Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Blacken]

As far as I can tell, this rating isn't for "people who just want to find out how wells work."  It's for contributors who are looking for articles that need improvement, and to them the labels are immaterial.  This fact is reflected by how unobtrusive the ratings are -- the people you're talking about aren't likely to notice them at all.

"Aren't likely"? They are a conventional tool in Mediawiki systems and are the barometer for whether somebody who actually knows how a wiki works should bother with the article. Unless you have demonstrable metrics to say they "aren't likely" to be looked at by end users, I think you can't make that statement.

[Footkerchief]

They should allow the user to take a glance at how good the article is considered to be.

Why?  To let the user know that the information may not be accurate?  Any warnings of that kind should be much more visible than the ratings are, like the big "This article needs cleanup" warnings on WP.

As far as I can tell, this rating isn't for "people who just want to find out how wells work."  It's for contributors who are looking for articles that need improvement, and to them the labels are immaterial.  This fact is reflected by how unobtrusive the ratings are -- the people you're talking about aren't likely to notice them at all.

"Aren't likely"? They are a conventional tool in Mediawiki systems and are the barometer for whether somebody who actually knows how a wiki works should bother with the article. Unless you have demonstrable metrics to say they "aren't likely" to be looked at by end users, I think you can't make that statement.

What does "somebody who actually knows how a wiki works" have to do with "people who just want to find out how wells work"?  And is it conventional for the ratings you're describing to be squirreled away in the top right corner, with small text and delicate colors?  To me, it looks like it's designed to be mostly ignored except by someone who's looking for it.  If there's a metric for calculating the objective visibility of HTML elements, I'd be happy to apply it here.

If you want to argue that we should have highly visible ratings for the reliability of an article, then I definitely see the merit of that, but I think that's somewhat orthogonal with the current rating system.

[G-Flex]

Footkerchief, quality measurements and templates are useful for everyone. If people think an article has problems, I want to know that before reading it, not just before editing it.

It's not just about inaccuracy either; it's also about incompleteness, poor style, etc.


And the current ratings are plenty visible. They're right next to the article header, in colored text. I'm not saying they should be extremely specific and descriptive, but at the very least, they shouldn't be misleading, and casual users are the ones most likely to be confused by calling a substandard article "fine-quality", because they haven't argued about it on here, read the descriptions and committed them to memory, or get whatever jokes we'd be trying to convey.

[Footkerchief]

And the current ratings are plenty visible. They're right next to the article header, in colored text. I'm not saying they should be extremely specific and descriptive, but at the very least, they shouldn't be misleading, and casual users are the ones most likely to be confused by calling a substandard article "fine-quality", because they haven't argued about it on here, read the descriptions and committed them to memory, or get whatever jokes we'd be trying to convey.

For my part, if I'm right that the ratings are just to help contributors find articles that need work, then any label that implies a value judgment should be avoided, because it'll give people the wrong idea about what the rating system is for (c.f. this thread).

Sure, I can get behind that.  It seems like the main questions now are 1) should the reliability warnings be separate from the grading, and if so, 2) should they use an implicitly numeric rating or non-numeric labels a la Wikipedia's "This article needs citations for verification"?

[Draco18s]

...

[Footkerchief]

e: nm

[SirHoneyBadger]

Wikis tend to have the ability for people to discuss the content of articles, on a separate page.

Are there good ways the "discussion page" could be better integrated and utilized for the DF Wiki, so that the specific merits/problems of a given article could be directly, in depth referenced and mulled over what have you, without directly impacting the readability of the article, itself? 

That seems like it would be more useful than any arbitrary ratings system.

[G-Flex]

I might be reading you wrong here, but aren't you referring to, say, the normal wiki templates that most wikis use (see: Any really bad article on Wikipedia)?

[Deathworks]

Hi!

First of all, I have to admit that I was very thankful that the elf-dwarf-human system got removed.

In addition, I feel that fine being the lowest quality article is actually the best way to go. Remember - no one is getting money for writing those articles. They are contributing their free time and energy for the community. And telling them that their effort was awful and rotten is not the way. If you really want a negative tag, that should only be with automatic stubs like those we had with the 2010 transfer where all the 40d articles were moved to their history section and the spaces initially remained empty. But as soon as someone makes an effort to give at least some help or information, that is something to be appreciated.

Personally, I don't think that the quality indicators are really that useful for non-editors in this particular case. After all, besides the data from the raws, everything else is just information we got from experience which is not necessarily 100% correct. In addition, no one but Toady One knows about all the hidden aspects and things that are in the game. So, even the best article has a chance of lacking interesting information. (Actually, I am trying to find the time to revise the trade article as it lacks information on varying prices based on racial preferences on decorations, sorry that I have not been able to insert that info yet).

Deathworks

[G-Flex]

They are contributing their free time and energy for the community. And telling them that their effort was awful and rotten is not the way.

And telling them that it's good no matter what is the way? If an article is problematic, then we should call it problematic; we shouldn't say that something is good when it's not. It's disingenuous for "good" to mean "bad". I mean, hell, it still is a negative label, yet it's done with a positive word. Are we afraid of hurting people's feelings or something by saying "the article has problems" instead of "the article is fine, which means that it has problems"?

Seriously, this doesn't even make any sense. At all.

[Deathworks]

Hi!

Does not the fact that it did not get a better rating already imply the problematic nature?

If an article is really containing incorrect information, should you not remove that information instead of putting a nice tag above it? And if the information is correct, then why can you not say, it is okay; not great, but okay?

Deathworks

[G-Flex]

Does not the fact that it did not get a better rating already imply the problematic nature?

No, it implies that it can be improved, not that it has problems. There is a difference. For instance, look at the actual description in the wiki.

If an article is really containing incorrect information, should you not remove that information instead of putting a nice tag above it? And if the information is correct, then why can you not say, it is okay; not great, but okay?

Deathworks

Huh? Now you're just arguing about using any templates, at all. Seriously, the logic you're using extends to every single possible rating system or template you could use to judge an article. "Instead of using tags, just fix it" only works to a certain degree, because people don't have infinite time to improve articles, and if people do want to improve articles, it's nice to be able to quickly see which ones are known to need improving. Also, as far as article quality is concerned, there is much, much more than the accuracy of the information to consider. There's accuracy, there's completeness, there's the way it's presented and written, and so forth.



And there's the simple fact that we should not have to say that every article is at least "Okay", because plenty of them might not be. I seriously do not understand what is hard to grasp about this.

[SirHoneyBadger]

I feel it's more important that any "ratings system" be there for the Wiki editors, not the visitors. And that the rating itself provides more than giving the author a pat on the back.

Random visitors are stuck with whatever's there--and what's there is what's there.
There's no private party backstage where the real content hangs out. The editors are the ones responsible for creating and improving that content. By telling visitors that the content's good or bad...what goal is being achieved, exactly?

"Sorry this article is shit, but deal with it because that's all we've got."
It's not like there's going to ever be a million similar articles on fishing to choose the best from.

Giving an article a gold star or a red mark, or whatever, doesn't really mean anything, if that rating doesn't also inform. Either: what needs to be fixed, or what can usefully be said about an excellent article that makes it outstanding, and that can be duplicated in future articles. Visitors don't need to know any of that, necessarily, because it can all take place behind the scenes.

They're being provided a service, not a writing course.

Ofcourse, there's no real reason it couldn't be more transparent, and there's every reason to reward good effort; but whatever the ends, the goal should be better articles, not some kind of Grammy awards.

[G-Flex]

I completely disagree with your opinion that ratings and other such comments aren't useful to readers.

They are, as long as they're informative. If there's something wrong with an article I'm reading, I want to know that. That's how Wikipedia works, and pretty much any other decent wiki. If there's any sort of consensus amongst the editors that an article has some factual problems, I want to know that before I read and take everything at face value. That sort of thing can serve as a really, really good warning sign.

Yes, if they don't edit, then what they see is all they get, but they might as well get a heads up to any problems that they may encounter when reading it (which they might not even be able to guess at otherwise), especially if there's some dispute concerning the validity of the information.

None of this should have to be said. It's all standard behavior on any remotely decent wiki.


Oh, and for the record, another thing that any decent wiki does is encourage its readers to edit them. Slapping a template on a page saying "hey, some of this information might not have been verified" or "this section could use some expanding" is a great way to prod people into contributing. Editors aren't born out of the aether, they're made from users.

[Draco18s]

I completely disagree with your opinion that ratings and other such comments aren't useful to readers.

They are, as long as they're informative. If there's something wrong with an article I'm reading, I want to know that.

In an attempt to not repeat myself and have Footkerchief yell at me again, the "rating" system is one that is blatantly obvious.  A short article (eg a stub) doesn't need a giant red-and-yellow label* at the top of the page informing the reader that the following article is a stub: they can see that.  It's eight lines long, has four red links, and there's no sidebar like other similar pages have.

The rating system is merely a categorization of articles so that contributors can find articles that need improvement.

For example, a Fine Quality Article has many of the following characteristics:

    * May be a stub article
    * Lacks information
    * Has a substantial number of redlinks
    * Contains inaccurate information
    * Contains information that needs to be verified
    * Is not categorized

All of the underlined characteristics are painfully obvious to the reader.  I mean, look at the Beak Dog article.  Stub, missing information (check the sidebar for question marks)...can't find an article with redlinks, as there aren't really any new pages needed anymore (though I did just fix on on the bloated tuber page, it was trying to go to [[gathering plants]] rather than [[plant gathering]]).

Inaccurate information and info that needs verification are basically the same and should be marked with {{verify}}, but an article that is mostly speculation (eg. information with {{verify}} on it) doesn't qualify as a higher quality article as there is almost no helpful information to the player, which the reader will see with the little blue ^verify rearing its head on nearly every line).

The article quality system is not one of "this article is well written!  Reads like a novel!" it's a measure of how complete it is.  And the reader doesn't really have a choice in this matter, what is on the page is what is on the page.  It's not a Game FAQs walkthrough where you have 10 or 12 to search from and get to pick the one that has the best rating.

*Not to mention that it might set the reader's eyes on fire.