a good number 10% misinterpretation but what about 0.1% interpretations? My challenge is based on body parts. I need to keep adding new body parts to the list as folks have bigger vocabularies than i.

Frankly I would rather get the notice it needs to be fixed or disabled than let junk stay out there.

---

Yes the publish date/hidden date is a problem. Groundspeek complicates i when they call the hidden date as placed when you edit the cache and nowhere else.
Alot of challenges uses placed in the text. I suspect that most CO that own caches with date requirement dont know the difference. It looks that way from the challenge description. And after tagging writing code and try to finds bugs i know that the number of incorrect logs are quite high. On more complex challenges i have not been surprise to see 4 incorrect finds of the last 10 finds.
I suspect that you will not notice the inconsistent/ambiguities in the requirements of a challenges before you try to think of how to write code that will meat all requirements.
Many of the incorrect log was that date requirement are nog at all looked at regardless of if publish/hidden is used.

Bur if you thin i dissegre with you I really dont. It would be nice as a CO to get mail of autochecks of challengs find. Bur i dont realy see the usecase as a scriptwriter or taget because the ratio bugs to incorrect logs will be so high. I would never have the time to do that
The one to find incorrect code the CO


If your challenge is for body part name i would never tag it for the challenge. the maps on PGC will show if you have made it and there would be alot of false negative.
If i whs my challenge i would create a checker without the gc code and include a link on the cache page to it. That will result in no false negative on the map. But there will also be no ok on the map.
Unforenally the checker is pass or fail an there is not a "not pass" that is not the same as fail