Topic on Talk:Article feedback

There is one major problem imo: The averages are destroyed by 90% fan and hate votes: w:Special:ArticleFeedback

I would adapt the IMDB rating system.

• Let every logged in user vote (votes by anons are useless)
• But for the displayed average output only consider ratings by "trustworthy users"
• "Trustworthy user" = If user has left X votes, check if he registered over X months ago + made more than X edits regulary + voted regulary. If so, give him status "trusted". Run regular script to check users.
• Don't tell how the average is calculated or how one becomes a "considered user". Well, as much as an open source code can keep it secret. At least don't add it to the FAQ :)
• Don't take the normal average or votes by "trustworthy users", but "weighted ratings"
• Run a daily script to calculate the averages ("Thank you for voting! The rating will be updated within 24 hours" or whatever IMDB used to return after voting). The script / cron job may be a problem or optional setting for the extension.

I don't agree that votes by unregistered editors (or logged-out editors) are useless. I think that we need feedback from the 99% of our users who aren't logged in. We need to be sophisticated about how we interpret that information (e.g., possibly using that to flag an article as a likely target of spam or vandalism), but I think that information is valuable.

I assume that most of the anons are not very familiar with the WP guidelines, thus their votes for the 4 criteria are not worth much. Furthermore it prevents hate/fan votes more than counting anons. And if anon votes would really be worth so much, IMDB would let anons vote and consider every vote blindly.

I completely agree with you Subfader. Without some drastic changes taking place, this feature is useless.

Do you need to be very familiar with WP guidelines to decide whether an article is complete? Or that it's well-written?

Do you think that before you became a regular editor, that your opinion on these two points would have been completely worthless?

Maybe not for well-written, sure. But if completeness was so obvious, there wouldn't need to be guidelines for writing articles. And you gently ignores teh other 2 criteria most anons have no clue about.

You're obviously a hardcore defender of the feature and ignoring my main point.

1. My main point is adapting the IMDB rating system to improve the average stats and not if anon votes should count or not.
2. In general there is no single abuse protection.
3. If the voting system was so perfect (as you say anon votes are worth countign), then why is w:Special:ArticleFeedback so fucked up with obvious fan and hate votes?

I think the dashboard should provide checkboxes and selection boxes (such as those from Special:RecentChanges) so that it allows one to get (and create links to) different reports from the ratings collected. For example, a user accessing the dashboard could have the option to generate lists after defining options such as these:

Consider ratings from:

• [ anonymous | registered | self-identified expert | all ] users
  • users registered [ N ] [ days | weeks | months ] ago
• users who have [ rated | edited ] more than [ N ] articles [ | since his/ her account was created | in the last [ week | month ] ]
• ...

Exclude articles having:

• Less than [ N ] ratings in the last [ N ] [ hours | days | months ] (see suggestion on this thread)
• [ Less | More ] than [ N ] bytes
• ...

So, once the users know which combinations of parameters generate the better metrics about the articles, these values could be used as the default on the dashboard.

I'd also like to have options like these—but not as the only option or as the default.

Requested on bug #30334.

All these options make perfect sense, I have requested that more data be made available via the toolserver so we can start experimenting with this.

If my goal was to find the one True™ answer about the article's status, then I could build a case for banning practically everyone except myself from rating articles. You say that you'd let anyone decide whether an article is well-written, but I might say that only people who share my views about comma splices and run-on sentences should be able to determine whether an article is well-written. If I want the True™ answer, then zero readers who are not native English speakers should be permitted to rate this item, and people who claimed to be native speakers would have to pass a written grammar test first.

But that's not my goal. I actually want to know what other people believe about our articles. I want to know this not because they're smart, or right, or because I've forgotten that half of them have below-median IQ, but because I want to know what these millions of users—nearly all of whom are unregistered—believe about the article. Wikipedia's future depends on keeping readers, not on keeping a couple of high-powered editors. If the article is written at the simplistic level that I associate with 12 year olds and our readers like that, then I want and need to know that. I want them to get the information they need and want far more than I want to show off my vocabulary.

I've already got a "trusted" user assessment system. I don't need AFT to find out what someone like you thinks about an article: you're going to tell me on the talk page, either by using the separate WikiProject assessment system or by leaving a written comment. AFT is my "untrusted" user assessment system. So long as I'm not such a fool as to confuse the two, then I'm getting the best of both worlds.

This means that I'm going to interpret "well-written" as meaning something like "I understood most of the stuff on the page", not as "This page should be submitted for a literary award". I'm going to interpret "complete" as meaning something like "Whatever fact I wanted to know happened to be on the page". I'm going to interpret "Trustworthy" as meaning "The names of sources, which the user probably didn't even look at, don't sound completely stupid" rather than meaning "The sources are actually high-quality and the material is actually supported by them". I'm going to interpret "Objective" as meaning "The page doesn't sound like it was written by a complete crackpot and the information sounds factual to the reader" rather than meaning "The page truly reflects the balance of high-quality sources".

And, by the way, I see no evidence of fan or hate votes at the dashboard. It's dominated by popular pages, but that's a function of its selection algorithm: unpopular pages are never displayed on the dashboard. That a page on a controversial figure like George W. Bush gets a 2.7 for "objective" (the dashboard apparently doesn't reflect the most current figures) tells me that we've probably hit the balance about right in the article—since one gets that kind of moderate score by pleasing neither the "pro" nor the "anti" readers. If that article was rated 4 or higher, I'd expect to find obvious bias in it—bias that just happened to line up with what most of our readers believed. Apparently unlike you, I don't expect "untrusted" users to have checked their personal biases at the door. I actually expect these ratings to reflect their biases and average ignorance, and consequently I account for that when I'm interpreting the results.

I see where you're coming from but I refuse to accept this for ratings.

I think the simple criteria for a "trustworthy user" would improve the ratings and effectivly kick out the fan and hate votes.
And yes there are fan and hate votes. And you just don't want to believe it? This is reality. You sound like every person on the internet is a perfect WP user...
Of course "trustworthy users" are fans of something and also hate things. But they're surely more objective on rating an WP article than those hate and fan anons.

"That a page on a controversial figure like George W. Bush gets a 2.7 for "objective" ... tells me that we've probably hit the balance about right in the article"
Sure :D Do you really believe this??? This is far from reality! It only tells me that more haters voted for Jimmy Wales than for Bush or that Bush has more fans voting than Jimmy Wales, who is of less interest than Bush. But it doesn't tell much about the quality of the article!

Actually I think the more votes a very low rated article has, the higher the chance that the article quality is actually rather good. Think about it.

How many anons and how many logged in users voted for the Jimmy Wales article? What's the rating when votes by anons are ignored? This basic example should be much closer to reality. Just the plain fact that Jimmy Wales pops up in one of those 2 lists tells me that the rating system is not working.

Talking of balance: I bet the ratings on w:Special:ArticleFeedback have extreme peaks on both ends like this:

Ask yourself if that's how it should be and what might be causing this.

To be fair, even the IMDB rating system can't avoid these: http://www.imdb.com/title/tt0411646/ratings
But that doesn't mean the rating is well balanced as you defended the 2.7 for Bush...

Plus, if the system was really balanced, the articles listed on w:Special:ArticleFeedback should reflect the diversity of WP articles in general with 95% of unpopular topics.
But in fact there are mostly popular topics popping up in these top and bottom ranking lists. Now my only question is: WHY? Please answer this. (I assume that all kinds of articles have been voted meanwhile and if not, it still won't be "balanced" after years...)

Jimmy Wales has all ONE ratings (from 1001 voters). Why is this rating system helpful again? There's no reason why his article should be one-point voted in all categories. Very helpful indeed. MikeAllen 07:39, 4 August 2011 (UTC)

And I bet not one "trustworthy user" among them... but wait, it means the article is bad, someone hurry up and improve it!

The odd rating for the Jimmy Wales page could be related to the possible bug reported below.

I think 1001 is the total of ratings for all revisions, and 1.0 is the average of non-expired ratings. See Article feedback/FAQ#How are the averages calculated?.

The dashboard (deliberately) reports only popular articles. To get reported, the article must have been rated ten times today. A page that doesn't receive at least ten ratings within 24 hours will never appear on the dashboard. Only our most popular pages will ever get reported there.

Because of the demographics (half of Wikipedia's users are under 22, and younger people have more time and interest in rating pages than middle-age people with jobs and families to take care of), we're always going to see a distinct age skew on the dashboard. This means that most articles on the dashboard are going to reflect the interests of younger people, like the latest movie or various entertainment celebrities. This will be the case because more teenagers and young adults will bother to rate the pages, not because these subjects are more important to the whole world.

Which, if you think about it, is really no different from you: Do you want to tell us how old you were at the time of your first edit, which was about a music festival? I'm going to guess that "middle aged" is the wrong answer.

We have to be smart about how we interpret this information, but we don't have to scrap it or restrict it to "perfect" users.

Well can this feedback tool be switched off for certain projects (like the Film project) if there is consensus (and it is among the project)? We just don't see any benefit for this feedback tool for the Film Project. Sorry.

The third question in the FAQ tells you how to do it.

You would officially need "community consensus" for any given page or group of pages, rather than "small group consensus", but a moderately advertised RFC hosted at the WikiProject's talk page would presumably be sufficient for that purpose. Given that the project claims 90,000 pages, you'd also want to file a bot request.

In the meantime, I suggest that each individual member who doesn't like it simply turn it off for his whole account. That some readers are rating articles doesn't oblige you to take the least bit of notice of their ratings.

Before you go down that path, let me also add that one of the goals is to eventually figure out whether certain kinds of pages attract inappropriate ratings. If you take all the film-related pages out of the study (and I agree that they might attract more fan/hate votes than other subjects), then you would effectively prevent us from gathering data on that point.

So you need "community consensus" for switching AFT off, but where was the consensus for turning it on?

WikiProjects are just groups of editors. They have no special rights over the pages they happen to work on. A discussion among one small group of editors is not sufficient for imposing their decision on 90,000 articles. It doesn't matter what the question is: they cannot demand that all of the articles they want to work on follow the members' favorite format, that they use only the members' favorite sources, that the navboxes be the members' favorite format, or—relevantly—that the articles be placed in a particular category. WikiProjects do not own the articles within their scope. The views of four members of a WikiProject are no more important than the views of any other four editors.

You talk like a politician. The question was: "So you need "community consensus" for switching AFT off, but where was the consensus for turning it on?"

Hi Subfader and everybody,

thanks for the great feedback. There are many cases in which ratings will display this kind of polarization, not just in the case of trending topics. This is a common issue in online rating systems and what this implies, among other things, is that averages do a very poor job as indicators of the typical quality score for these articles. There are several possible ways of addressing this issue, but the polarization per se is not necessarily a "bug" in the system, it just reflects a strong diversity of opinions on the quality of the article among raters.

There are different issues that you and others have raised in the AFT threads that should not be conflated:

1. how to represent scores of articles that display highly polarized ratings compared to articles that don't
2. how to tell apart genuine ratings on the quality of an article from hate/love ratings about the topic of the article
3. how to cope with the skewedness produced by trending topics or controversial articles on ratings (assuming they attract visits from users that are substantially different from visitors of non-trending/non-controversial topics)
4. how to increase the number of observations for non-trending articles
5. how to cope with explicit gaming

Each of these issues needs a dedicated solution and we shouldn't assume, for example, that 1) necessarily follows from 2) unless this results from the data.

As for the idea of arbitrarily switching off the tool on some articles, WhatamIdoing hit the nail on the head: this won't allow us to gather data that we need to address the problems AFT is currently facing. As I said before, it's not by reducing the amount of data we collect that we can analyze and solve these problems, but by deciding what we do with this data once it's been collected (how we generate aggregate scores from raw rating data, how we use this data to rank/filter articles etc.). These are problems for which expertise from editors familiar with community dynamics would be extremely useful.

We definitely do not want to exclude anonymous ratings as the tool is designed to get a broader source of data about quality and to work as a vector of new user engagement. There are ideas we can experiment with at the moment, such as looking at rater consistency across multiple ratings, but there's definitely value in anonymous ratings and we shouldn't throw out the baby with the bath water.

One last word: there's a lot of concern expressed on this page about negative ratings, but no one seems to consider the fact that Article Feedback could (and should) be used to praise editors who do a great job. Our community currently works in a gratification void, the only kind of explicit praise (a very small part of) our community gets is via barnstars, personal messages on user talk pages and other community-centric mechanisms such as WP1.0 ratings. But shouldn't we allow readers to express gratitude to editors for their hard work? Shouldn't we let editors know, for example, that an article to which they substantially contributed was visited by 100K users every day and was consistently rated high on quality scales? I believe there are smart ways of using AFT for this purpose.

I agree with WhatAmIDoing. Frankly, feedback from "trusted Wikipedian editors" is useless since they represent an infinitesimally small percentage of the people who are actually going to use the article. From a market research perspective, it is like doing a taste test for a new line of microwaveable lunches and instead of asking the office-cubicle crowds that are going to be using your product, you only ask "trusted users" like professional chefs.

The real value is from the actual users of the articles, the drive by anon IPs who came to the page via a google search or blog link. Sure, any open ratings system will have crap responses but you have to take the chafe with the wheat. But above all it is the readers, not the editors who shape public opinion about the accuracy and usefulness of Wikipedia. THESE are the people that you want to reach and you're not going to reach them by 1.) asking them to create an account before ratings or 2.) hiding the ratings system on the talk page where most readers don't even go to anyways

Besides, if anyone wants to know what the "trusted users" of Wiki-Illuminati think about an article it can always be submitted to FAC. But that will still tell you jack squat about what the majority of Wikipedia readers think about it.

You're right but this is no market research! We don't want to sell any articles.

To stay up with that example we don't ask the average John Smith if he likes the new product, but we also ask questions that only invloved people can fairly judge. If John Smith never read the WP guideline for the completeness of articles, how can he judge that criteria?

In your example: "Is this new coffee machine trustworthy?" Answer: "Totally. The coffee is hot. 5/5!"

The principles outlined by the anon still apply. The opinions of the elite (like you) do not tell us anything about our audience's view of the product. They give us a specialist's view, which is useful for identifying (for example) whether it complies with certain, specific, technical standards, but they do not tell us what our audience thinks.

John Smith doesn't need to read "the WP guideline for the completeness of articles" (which, BTW, is a poor example, because no such guideline exists) to provide useful and accurate information to us about his opinion of whether the article is complete according to a real-world, dictionary-definition kind of completeness.

You seem to be stuck in the wikibubble. The tool is trying to use plain English to get normal-person, real-world responses. It is not trying to correlate perfectly to wikijargon and the quirky details of official policies and guidelines.

Subfader, you may not be aware of it but market research deals with FAR, FAR more than just selling products. Pretty much anything that involves getting people's opinion on the matter has its root in market research principles.

But to go with your coffee example, you could gleam a lot from such a glib answer to the question of whether the machine is trustworthy. If you see a pattern of 5/5 stars given simply on the basis of the coffee being hot than you obviously know that "hotness" is intimately equated in the consumer's mind with a trustworthy machine. So you better believe that coffee machine engineers would focus on that angle.

Similarly, if we see a pattern of "trustworthiness" for articles with only news article internet sources (versus offline sources or academic journals) being consistently rated low, we may be able to gleam that readers put more faith in articles that appear to be sourced to books published by reputable publishing houses or recognizable academic journals, etc.

The above is just a hypothetical and again, patterns are what we are looking for. But you can't see any patterns unless you start asking the questions.

You would't know he likes coffee hot because you only ask him to rate X stars out of 5...

LOL! Well I assumed we were including your comment but okay, I'll play that game ;)

Even if we don't know that the "hot" coffeemaker was getting 5/5 stars because it is always makes hot coffee, the coffeemaker is still fulfilling the user's expectations even though the exact reason is not detailed. Ergo a good thing.

But when you get several hundred results over many different coffeemakers eventually patterns emerge where you will see that the "hot" coffeemakers trending with higher trustworthy ratings than coffeemakers who don't consistently make their coffees as hot. This is the parallel I drew above with looking at the pattern of how Wikipedia articles are rated.

As a member of the Wine Project on en.wiki, I'm pretty familiar with all the high traffic wine articles and how they are formatted and reference. While the majority are referenced to offline book written by credible wine experts we have a few that utilize various websites and wine magazines. If I see the former receive consistently 5/5 trustworthy ratings while the later are getting 2-3/5 then I can start seeing a pattern that tells me either A.) The sources that we are using are not reliable or B.) The sources that we are using are not perceived to be reliable and maybe I need to take a closer look.

But again, you can't start seeing patterns until you start asking questions and getting results.

I agree that this is a big problem with the tool. An article I wrote had around 80 ratings, with all the parameters highly rated. One evening in London, an activist group held a meeting that was related to the topic of the article. One of them arrived on the talk page during the meeting to change one minor aspect of it, in accordance with the meeting's views. Several Wikipedians resisted the change. We watched as the ratings suddenly fell, as members of the activist group expressed their displeasure at not being allowed to make that one change. Now, a good and neutral article is rated as biased, with only moderately trustworthy sources, when in fact it's as neutral as we can make it, and uses the best sources available. This kind of activism seriously reduces the tool's value. SlimVirgin 21:16, 17 August 2011 (UTC)

How frustrating for you to feel like they were taking their revenge that way. I think that we can reasonably expect that to be a rare occurrence, however, since most POV pushers are working "solo". Fortunately, their ratings are only temporary: Thirty edits from now, their biased ratings will be out of the calculation, and you'll be back to your usual group of readers.

But it's not "rare". I see it happening all the time. Maybe not specifically for "revenge" per say, but my area of "expertise" (if you can call it that) is actors (more specifically, former and current child stars). My dedication to the subject of child stars arose from the level of vandalism I found on younger stars' pages, so I took it upon myself to become one of the few editors dedicated to monitoring the pages most of the trusted adult Wikipedia editors, understandably, don't know anything about and/or couldn't care less about. As others have said - I've seen an extremely high number of fan/hate "votes" that have absolutely no reflection whatsoever on the merits of an article. Granted, this occurs more often on "celebrity" biography pages than on articles about astrophysicists, but there are a lot of highly trafficked "celebrity" Wikipedia pages, so the idea that this is "rare" doesn't hold water with me.

On the subject of IMDb's ratings system, I have to agree that if we are to keep the tool, then we could take a cue from IMDb. #1 - I've noticed IMDb does NOT allow "voting" on pages about people. #2 - On the pages where they DO allow "voting", the votes are invisible until there are at LEAST 10 votes. If people are absolutely dead-set on keeping this tool (which I've found completely useless on most of the "celebrity" pages I'm watching), then I would suggest following IMDb's lead on some of their guidelines.

But in the grand scheme of Wikipedia's 3 million + English articles, the small scope of celebrity articles that may get "silly ratings" is still relatively rare and infinitesimally small.

Useless extension, useless discussion. The AFT mafia ignores all of the critics anyway. Epic fail.

I fail to see the constructiveness in characterizing any editor who happens to disagree you as part of a "Mafia". I have no connection to the AFT tool outside of being a regular Wikipedia editor who is excited about the potential usefulness of this tool.

Fan votes/hate votes are not the problem SlimVirgin is describing. She's describing a one-time problem with an organized group of activists trashing the ratings on an article specifically because the specific POV-pushing change that they approved at an face-to-face meeting in London was rejected by Wikipedia editors as violating policies. This is not the individual fan expressing an opinion: this was an organized campaign to harm the article, and, when that failed, to "punish" Wikipedia for not accepting the POV pushing.

I seriously doubt that this is a common problem for any kind of article, simply because so few organized, non-Wikipedia-related groups bother to vote on what changes to make in a Wikipedia article. If you've ever seen a similar situation, please let us know the name of the article.

I guess you missed the point of what I said. I think I made it clear that the "revenge" circumstance was possibly unique, but the outcome of 12-year-old girls voting down someone they don't like or voting up someone they DO like is essentially the same. They're voting on the SUBJECT of the article, and not on the merits of the article itself. And I'm not buying this idea that the "celebrity" bio articles on Wikipedia represent an "infinitesimally small" portion of the daily traffic on Wikipedia. I would think anyone who has taken a look at the Article Feedback Dashboard would see celeb bio pages make up at LEAST 50% of average daily traffic, with articles like Floyd Mayweather Jr (rated B-Class), Justin Bieber (rated B-Class) and Adolph Hitler (once rated GA-Class, now one would assume B-Class), all pages that are LOCKED from unverified users, by the way (presumably from too much vandalism), getting some of the LOWEST ratings for the day??? What a "coincidence"... Celebrity pages obviously make up a HUGE portion of the daily traffic on Wikipedia. Anyone who tries to pretend like they don't is in serious denial. Like it or not, we live in a celebrity culture. I guarantee you Charlie Sheen will get 10x more hits today than this year's Nobel Peace Prize winner.

By the way - I just noticed the #1 HIGHEST rated page today is the Liberty Bible Academy that boasts a grand total enrollment of 155 students (grades PreK-8). Hmmmmm... A "Start-Class" article (at best) with two broken reflinks, rated as providing "Comprehensive Coverage" with "Great Reputable Sources".. Is there anyone who honestly believes that's not a ratings flash-mob descending on the page??? Gimme a break. The ratings are a big joke and basically exist as a new (and now sanctioned) way for anonymous users to vandalize pages.

You might like to look at the list of 1000 most visited articles. From my quick scan, the first "celebrity" bio is Justin Bieber at #29, meaning that 28 other pages are more popular at the moment. "Most rated" and "most viewed" are not the same.

Furthermore, the dashboard doesn't show "most rated". It shows "most extreme ratings". It is deliberately designed to help us find strange ratings. Basing your view of the whole program on this is like believing that America's Funniest Home Videos shows normal, everyday American life.

As for the existence of ratings that editors disagree with: Fan votes and hate votes are not vandalism. They are also harmless. You're not going to screw up a decent article to get a better rating, right? No experienced editor is going to do that. So why do you care what the rating is at any given moment?

Basing your view of the whole AFT feedback here and on WP is like believing that everyone who did not post negative feedback yet accepts and believes in the usefulness of AFT.

Yes, I have viewed the list of most trafficked pages. But since we're talking about the rating tool, I thought the most rated numbers were the most relevant to the conversation. What looking at the top 30 most viewed articles does NOT show is the AVERAGE rate of views on the other 3+ million pages that another user previously cited. From my experiences on Wikipedia I would bet that if you average ALL daily views, at least 50% would be celebrity pages - throwing out some arbitrary .000001% sample doesn't show anything. In fact, my concern is the articles that do NOT show up in the "most rated" results - where kids are vandalizing pages using the ratings tool (yes, I said VANDALIZING, deal with it) and the ratings stay up for MONTHS, because the pages are locked (due to excessive vandalism), so they're not being edited every day, or even every week.

"So why do you care what the rating is at any given moment?"

Thank you!! You just made my point. What is the point of having a ratings system if the working philosophy is "Why should we care what the ratings are?". That's EXACTLY what I and others have been saying!! A large portion of the ratings are meaningless so what's the point of having them? And before you try and cook up ANOTHER condescending answer - It's a rhetorical question, so you can save yourself the trouble. I thought the point of this page was for users to give FEEDBACK about how the tool was working (or, in this case, NOT working), not to have one or two users repeatedly try and "sell" the tool to us every time anyone disagrees with you.

As I said in my original post - I'm coming to this page from a segment of Wikipedia where the rate of fan/hate votes are extremely high - specifically, current and former child stars - From previous experience, I'm well aware that most Wikipedia editors couldn't care less about the young Disney / Nickelodeon stars' pages (where the average age of readers ranges from 8 to 14, and, as a result, the highest percentages of mischievous edits and ratings per user occur), so I thought it might be useful to give some feedback from the perspective of those pages. Obviously there are people here who have trouble reading, but in my original post I clearly stated that if the tool were to be kept I would suggest following IMDb's format - NOT using ratings on bio pages (where vandalism voting is most likely to occur), and NOT showing ratings until there are at least 10 votes to allow at least SOME chance of there being something useful to discern from the results. Again, I thought this page was for users from different areas of Wikipedia to give feedback about their experiences with the tool - not to be subjected to obnoxious and condescending feedback on our feedback. This was clearly a mistake on my part. It won't happen again.

Editors can remove the tool from any set of pages (e.g. bio pages) by adding them (or a template which is transcluded on them) to the black list (as someone did for the wiktionary redirects).

As for the specific suggestions such as "NOT showing ratings until there are at least 10 votes" and others in the ideas log, I think they should likely be requested on bugzilla before it get lost on the threads of this talk page, so it will be on the list of open AFT requests and may be acted on once we have developers for the next-generation of features for the Article Feedback tool.

Thanks, that's a good idea. I didn't know about those pages. I thought this was where we were supposed to give our feedback, but it's good to know there's a more organized way to share our problems/ideas on the subject. I'm glad there's somebody interested in actually giving helpful advice instead of just trying to convince everyone who comes here that we don't know what we're talking about.

What is the point of having a ratings system if the working philosophy is "Why should we care what the ratings are?"

The goals of the tool are explained in the FAQ. Have you read it?

Yes, I've read it. The goal was to get readers interested in editing. I've also read your own admission that (paraphrasing) only the biggest morons visiting Wikipedia would give any weight to the ratings. You've repeatedly undermined your own defense of the tool. If the purpose of the tool is to lure only the biggest idiots to edit Wikipedia - then I agree with you 100% - I'm sure it'll be a smashing success.

By the way - I'm STILL waiting for one of the "experts" on here who claims to LOVE this tool's ability to cultivate well-written articles, to link to something you've actually written...........*crickets*...........yeah, I didn't think so.

w:en:Champagne (wine region) w:en:Cabernet_Sauvignon w:en:Chardonnay w:en:Sangiovese w:en:South_African_wine w:en:Barolo w:en:History_of_French_wine w:en:History_of_Champagne w:en:Oak (wine) w:en:Ripeness_in_viticulture w:en:Ancient_Rome_and_wine

....are a few of the articles that I've written.

And, yes, when the AFT Dashboard becomes something that can be sorted by categories and/or Wikiproject tags, I highly look forward to being able to use this tool to identify wine articles that consistently receive low scores so that I can work on improving them.

Vicious circles alert!!!

• Improving articles is not the purpose of this tool. We were advised to read the FAQ several times, so should you do.
  • Even if that was the purpose, you would only find abusive fake / fan / hate votes like now. Go improve Bush and Justin Biber and all the oher stuff popping up there. Go improve! According to AFT these are super bad articles.
    • Even if improving WP was the purpose, AFT would need to work properly. But it doesn't. If it would, you would face another problem: Only a few percent of WP articles are quality. There are not enough editors to improve all those and AFT can not create new editors with abusive ratings.

More like a straw man alert.

Improving articles is, indeed, not the primary and stated purpose of the tool. Using a index card to squash the mosquito that landed on my desk and wipe it into the trash is not the primary and state purpose of index cards but...like most tools, it have its multi-tasking moments.

However, like any tool, it does require some common sense to use (Like for instance, while index cards are great bug killers, they probably wouldn't be very helpful for stuffing your bra). So I may see an article like w:en:Wine_tasting get relatively low scores, especially when it comes to completeness and objectivity. Now,using common sense, I can probably assume that these are not "fan/hater" ratings for a topic like Wine tasting (though I may wonder about that for an article like White Zinfandel perhaps).

I'll look at the laundry list of "further readings" and the uneven distribution of refs and figure that may have a role in the reader not having much confidence in the objectivity of the text and I could see that the several tiny sections on glassware, wine color, etc probably could use more info to help with completeness.

Eventually an editor like myself will come across this article and see these things that need to be improve. Indeed, it has been tagged with needing ref improvement since June 2010. However, I've got a lot of articles on my list that need to be worked on and this may not be a priority.

BUT...if my wikiproject was doing a dedicated article improvement drive and we had a tool like the AFT dashboard that we could see a cumulative rating of wine articles for the month, we could see that perhaps that Wine Tasting was one of our lowest rated articles for the last month and then use this information to readjust our priority list.

Sorry, this thread is like a maze to navigate with all the hidden replies, etc, so I didn't even see this until just now.

Nice try pretending to take credit for those articles, but I checked and they've CLEARLY been edited by HUNDREDS of people. Fixing a typo, or removing a dead reflink isn't exactly what I call "writing". I would expect someone who actually "wrote" an article on Wikipedia to be able to link to their Wikipedia USERNAME and the revision history page that clearly showed where they took a page from nothing to something. Nice try, but EPIC FAIL.

As far as asking to see someone's work who characterizes others feedback here as "infinitesimal". I have to disagree that it's a "pissing contest". The article a person has written demonstrates for EVERYONE here you're level of dedication and "expertise" when we evaluate your snotty little "opinions" about our opinions. Unfortunately you've failed to show anyone here a page you've written, but the articles that you TRIED to take credit for DO tell a lot about you. I suggest you put down the bottle of wine and take a look around at articles that aren't the #1 favorites of alcoholics across the English speaking world. Talk about "infinitesimal" - I guarantee you Justin Bieber gets more hits per day than ALL the Wine Wikiproject pages put together.

Lol...yeah, you've definitely have given up on trying to support your contentions and make some valid points. Now it's just petty mudslinging to see if anything sticks.

You've also demonstrated some deficiency in knowing how to read an article history page because if you read one you would see that 90%+ of those articles were written by one user---Agne27 which is who I am "claiming to be". But, whatever, if you don't want to deal with disagreeing opinions and have civil discourse on how to improve the tool then I wish you the very best in your wiki endeavors and hope that Justin Bieber never opens a winery so that I'll have to deal with that article.

I think the truth just came out. Albeit most likely out of some alcoholic haze, but the truth nonetheless.. You PERSONALLY don't care about the celebrity articles, so you wish everyone else would simply dismiss them as "infinitesimal", no matter how blatantly ridiculous your worn-out 1% assertion is..

You've also once again demonstrated some serious deficiency in your reading abilities. NOWHERE did you post your username, and NOWHERE did you post to the revision pages where you took these pathetic little "stubs" to sterling *GA* articles. Aside from lacking some crucial mental faculties (perhaps a long-term side effect of your apparent love affair with alcohol), you expect everyone here to sift through hundreds of edits on the edit history pages where you're trying to take credit for other users' work.

Wow....you really are grasping at straws. Though I guess when your attempts at making any sound argument against AFT collapsed so ingloriously, what else do you got but incivility, naming calling and climbing the Reichstag?

But hey...if that is how you roll, Crakkerjakk, so be it.

You sound like a teenager, that's why nobody wants to argue with you ;)

Eh.... given Crakkerjakk's weird association between wine editing=alcoholism and his reduction to just name calling and taunting, it would be uncivil and patronizing to make cracks about his age. We can just let his comments on the page speak for themselves.

Oh, wait, did you want to direct that ad hominem at me? Oh well.

What you fail to understand is that when the numerous people who have come here have "given up" trying to explain things to you, it's not because they lacked the ability to make sound arguments - It's because you've repeatedly demonstrated a complete lack of ability to grasp even the simplest of concepts: Remedial math, Remedial English, etc, etc.. You didn't "back them into a corner" or "win" any arguments - they simply grew tired of you and left.

I have to give you credit though. You've managed to achieve what I would have thought to have been the impossible. You've made WhatamIdoing look like a rocket scientist next to you. WhatamIdoing is just your average, garden-variety, self-appointed know-it-all (Wikipedia is filled with them). You, on the other hand, are a bona fide troll, as you've made painfully obvious to everyone who's visited the page over the last month. It all just makes a lot more sense now that this apparent "fixation" you seem to have with your alcohol has been exposed.

Crakker,

If you're over 21 and are ever in Seattle, do consider stopping by Woodinville Wine Country and maybe even drop me a note on my talk page. w:en:User_talk:Agne27

I'll you introduce to the basics of wine tasting-sniff, swirl and spit--yes, spit :) You'll probably be surprise at what wine appreciation is really about.

Lol.. I live so close to the Napa Valley a desperate wino like you would probably cream your pants, but I'm proud to say I've never touched a single drop of it. I know misery loves company, but I'm sorry to have to say that sloppy alcoholics make me sick.

Actually Napa is fairly overrated. Everyone wants to make a $100+ "Cult wine". It's all about money and prestige and not the beauty of how man and nature can take a simple little grape and turn it into something poetic.

Sonoma, on the other hand, is awesome! Russian River Pinots, Alexander Valley sparkling wine, Rockpile and Dry Creek Zins, Knights Valley Cabs...ah, heaven!