Notice the Silence in Site Feedback Forum????

[seanb]

Since the subject of IP addresses has come up as a means of identifying posters, can I just point out that a large number of posters will be behind a NAT firewall at their organisation. For example, I share my internet IP address with several hundred other people; all it takes is for one of those people to be on photo.net at the same time as me, and hey presto, two posts from genuinely different people at the same time from different IP addresses.

 

For organisations such as IBM or Microsoft, there must be thousands of potential users behind a single internet facing address. The same applies to schools and colleges.

[yongbo]

Sean, unfortunately, Brian doesn't have time to teach computer forensics.

[mottershead]

We don't just rely on IP addresses, since proxies, NAT, etc, make them inconclusive. However, there is other information. A combination of technical information plus typical behaviour patterns can make things quite conclusive. In the particular cases under discussion, which is one photo.net user giving ratings predominantly to one other user, and all high, it doesn't actually matter if they have the same IP address, etc, since the ratings are disqualified even if both parties are real people.

[vincetylor]

"I carefully read what he wrote and did NOT come to the conclusion, then or now, that he was doing anything more than making a point about logic, and not a point about your ratings tendencies or your character" -- John Crosley

 

Well I certainly appreciate your thoughts John. I can see you are trying to lighten the situation here, which is always nice to see. However, I respectfully have to disagree with you. Brian has ALREADY removed 378 of my ratings based on his flawed system. He also stated here that I would be among the top 25 mate-raters/abusers or whatever else you call them, if heads (ratings) were to be "chopped off" based on his flawed system as well. His comments were not just to present some hypothetical viewpoiint in my opinion, but were made to state his case against directly going after abusers and to also insinuate that I am a hypocrite.

 

IF I were in fact a mate-rating gaming-hypocrite, then I would be more than willing to admit that here and humbly make the adjustments after receiving what I'd consider a fair warning. If all of my ratings were deleted as well as the other 25 or so ahead of me, I'd still consider it a victory for the site as a whole. As I already stated above though, I disagree adamantly with Brian trying to lump me with these others. I also specifically explained not only why I disagree with his assessment above, including the facts as they really are, but how in fact the problem here truly lies with the actual system he is using. Calvinball was basically something he threw together in which he did not want to share the details or specifics with others. I can now see why. What little he did share has exposed his system as one that penalizes those who have more time/involvement regarding rating, rather than just targeting the true abusers themselves. Like I already stated though, it just needs a little more intelligence or refining to take some of these other things into account. No big deal, no hard feelings, just hope he gets it done for a change. Those of us that have PUT-IN so to speak with involvement in the Photo.net gallery, have been penalized by having ratings unnecessarily removed and by being assured little visibility if we happen to post images right now. This is not exactly you want to see happening here especially since it is completely avoidable with better decision-making.

 

"Is anyone else chuckling to themselves that now, instead of people complaining about how people got into the TRP by having high rates from their friends, they're complaining that people are being kept out of the TRP because of rates from their friends?" Good ole Mike Dixon

 

One day (hopefully) you will either take the time to read the specific points actually written, or find something meaningful to contribute in the forums. I'll put my money on the former.

 

Brian has used a blanket *ten 7s disqualifies all ratings exchanged*. In fact it's probably more like, if any photographer even RECIEVES ten 7s from another person, they are ALL discounted. How else can you account for photographers that do not even rate images losing ratings?? Again (it's all posted above Mikey) if a photographer has been around a while, and has MANY images posted, eventually all of his images-ratings will be penalized, because eventually people will go beyond the ten limit. It's just a matter of time. And for many that have already been around for some measure of time, it has already happened. IF Brian takes some of these factors into consideration, he can refine his process so as not to discount honest rates, as is currently the case. He can also see who is really abusing the system today which has caused him to limit 7s, introduce Calvinball, remove raters names, and change the TRP default pages several times.

[mottershead]

Well, Vincent, at present I'm not lopping off heads based on reciprocal high ratings. I'm just disqualifying the mutual ratings. That means only that whether the people are mate-raters, or just long-time-members-and-great-photographers (such as your illustrious self), the standing of their photos in the TRP has to depend on the ratings of people who are not fellow mate-raters or other long-time-illustrious-people. Assuming that there isn't some reason for new-and-not-so-illustrious or non-mate-raters to appraise the photographs of the illustrious ones differently, the photos should still come to the fore, and only the mate-raters depending on the exchange of ratings will see a difference.

[jay_patel]

Hello Brian, Carl, <br><br>

I have a suggestion...we should identify diversity of rating for everyone (not just the totals). For example, out of 20 7/7 ratings I gave out, 10 of them are for Vincent's image, 5 on Walter Images and 5 on Carl's images, then my distribution of ratings would 7/7 ratings would be fairly narrow. On the other hand, if I have given out 10 7/7 to Vincent's images, and 1 each for other photo.net memeber's images my diversity of ratings would be fairly high. This would indicate, that I am a big fan of Vincent's work, but at the same time, I judge other photographs on their own merit. Perhaps everyone should have their diversity scores published. <br><br>

I don't think that the absolute diversity score would mean anything, but the diversity score when compared to the average diversity score on photo.net should be able to tell volumes about someone's rating behaviour. Is there a flaw in my thinking? I would love to hear from everyone who has commented. Brian, is there anyway we can test this theory out on some know mate raters?<br><br>

 

-Jay

[root]

Brian, you've just described a TRP sort that doesn't exist (unless Calvin has been hard at work during the last week or so.)

[root]

Jay, in Brian's example he gave the impression that only total volume mattered. Vincent raised the issue of frequency. Now you've mentioned diversity. They all matter, but the catch-22 in all this is that as soon as you try to define any of these parameters with any kind of clarity, you've given the gamers information that will help them circumvent the system.

 

For example, you could give 7s to your mates in a timely fashion (within three days), then give 7s to random older images.

[johncrosley]

Vincent K. Tylor

You made the statement in a post above 'My total of images rated 7/7 Brian is 64.' Yet later you made a statement that Brian had deleted 378 (if I am correct) of your ratings. I presume if Brian is deleting alleged mate-rating rates that a substantial portion of those were 7/7s and not part of the '64' you wrote about.

 

My remarks above were predicated, as I stated, on the assumption that you were not withholding any 7/7 rates that had been deleted by Brian. I cannot be sure if the '64 7/7 rates' is the grand total remaining not deleted after Brian's deletions or if some of this 378 deletions were 7/7s. If some of those 378 deletions were 7/7s, then the logic of my remarks is faulty, as I noted when I made them, and the conclusion would simply be wrong.

 

I don't have enough information to make a judgment one way or the other, but I have a strong inkling that the 64 7/7s is the number after Brian's deletions, and that your prior post was a little misleading. Maybe you can clarify, and if my assumption is wrong, I apologize in advance for any innuendo.

 

On a more global note: I think Brian moves at a somewhat glacial speed. That is not always wrong in dealing with a very large organization with a very large membership, not all of whom sign on and off all day, every day. I think he keeps mind of what I noted before: The Law of Unintended Consequences and he likes to see a change work its way through the system before changing or tweaking it; I don't think he has his head in the sand. I think he pays careful to the 'site feedback forum' even though it's primarily the home of a very vocal but very small minority of users.

 

Remember, the major problem a year ago, was just keeping the site running, and now the major problem has devolved into how we keep/don't keep our friends numbers on our photographs.

 

That sounds like progress to me.

 

John (Crosley)

[vincetylor]

John, I have given only 63 7/7 ratings altogether with or without his deletions. 63 7/7 TOTAL. They are in fact still posted on the images I gave them. However, they are just not counted when the ratings are tallied up.

 

"I'm just disqualifying the mutual ratings." B Mottershead

 

Then why even bother with names on the rates at all? Why also promote an interested persons list? If my ratings on Richard V Hoesel, or Dave Nitsch, or Jay Patel, Walter T. some new guy or anybody else are not even going to be counted, why bother taking the time to rate at all? Is this the actual goal now? I may not have as much interest in Carl Roots type of images as I do landscapes, but still throughout the course of time have given his images a few 7s as well. If enough time elapses, eventually all of those ratings will be discounted. How reasonable is that? You have already taken the names of the raters away from the rates themselves Brian. Now you wish to discount ALL rates once somebody hits a total of ten? If I give ten 7s over the course of a few years even though they may have posted close to 50-60 or more images they all get thrown out.

 

 

And remember, all of this is your chosen course of action in trying to deal with the MATE-RATING. Well, they're still quite alive and well. In fact, I just took a look at the older default "AVERAGE" view. The exact same names are plastered all over page one. Take a look right now. All this has done Brian is really penalize honest people, and slowed the raters NONE.

 

I am sorry if I have sounded like a broken record, but you have targeted the wrong people here. That first page is a who's who of mate-rating. Go after those ones with warnings and eventually delete all of their ratings IF they continue. It is the only way to succeed here.

[jay_patel]

Carl,<br><br>

This is interesting...Now we need to define objective for the ratings system. What are we trying to accomplish? It is diversity we are trying to promote, then we can target a statistical measure to promote diversity. But, it the game it to get on the front page of TRP, other measures would be far more effective. I guess, if I ask a question as to what you want ratings system to accomplish, I would get different answers from different people. <br><br>

So, lets clarify...What are we trying to accomplish with Ratings? In other words, how should the photo be rated? How should it get on TRP?<br><br>

-Jay

[mottershead]

No, Vincent, AT PRESENT, if two people EXCHANGE ten 7 ratings, then their mutual ratings are disqualified. EXCHANGE means A gives B ten 7 ratings, and B gives A ten 7 ratings. Also, this is Calvinball. The last time I did it was ten 7 ratings. The next time it might be 15, or it might be 8. Or it might bring 6 ratings into the mix. Or it might take into consideration other factors like the length of time over which the ratings were given, since you seem to think this is all-important. The ratings aren't deleted, by the way. So if you feel you are communicating something by your ratings to the recipient, you still are. And if Calivball disqualifies some ratings that shouldn't have been disqualified, then nothing stops them from being dis-dis-qualified.

[mike dixon]

<i>One day (hopefully) you will either take the time to read the specific points actually written, or find something meaningful to contribute in the forums. I'll put my money on the former.</i><P>

Vincent, I didn't go into addressing every specific point because your overall thesis had a very significant problem: it seems to define mate rating and problems in terms of what <b>you</b> do or don't do rather than objective criteria. As for my contributions to the forums, if you bothered to venture outside the feeback forum, you might realize how utterly baseless your accusation is. Some of us contribute something other than constant complaints about how people put numbers on other people's photos.

[root]

" . . . it seems to define mate rating and problems in terms of what you do or don't do rather than objective criteria."

 

OK, Mike, what sort of objective criteria can we get out of you?

 

. . . . .

 

I thought not.

 

Since you don't really care, why are you here?

 

I think it's an interesting problem in social interaction.

 

May I remind you that the gallery drives this site a whole lot more than any of the forums do.

[bens]

I have utilized my extensive legal training to carefully read this forum and hereby conclusively conclude and definitively determine that the horse in the middle of the room is indeed now dead.

[vincetylor]

Well Brian, I definitely do appreciate your last sentence:

 

"And if Calivball disqualifies some ratings that shouldn't have been disqualified, then nothing stops them from being dis-dis-qualified."

 

Fair enough. I realize this is not some proven scientific formula. So, if you do see some problems with a few of your chosen algorithms to start, it is good knowing you can and are willing to change them. I just did look at three photographers that had lost ratings when you first implemented the Calvinball system, even though none of them have rated a single image. All of their rates have been since put back. That was the fair thing to do. I also hope that certain criteria be implemented regarding some of the issues written above. Especially regarding the general blanket ten 7s and boom all are disqualified. Other considerations could and should be factored in that equation.

 

.."your overall thesis had a very significant problem: it seems to define mate rating and problems in terms of what you do or don't do rather than objective criteria". Mr. Mike Dixon again

 

Mike, if I wanted to mate-rate or stack the deck so to speak, I sure don't need to come here now do I ? In fact none of these obvious abusers ever bother coming here. Well, one did several months ago, and if memory serves me correctly he was roasted out of the site altogether. Any of us could easily just do the "Great Image". "good colors", or my favorite nowadays...(I see it's really the new catch-phrase with mate-raters) 7/7 on the heading, and then 7/7 for the comment. I think I could learn to do that Mike. No, if trying to manipulate this system was my goal, there would sure be easier ways to go about doing just that... with no typing practically and no time away from work. Yes, I probaly should had phrased my comment a little differently. Here I'll try it again, this time with a marked improvement.

 

One day (hopefully) you will either take the time to read the specific points actually written, or find something meaningful to contribute in the *SITE FEEDBACK FORUM*. I'll put my money on the former.

 

There, is that better??

 

The bottom line Brian, Mike and anybody who gives two cents; mate-rating has not been stomped out at all. In fact the first page of the "average" view is filled with these very same people. You've done the anonymous rating thing, the Calvinball, the rate recent sum, limiting 7s etc etc...

 

So now what??

[mike dixon]

<i>OK, Mike, what sort of objective criteria can we get out of you?

<br>

. . . . .

<br>

I thought not.<br>

Since you don't really care, why are you here?<br>

I think it's an interesting problem in social interaction.<br>

May I remind you that the gallery drives this site a whole lot more than any of the forums do.</i><P>

I agree that it's a somewhat interesting problem in social interaction. However, it's important to distinguish between satisfying the desires of a very small but very vocal minority of users and satisfying the desires of the vast majority of users. A system that makes a few people unhappy because they think it's unfair isn't necessarily a systems that damages the site's viability.<P>

I think the current default TRP is actually very fair in the sense that it provides an relatively even playing field among both new contributors and long-time users who would like to see their images in the TRP. A random assortment of a wide variety of people get to "cast their votes," and the images that have the most appeal to them are shown in the TRP. (As an aside, I don't think it's a particularly meaningful way of saying which images are <b>best</b>, but it is a fair method of determining mass appeal.) I don't have much sympathy for those who are unhappy because they no longer benefit from having a (non-random) group of people give them good ratings (whether those ratings are "fairly" assigned or not), not after hearing months of complaints about how horrible it was that other people benefitted from a similar situation. As Brian noted above, and as I've noted before in rebutting the arguments from one of Paul Greenwood's friends: if your images really do have a high level of mass appeal, having the ratings from a small, select group of raters removed or ignored will not significantly affect your "score."<P>

Vincent, as John Crosley pointed out, Brian's point about your rating behavior was that it was difficult to discriminate between it and the "mate raters" you so despise based on objective criteria. Rather than recognizing that conclusive identifying the abusive mate raters isn't as simple a matter as you claim, you instead act like you're being horribly persecuted. Give it a rest already! John started this thread to note the improvements that have been made, but you couldn't pass up the chance to fire off still more criticism of the job that's being done by the admins. And you act as though your constant harping is a valuable contribution to this site! Tell me, Vincent, what makes your constant criticism and insults of the management any more valuable than my criticism of your constant whining?

[vincetylor]

Because I have been laying specific suggestions on the table Mike. And because I AM involved with the gallery, I AM involved with rating (YOU have zero ratings). I DO comment (last I checked you had eight total) and I have been active posting images, sharing information with many others. Something you have little or no involvement with at all in the gallery. All you have done is complain about our doing just that, making suggestions to help improve things here. If you really knew what you were talking about regarding these issues, you too could perhaps offer SOMETHING constructive. Instead, you only complain about those trying to offer suggestions.

 

You say the current TRP is very fair. I would suggest you are dead wrong. In fact, Brian Mottershead has already mentioned it's NOT fair the way it is now, and that he plans to change it. Care to comment about that now??

 

A level playing field is all people are asking for. One where mate-raters do not cheat their way to the top pages, and one where people who've been involved are not penalized for doing just that. Rate recent SUM as TRP default does just that. Give me a break. You are way off! This too however is nothing new.

[root]

" . . it's important to distinguish between satisfying the desires of a very small but very vocal minority of users and satisfying the desires of the vast majority of users."

 

By your logic, only the very vocal minority count as being dissatisfied. The turnover in the gallery is quite high. Maybe they expected to post a few images and then leave. . . . but I doubt it.

 

"I think the current default TRP is actually very fair in the sense that it provides a relatively even playing field among both new contributors and long-time users who would like to see their images in the TRP."

 

See Vincent's comment above. It's primary purpose is to attempt to change the behavior of mate raters. It's not about quality, variety or fairness.

 

"A random assortment of a wide variety of people get to "cast their votes," and the images that have the most appeal to them are shown in the TRP."

 

Not true. It's about volume of rates of the RFC queue and features images that somehow managed not to be rated in any other way. (How do you explain images with a below 5 average on page one?)

 

(to Vincent)

" . . it was difficult to discriminate between it and the "mate raters" you so despise based on objective criteria"

 

No it isn't. You simply define abusive rating as exceeding a specific number of 6s and 7s in a given time frame offered towards both images in the current rotation and individual photographers, reciprocated or not.

[jay_patel]

Ok...After reading most of the resposes. It seems that talneted photographers rating other talented photographers may find themselves classfied as "Mate" Raters. If I give Dave N. 7/7s on bunch of his photographs and he does the same to me, then we our ratings are disqualified? Or did I get it wrong? <br><br>

If this is the case, to me it just seems that we are discriminating against the talented photographers on this site. This should not be the case. <br><br>

-Jay

[mottershead]

Jay, not really. The talented photographers don't need the ratings of the other talented photographers. Their work will get plenty of visibility based on the opinions of non-talented photographers. The only "talented photographer" who is penalized by this approach is one whose work can only be appreciated by the other talented photographers to whom he happens to have given many seven ratings.

 

Yes, there is a problem. It is that "talented photographers" exchanging high ratings are more or less indistinguishable from mate raters, and their mutual ratings get disqualified if they mutually lean too heavily on the seven key when rating each other. However, it isn't really a problem because, unlike the real mate-raters, they don't NEED those seven ratings. If they do need them -- well, maybe they are mate-raters after all.

[root]

Jay, there are several problems.

 

1) Volume. Some photographers upload quite a few images and they all rise to the top because of the social pressure on friends to rate them all equally high. If instead, I only rate a few of yours high, you are much more likely to only rate a few of mine high in return. The flip side of mate rating is that most mate raters tend not to pay too much attention to other photographers unless they respond in kind. So it ends up being about back scratching rather than talent.

 

2) Mediocrity. There are a number of well known examples of people who cultivate reciprocal rates by offering a high volume of high rates themselves. The problem is that by any objective standard, their work doesn't warrant anywhere near the high rates that are returned. So again, it looks more like back scratching than objective evaluation.

 

The bottom line is that there is very high correlation on this site between the volume and numerical reciprocal rates that people get and the ones they receive, both individually and on average. It becomes obvious that those who are the most generous and least discriminating will become the most popular. Making ratings semi-anonymous has probably reduced this correlation, but only slightly.

[joe604]

Carl, I suspect that making ratings semi-anonymous had <b>zero</b> impact on mate-rating. The mate raters have no trouble making there actions visible to their beneficiaries. The rationale for semi-anonymous ratings is to inhibit mean-spirited retaliation. --Joe

[jay_patel]

Brian, <br><br>

Glad that you acknowledge that there is a problem that we cannot distiguish between Mate Raters and talented photographer giving 7/7s to each other. But, still my point remains valid that good photographs by talented photographer are discriminated against in the ratings system.<br><br>

Now whether a talented photographer needs or does not need a ratings 7/7 depends upon what the photographer is trying to accomplish. For someone like me who is semi professional, I tend to view ratings (averages as well as number of ratings) as an indicator of how much the photograph impacts others. I also believe in what my peers think about the photograph...sometimes they will leave a comment and other they will leave a rating. Sometimes, I may love the photographs, that other talented photographer may have difficuly understanding...I want to know that before I submit a photograph to my clients. So, to extend a philosophy that "talented photographers" do not need ratings is does not seem appropriate.<br><br>

Having said, I am in support of establishing a policy that give everyone a chance to get visibility. I agree with Carl, that volume is a problem with some photographers. For some photograph their rise to the top heavily depends upon the volume of photographs uploaded at the same time. So, why not limit only one photograph from every photographer on TRP per week (or some other duration)? i.e. you can upload zillions of photographs a day, but only the top rated one will show up in TRP for a given time period. This would give everyone an ample opportunity to get their photographs visible on TRP...and you may not need to enforce the policy that seems to be unfair to works of talented photographers.<br><br>

-Jay

[root]

Joe, I was referring to reciprocal rates of all kinds. If you give a lot of 5s, that's what you'll get in return in far greater numbers than if your average was 4 or 6.