Reminder: Use of Flags on Reviews

To all: please note the wording that appears when reviews are flagged:

Important: Do not abuse this feature. "Not a review" is not intended for short reviews, poorly-written reviews or reviews you disagree with. Links to the reviewer's blog are also acceptable. "Abuse of terms of service" includes violations such as spam reviews and posting copyrighted material without permission. See the terms of service for more guidelines.

In particular, please note that per our Terms of Use, reviews consisting solely of links to the reviewer's blog are allowed and are not blue-flaggable.

Please note the important exception to this rule: LibraryThing Early Reviewers reviews, as per the LTER Terms:

"A book will be considered "reviewed" if a review is posted to LibraryThing and is composed of at least 25 words. Reviews may include but not solely consist of a URL."

In this case, bloggers and others must post some text (the whole thing, just the first paragraph, some substantive portion) of their reviews onto LT, although as always they are welcome and encouraged to also post it to their own blogs, &c.

Are you saying we should blue-flag LTER reviews that are acceptable for the site in general but just don't count for LTER?

I suspect it's more to NOT flag reviews that are links only. Early Reviewer reviews are the only place the text is required.

I think what he's saying is that it's OK to flag LTER reviews that don't meet the 25-word threshold (including LTER reviews that consist only of a link), though I don't see the message as necessarily encouraging flagging (i.e. saying we should flag).

He's also saying -- and I think this is the main point -- that it's not ok to blue-flag short or link-only reviews on normal, non-LTER books/reviews.

It sounds like this message is responding to a problem of people misusing flags.

4 > I can't imagine that it would be OK to flag an LTER review missing text. It's program specific and the reviewer will get dinged in the algorithm for excluding text, thus being self correcting.

>3 readafew:/4 - exactly
>5 readafew: - It is okay to flag ER reviews missing text, or in which the text is not actually a review.

>6 jbd1: So, what's the point of blue flagging those ER reviews? To get the review to sort to the bottom, or to notify the reviewer in case they happen to look back at their reviews?

6: jbd1: I know link-only general reviews are allowable, but I've never understood why. To me they provide no benefit to LibraryThing; they're just ads for someone's blog. I never click through on them, and judging by past comments many (most?) other people don't either.

I raised the issue about the ER reviews here:

http://www.librarything.com/topic/116038

Adding the link so folks can see the discussion.

Maybe adding the exception about the ER reviews to the bold text copied above on the flagging dropdown would be helpful.

Did we ever get any follow up though on the commercial and/or affiliate links within reviews? Yes, Jeremy, I'm bugging you about that again. :)

>7 _Zoe_: - I'm not suggesting that anyone go around looking for these things and and flag them, I'm saying that in those cases, it is acceptable to do so. Anyone can or cannot blue-flag non-review LTER reviews as they see fit. They may not flag link-only non-LTER reviews. The point is the same as any flagging of reviews. I don't generally do it myself unless it's a particularly egregious example, but I know some feel strongly about it.

>8 gwernin: - I don't click them either. But they're okay as per the Terms of Use, and Tim continues to feel that they're acceptable, so they're allowed.

>9 theapparatus: - I've not crossed that bridge yet, no. It's a fine line there and frankly not one I think we want to spend a great deal of time trying to make judgments about (what links are allowed in reviews and which aren't). There are a finite number of hours in the day.

I have a question regarding the linking to an outside source review:

- A group of bloggers has two LT accounts. They link to their reviews, but don't necessarily list their blog in their LT profile. Its multiple bloggers, not one person. What if people have flagged these reviews?

- Add to this issue, they link to the same review for both accounts for the same book. How does one deal with this?

9: Did we ever get any follow up though on the commercial and/or affiliate links within reviews?
11: It's a fine line there and frankly not one I think we want to spend a great deal of time trying to make judgments about (what links are allowed in reviews and which aren't). There are a finite number of hours in the day.

I agree with Jeremy that it's probably not worth the effort to police or define rules for individual reviews (how many clicks could an individual review get anyway). I could imagine someone abusing this, and posting just to get the affiliate links, but that seems a rare case, and one best decided case by case.

Slightly OT - I did, however, once find that someone had edited a "quick link" (for amazon or a similar site) to add her affiliate code. That's a link lots of people click on, and I thought it was a very shady thing to do. I changed the link back, of course, to get rid of the affiliate link.

>but that seems a rare case

I keep finding these "rare cases" though. What's annoying is, no, I'm not going looking for them. I run down the Zeitgeist -> Reviews -> Page 2 list once a day and find these things.

Re: OT, I've found those as well. I'm sure you realize that you can pull up the history to find out who did them.

And I just found the user posting the outside link to their reviews in question. I have to admit that I wouldn't follow such a link either. I'm more concerned though that the user in question is posting comments consisting of "DIE HIDEOUSLY" on other people's profiles.

edit: The poster is also getting around the 25 word requirement also by including 'Edit: added to meet the following requirement: "A book will be considered "reviewed" if a review is posted to LibraryThing and is comprised of at least 25 words."' in their reviews. I think that's bending the rules slightly but if they're still getting ER books, it must be acceptable.

Yes, I monitor quick links regularly through the zeitgeist/helpers log (there's no history for quicklinks beyond that). However, that page doesn't necessarily indicate who edited a quick link. It always lists the name of the first person who added it; if someone edited a link that was already there, it will display the name of the person who initially added it, not the person who edited it.

On the reviews thing: Can you post an example of one of the cases you're talking about of affiliate links in reviews? Why do you run down the zeitgeist>more reviews page?

edit: The poster is also getting around the 25 word requirement also by including 'Edit: added to meet the following requirement: "A book will be considered "reviewed" if a review is posted to LibraryThing and is comprised of at least 25 words."' in their reviews. I think that's bending the rules slightly but if they're still getting ER books, it must be acceptable.

That's clearly not acceptable, and I think I've read elsewhere that LTER word limits also take into account content. You couldn't post "didn't like it" 7 times just to get over the word limit.

>14 theapparatus: - See the first post, where I note of LTER reviews "In this case, bloggers and others must post some text (the whole thing, just the first paragraph, some substantive portion) of their reviews onto LT, although as always they are welcome and encouraged to also post it to their own blogs, &c." Straight links, or straight links with filler text, are not consistent with LTER rules.

I was thinking authors. Never mind. One of those 40 tabs open in the browser days.

You'll have to forgive me for not posting example links. The last time I did such a thing, I received quite a lot of negativity from both other users and staff for doing so.

What has me concerned is the continous red-flagging of this review - I think Jeremy lifted it once or twice and the flag is back.
http://www.librarything.com/work/10714719/reviews

I messaged the reviewer and she was trying to link to the author's website as a way of getting more information.

Is this not allowed?

>19 fdholt: - I don't see any reason that review should be flagged. Linking to a relevant outside website seems entirely reasonable to me.

All this confusion is a darn good reason why we should get rid of all links in reviews. I know others will disagree with me (Heck, I'll be the first to say some links are fine, I've done one or two in my reviews) but there are so many gray areas and confusion in what's allowed, what's not and when the rules should be applied.

20 - Ditto. I think people need to be more cautious in flagging, and actually check out the user's profile to see if the person seems to show some intent to use LT as an advertising mechanism, and give people the benefit of the doubt. In this case, there's absolutely no sign that the user is doing so (advertising I mean).

True spam and true author or publisher promotion are almost always VERY easy to spot; it just requires a quick check of the profile and the pattern of reviews. (The only really difficult cases are the ones where authors or publishers create sockpuppets, and those require a lot more investigation and staff intervention rather than the flagging route.)

jbd1, looks like users here disagree with you. Take a look at the user's review history and all of their removed reviews:

http://www.librarything.com/profile_reviews.php?view=audreyl1969

Most of the removed reviews include a link to ezinearticles articles.

The first thing that comes to my mind with those links is that they're actually landing page for affiliate accounts to but those books. Looking at the whois, it appears that this is not the case. For example:

http://whois.domaintools.com/matthewgollub.com

The examples that I looked at, they're registered to different people and scattered around the net.

In this case, it does look legit. edit: Well, the more recent ones. I know in our antispam database, anything with an ezinearticles article link currently scores a 11.7 on a 5 point scale for spam. (Which means it's automatically tagged as spam.)

Actually, in this particular case, it looks like a few legit reviews might have got mixed up with a bunch of not legit reviews. The hidden reviews look to be copyright violations (and flagged properly for that reason, not for the reason that she included links). It's possible some people, once seeing so many copyright-violating reviews, accidentally flagged some legitimate reviews while going down the list.

Edited to add: the problem is not that they include a link to ezinearticles, but that they reproduce ezinearticles, against copyright.

Hmmm, might be the case.

I didn't go through them all but I believe the ezinearticles were all done by the same person. I was thinking that it was in this case the user here as well.

The more recent reviews do not include such a link. If they are intermixed, I;m not seeing it but I didn't click on every link to check.

You know, much as I'm not a fan of over-flagging, I wonder if it might be better to have a separate flag just for copyright violations. One problem now is that if something's flagged, you have no idea why. Often, it's not that hard to figure out, and of course we can ask in this group. I wonder if it would be useful to have a separate category for copyright violating reviews, which would also mean that they could be dealt with differently (hidden with an explanation of copyright, a message sent to the member, etc.). Just a thought.

25 - No, the reviews are by various people, none of them -- in the reviews I've looked at so far -- the member herself.

26 > isn't that the primary use of the red flag?

If they are intermixed, I;m not seeing it but I didn't click on every link to check.

What I mean is that, since the vast majority of this member's reviews are copyright violations, it's not much of a stretch to imagine that someone, having seen dozens of copyright violating reviews, would just go down the whole list of reviews red-flagging most or all the reviews (and stop checking individual ones), and thus accidentally flag some legitimate, non-copyright-violating reviews in the process.

>I wonder if it might be better to have a separate flag just for copyright violations.

We were told previously to take that up with Jeremy directly if we saw a pattern I believe. He has stated a couple of times though he is limited on time. I say that in his defense and not as a negative.

*sigh* I'm not getting anything done today.

I believe I flagged a review for a work that had five reviews, all by the same user, all same link. The linked blog entry consisted of something like 'read'.

I do not remember which work this was on, though I think I might have flagged only one of the reviews.

I disliked actually deciding to read a review, finding 1 link five times as the only 'reviews', and actually clicking on link to find ... nothing much. So I flagged it. I won't do it again.

> 8

they're just ads for someone's blog

Not necessarily. It's often just easier for a person who reviews on multiple forums to use a link, rather than cut-and-paste the entire review multiple times, particularly when some forums allow html or images and others don't.

...And if they do allow images and html, some do it in different formats like bbcode while others do it with normal html.

re> 14: I got one of those "DIE HIDEOUSLY" threats on my home page, and only for telling a member why others were red-flagging his false ER reviews. It's on my page if anyone cares to see it. I hope this is not considered acceptable behavior here

No, it's not acceptable behavior, and the member has been warned; that was the whole impetus behind the discussions Tim and I had which resulted in this post.

34> Yup, that was the one that I was looking at.

>10 jbd1:
I'm not suggesting that anyone go around looking for these things and and flag them, I'm saying that in those cases, it is acceptable to do so. Anyone can or cannot blue-flag non-review LTER reviews as they see fit. They may not flag link-only non-LTER reviews.

Is there some type of obvious notice whether a review is for LTER? I know I've counter-flagged link only or short reviews. Maybe the reason for the flag was that it doesn't meet LTER requirements and I shouldn't have challenged the flag.

I just ran across an author with this review of his own book:

http://www.librarything.com/work/11213805

Is this flaggable? At least he didn't rate it - most authors rate their own books as 5 star.

At least he didn't rate it - most authors rate their own books as 5 star.
He has now.

If I remember from past discussions, an author reviewing her/his own books is not, in itself, flaggable. (One author weighed in once to say that he'd actually given either a critical review or a less than 5-star rating evaluating his own work, so it is possible that author-self-reviews can be done in good faith.) If the author crosses the line to clear selling, then it is. This one seems borderline to me, not clearly flaggable. Not much of a review, though.

There were plans, at one point, to make authors' reviews more transparent, by having them marked as such. What ever happened that?

ETA: Here's a line from the rules for authors: "While you can review your own book, be aware that the review will be marked as the author's."

38>

Tim has ruled that it's acceptable for authors to review their own books. At the time he dismissed our concerns with "Oh, I'll put some sort of indicator on them", but that has obviously never happened.

38/39/40 Considering that the author's account was created just to review their book as well as promote their website on their profile page, I'm almost tempted to click on that Report for Spam link. There's little there that differentiates them from the common spammer around here.

edit: As an aside, the user profile needs one of those LT Author flags.

reedit: Us Dr. Mike's have to stay ethical after all. :)

Authors get the LT Author flag by requesting LT Author status.

There's little there that differentiates them from the common spammer around here.

Except that LT spam rules on profile flagging distinguish between authors and other commercial spam. Authors promoting fall under "overzealous author," reported by email rather than the profile flag. In this case, I'd call him zealous perhaps, though not overly so. ;) There are actually very many authors on here with only their own books and a link to their publisher's book page. That in and of itself hasn't been considered spam.

Just to add to my earlier message in 38: the review is now red flagged. I thought the group decided that this was not a flaggable review. Jeremy, could we somehow unflag?

44 - Anyone can unflag. Click flag, then click "abuse, my foot!" That registers a vote against the flag, and if there are enough votes against (i.e. counter-flags) to outweigh the flags, the flag will go away. I can't remember if it's a 1:1 or a 1:2 ratio.

> 38. Frankly, this case strikes me as problematic. For an author to use this site to promote his book is one thing, but for him to do so with fake reviews of his own work is corrupt. As others have noted, the author has only one book in his LT "library" -- his own -- he gives it five stars and writes a so-called review that is nothing but a puff piece (spam). Fortunately in this case, no one is likely to be fooled, but it does make a mockery of the "review" process that others take seriously. I suppose there would be nothing to prevent a self-published author from flooding the site with spam reviews of their own book?

Authors are allowed to review their own books. The name of the user makes it obvious that the review was written by the author. LT should definitely implement a label to highlight author reviews, sure. But the author has done nothing against the TOS, even if it's somewhat beyond good taste.

If an author was to truly flood the site with spam reviews -- and authors have -- that would have to be through (a) adding and reviewing multiple copies of the work (against the TOS, I believe, especially if used to skew the database) or (b) through sockpuppets, definitely against the TOS.

One review, transparently by the author, might be a bit gauche, but it's within the rules.

(ETA: maybe the rules should be changed, and authors shouldn't be allowed to review or rate their own books. That's contestable. But staff have made clear that they are allowed to do so, currently.)

46>

It's not fake, and Tim has ruled that he does not consider it corrupt. Pretty much everyone thinks there should be an obvious indicator that a review has been written by the author, but this has not been a high priority for the staff to create.

I suppose there would be nothing to prevent a self-published author from flooding the site with spam reviews of their own book?

Not as the rules stand; it has been ruled that (a), it is acceptable for authors to review their own books, and (b), it is acceptable for people to review the same book multiple times; if they tried doing so with multiple accounts, that would probably fall under sock-puppet rules (though IIRC the current sock-puppet prohibitions only apply to Talk), but if they're all on one account it falls within the rules.

though IIRC the current sock-puppet prohibitions only apply to Talk
No, several authors have tried doing this in order to add, rate, review, and recommend books -- including one author who has been banned for life as a result. Tim has come down very hard on these cases, making it very clear that such sockpuppets are against the terms, and deleting and suspending accounts.

It's true that multiple reviews are allowed, but there used to be an explicit rule in the terms (before they were revised) that said this: "Users may not use LibraryThing for unfair promotion. For example, authors may not add 500 copies of their book to LibraryThing in order to become the top book, even if they actually have 500 remaindered copies in their basement." In the current version, I think that would simply fall under "Do not subvert or misuse site features." If it were 2-3 reviews of the same book that might be one thing. If it were truly a flood of reviews, I'd venture staff would consider that misuse, though. 500 reviews by the same named member, though, would be very obvious, and isn't a very likely strategy for authors promoting their books; creating sockpuppets to do so has actually been relatively common.

Any comments on my questions in post #12?