Jump to content

What kind of user ratings should be available?


SgtKlaos

Recommended Posts

it is interesting that you assume the "no" voters do not understand the implications, and that we could only possibly be thinking of caches near our homes.

 

i'm a "no" voter, and i am not thinking of caches near my home.

 

i am thinking of caches eight or nine of hours from my home. when i go on the road to cache, i typically stay out for upwards of twenty days and i have to sort through a LOT of caches over several states.

 

the best and quickest way for me to sort through the caches that are there is to GO AND SEE FOR MYSELF.

 

i do not care which caches most people like; i am about the business of drawing my own conclusions. my experience of a cache rests entirely on MY VISIT, and not someone else's visit. my enjoyment of the cache relies on my personal experience of the hunt rather than the container or even the location.

 

not only will i not come up with a way to "sift through 283 squintillion caches to find the ones that most finders thought were really good", but i don't support this approach to caching and if i had any ideas on how to do it, i would take them with me to my grave.

 

i can't think of a single reason why a rating system would be good in any way.

 

please do not assume that those of us who think the idea sucks simply don't understand.

Link to comment

it is interesting that you assume the "no" voters do not understand the implications, and that we could only possibly be thinking of caches near our homes.

 

i'm a "no" voter, and i am not thinking of caches near my home.

 

i am thinking of caches eight or nine of hours from my home. when i go on the road to cache, i typically stay out for upwards of twenty days and i have to sort through a LOT of caches over several states.

 

the best and quickest way for me to sort through the caches that are there is to GO AND SEE FOR MYSELF.

 

i do not care which caches most people like; i am about the business of drawing my own conclusions. my experience of a cache rests entirely on MY VISIT, and not someone else's visit. my enjoyment of the cache relies on my personal experience of the hunt rather than the container or even the location.

 

not only will i not come up with a way to "sift through 283 squintillion caches to find the ones that most finders thought were really good", but i don't support this approach to caching and if i had any ideas on how to do it, i would take them with me to my grave.

 

i can't think of a single reason why a rating system would be good in any way.

 

please do not assume that those of us who think the idea sucks simply don't understand.

 

So you are saying that you like the way you cache and everyone should cache that way too.

 

If there were a rating system would it stop you from caching as you currently do? I can't see how it would. Your method seems to imply that you probably don't even need to read the logs to weed out caches because, for you, every cache is an experience that you want to draw your own conclusions from. That's fine. It works for you.

 

But currently many of us are frustrated and don't enjoy playing that way. We don't have all the time and money at our disposal to visit every single cache while travelling. Why can't we have a system that would help us and still will allow you to continue to play as you do now?

Link to comment

I vote no. The cache pages and logs give me all of the feedback I need. Of course my method doesn't work for the paperless hordes who never read cache pages or logs. :D

This works if you have time to read all the description and logs in the area, especially if it's your home area. I cache paperless, but I almost always read the description and logs before visiting; if not, I'll read them on-site on my PDA.

 

I periodically visit the left coast on business trips. There are approximately 283 squintillion caches in the LA basin, so reading that many descriptions is not feasible. So I typically look for places to hike and hit caches along the trail. Last trip I hiked in Griffith Park, where the Hollywood sign is; time before that was Puente Hills. But what if I'm driving past half a dozen truly awesome caches (surrounded by 5,000 park-n-grabs) in less prominent parks on my way to my pre-chosen destination? Sometimes I don't have time for a long hike, but have time to grab a few caches. Where do I start?

 

I would love and regularly use some way, any way, to narrow down the list of potential caches to seek. This can be a straight rating system (1-5 stars or whatever); it can be official top-ten-percent lists (each cacher gets to nominate 10% of finds as exceedingly awesome); it can be a cache categorization system (park-n-grab, hike, view, just-for-the-numbers, etc.). Anything.

 

Exactly J-Way.

 

I think the 'no' voters are focussing on their home area. It's easy to hunt everything with 25 miles of your house. But what about trips - I have yet to see a post from a 'no' voter that explains the best and quickest way to sift through 283 squintillion caches to find the ones that most finders thought were really good (exceeding average standards).

I do cache paperless when on a trip and I read every cache page and recent logs before choosing to hunt said cache. It may not be the "quickest" way but it works for me. Since I don't try to hunt hundreds of caches when on a trip I am able and willing to spend a bit of time online having a look at caches in the areas I will be visiting. This means I likely miss seeing caches that others thought were really good but I am ok with that. If I arrive at the area for a cache and don't like it I simply move on.

Link to comment

 

Why can't we have a system that would help us and still will allow you to continue to play as you do now?

 

because i believe that a rating system would quickly devolve into one more sucky thing that cheapens the sport.

 

i don't object to an affinity rating that isn't public.

 

as for assuming that i have some kind of unlimited funding and that's why i prefer to see for myself, that's just rude.

 

and wildly inaccurate.

 

but here's the really terrific thing: there isn't a stupid rating system to cheapen the sport and there isn't likely to be one anytime soon. so yes, you should continue to play that way.

 

..or not.

 

ça m'est égal.

Link to comment

I believe that a rating system would be perceived as a benefit by those who are most calling for one. The issue is that if we have a system why not develop one that benefits the most people and at the same time gives a measurable benefit rather than just a perception of a benefit. What exactly do you want a rating system to achieve? Do you want it to recommend some caches when you are visiting a new area? Do you want it to look at what caches you have said you liked and have it find caches recommended by people who liked the same caches? Do you want it to be a popularity contests to see which caches are liked most among the people who have found those caches? Do want people to try to make caches that will score higher hoping that it will result in fewer of cache you consider lame? Do you want a system that tells you to avoid certain caches because someone else thinks that cache is lame?

 

I am certainly sympathetic to the idea of getting caches recommended when traveling. You generally are going to have limited time for caching and you might want to spend it looking for the best caches. (Of course some people would rather plan their vacation around looking for the most caches). That's why I like Markwell's favorite list suggestion. Premium members can already recommend cache by adding them to a favorites list. Simply allow searches for caches that appear on some number of favorites lists and you'd get a list of recommended caches.

 

Now how would a simple 1 to 5 star rating help? Well those that call most for this are probably concerned that that they have limited time for geocaching. First of all that means they have limited time to prepare for geocaching. Looking at log and cache descriptions is just too time consuming for them. So they either look for everything or they use a less exact filter method like eliminating all 1/1 hides. This is unsatisfactory since they end up still finding too many caches they consider lame or they fear that they have eliminated too many caches that might be really enjoyable. A rating system would provide them with another method to filter. It probably wouldn't matter that the filter is no better than the ones they use now. It would be perceived to be. One certainly would not know if they missed any really exceptional caches. They will have looked at anything with a rating above 4 stars and all exceptional cache would certainly have gotten at least that rating. And while they may still find lame caches, they will undoubtedly believe that they would have found far more but they didn't look for any cache rank below 2 all of which surely were lame. They will perceive the system to have help them improve their caching enjoyment.

 

Now this argument is still hard for some on the 'No' side to understand. After all any time spent caching is enjoyable so how could eliminating some caches make it more enjoyable? While the occasional really clever or original hide adds to the enjoyment there is no expectation that every cache be exceptional. What is most enjoyed about caching isn't the cache itself after all. It is the adventure and experiences that occur when out caching. Someone already posted in this thread about finding a hornets' nest when they lifted a lamppost skirt. Perhaps not the most enjoyable experience but they had a good story to tell when they logged that LPC. The oft stated mantra "If your not having fun stop looking" makes me wonder some times about the people who lobby most for the rating system. If they are not having fun and consider their time too important to do preparation, why do they continue to search when they get to a parking lot and the GPS is pointing to the lamppost? I'm guessing they are enjoying themselves. That might include the enjoyment they get by complaining later in the forums or writing DPM in the log. If there was a rating system, they could very well use it and believe that it is helping. But I'm certain the same people will still be complaining in the forums that they are finding too many "lame" caches that they don't enjoy.

Link to comment

I'll chime in quickly here to say that Groundspeak actually has been tossing around the idea of badges or awards that can be given by users to other users or caches. The nature of those awards are not yet known but will be intelligently discussed and implemented in a positive way that promotes caches/users instead of criticizing "bad" caches/users.

 

Something that has been successful in other areas of the interweb is the suggestion engine (Since you like X, maybe you'll like X). There must be some way to integrate awards/badges with a suggestion engine that would offer very personalized results and accomplish what the OP is seeking.

 

I have to stress that we have no specific plans yet for anything like this. As with any other feature we will first explore in excruciating detail all of the ins and outs a feature like this entails. We're definitely open to feedback from the community, so feel free to continue your healthy debate. :D

Link to comment

I'll chime in quickly here to say that Groundspeak actually has been tossing around the idea of badges or awards that can be given by users to other users or caches. The nature of those awards are not yet known but will be intelligently discussed and implemented in a positive way that promotes caches/users instead of criticizing "bad" caches/users.

Great to hear. :D I'm in favour of an reward system.

 

Something that has been successful in other areas of the interweb is the suggestion engine (Since you like X, maybe you'll like X). There must be some way to integrate awards/badges with a suggestion engine that would offer very personalized results and accomplish what the OP is seeking.

 

I have to stress that we have no specific plans yet for anything like this. As with any other feature we will first explore in excruciating detail all of the ins and outs a feature like this entails. We're definitely open to feedback from the community, so feel free to continue your healthy debate. :D

 

I'm really happy to hear from TPTB that they are reading and responding to the rating system threads. Thank you.

Link to comment

i believe that a rating system would quickly devolve into one more sucky thing that cheapens the sport.

 

i don't object to an affinity rating that isn't public.

Flask, sorry about my comment before. You do question the status quo, quite often apparently.

I now fully agree a 5 star type rating system could cheapen the sport and cause people to abandon their low-rated caches, creating geo-trash (besides, it's already available at GCVote). However, I do believe a user award type system would provide many benefits. And I don't think it would conflict with Markwell's proposed system.

 

What exactly do you want a rating system to achieve? Do you want it to recommend some caches when you are visiting a new area? Do you want it to look at what caches you have said you liked and have it find caches recommended by people who liked the same caches? Do you want it to be a popularity contest to see which caches are liked most among the people who have found those caches? Do want people to try to make caches that will score higher hoping that it will result in fewer of cache you consider lame?

Yes. All of the above.

Something that has been successful in other areas of the interweb is the suggestion engine (Since you like X, maybe you'll like X). There must be some way to integrate awards/badges with a suggestion engine that would offer very personalized results and accomplish what the OP is seeking.

I think some suggestion engines are based on text such as descriptions of products. That might work using cache details, but may not be very useful due to distance limitations. Another type I remember is Amazon's "after viewing this product, people ultimately bought these:". I don't see that being feasible.

I'll chime in quickly here to say that Groundspeak actually has been tossing around the idea of badges or awards that can be given by users to other users or caches. The nature of those awards are not yet known but will be intelligently discussed and implemented in a positive way that promotes caches/users instead of criticizing "bad" caches/users.

 

I have to stress that we have no specific plans yet for anything like this. As with any other feature we will first explore in excruciating detail all of the ins and outs a feature like this entails. We're definitely open to feedback from the community, so feel free to continue your healthy debate. :D

If Groundspeak wants intelligent discussion and feedback from the community, it might be prudent to sticky (pin) this topic in order to keep the intelligent discussion going in one, concerted, progressive, flow. Otherwise, any temporarily low readership will drop the topic off the radar, and uninformed noobs like myself will bring it up again with all the bad ideas. And if Groundspeak is settled on the award/badge system, maybe you could start a new pinned thread and frame the situation like you did in your last post. Can you help the community develop this dialog to an actual conclusion?

Edited by SgtKlaos
Link to comment

As has been brought up almost every time this topic arises, what makes a "wow" cache?

 

If you don't understand you have every right not to participate. Those that have an opinion can can express it and those that don't care can ignore the data. I don't see any problems for anyone. There could even be an opt out button for cache owners that are too thin skinned to accept criticism of their hides.

Link to comment

Subjective qualities are suited to be rated by the community

I found this chart on the internet. I think I will use it next time I go to the store to buy ice cream :D

Favorite ice cream flavors
Rank 		Flavor 		Percent 
1. 	Vanilla 		 29.0% 
2. 	Chocolate 	   8.9 
3. 	Butter Pecan 	5.3 
4. 	Strawberry 	  5.3 
5. 	Neapolitan 	  4.2 
6. 	Chocolate chip   3.9 
7. 	French vanilla   3.8 
8. 	Cookies 'n' cream 3.6 
9. 	Fudge ripple 	2.6 
10.	Praline 		 1.7

 

interesting... 'cause I rank Vanilla right up there with LPC...

Link to comment

Firefox users with GreaseMonkey installed already have the solution to this issue if they choose to employ it.

 

GCVote

 

While it may not substitute for reading some of the past logs, it can certainly steer you away from caches that nobody likes, and toward caches that everybody likes. Certainly a valuable tool when visiting an unfamiliar area, as when vacationing.

 

It could even be a hoot to visit an area and only go for caches rating 2 stars or less! The sordid underbelly tour!

Link to comment

Firefox users with GreaseMonkey installed already have the solution to this issue if they choose to employ it.

 

GCVote

 

While it may not substitute for reading some of the past logs, it can certainly steer you away from caches that nobody likes, and toward caches that everybody likes. Certainly a valuable tool when visiting an unfamiliar area, as when vacationing.

 

It could even be a hoot to visit an area and only go for caches rating 2 stars or less! The sordid underbelly tour!

 

Although a good tool there are issues with GCVote.

 

1. It's not widely used.

2. Only works with Firefox.

3. You have to send the owner a password to get it to work - i.e. you can't set your own password

4. If you can just barely understand how your GPS works and how to download waypoints, you'll probably get scared away from downloading the script

5. You have to download it to every computer you use in order to vote from anywhere (home, office, laptop, public library)

6. It doesn't work with my laptop (works fine on my desktop computer). But now it doesn't work for me because recently the GCVote creator requires that you email him through his GC profile with a password. The privacy issue bothers me a bit.

Edited by Lone R
Link to comment

1. It's not widely used.

2. Only works with Firefox.

3. You have to send the owner a password to get it to work - i.e. you can't set your own password

4. If you can just barely understand how your GPS works and how to download waypoints, you'll probably get scared away from downloading the script

5. You have to download it to every computer you use in order to vote from anywhere (home, office, laptop, public library)

6. It doesn't work with my laptop (works fine on my desktop computer). But now it doesn't work for me because recently the GCVote creator requires that you email him through his GC profile with a password. The privacy issue bothers me a bit.

I am the author of GCVote and would like to comment on your points.

1. There are 2000 registered users and 500,000 ratings for 150,000 caches. But most of the rated caches seem to be in Germany.

2. GCVote will support Internet Explorer soon and you can used it with GSAK and other tools.

3. What is the difference between sending me a password and setting it yourself? As the service's administrator, I can easily access all your data for GCVote. But since a lot of users seem to have a problem with sending this password by mail, I will add a feature to change your password soon.

4. The installation requieres several steps, but they should be documented well enough to execute them without deeper knowledge.

5. True.

6. I would be glad to help you with your installation. And I cannot understand your privacy issues. You are not supposed to send a relevant password (like the one for geocaching.com) to me. Every administrator of every service can read your password for that service. (If the password is encrypted on the client, you can regard the encrypted password as the real password.)

Link to comment

Although a good tool there are issues with GCVote.

 

1. It's not widely used.

2. Only works with Firefox.

3. You have to send the owner a password to get it to work - i.e. you can't set your own password

4. If you can just barely understand how your GPS works and how to download waypoints, you'll probably get scared away from downloading the script

5. You have to download it to every computer you use in order to vote from anywhere (home, office, laptop, public library)

6. It doesn't work with my laptop (works fine on my desktop computer). But now it doesn't work for me because recently the GCVote creator requires that you email him through his GC profile with a password. The privacy issue bothers me a bit.

1. There are new users every single day!

2. Yep, FF for now and heard that it works under Opera and Guido is looking for IE7+. Also works for GSAK users!

3. Yes, you send a single message with the password you want then you are set.

4. I gave the website a complete overhaul with better instructions and many screenshots.

5. Not so many people use so many different computers. Also, you can always add ratings later once at home, there is a single page to visit to do that. It takes what, 20 seconds to rates 10 caches?

6. It is explained that it has to be done that way to ensure it's YOU and only you who can give rating with your name. Again, once set you never touch that again.

 

Hope that helps.

 

There are some stats of it growing rapidly in the last months :

 

   May 9th :  91000 ratings on  51000 caches worlwide.

  June 1st : 300000 ratings on 110000 caches worlwide

  July 1st : 533000 ratings on 150000 caches worlwide.

August 1st : 678000 ratings on 176000 caches worlwide.

Edited by Rhialto
Link to comment

If Groundspeak wants intelligent discussion and feedback from the community, it might be prudent to sticky (pin) this topic in order to keep the intelligent discussion going in one, concerted, progressive, flow. Otherwise, any temporarily low readership will drop the topic off the radar, and uninformed noobs like myself will bring it up again with all the bad ideas. And if Groundspeak is settled on the award/badge system, maybe you could start a new pinned thread and frame the situation like you did in your last post. Can you help the community develop this dialog to an actual conclusion?

 

I agree with SgtKlaos - probably a good idea to pin the topic and encourage a constructive discussion about how to create a decent rating system. And hopefully, discourage posters whose only contribution is... "a rating system, no matter what you come up with, is a bad idea".

 

Regarding implementing a rating system, here's my input:

 

  1. Make it anonymous in order to get more honest assessments and not to hurt the owners feelings.
     
     
  2. Make it a cream of the crop system - e.g. a 10 star system. Those that get rated 8 or more stars get a special mark indicating that most raters liked this cache. This ties into the anonymous criteria - owners know only when their box is considered exceptional, not when it's average or below average.
  3. Something that you might want to consider, but may not be practical -
    Weight the quality of rater votes - I'm a member of a website where my vote is weighted. I can't mark everything as 5/5 - that shows I put little thought into my vote. A variety of factors is taken into account to determine rankings such - voting history, experience level, and even the standard deviation of how people vote on a given letterbox.


    Here's what my ratings stat page says:

     

    You have rated 142 [items] with an average rating of 3.02 and a standard deviation of 0.92.

     

    You've distributed your votes as follows:

    1. 5% (7 votes)

    2. 21% (30 votes)

    3. 46% (66 votes)

    4. 22% (31 votes)

    5. 6% (8 votes)

     

    The database tries to determine the quality of your votes when it comes time to calculate which items should get diamond ratings. The higher quality your votes are, the higher their weighting will be in the calculations.

     

    You have rated plenty of items within the expected distribution.

     

    Only those that have visited can vote. Those who rate caches must rate all of their finds for the system to work fairly and accurately. This should weed out those who just want to rate their friends' caches only and give them high marks -- there's more of a commitment to the website and appropriate ratings when you need to rate all your finds and distribute the votes to a standard deviation. More than one visitor vote is required for calculations. Once every few weeks the database runs these calculations and assigns diamonds to those items that rank the highest. I believe that only the top 5% of items in the database get a diamond rating. In my area, those that received the diamond rating deserved it. The items are high quality.


  4. Make the ranking system optional. If you are an owner that, on principal, doesn't like the idea of people ranking your caches, you can opt out.

Link to comment

3. What is the difference between sending me a password and setting it yourself? As the service's administrator, I can easily access all your data for GCVote. But since a lot of users seem to have a problem with sending this password by mail, I will add a feature to change your password soon.

6. I would be glad to help you with your installation. And I cannot understand your privacy issues. You are not supposed to send a relevant password (like the one for geocaching.com) to me. Every administrator of every service can read your password for that service. (If the password is encrypted on the client, you can regard the encrypted password as the real password.)

You misunderstand proper secure password storage. Passwords should be encrypted via a one-way hash so they cannot be recovered by anyone. When authenticating the user, the entered password is hashed and the hash compared against the stored value. If they match, then the user has entered their password correctly.

 

This is how UNIX systems have been doing it for decades. It's pretty well-established and accepted.

 

As administrator, you should not have access to anyone's password in plaintext.

Edited by dakboy
Link to comment

1. It's not widely used.

2. Only works with Firefox.

3. You have to send the owner a password to get it to work - i.e. you can't set your own password

4. If you can just barely understand how your GPS works and how to download waypoints, you'll probably get scared away from downloading the script

5. You have to download it to every computer you use in order to vote from anywhere (home, office, laptop, public library)

6. It doesn't work with my laptop (works fine on my desktop computer). But now it doesn't work for me because recently the GCVote creator requires that you email him through his GC profile with a password. The privacy issue bothers me a bit.

I am the author of GCVote and would like to comment on your points.

1. There are 2000 registered users and 500,000 ratings for 150,000 caches. But most of the rated caches seem to be in Germany.

2. GCVote will support Internet Explorer soon and you can used it with GSAK and other tools.

3. What is the difference between sending me a password and setting it yourself? As the service's administrator, I can easily access all your data for GCVote. But since a lot of users seem to have a problem with sending this password by mail, I will add a feature to change your password soon.

4. The installation requieres several steps, but they should be documented well enough to execute them without deeper knowledge.

5. True.

6. I would be glad to help you with your installation. And I cannot understand your privacy issues. You are not supposed to send a relevant password (like the one for geocaching.com) to me. Every administrator of every service can read your password for that service. (If the password is encrypted on the client, you can regard the encrypted password as the real password.)

 

Thanks for the quick reply.

 

1. How many of people in Ontario, Canada use GCVote. I think I was the only one. But the more we get the word out the more people will use it. Still I'd like to see something implemented on GC so that more people in my area will use it.

 

2. Nice to hear, that will likely up the number of users

 

3. If I send you a password I'd want it to be different from the password(s) I normally use since it's not private if I send it to you (although, as you say as admin you can probably see my password anyway, right? - which I didn't realize). And if I create a newish password I might forget it and then what do I do? I suppose I'd have to contact you again to get you to send me the password.

 

4. I was hesitant at first, not sure that I'd download it correctly, but it worked fine - wasn't difficult. I have some experience with downloads. I can see how those who have little experience may be hesitant, for fear of viruses and such. If GC implements something that doesn't require a download I think more people will use it.

 

Anyway, it is a decent system and the best thing out there at the moment. Kudos for providing a rating system.

Link to comment

You misunderstand proper secure password storage. Passwords should be encrypted via a one-way hash so they cannot be recovered by anyone. When authenticating the user, the entered password is hashed and the hash compared against the stored value. If they match, then the user has entered their password correctly.

 

This is how UNIX systems have been doing it for decades. It's pretty well-established and accepted.

 

As administrator, you should not have access to anyone's password in plaintext.

 

As an Unix administrator with full access, I can easily alter the system to log your password when you enter it. The hashed storage only prevents extraction of the passwords in the case that the disc is stolen.

 

In case of a web based service like GCVote, it is even easier: I only have to log the communication between your computer and my server.

 

As I tried to explain, GCVote could encrypt the password on your computer, and only send a hash to my server. But in that case the encrypted password is the *effective* password. There is no way to make this really secure.

 

I could also do what every Unix admin does: Copy the encrypted password, set a new password, do my stuff and copy the first encrypted password back into the password storage.

Link to comment

3. If I send you a password I'd want it to be different from the password(s) I normally use since it's not private if I send it to you (although, as you say as admin you can probably see my password anyway, right? - which I didn't realize). And if I create a newish password I might forget it and then what do I do? I suppose I'd have to contact you again to get you to send me the password.

 

4. I was hesitant at first, not sure that I'd download it correctly, but it worked fine - wasn't difficult. I have some experience with downloads. I can see how those who have little experience may be hesitant, for fear of viruses and such. If GC implements something that doesn't require a download I think more people will use it.

 

3. Please do not use the same password for different services. Administrators really can read them! If you forget your password, you can contact me or you can take a peek into about:config, where the password is stored on your computer. (Again in plain text, as there can be no added security in encrypting it. I am very willing to discuss that point.)

 

4. One could set up a kind of proxy website that would read the information from geocaching.com and alters it. But Groundspeak does not allow that. Plus, the proxy server would know your password for geocaching.com...

 

I still hope that Groundspeak implements its own rating system, as that is the only perfect solution.

Link to comment

You misunderstand proper secure password storage. Passwords should be encrypted via a one-way hash so they cannot be recovered by anyone. When authenticating the user, the entered password is hashed and the hash compared against the stored value. If they match, then the user has entered their password correctly.

 

This is how UNIX systems have been doing it for decades. It's pretty well-established and accepted.

 

As administrator, you should not have access to anyone's password in plaintext.

 

As an Unix administrator with full access, I can easily alter the system to log your password when you enter it. The hashed storage only prevents extraction of the passwords in the case that the disc is stolen.

Only an unethical adminsitrator, with a lax/non-existent security policy and no oversight would do such a thing.

 

In case of a web based service like GCVote, it is even easier: I only have to log the communication between your computer and my server.
Not if you're using HTTPS authentication. If that's the case, then you could only "sniff" the password within the code of your application. See above re:ethical vs. unethical administrators doing such a thing. There are authentication methods that can be used external to your application which would completely isolate your application from the password - your app would only know that the person has authenticated successfully.

 

As I tried to explain, GCVote could encrypt the password on your computer, and only send a hash to my server. But in that case the encrypted password is the *effective* password. There is no way to make this really secure.
There are plenty of ways to make it secure, you've simply chosen not to. The "encryption" you describe here is merely creating a password of "random" characters.

 

I could also do what every Unix admin does: Copy the encrypted password, set a new password, do my stuff and copy the first encrypted password back into the password storage.
Or create a proper authentication/password management process which:

A ) Uses SSL for all communication.

B ) Stores the password hashed in the database or in an otherwise protected manner.

C ) Removes you personally (or any other human) from the password-setting/retrieval process.

 

Even better, use something like OpenID so you don't have to get involved with it at all - just use OpenID's services.

 

I had considered installing GCVote to see what it's all about, but have decided against as a result of this thread. Even if it's a "throwaway" password which I don't share with any other online account, I don't like how this is being done.

Link to comment
3. Please do not use the same password for different services. Administrators really can read them!
Please stop spouting this. My bank website password is not readable by anyone (if it were, I wouldn't have to go through a lengthy process to reset my password, they'd just email it to me after I provide my user ID (not my email address)). My LDAP password and myriad other application passwords in my company are not recoverable by my system administrators or anyone else, because those systems were designed securely.

 

What you are saying is "I'm not going to install a lock on my door, because there isn't one on the house next door. But there is one 3 doors down."

Edited by dakboy
Link to comment
[*]Make it anonymous in order to get more honest assessments and not to hurt the owners feelings.

It is anonymous... unless you use the button that paste text at the end of a log and that text disclose your rating. The text can be modified to remain anonymous, it is explained in the instruction.

 

[*]Make it a cream of the crop system - e.g. a 10 star system.

Personally after a few months using it like it is now, I think it's perfect. We rate 3 most of the cache we do (it's a cache and nothing else) and you either go with a 2 if it was 'déjà vu' (there are 10 like this around my house!) or is some rare case a 1 when it was a pain, no fun, nothing to see, boring, whatever can make you really dislike a cache and on the other hant, you can put a 4 for a fun cache or a 5 for a MUST DO, MUST SEE.

 

Half star (10 steps system) is already in the code but not enabled yet (Guido could tell you more) and again, personnaly, I'm not really looking to it but that's my opinion.

Link to comment

dakboy, only unethical administrators can read passwords? Yes, that is true. Only unethical administrators would *want* to read passwords. But you cannot be sure of the ethical qualities of any service's administrators.

 

I could implement additional security layers to prevent me from reading passwords. But even after that, I could still disable those layers and sniff passwords anyway.

 

Please explain how to make this system secure without delegating the security to an external service, as that service could break the security itself.

 

HTTPS/SSL does not help as long as I have access to the server's processes.

Hashed passwords in the DB do not prevent me from sniffing your input.

 

Even if there was a way to implement perfect security (and there is none), it would be overkill for a system like GCVote. Did you check if Groundspeak implements any of your ideas?

 

Let me "spout" this again: Your bank, as an entity, can sniff your password, only the individual employees can not. If all employees worked together, they could sniff your password. And there is no way to prevent that.

 

Password *recovery* is a totally different matter. But as long as you are using your password, it can be sniffed.

 

You seem to know a lot of good practices, but you should realize that they do not provide ultimate security.

Link to comment



  1. Something that you might want to consider, but may not be practical -
    Weight the quality of rater votes - I'm a member of a website where my vote is weighted. I can't mark everything as 5/5 - that shows I put little thought into my vote. A variety of factors is taken into account to determine rankings such - voting history, experience level, and even the standard deviation of how people vote on a given letterbox.
    Only those that have visited can vote. Those who rate caches must rate all of their finds for the system to work fairly and accurately. This should weed out those who just want to rate their friends' caches only and give them high marks -- there's more of a commitment to the website and appropriate ratings when you need to rate all your finds and distribute the votes to a standard deviation. More than one visitor vote is required for calculations. Once every few weeks the database runs these calculations and assigns diamonds to those items that rank the highest. I believe that only the top 5% of items in the database get a diamond rating. In my area, those that received the diamond rating deserved it. The items are high quality.[/indent]
     
  2. Make the ranking system optional. If you are an owner that, on principal, doesn't like the idea of people ranking your caches, you can opt out.

1, 2: I agree

3: Terracaching.com uses a similar system, and it is an unending source of angst. Instead of assigning a "quality rating" like you described, the TC system calculates the average rating you've assigned, adjusts that average to 5 (it goes from 0 to 10), and adjusts all the individual ratings accordingly. I agree in principle, because it attempts to negate the affect of someone giving every cache a 10 or every cache a 0. Unfortunately, most people don't understand statistics/bell curves. They want a rating of 8 to mean a rating of 8, not be adjusted to a 6.5 because they tend to rate caches higher than average.

 

Another problem with this method is that it doesn't take into account the quality of the caches themselves, just the quality of your ratings. Lets's say the rating system goes into affect, and I decide I only want to hunt for caches with a rating of 3.5 or higher (using your 1-5 scale). So now the average rating of all the caches I hunt is around 4. If I assign ratings that exactly match the typical rating for each cache my rating average is also 4. Using a bell-curve system, my quality rating would be low because it appears that I over-rate caches. Solution: don't compare individual ratings with a generic statistical curve, rather compare individual ratings with a statistical curve generated by the average ratings for all the caches found by that person.

 

4: This would defeat the system if a crappy cache hider could just opt out of allowing his caches to be rated. Unless ALL crappy cache hiders refused to participate, then you could just ignore caches with no ratings.

Link to comment



  1. Something that you might want to consider, but may not be practical -
    Weight the quality of rater votes - I'm a member of a website where my vote is weighted. I can't mark everything as 5/5 - that shows I put little thought into my vote. A variety of factors is taken into account to determine rankings such - voting history, experience level, and even the standard deviation of how people vote on a given letterbox.
    Only those that have visited can vote. Those who rate caches must rate all of their finds for the system to work fairly and accurately. This should weed out those who just want to rate their friends' caches only and give them high marks -- there's more of a commitment to the website and appropriate ratings when you need to rate all your finds and distribute the votes to a standard deviation. More than one visitor vote is required for calculations. Once every few weeks the database runs these calculations and assigns diamonds to those items that rank the highest. I believe that only the top 5% of items in the database get a diamond rating. In my area, those that received the diamond rating deserved it. The items are high quality.[/indent]
     
  2. Make the ranking system optional. If you are an owner that, on principal, doesn't like the idea of people ranking your caches, you can opt out.

1, 2: I agree

3: Terracaching.com uses a similar system, and it is an unending source of angst. Instead of assigning a "quality rating" like you described, the TC system calculates the average rating you've assigned, adjusts that average to 5 (it goes from 0 to 10), and adjusts all the individual ratings accordingly. I agree in principle, because it attempts to negate the affect of someone giving every cache a 10 or every cache a 0. Unfortunately, most people don't understand statistics/bell curves. They want a rating of 8 to mean a rating of 8, not be adjusted to a 6.5 because they tend to rate caches higher than average.

 

Another problem with this method is that it doesn't take into account the quality of the caches themselves, just the quality of your ratings. Lets's say the rating system goes into affect, and I decide I only want to hunt for caches with a rating of 3.5 or higher (using your 1-5 scale). So now the average rating of all the caches I hunt is around 4. If I assign ratings that exactly match the typical rating for each cache my rating average is also 4. Using a bell-curve system, my quality rating would be low because it appears that I over-rate caches. Solution: don't compare individual ratings with a generic statistical curve, rather compare individual ratings with a statistical curve generated by the average ratings for all the caches found by that person.

 

4: This would defeat the system if a crappy cache hider could just opt out of allowing his caches to be rated. Unless ALL crappy cache hiders refused to participate, then you could just ignore caches with no ratings.

 

This is great - fleshing out some ideas and getting some thinking going about creating an optimal system.

Re: #4 I would probably ignore caches that have opted out of the rating system (especially while on vacation, maybe not so much within 25 kilometers of my home base when I have more time to thoroughly read through the logs).

Link to comment
A request is made almost weekly for some kind of rating system. It is often made by new cachers who see that there are an awful lot of caches to find - more they they can ever find - and they are looking for ways to tell which are the really good caches that they should be looking for (and sometimes what are the real stinkers the ought to skip). The problem is they might not realize at first that not everyone enjoys the same kinds of caches. Some people prefer hikes in the wilderness while others want to find caches they can drive right up to. Some people enjoy finding tiny well hidden micro caches, while others like regular sized containers with lots of items they can trade for. Some want easy finds, other want difficult challenges. Simple rating systems fall apart because there is no average cachers. Different people like different caches.

Color me surprised that this same fallacious argument keeps showing up every time somebody wants a rating system. The argument is completely bogus.

 

Why do we have user ratings for restaurants? Movies? Books? After all, not everybody likes the same kind of food, or movies, or books. Clearly, according to your argument, there is no need for any kind of review of anything.

 

The fact of the matter is that the vast majority of people prefer good food, good movies, and good books. User ratings have proven very valuable all these things. Look at the nice review system Amazon has for books. If there were no value in it, do you think they would spend all that money to develop and maintain the system?

 

Likewise, a good review system for geocaches could be very useful.

 

I am really, really, really tired of the argument that if everybody doesn't like exactly the same things then user reviews are useless. I'm a little discouraged that such a weak argument is supported by so many here. If you want to argue against ratings, please come up with something that actually, you know, makes sense.

Edited by fizzymagic
Link to comment

(............)

The problem is they might not realize at first that not everyone enjoys the same kinds of caches. Some people prefer hikes in the wilderness while others want to find caches they can drive right up to. Some people enjoy finding tiny well hidden micro caches, while others like regular sized containers with lots of items they can trade for. Some want easy finds, other want difficult challenges. Simple rating systems fall apart because there is no average cachers. Different people like different caches.

 

Why do we have user ratings for restaurants? Movies? Books? After all, not everybody likes the same kind of food, or movies, or books. Clearly, according to your argument, there is no need for any kind of review of anything.

 

The fact of the matter is that the vast majority of people prefer good food, good movies, and good books. User ratings have proven very valuable all these things. Look at the nice review system Amazon has for books. If there were no value in it, do you think they would spend all that money to develop and maintain the system?

 

Likewise, a good review system for geocaches could be very useful.

 

I agree with your last sentence. In my opinion, however, tozainamboku is also right in some aspects.

The key issue is how the review system works and is set up. If it is very simple, I doubt that it is helpful for cachers like myself who do not share the mainstream preferences.

 

A system like GCVote (using the scale from 1*-5*, 1* being the lowest) does a very bad job for me while I regard the system at Amazon as helpful for me.

 

I provide you with some examples:

 

Consider a 1*/1* micro cache A which is hidden at the parking lot of a supermarket and easily reachable and quickly loggable even in presence of muggles. Typically such caches receive a 2* rating on average at

GCVote.

 

Now consider a 3*/5* micro cache B also hidden at the area of a large shopping mall, but this time placed highly up at a column which carries the shops that are present at the centre. (Of course it is not allowed to climb up there and such caches only work as long no one searching for it is detected by the police or by private security officers). Typically such caches receive a 4-5* rating on average at GCVote.

 

I neither like A nor B (for me both at the lowest level of my personal enjoyment scale), but I prefer A by far to B.

 

In general I made the experience that hiking caches in a nice, but not spectacular landscape where nothing spectacular has to be achieved and which are hidden at a standard hideout, typically end up with ratings at GCVote around 3* which is a lower rating than boring drive-in caches get that have some special sort of container which end up with 4* or higher.

 

If I am searching for example for nice hikes in a scenic landscape, it does not help me to look at ratings where aspects like the originality of the hideout, the container and the stages (in case of a multi cache) are rated and the aspect I am interested in does not even play a role.

 

In the case of a typical Amazon review the text of the review and the rating are linked to each other. This helps me to understand for which reasons someone rated a book or another item in the way he did.

 

Moreover, an essential between a system like the one at Amazon and a cache rating system is that

the contact between cache hiders and the cache raters is much more close than, for example, between the author of a novel and customers of Amazon. Even if the author checks the reviews, he/she will be usually aware of the fact that there is no book that pleases everyone. If someone writes that he/she did not like a cache and assigns a low rating to it, this often ends up in a situation where the cache owner is annoyed.

One of the main reasons for this in my opinion is that systems like GCVote do not distinguish between

rating the enjoyment a cache has provided to someone searching for it and the quality of the cache.

I often encounter arguments like "This is a well done-cache at a nice place with hours of work invested constructing the container. This cannot be a 1* cache." Of course it can be a 1* cache for a particular cacher when he did not enjoy the cache at all. The cache can be a good cache nevertheless. Many cache hiders regard ratings of their caches however like a school report. As long rating systems are set up in such a way and like systems such as the one used by Amazon which are mainly designed to help customers to make their decisions, I feel that rating systems do have more negative effects than positive ones. Nevertheless, it is clear that for cachers with mainstream preferences, the average GCVote ratings might be of help. This does not imply, howevere, that systems like GCVote are good cache rating systems.

 

 

Cezanne

Link to comment

I'm happy to know that Groundspeak is thinking on a ratings system. I'm not enthused about a 1-5 star system, would prefer either the Favorites bookmarked list, coded such that I could PQ for caches that appear on 2 or more lists, or some type affinity system.

 

I'll mention (again) that there is a 1- 5 star rating system for Wherigo carts. Here's the link to the list of California Wherigos

Note that the majority of carts are unrated, and the the lowest rating is 3.5 stars. Most rated carts are 4, 4.5 or 5. Is this helpful?

 

To be fair, you must complete a cart to rate it (this is different from signing the cache log). Many cachers won't bother to log a cart completion, which takes either uploading a saved completed cart, or saving a completion code, where the cart owner wrote such a code into the cart.

Link to comment
Subjective qualities are suited to be rated by the community
I found this chart on the internet. I think I will use it next time I go to the store to buy ice cream :)

Favorite ice cream flavors
Rank 		Flavor 		Percent 
1. 	Vanilla 		 29.0% 
2. 	Chocolate 	   8.9 
3. 	Butter Pecan 	5.3 
4. 	Strawberry 	  5.3 
5. 	Neapolitan 	  4.2 
6. 	Chocolate chip   3.9 
7. 	French vanilla   3.8 
8. 	Cookies 'n' cream 3.6 
9. 	Fudge ripple 	2.6 
10.	Praline 		 1.7

interesting... 'cause I rank Vanilla right up there with LPC...

I don't. Vanilla, while can be eaten by itself, is more of either an ingredient or a condiment. That's why it's more "popular" or was it the most sold. Big difference. Anyway, I like vanilla because I like it on pie or so I can add my own extras. If I'm eating ice cream by itself, I like a variety, especially things with chips and nuts.

 

So, in a way, vanilla is more of a 1.5/1.5 regular traditional that's just a tad more than a P&G. Kind of plain. Praline and Cream is a mystery with a good theme and back story. Yummy.

Link to comment

We all agree that we all have our opinion on a great and less great cache.

 

What could be confusing for some is that if you are the only one rating a cache yet and you gave it a 2 this may not be your opinion as you would have given it a 4. Thist is NORMAL it happen just like it is normal there will never be 50 identical ratings for a cache that counts 50 ratings. We can guess it would range from 2 to 4 for that cache with maybe 15*2, 20*3 and 15*4.

 

What you will see on that cache is 3 stars, unless you click the stars to see detailed information. This mean that it's probably an everyday cache that some likes a tidbit better than average and some that less like it than average so basically it doesn't really matter what the rating is, you just go and search it (if the cache description is fine with you).

 

On the other hand, a cache with detailed info who shows 15*1, 25*2 and 10*3 will be displayed as a 2 stars so it gives you an idea that generally people less liked it for whatever reason (read logs to find out?).

 

What's great and to avoid confustion, you can instruct GCVote to not show ratings for a cache UNLESS there is a minimum of X ratings for a cache. When that minimum is reached, we agree that you have a pretty good idea of the cache's rating.

 

Simple and very effective to me once you stop thinking the system is not perfect.

Link to comment

I'll chime in quickly here to say that Groundspeak actually has been tossing around the idea of badges or awards that can be given by users to other users or caches. The nature of those awards are not yet known but will be intelligently discussed and implemented in a positive way that promotes caches/users instead of criticizing "bad" caches/users.

 

Something that has been successful in other areas of the interweb is the suggestion engine (Since you like X, maybe you'll like X). There must be some way to integrate awards/badges with a suggestion engine that would offer very personalized results and accomplish what the OP is seeking.

 

I have to stress that we have no specific plans yet for anything like this. As with any other feature we will first explore in excruciating detail all of the ins and outs a feature like this entails. We're definitely open to feedback from the community, so feel free to continue your healthy debate. :)

 

I have an "angle" to consider. I've noticed that the caches I prefer to find almost always have large photo galleries, multiple "top finds" bookmarks, and logs that feature multiple paragraph "found it" logs. The current system available to us, has no mechanism to search for caches that have large galleries, long logs, and multiple "top finds" bookmarks. The ability to search for caches like this would be a positive addition.

 

Examples:

 

Any Gun Can Play

 

Big Fig

 

return to scab island

Edited by Kit Fox
Link to comment

What could be confusing for some is that if you are the only one rating a cache yet and you gave it a 2 this may not be your opinion as you would have given it a 4.

 

I definitely do not belong to the group of people who are confused by this aspect.

The weaknesses of GCVote from my point of view have nothing to do with the number of cachers who have rated a cache and are not related to the usage of a certain statistical parameter. I know that I can look at the distribution of all ratings and I would be able to compute every type of parameter I would like to.

It would not change my opinion on the system. The main weakness is that different people attribute completely different meanings to the steps of the 1-5* scale and that there is no system-immanent way to find out why a certain rating has been given to a cache.

 

A cache rated 2* on average because the majority of cachers feel that the hideout is standard and they are not interested into the beauty of the walk to the cache location, is a valuable candidate for me for being visited. A cache rated 4* on average because it has an ingeniuous hideout, but is a boring drive-in from my point of view, should not appear on my candidate list.

 

The only cache rating system that would be of any value to me would be one where I either could look at recommendations of cachers with similar caching preferences or whether I could select some criteria like nice hike and then search for caches who get high ratings due to the hike which is involved.

 

I am well aware of the fact that there does not exist a perfect system. As a consequence there will always be caches that end up on recommendation lists though they do not fit to one's own selection criteria and there also always will be caches that are missed. A good system should manage, however, to come up with a reasonably high percentage of caches correctly classified (according to the chosen criteria). GCVote does not serve this purpose at all for me and the same will happen for any 1*-5* system with no clear criteria. As I said, for mainstream cachers it might work quite well.

 

What's great and to avoid confustion, you can instruct GCVote to not show ratings for a cache UNLESS there is a minimum of X ratings for a cache.

 

I do not need such approaches - my brain can do the job much better (I can read and understand the logs and try to estimate how much value I attribute to the opinions of the finders of the cache).

 

When that minimum is reached, we agree that you have a pretty good idea of the cache's rating.

 

No, I do not agree on that at all. Typical adventure caches where logging them involves actions that are idiotic from my point of view and forbidden from the legal point of view, typically get very high ratings as most people you visit such caches loves them. (This is just one of many counterexamples I could provide to your statement.)

 

 

Cezanne

Edited by cezanne
Link to comment
I've noticed that the caches I prefer to find almost always have large photo galleries, multiple "top finds" bookmarks, and logs that feature multiple paragraph "found it" logs. The current system available to us, has no mechanism to search for caches that have large galleries, long logs, and multiple "top finds" bookmarks. The ability to search for caches like this would be a positive addition.

 

Yes, I believe CR has written a "length of log" macro for GSAK that he uses.

 

I just provided my "top ten newer Florida caches" list to the FGA (I don't know when that bookmarked list will be available). I avoided providing another list of "big" multicaches, which have dominated those FGA "best of" lists in the past and are, truthfully, my own favorites. While those are fine caches, they're out of reach for many, particularly short term or summer visitors. Those kinds of caches which involve considerable time and effort on the part of the seeker, tend to generate long logs with pics.

 

So I tried to work with terrain 2.5 or less caches. Some of those don't have especially long logs, or a bunch of photos, but they were all scenic or interesting in some way. One I included largely because of the lovely custom made logbook. Otherwise it's just one of many nice containers in that patch of woods.

Link to comment

We all agree that we all have our opinion on a great and less great cache.

 

What could be confusing for some is that if you are the only one rating a cache yet and you gave it a 2 this may not be your opinion as you would have given it a 4. This is NORMAL it happen just like it is normal there will never be 50 identical ratings for a cache that counts 50 ratings. We can guess it would range from 2 to 4 for that cache with maybe 15*2, 20*3 and 15*4.

 

What you will see on that cache is 3 stars, unless you click the stars to see detailed information. This mean that it's probably an everyday cache that some likes a tidbit better than average and some that less like it than average so basically it doesn't really matter what the rating is, you just go and search it (if the cache description is fine with you).

The main weakness is that different people attribute completely different meanings to the steps of the 1-5* scale and that there is no system-immanent way to find out why a certain rating has been given to a cache.

Right, this is the error I referred to in my other post that is inherent in star rating systems. Like others have said, people enjoy very different aspects of caching. While the general quality of a cache can be generally rated on a general 5 star scale and therefore be moderately useful, a rating system can be much more useful if it focuses on those different aspects of caching.

 

Enter the award/badge system. All you have to do is decide what awards you would give a cache (e.g. an award for scenery, creative site setup, overall cache design (which includes the details page), the hike/trek/trip/drive to the site, and/or the container camouflage). The cache is awarded bronze, silver, or gold medals in each of the categories if the recommendations vs. no recommendations are of a high enough percentage (e.g. bronze for 70%+, silver for 80%+, gold for 90%+). That way people can find what they are looking for.

Edited by SgtKlaos
Link to comment
I am really, really, really tired of the argument that if everybody doesn't like exactly the same things then user reviews are useless. I'm a little discouraged that such a weak argument is supported by so many here. If you want to argue against ratings, please come up with something that actually, you know, makes sense.

Yep!

 

To simply state a rating system would be bad because of diversity simply doesn't make sense. It's not as if folks would be looking only at finders' ratings of the cache. Folks will still be looking at size, type, and location.

 

My biggest issue with some of the proposed schemes are the requirement of having a find it log that will "stick." The reason I say it that way is that until the rod wielded by the cache owner used to force compliance by the finder is removed, ratings will be pretty much useless. Folks will fear logs deletions and therefor artificially bump their ratings.

 

Also, why should have to visit a crappy area and do everything needed to write a legitimate find log in order to have an opinion? I can show up at the location of a traditional and know I don't want to continue, yet I'm not allowed to have an opinion because I believe I must find the cache and sign the log in order to log a find? I don't think so.

 

I think with any rating system it should emphasized that the rating is the opinion and preference of rater only and not an indictment of the cache. I rates things how I like them and not whether I think they should exist. A rating system should not be viewed as to whether the cache should exist, but whether it should exist on my radar, and nothing more. I think a good system would allow the consumer of the rating scheme to rank the caches and dial whatever percentage of the top rated cache he wants. It could be the top 5% or 100% (all caches)--whatever is useful for him or her. Some folks may only want to hunt the cream of the crop and some may only want to toss out the stinkers.

 

I don't think "wow" would be the right criteria. There's actually so few real "WOW!" caches that it's pathetic. I think if one thought it more along the lines of "if a friend could only hunt half the caches in this area would you recommend this cache be on that list?" A 1 star would be "skip it." 3 stars would be "eh, either way." 5 stars "this is one you simply must do." This is really what folks are trying to do with a rating system, find out if others are recommending we find that cache.

Link to comment
Why do we have user ratings for restaurants? Movies? Books?
Because people like numbers? I find such simplistic ratings pretty useless myself.

 

Likewise, a good review system for geocaches could be very useful.
So are we talking about reviews, or are we talking about a 5-star rating system? I find reviews useful. I find "you may also like" recommendations useful. I find personal recommendations useful. It's simplistic 5-star (or n-star) rating systems that I find useless.
Link to comment
A request is made almost weekly for some kind of rating system. It is often made by new cachers who see that there are an awful lot of caches to find - more they they can ever find - and they are looking for ways to tell which are the really good caches that they should be looking for (and sometimes what are the real stinkers the ought to skip). The problem is they might not realize at first that not everyone enjoys the same kinds of caches. Some people prefer hikes in the wilderness while others want to find caches they can drive right up to. Some people enjoy finding tiny well hidden micro caches, while others like regular sized containers with lots of items they can trade for. Some want easy finds, other want difficult challenges. Simple rating systems fall apart because there is no average cachers. Different people like different caches.

Color me surprised that this same fallacious argument keeps showing up every time somebody wants a rating system. The argument is completely bogus.

 

Why do we have user ratings for restaurants? Movies? Books? After all, not everybody likes the same kind of food, or movies, or books. Clearly, according to your argument, there is no need for any kind of review of anything.

 

The fact of the matter is that the vast majority of people prefer good food, good movies, and good books. User ratings have proven very valuable all these things. Look at the nice review system Amazon has for books. If there were no value in it, do you think they would spend all that money to develop and maintain the system?

 

Likewise, a good review system for geocaches could be very useful.

 

I am really, really, really tired of the argument that if everybody doesn't like exactly the same things then user reviews are useless. I'm a little discouraged that such a weak argument is supported by so many here. If you want to argue against ratings, please come up with something that actually, you know, makes sense.

I'll concede at least partially to fizzymagic's argument here. A simple 5 star ranking system will assign a rank to each cache that many people might decide is helpful in selecting which caches to hunt. Most caches are likely to be rated somewhere in the middle but a few might be outstanding and others may rate very low. Several people have pointed out that they will look at outstanding caches first and based on the cache description, logs, and other information will choses whether or not to hunt these caches. On the other hand they will probably avoid low rated caches altogether.

 

Ratings of course depend on a getting a good sample from the population. Unlike some books, movies, and even restaurant ratings some geocaches get relatively few finds and ratings might be subject to some manipulation. Imagine a restaurant where 90% of the ratings were made by the owner and employees. Would that be a meaningful rating.

 

My main reason for making the comment that fizzy quoted above it that we often get suggestion for a rating system expecting to get from it more than it can really deliver. They often feel it will identify the caches the find as "lame" - usually certain types of urban park and grabs. The popularity of these hides indicates that many people actually enjoy these caches. I am not convinced that these caches will be ranked low on a 5 star system. You need to understand what ratings means and just what you can expect to use them for.

 

People go to Amazon or Netflix to buy or rent specific books and movies. These companies want to encourage these people to buy more books or rent more movies. They invest in sophisticated systems that try to recommend other products to the customer. Most people, when confronted with the huge choice of products, decide not to look at anything else at all. By presenting reviews and ratings that are given by other consumers, Amazon and Netflix are able to get the consumer to spend a little more time looking at the recommendations and some of them will decide to make additional purchases based on the user ratings and reviews. Netflix had a competition where it is awarding $1M for improving the quality of its recommendation engine by 10%. It is clear that for a website that sells or rents something there is value in having users rate what they have bought or watched.

 

Geocaching.com probably has an interest in keeping people geocaching. They will get a certain revenue from premium membership and from selling trackables and other gear, and they will get traffic to the website that influences the rates they get paid for advertising. Like Amazon and Netflix, some people will find a few geocaches near where they live (or where they went on vacation). Afterwards they may find the work in selecting additional geocaches to find too daunting. Or they may have discovered that some geocaches are really enjoyable and others are pretty lame and freeze up at that thought that they may have to find 10 LPCs in parking lots before they find something hidden some clever way that they find amusing. My guess is that any kind of recommendation system is useful for people who are confronted with a choice that seems too hard to make. It should be clear that different people will get different added value from different methods for getting recommendations and selecting caches. For some the status quo of reading logs is sufficient. Others may even enjoy the serendipity of simply looking for all the cache in a given area and finding that they are more often amused than disappointed. But some will want additional tools to help them make selections. I don't think that Groundspeak will have a $1M prize for inventing a better recommendation system like Netflix did. Instead they will consider various alternatives and look at the pros and cons of each and will cautiously proceed with some kind of system in the future.

Link to comment
The main weakness is that different people attribute completely different meanings to the steps of the 1-5* scale and that there is no system-immanent way to find out why a certain rating has been given to a cache.

Right, this is the error I referred to in my other post that is inherent in star rating systems. Like others have said, people enjoy very different aspects of caching. While the general quality of a cache can be generally rated on a general 5 star scale and therefore be moderately useful, a rating system can be much more useful if it focuses on those different aspects of caching.

 

Enter the award/badge system. All you have to do is decide what awards you would give a cache (e.g. an award for scenery, creative site setup, overall cache design (which includes the details page), the hike/trek/trip/drive to the site, and/or the container camouflage).

 

I agree with on that aspect. Some cachers have proposed introducing a few categories to GCVote, but the

cacher who has implemented GCVote strictly objects to that idea. The main argument brought up is that only

a simple rating system with only 1*-5* and a single vote will find acceptance. My claim is just that such a simple system does not work properly for me. My claims do not hold for other types of rating systems in general. It will always depends on the details. My objections against GCVote are not theoretical ones - I really have tested the system in an area that are know very well and where I can judge whether the results are helpful for me or not.

 

Cezanne

Link to comment
Likewise, a good review system for geocaches could be very useful.

So are we talking about reviews, or are we talking about a 5-star rating system? I find reviews useful. I find "you may also like" recommendations useful. I find personal recommendations useful. It's simplistic 5-star (or n-star) rating systems that I find useless.

I'd like to point out that the cache logs are a review system, they're just time consuming to use compared to short, concise indicators. The logs are great and have their purpose, but for quick reference, nothing beats short descriptors - like the information already provided on the search page (distance, cache type, difficulty and terrain). But us particular types are looking for particular qualities. We're asking for more quick descriptors so we can find the types we like.

 

Listen to the arguments on each side: the no ratings people argue against the n-star system people because it'll cheapen and discourage, the n-star system people argue against the no ratings people because they just want a better way to narrow down the "good" caches, and the more searchable people won't settle for an error prone n-star system that doesn't help them find their different preferences. Notice no-one has made an argument against a system that provides more ways to search or more descriptors to search by.

Link to comment

Listen to the arguments on each side: the no ratings people argue against the n-star system people because it'll cheapen and discourage, the n-star system people argue against the no ratings people because they just want a better way to narrow down the "good" caches, and the more searchable people won't settle for an error prone n-star system that doesn't help them find their different preferences. Notice no-one has made an argument against a system that provides more ways to search or more descriptors to search by.

 

Most of us have been asking for improvements regarding search descriptors for years, and nothing has changed. The best thing that Groundspeak ever made in my five years of geocaching was the ignore list.

 

The best suggestion i've seen so far was KBI's Feature enhancement request: 'Interesting Location' attributes

 

My suggestions for some specific attributes to be made available:

Historic Place

Educational Location

Beautiful View

Strange Spectacle

Nostalgic Setting

Engineering Marvel

Natural Wonder

OMG You Won’t Believe This Until You See It

Edited by Kit Fox
Link to comment

Notice no-one has made an argument against a system that provides more ways to search or more descriptors to search by.

 

Maybe not in this thread, but in other threads (also in other geocaching fora) on this classic topic.

One of the main objections appears to be that it is claimed that only very few cachers would be willing to provide data for such a system. I would be willing personally if the system appears to make sense to me.

Otherwise, I would not use it.

 

The main claim of supporters of a 1*-5* system (I do not belong to that group) is that only such a simple system will be used by a sufficiently large group.

 

Cezanne

Link to comment

My biggest issue with some of the proposed schemes are the requirement of having a find it log that will "stick." The reason I say it that way is that until the rod wielded by the cache owner used to force compliance by the finder is removed, ratings will be pretty much useless. Folks will fear logs deletions and therefor artificially bump their ratings.

 

That's where anonymity comes into play. The CO shouldn't know who rated and how they rated.

Link to comment

I think if one thought it more along the lines of "if a friend could only hunt half the caches in this area would you recommend this cache be on that list?" A 1 star would be "skip it." 3 stars would be "eh, either way." 5 stars "this is one you simply must do." This is really what folks are trying to do with a rating system, find out if others are recommending we find that cache.

 

No, that does not apply to me at all. The list I would supply depends heavily on the caching preferences of the friend. A cache that is 1* for myself, might well be among my top recommendations for the friend and vice versa.

 

The attempt of trying to make recommendations for everyone has to fail. The only way it can be turned in a usable approach is if additional information is added that makes clear to which group of cachers a certain cache is recommended.

 

There are many caches that I would highly recommend to about 90% of my caching fellows, but definitely not to the remaining 10% (including myself) for whom these caches might be very bad choices, much worse than taking arbitrarily some 0815 cache.

 

Cezanne

Link to comment

Most of us have been asking for improvements regarding search descriptors for years, and nothing has changed. The best thing that Groundspeak ever made in my five years of geocaching was the ignore list.

 

The best suggestion i've seen so far was KBI's Feature enhancement request: 'Interesting Location' attributes.

Maybe, but I'm excited to hear OpinioNate's news and I note that an awards or badges would serve as descriptors of (hopefully) popular cache qualities. Also, there are many more elements of the cache that people enjoy and therefore deserve descriptors. The experience can be broken down into the cache details page, the trip to the cache, and the cache site. And the site itself has many more enjoyable qualities besides location.

I'll chime in quickly here to say that Groundspeak actually has been tossing around the idea of badges or awards that can be given by users to other users or caches. The nature of those awards are not yet known but will be intelligently discussed and implemented in a positive way that promotes caches/users instead of criticizing "bad" caches/users.

 

Something that has been successful in other areas of the interweb is the suggestion engine (Since you like X, maybe you'll like X). There must be some way to integrate awards/badges with a suggestion engine that would offer very personalized results and accomplish what the OP is seeking.

 

I have to stress that we have no specific plans yet for anything like this. As with any other feature we will first explore in excruciating detail all of the ins and outs a feature like this entails. We're definitely open to feedback from the community, so feel free to continue your healthy debate. :)

Edited by SgtKlaos
Link to comment

So are we talking about reviews, or are we talking about a 5-star rating system? I find reviews useful. I find "you may also like" recommendations useful. I find personal recommendations useful. It's simplistic 5-star (or n-star) rating systems that I find useless.

Ratings will always be combined with reviews since no one is suggesting that online logs (reviews) be eliminated if a rating system is adopted. The cache gets a good rating or award, you read the logs and decide if you agree with the consensus. A rating system should in no way affect your ability to use reviews (logs/bookmarks) to make a decision on whether the cache is something you'd like.

Link to comment

For all you geocachers that are also looking for an extra-ordinary experience while you are caching, there is a simple answer: WOW Factor ratings! When a cacher logs a find, he has the option to give a 1-5 star rating (with half stars), concisely characterizing the positive feelings or experience the cache provided. The WOW Factor should be a column on the search results page and shown on the cache details page.

 

It seems like it would be relatively simple to institute and would prove useful to many users. I count the following as benefits that don't take away from any other aspect of the game. User ratings would encourage cache placers to develop their skills. And many would also enjoy the competition and recognition!

 

So consider this a poll. Would you like to see, search by, and rate the WOW Factor of your local caches?

Bad idea! Groundspeak thought that ratings for bookmark lists were a good idea too! It only provides disgruntled geocachers a way to vent their spleens without owner control. :)

Link to comment

Most of us have been asking for improvements regarding search descriptors for years, and nothing has changed. The best thing that Groundspeak ever made in my five years of geocaching was the ignore list.

Maybe, but I'm excited to hear OpinioNate's news and I note that an awards or badges would serve as descriptors of (hopefully) popular cache qualities.

I'll chime in quickly here to say that Groundspeak actually has been tossing around the idea of badges or awards that can be given by users to other users or caches. The nature of those awards are not yet known but will be intelligently discussed and implemented in a positive way that promotes caches/users instead of criticizing "bad" caches/users.

 

Something that has been successful in other areas of the interweb is the suggestion engine (Since you like X, maybe you'll like X). There must be some way to integrate awards/badges with a suggestion engine that would offer very personalized results and accomplish what the OP is seeking.

 

I have to stress that we have no specific plans yet for anything like this. As with any other feature we will first explore in excruciating detail all of the ins and outs a feature like this entails. We're definitely open to feedback from the community, so feel free to continue your healthy debate. :)

 

I'm in favour of an award system - it's positive, it does not criticize or discourage.

The awarding by category idea is intriguing. Makes things more complicated possibly, in terms of programming (but maybe not).

Link to comment

For all you geocachers that are also looking for an extra-ordinary experience while you are caching, there is a simple answer: WOW Factor ratings! When a cacher logs a find, he has the option to give a 1-5 star rating (with half stars), concisely characterizing the positive feelings or experience the cache provided. The WOW Factor should be a column on the search results page and shown on the cache details page.

 

It seems like it would be relatively simple to institute and would prove useful to many users. I count the following as benefits that don't take away from any other aspect of the game. User ratings would encourage cache placers to develop their skills. And many would also enjoy the competition and recognition!

 

So consider this a poll. Would you like to see, search by, and rate the WOW Factor of your local caches?

Bad idea! Groundspeak thought that ratings for bookmark lists were a good idea too! It only provides disgruntled geocachers a way to vent their spleens without owner control. :)

You're way behind in the discussion, Konnarock Kid. You can't read the first post of 98 posts and expect to say anything that hasn't already been said 20 times. I've learned a lot since my first post and it seems there are a few clearly opposing, well numbered sides. What I'd like to hear now, is Groundspeak's ideas and intentions. Do they respond in the forums or is there another place we can see their plans?

Edited by SgtKlaos
Link to comment
No, that does not apply to me at all.

...and that's where the failure for you to understand comes into play.

 

It isn't all about you. It isn't all about your friend. It isn't all about me. It's not about trying to create a list that is a perfect fit for any one individual. It's about creating a number that represents a group of people who think their friends (not the plural) will likely enjoy the cache.

 

It appears so many folks are trying to create something that will pinpoint a recommendation. It's not something that will say "You will like this cache." It's to say "based on what other folks are saying you're likely to enjoy this cache." (...or not depending on the rating.)

 

Maybe I simply don't understand considering a like all types of caches, just not all caches. There even some caches I didn't like of the type that I do generally enjoy. Go figure. There are urban micros P&Gs that I've enjoyed while some hiking caches I've not enjoyed. Again, go figure.

 

Now, if you have a friend that, say, simply hates micros simply because it's a micro, then sure, they'll hate a 5 star micro that almost everyone loves--except for your friend. Can't help that, but that is not even close to a reason to toss out a rating system. Besides, I don't think anyone is going to base a hunt solely on the user ratings when they can't stand a type that is easily filtered out. That's simply a dumb proposition.

 

The rating system is not a pinpoint saying you will either love or hate a cache. You're right, that will not work. That goodness such a system is not in place. A good rating system will provide a percentage of chance that you will either like or dislike a cache. Again, a 1 star is likely to disappoint you with that "genre." A 3 star cache could go either way--you might like it, you might not. 5 stars would mean you're most likely to like the cache--even if it's the type of cache you might not generally enjoy.

 

Here's an example: how many did Tube Torcher and didn't like it? The sequel? I'd think one of the most celebrated caches in the early years of caching would rate a solid 5 stars. But guess what, a lot of folks don't like urban caches. I'd venture to say that a lot of folks who despise urban caches loved Tube Torcher. Guess what. Tube Torcher was very much an urban cache. It also was a multi and had many puzzle elements to it. See a good cache of a type that is generally not likely can very well be a favorite.

 

I filter out micros beyond a certain distance from home as they generally aren't worth hunting. Yet, some of my all time favorites are micros.

 

I think you're wrong in saying that a highly rated cache of a type that is generally disliked by someone would be university disliked, by that same person, regardless of how well it is done--some examples excluded like ones that require skills beyond that person's capabilities.

 

Me, I'd probably include micros back into my distant PQs if I could get only the top 25% rated caches. (or whatever percentage that it works out to work for me.)

Link to comment
Guest
This topic is now closed to further replies.
×
×
  • Create New...