Mist of Avalon flagging conspiracy - any way to prevent

I was somewhat surprised to see my review of this book flagged - yes it is short and meaningless - but it is also innocuous. I was curious to see why it was flagged - so I went to the book page and lo and behold someone went through and flagged all short reviews and (even worse) all reviews linking to a website. The former is a matter of opinion, the latter is a misunderstanding of LT policy. Is there anyway to monitor something like - percentage of reviews flagged per book - to prevent this?

Unfortunately we don't have any way to force people to read instructions. Of course, this is true of more than just LT review-flagging...

READ, people! This is a book cataloguing site, don't you like to read?!?

Ahh, end rant.

I really think people who blatantly disregard the instructions should just have their flagging privileges turned off.

Supposedly Tim et al are thinking of implementing a system whereby a user can flag a flag.

http://www.librarything.com/talktopic.php?topic=17366

But, like nperrin said, I don't know why people can't read the instructions. Flags upon flags upon flags probably won't change that.

I think Tim was willing to turn off users flag ability if they are abusing it...

I know I said "end rant," but then I went and looked at how many reviews got flagged. This really bugs me. The short reviews are bad enough but flagging a link to someone's blog of book reviews would, in my world, be grounds for swift and serious punishment. Some of the best reviewers around here have book review blogs where they actually post long and thoughtful reviews, and do so often. The review field specifically says, on the edit page, review OR URL.

What part of "Important: do not abuse this feature" is so hard to understand?

Maybe we can have a 'flagging log' like the combiners one, so we can leave polite comments, explaining why a given review was OK.

After all that feature wouldn't be abused either would it?

I do not understand why there is a feature that labels a user's review with "flagged as not a review". Why would you care if a review was not to your standards...it is not your review. Simply ignore it.

Also, to all those review purist who seek and find those one sentence reviews...many of the greatest reviewers have printed reviews with one sentence, to name a few: Quentin Crisp, Dorothy Parker, Pauline Kael, Roger Ebert, Clive Barnes. And Alexander Woollcott's review of the play "Wham!" contained only one word: "Ouch!" Enough said.

reading_fox: No, not at all...

There could be a flaggers' group in talk though. :)

#8: My understanding is that 'not a review' is meant for the ones that list, say, the price of the book, or where it was purchased. 'Ouch' is a short review, but is '$7.99 from Borders'? I'd guess that was meant to go into the comments field and got put in the wrong box... I'm surprised I've never done that, in fact, especially with the manual entry form where the boxes are in the opposite order than they are on the edit page.

The flagged reviews of Mists of Avalon and The Brothers Karamazov did not contain price nor format information, just short opinion. Those reviews were flagged because a purist user deemed them too short.

Yes, I realize that... I was responding to: I do not understand why there is a feature that labels a user's review with "flagged as not a review".

The fact that some people may misuse a feature is not, in my opinion, a reason for the feature not to exist.

#9 good idea.

Flaggers! please join and discuss!

kingkama:

What antqueen said. Yes, the "not a review" feature gets abused. So does the review feature, by people posting nonsense like "7.99" or "n". (I have seen both of these as 'reviews'.) That doesn't mean they shouldn't exist.

And as long as there's a thumbs-up and no thumbs-down, people WILL use the only tool they have -- the flag -- to indicate a crappy review. I use it only for things with zero content, but I can understand why people use it for things like "this was terrible" or even "this was great".

kingkama: I flag reviews that have "Book 3 of a series" or "$1.25" or "haven't read yet". Things that have no real content.

#14

Well it seems that a few people are. Maybe a public warning to everyone that abuse of the flagging system is an abuse of the TOS and as such could lead to further action like suspension could work on lessening this tendency in some.

I think the "not a review" flag feature should be removed because it's too open to interpretation and it will continue to cause problems like the examples mentioned here. Anyone agree/disagree? For those who disagree, I'm curious why - what value doe this flag provide you?

Well, a lot of us asked for the 'not a review' feature!

However the directions do clearly state that short reviews should not be flagged. We probably do need to be able to unflag such reviews.

When flags for posts were introduced there was a period when some people went wild flagging stuff but there were modifications to the system and the kinks have mostly been worked out. So, I'm taking something of a wait and see attitude.

It's not at all open to interpretation; people are just too lazy to read what the flag is for.

@17

Er, exactly the value that I already described. It lets me indicate things that AREN'T REVIEWS. Quirkily enough, when I go looking for reviews, I'm looking for actual reviews, and the letter "n" doesn't quite give me what I'm looking for. So I flag that lonely letter, in the hopes that it will give that 'n' review a little less weight so it won't be on the top of the list. "Just ignore it" is like "just hit delete" as a spam response; okay in the individual case, unsustainable in general. I don't want to have to sift through half-a-dozen instance of "7.99", "n", and "345 pages" to find the single review for a work.

In my ideal world there would be a "short review" flag, so that users could choose whether or not to see one-sentence reviews, but I'll settle for being able to downweight the one-character reviews. (Dorothy Parker could get away with a one-sentence review because she had a regular readership for her reviews, and a reputation; J. Random LT User doesn't have that luxury.)

dchaikin, kingkama: what concievable justification could there be for calling a single character or a price a "review"? If it can be universally agreed that something is not a review, shouldn't there be some way of *indicating* that?

Conversely, in my ideal world there would be a "short long review" flag, so I could more quickly scroll by someone's thesis and/or diatribe and get to a review that will tell me something before the seasons change and the book is out of print.

Yeah, it would be neat to have a "short review", "long review" categorization method. Not flags, per se, since the content is fine.

forgive me if I'm missing something, but you can tell by looking at it if it's short or not...

Oh! Do you mean there's one page for short reviews, and one for long ones? That's a good idea. Alhtough where would the cut off point be, and who would have the final say as to which they belonged on?

>23 LadyN:
I meant if there's call for short review flags, why not flags for long reviews? Why single out short reviews?

Would it be helpful if reviews could be sorted by length (number of words, characters) or complexity (using average sentence length, word length, etc.)?

#24 - Sorry, fleela, I got what you meant, I should have been clearer in my post, and numbered it for 22.
I have a problem with the overly long reviews too. I think they can be as equally un-reviewish as the one sentence, price tag ones.

#20 dchaikin, kingkama: what concievable justification could there be for calling a single character or a price a "review"? If it can be universally agreed that something is not a review, shouldn't there be some way of *indicating* that?

My thoughts are that these are self-evident as "not reviews" and easily skipped - in other words they are kind of harmless. If they were long, then that would be bothersome. But, short misc. stuff is easy to skip past.

I'm certainly OK with flagging "not reviews" in theory, but I'm not sure I like it in practice.

#19: This was discussed in another thread where it was pointed out that under some definitions of review a summary counts. So if a summary is a valid review then "Book 3 of a series" tells you something about the book and should be allowed as well. So it seems that there is room for interpretation.

#17: I agree and I already stated that opinion on that other thread. Thumbs up should be enough. The short reviews and non-reviews will get sorted to the bottom because they won't get any thumbs up.

>28 jjwilson61:

Why won't short reviews get any thumbs up?

I meant "too short" reviews. Those reviews that people are trying to misuse the not a review flag for now.

I've never flagged a review. The only time I've been tempted to hang a flag is when I see massive cut-and-paste jobs from, for instance, Publisher's Weekly; would that be out of line or would it be helpful in helping LT steer clear of copyright infringement trouble?

If anybody flagged one of mine, how would I know? And would I know why? (I just passed 140 and am bound to irritate somebody one of these days.)

In one of the first discussions of flagging, Tim said that cut and paste reviews (even if credited) should be flagged, as they are copyright violations. Generally, these are the only ones I flag. For some reason, they just really bug me.

Actually, the ones that bug me the most are when people c&p reviews from Amazon customers; I've even seen people post kids' reviews from there!

#31: There are two kinds of flags, a not-a-review flag and a copyright-violation flag.

#31 - You wouldn't neccessarily know unless you checked back over all your past reviews. And even then, you wouldn't have to care if someone had flagged one of them. I don't.

#34 I care. I have this idealistic notion some users will actually read my reviews and find them useful. I would prefer them not to have flags (at least not unless they deserve it).

#31 i think the only way to check is to go to your reviews and reverse sort (ie click twice) by votes. I'm not completely sure this works properly yet, but flagged reviews are supposed to sort to the bottom.

I've just checked and founf this does work. AND two of my reviews have been flagged, ok they are short but they are complete sentances, and say something useful about the book. grrr.

#34 I guess I care too.

I've been thinking more about the "not-a-review" flag. I think reviews are part of our personal libraries and that our personal preferences for how we use reviews should be respected. Each user has their own philosophy. As long as users follow the Terms of Service, I think they should be able to use the reviews as they see fit, without penalty.*

Also, I think the cost of the flags outweighs the benefits. The main value of these flags is to knock these reviews to the bottom of the list. But, I’m not sure how much this will gain over simply letting these reviews drop due to a lack of thumbs. Also, I don’t find a problem in most of the reviews complained about.** Books prices and one letter/word reviews are so easily skipped over, they are really harmless. The cost of the misplaced flags however is high because it really upsets and pisses off users. At best it will make users hesitate to post reviews and therefore reduce the reviews. At worst it will maybe chase users away from LT.

One final note stated elsewhere, a potential benefit of the flags is to get users to improve their reviews. However, as editing the review won’t remove the flag, this won’t actually work.

* IMO flagging would be OK if:
1. LT introduces a bad review flag (unlikely, and I hope this does not get introduced)
or
2. LT introduces a spoiler flag (This I would like. It should not be a criticism, simply a notice for readers. Reviewers should be allowed to flag their own reviews)

**Offensive reviews are a different story. Does the TOS have a policy against particularly offensive reviews – for example, racist or misogynist or other hate-themed reviews?

>Does the TOS have a policy against particularly offensive reviews – for example, racist or misogynist or other hate-themed reviews?

Doubtful since they allowed religious hate speech in Talk.

Some people probably don't even realize that what they put in the Review box shows up anywhere other than in their catalogs. (There are people who have been around for over a year who have never looked at a lot of features, and who don't read the Groups that discuss the site.)

#37,

The "offensive" issue is problematic. There are tags used that could qualify as well. And there are few things that won't offend someone. (One of my friends regards it as a personal attack if someone says that her favorite sitcom is silly.) It could even be tricky to ban racist epithets; I noticed a post inquiring about the origin of one.

A review of that sort could actually be helpful --- if the reviewer likes it, you probably won't. If he denounces it, you may want to get a copy. :-)

dchaikin: This suggestion totally devalues the thumb. It's grade inflation at its worst.

I want to be able to use the "thumbs-up" to indicate an exceptional review, not just something that is, in fact, a review. No thumbs should be the *default*, not an indication that 'this is a random string generated by someone's cat walking across the keyboard'. Unless we get a thumbs-down, the thumbs-up feature makes review-rating nothing more than a feelgood exercise.

Maybe on something like Harry Potter the one-letter reviews are easily skipped, but if I see four reviews for something I'm considering buying, and two are one-letter or price reviews and a third is a broken link to a blog that no longer exists, the review feature is basically useless. I want to be able to hide that sort of noise. Reducing the number of reviews, without reducing *reviews that contain actual content*, is a GOOD thing, not the horror you make it out to be. Maybe someone who gets all their 'reviews' that consist of the book's price flagged will discover the comment field as a result, and everyone is happy.

I know I'm outnumbered, but I don't understand *why* the prices and single-character reviews have such passionate defenders!

#41

I would agree with you.

As I posted elsewhere it is a tragedy of the common issue. One poor review might not be offensive enough to introduce flags. But what about five? What about a dozen? The book page only shows 6 reviews by default - if all of them are prices and single letters how is that adding value? It means that every user will have to click a link (and retrieve the next page or all the reviews) in order to see a single useful review. Not very considerate to the community at all.

By flagging these and moving them to the end we are not preventing such users from entering price info or single letters - I doubt Tim will ban them (or even warn them) for it. They can still see their 'review'. We can still see their 'review' if we really care. All we are doing is saying that we would rather read a review than their single character 'reviews'.

#41/42 - I agree with you in theory. I haven't come across examples andyl describes. That would be an annoying problem. But, if a book has only four reviews and three are from the cat, sorting doesn't really help.

Maybe the blue flag should be re-defined. Instead of "not a review" which is (too?) open to interpretation, perhaps it should be narrowed down to specifics and read something like "Review is either a broken link, random text or only a price"

37:
One final note stated elsewhere, a potential benefit of the flags is to get users to improve their reviews. However, as editing the review won’t remove the flag, this won’t actually work.

There's also the problem that not enough information is conveyed by a flag. Did the flagger think my review was too short? Too long? Grammatically incorrect? Did I accidentally give a spoiler? Should I give up and try a different web site?

The problem is that the not-a-review flag is being used on more than just the one-letter and price reviews. I see your point that if the thumbs up just means that it meets the minimum qualifications to be a review then it doesn't mean much. I think a better solution to that problem than the not-a-review flag would be the ability to rate reviews. If you made 1 be a minimal review, 2 average, and 3 very good, then no thumbs would be basically not-a-review.

I think the directions would be crucial though. It should be clear that 1 shouldn't be used for reviews that you disagree with, just that they don't contain much info.

#43

That may cut down the number of people who are incorrectly flagging.

I think one of the issues at the moment is that the large number of blue flags that have been incorrectly applied could be the result of one or two users or it could be the result of lots of users. If it is the former the problem may well be controllable without rewording.

get users to improve their reviews

If my reviews are flagged because someone thinks they should be "improved", it will do nothing but make sure I continue to put up short, succinct reviews.

I only use a flag on things that are not reviews. (unread, prices, etc.) Anything that's actually about the book, I don't flag.

Ok, so I fixed my "anticipatory review" (I gave the book a rating before I had finished it, and said so.). Now the review is "corrected" how do I get the flag off?

I tend to agree with #37. Flagging "not a review" seems good in principle, but it's being misused to the point that it does more harm than good.

I said before that I thought people who were misusing it should just have their flagging privileges turned off. But it seems like that's not going to happen; Tim even said once that one of the people who was incorrectly flagging a lot of things as non-reviews was also correctly flagging a lot of things as non-reviews, as if that makes up for it.

I think the thumbs-up feature on its own is sufficient, in that it will get the best reviews to the top. I don't think it's important to have all the reviews precisely ranked from best to worst, as long as the ones that show up on the first page are the best. If you want to take the time to read more than six reviews, it's not much more effort to skip over the ones that just include a price.

At this point, I think the best move would be to eliminate the not-a-review flag, and also reset all the thumbs-ups at zero to make up for people giving thumbs-ups just in compensation for incorrect flagging. Then we could start over again without all the negativity.

> 45

If you made 1 be a minimal review, 2 average, and 3 very good, then no thumbs would be basically not-a-review.

The problem with that is that just as many (probably most) people don't rate books, you're probably not going to get most people to read, much less rate, all the reviews.

And, frankly, no matter how often you tell them they shouldn't, people will rate reviews based on whether or not they agree with the reviewer's assessment.

Then we could start over again without all the negativity.

I think the problem is we started the first time without any negativity. This clearly all happened because some people didn't want to allow a thumbs down, while many (most?) people want to be negative - and have found an alternative way to do it. What are you supposed to do if there are five reviews on a page and four are mediocre, and the fifth one is really crap? Give four thumbs-up to so-so reviews, or just leave them all equal? Blah. I go with the second one but I want a way of saying that a review is just completely vacuous and worthless "to anyone but the reviewer's soul mate" as someone put it on another thread. Yes, "Great!" is a "review," but it sucks. People clearly felt they needed a way to express that. I don't think there are enough people giving thumbs up to make that sufficient in itself, and I don't think there are enough reviews for most books to make any of this matter much at all.

If there are only five reviews on a page, they don't need to be sorted anyway. I don't see the point of allowing negativity just for the sake of negativity. So someone wrote a review that's worthless to anyone but their soulmate--so what? Send them a private comment if you don't like it, and see what kind of reaction you get.

I understand the reasoning behind the thumbs-ups: it makes sense to bring the best reviews to the top. But if there are no good reviews, why do you need to tell someone that their review is crap? It wouldn't make the work page at all better, and would just make someone upset.

I give a thumbs up to any review I see as helpful, especially if it's the only review that a book has at the moment. I consider it as a way to say thank you for taking the time to write something.

I'm fairly new here.I have written only three reviews but am beginning to notice flags on quite a number. Am I in danger of being flagged as "not a review" if I were unable to write more than "Excellent" or "Boring" for example? Or for a single sentence?

#55

From the three reviews you have done I would say that none are in any danger of flagging. Personally I don't like seeing reviews which are much shorter than your review of By A Slow River but I don't flag them.

#55

Your reviews are fine, and I wouldn't worry about them being flagged.

If all you have to say is "Excellent" or "Boring" you may want to consider just rating the book, rather than reviewing it. Left to my own devices I'd flag those but the general opinion is so strongly against such actions that I'm not doing so in practice, lest blue-flagging go away entirely and leave us without any way to flag things like "$7.99".

I flagged my first review today. It said (in German): 'I can't believe there are no other reviews of this book.'

On what basis would you flag it? "Excellent" is a review whether you like it or not.

I consider "Excellent" to be a rating, not a review. It doesn't give a potential reader any information they could use to decide whether or not to read the book -- one person liked it, but without knowing about their tastes or criteria that's all but meaningless.

However since consensus is, as I said, so strongly against this opinion I refrain from flagging such reviews -- one-word reviews are safe from me. One-character reviews, however, I will continue to flag.

I second the flagging of one-character reviews. :)

To jump in, we plan to allow people to flag flagging—to say, "the flags on this are wrong." It's going to happen, but it's not going to happen in the next week. Sorry if this seems like me saying "patience is a virtue." My mother was fond of that one...

!

p

So, Tim, will we then be able to flag the flagged flags, if the flagging of flags is in error?

:oS

Now I'm all discombobulated.