Wikipedia talk:Readability/proposal discussion

Introductory comments

I have proposed a policy proposal on Wikipedia:Readability that contains two provisions. The proposal is here. Feel free to discuss them on this page.

The policy proposal is also posted on Village pump. Wooyi 22:23, 25 March 2007 (UTC)

I don't really like the first provision. Some articles/sections need a lot of work. I would support it if the number was higher, like no more than 3. If an article needs 4 cleanup templates, it could just use {{rewrite}}. As for the second provision, I can't think of many examples of this. Most articles that have really long lists are lists and most articles with a ton of external links can usually have most of them removed as spam. If most of the article consists of external links, the whole article could probably be PROD-ed or taken to AfD as its probably all spam. Mr.Z-mantalk¢Review! 22:31, 25 March 2007 (UTC)

The second seems sensible (with the obvious exception of "List of ..."). The first, however, is totally unworkable. You're essentially saying that if I come across an article with two cleanup tags on it, and it happens to fail NPOV, I can't tag it as such. Chris cheese whine 23:05, 25 March 2007 (UTC)

No, he is saying that, if there are many problems in a page, they should be merged in only one box. Just to make the whole thing more readable. --189.12.139.198 10:02, 11 October 2007 (UTC)

I agree that too many boxes are distracting, but I think that 3 is probably a better limit than 2. I would suggest that if more are needed that they should be directed to the talk page of the article. One other thing to think about is the use of these on stubs vs. more established articles. Often stubs have many boxes early on that are placed by new page patrollers, so if the goal is to impact this behavior, the NPP community should probably be involved in this discussion. --After Midnight ⁰⁰⁰¹ 00:17, 26 March 2007 (UTC)

I agree with After Midnight; some articles need more than 2 boxes. The second suggestion above is a good one which should be followed. V Schauf (talk) 05:52, 26 September 2009 (UTC)

Strong Oppose on both counts; misuse of template

1. Dispute and cleanup tags exist for a reason, to vividly identify article problems and attract readers to fix them. There is no reason to limit the number of problems to be identified. This is like saying that if you get robbed 5 times in one month, you just have to STFU after 2 police reports. The argument that "it makes the articles" ugly is of no consequence. The point is to make the articles ugly to inspire action to clean them up, and to alert readers potentially depending on the article that they should be cautious, about what, and why.
2. Many articles are lists. I don't know what else to say. If this proposal went through, List of World Snooker Champions would have to be deleted or reduced to a couple of sentences about the concept, but no lists of the champions, or filled with endless streams of non-encyclopedic blather to meet the proposed ratio requirement.
3. I have removed the {{Proposal}} tag. That is for entire documents with well-formed, compelete contents and rationales, that have already had broad community input and are believed ready by a significant number of editors for designation as a policy or guideline, not for random ideas-of-the-moment embedded somewhere in one-editor essays. Things like this should simply be discussed as topics in WP:VPP.  :-)

— SMcCandlish [talk] [contrib] ツ 00:28, 26 March 2007 (UTC)

Comment sorry for the misuse of template, I wasn't aware of that. Regarding to the list articles, I have just made that exception in the proposal. And also while some warning boxes are needed, it impairs a reader's interest to read it, as that he would think "oh this article is so bad i will come back and read it after it gets revised", and nothing gets done. I do agree that the limit number might raise to 3 though. Wooyi 00:35, 26 March 2007 (UTC)

Comment: I don't think this really addresses the issues. Many articles are not lists but have many, or long, lists in them by their very nature, and can't be any other way (other than by forking them into a prose stub that will never expand, and an out-of-context list on a separate page, which would be far worse.) For example, many sports organizations cannot have a reliable sourced article with encyclopedically valuable information written about them that are very long in the prose (when it was founded, where it is located, who its boardmembers are, controversies and other interesting things about the entity's history, etc. Might only be two or three paragraphs). But it might necessarily have 8 tables of data about sanctioned event in the history of the organization and their winners and runners up, etc. The article really can't be written any other way. I am aware of the kinds of articles that you are actually trying to address, but there is no policy deficiency here. Such articles get AfD'd on a pretty frequent basis (often converted into explicit "List of..." pages instead of just deleted). Your proposal is a solution looking for a problem, a hammer without a nail.

If your other complaint is simply about too many headers at the top of the page, there are often ways to resolve that, e.g. by using {{Totallydisputed}}, by convering some to inline citations, and changing some to section-level tags intead of page-level ones. If your complaint is with regard to total number of tags, then I don't see the point, as I've already covered above (though, some of those could also possibly be converted into inliners, like {{Fact}}. This isn't a policy matter, however, it's an editorial preference one to be determined by editors on an article-by-article basis. I don't buy the "some readers will just walk away from the article" part; if an article is problematic enough that it has that many tags, the reader probably should walk away from it, and the article may well be a good AfD candidate. Your proposal would encourage readers to read and trust bad articles, and make it harder to identify and get rid of them. — SMcCandlish [talk] [contrib] ツ 01:32, 27 March 2007 (UTC)

I once saw a very nice article destroyed (for all intents and purposes) because one zealot plastered 13 banners over it.(Yes there was a vote going on, how did you guess?) When the admins were called in, the guy was (believe it or not) able to make the case that he was actually trying to help. Of course once all the votes were in, the "helper" disappeared and the banners were removed. Using banners in this way amounts to vandalism. Now with that having been said, I find myself forced to agree with SMcCandlish. (Though I still disagree about people leaving a mangled page) The proper places to discuss this are elsewhere. This is not a policy issue at this time. Sue Rangell 02:44, 27 March 2007 (UTC)

Tags telling people an article is a problem is a good thing

A haven't seen any evidence of the assumed premise that multiple cleanup tags are so bothersome they would drive people away from the article; tags that render the article "inconvenient to read." I imagine if the person is at a particular article for encyclopedic information and the article has many tags, then they would leave not because a few tags can't be scrolled past but because they read those tags (unreferenced; not factually accurate; hoax; poor tone; not written in formal tone...); and come to the very likely correct conclusion that the article is not a reliable source of information—that is the proper result. A person on a hunt for reliable encyclopedic information is well warned that a multi-tagged article does not fit the bill. If, on the other hand, the person is a user, or someone who understands the purpose of the site and is of a mindset to become a user, then the tags serve the purpose of marking the article's problems for what we do as users or soon to become users. In either case, the tags tell us, here be demons, and it should be no other way. So, if the intent of the reader is to edit, I can't see how the tags will make them lose interest, and if the intent of the reader is to read an proper encyclopedia article, they shouldn't read that article; we should want them to lose interest and move on for real content.

I also think not having the tags could be actively harmful for our imaginary "encyclopedic reader." Let's imagine a person who has never heard of Wikipedia and has just come across one of the numerous online articles trashing Wikipedia for its unreliable content. So they make it here for the first time, hit random page, and land on one of the hundreds of thousands of unsourced, poorly-written, stubby articles that haunt this site and which happens to have no cleanup tags. If no tags are present, they might take that untagged article as a perfect confirmation that Wikipedia is as a whole as they read it described— unreliable, unsourced, etc. However, with the tags in place the natural conclusion is "oh, this is an unfinished article in development" and that reader moving on to look at other articles (maybe finding an FA or other great content thereafter) is more likely. Accordingly, I disagree with the basic assumptions underlying this essay.--Fuhghettaboutit 12:46, 26 March 2007 (UTC)

I understand the problem. I did not say cleanup template are not needed. I only propose that their numbers should be limited within reasonable scope, and a numerical restriction would be reasonable (I proposed 2, but some advised to change to 3, which I am still considering). And there is one editor below suggesting a similar approach like {{WikiProjectBanners}} to contain multiple boxes, I like that idea too, just need a template for that. :-) Wooyi 21:55, 26 March 2007 (UTC)

Proposing that their number be limited is proposing a limitation on the number of flaws that can be adequately identified in the article, in effect. See my reply comments above in "my" section for existing ways of reducing template clutter. We don't need a Rule about this, just common sense. — SMcCandlish [talk] [contrib] ツ 01:34, 27 March 2007 (UTC)

Not such a good idea

Apart from the fact that making policy proposals in an essay is not all that suitable, both of your proposals are worded prescriptively, overly legalistic, impose arbitrary limits, and do not solve an actual problem but rather an alleged problem. This is almost a textbook example of m:instruction creep. >Radiant< 12:46, 26 March 2007 (UTC)

Not all arbitrary limits are instruction creep. Sometimes numerical restriction with leeways (like in extreme cases you can use WP:IAR) can be conducive to Wikipedia by mandatorilly making the encyclopedia article actually look like an encyclopedia article, instead of just a bunch of useless words combined. Wooyi 21:50, 26 March 2007 (UTC)

• Er, no. All arbitrary limits are instruction creep. If you don't want creep, propose limits that aren't arbitrary. >Radiant< 08:29, 27 March 2007 (UTC)

3RR is arbitrary and isn't instruction creep. --189.12.139.198 10:06, 11 October 2007 (UTC)

Perhaps a multi-template would be useful

Perhaps someone (not me - I don't code templates) could create a template something like {{article-issues}} which would look like {{article-issues|POV|sources|notability|etc|etc}}, which takes a number of variables, and will make one box with some boilerplate text for each problem. That way, when an article does have lots of issues, there's just a somewhat larger box instead of a stack of unequally-sized boxen cluttering the top. Αργυριου (talk) 17:09, 26 March 2007 (UTC)

This is a good idea. If someone can code such a template you are more than welcome to do so. In fact, there has been an precedent to this idea, the {{WikiProjectBanners}}, which is used on talk pages to contain huge numbers of WikiProject templates. Wooyi 21:52, 26 March 2007 (UTC)

There is a similar template used on talk pages, Template:ArticleHistory, used to list multiple events in the history of an article (peer review, GA, FA, AfD, DYK, etc.) Are we thinking something like this, to save a lot of space, or something that would just consolidate tags without the show/hide feature? Mr.Z-mantalk¢Review! 21:57, 26 March 2007 (UTC)

While {{ArticleHistory}} has a lot of buy in, please be aware that many, many people absolutely hate {{WikiProjectBanners}}, and its future is quite uncertain. To do something like this it would need to be done in a way that none of the actual issues about the article were hidden. — SMcCandlish [talk] [contrib] ツ 01:36, 27 March 2007 (UTC)

I suggest we can have a box that the editors can list problems in concise text form by their own instead of template with fixed text. Wooyi 01:41, 27 March 2007 (UTC)

Okay, I have a draft box in my userspace at User:Mr.Z-man/Sandbox‎. Right now, I only have the text for cleanup and NPOV in it but more could easily be added. This is a very rough draft. If anyone can think of a way to simplify the code or generally improve it, please tell me. For example, {{User:Mr.Z-man/Sandbox‎|npov|cleanup}} gives:

--Mr.Z-mantalk¢Review! 04:17, 30 March 2007 (UTC)

I personally support this approach. Now I might go to ANI to inform admins about this. Wooyi 03:01, 1 April 2007 (UTC)

Would such a template be able to include an appropriate category for each problem listed in the template? Each one of the maintenance templates places the article is a specific category. For example, {{npov}} carries with it Category:NPOV disputes; {{hoax}}, Category:Suspected hoax articles, {{wikify}}, Category:Articles that need to be wikified and so on. Each maintenance parameter invoked would have to also place the article in the associated category.--Fuhghettaboutit 03:13, 1 April 2007 (UTC)

Yes, that is no problem. The wikiproject banners used for WP 1.0 do exactly the same thing. CMummert · talk 03:24, 1 April 2007 (UTC)

I'll get right on that. Mr.Z-mantalk¢Review! 03:35, 1 April 2007 (UTC)

Okay, I've added in a date parameter. Since the issue parameters use the numbered ones, you have to use it as {{User:Mr.Z-man/Sandbox‎|npov|cleanup|date=April 2007}} I've changed the above example to reflect the new changes. Should we move this into template namespace, something like {{articleissues}}? Mr.Z-mantalk¢Review! 03:50, 1 April 2007 (UTC)

There is no need to hurry. I would recommend syntax such as

{{articleissues|npov=y|unreferenced=December 2006|cleanup=March 2006}}

Each issue may have a different date. This syntax also makes it easier to add new parameters to the template because each one has its own block instead of repeating in the switch statement. It will take two if statements to per issue to achieve the syntax, as far as I know. CMummert · talk 04:02, 1 April 2007 (UTC)

That is the altenate way I was thinking of, sort of an infobox style. The current way is a little simpler, but as it gets bigger, it will grow exponentially. I'll convert it over later. Mr.Z-mantalk¢Review! 04:20, 1 April 2007 (UTC)

I'm trying to think how to do it now. I could make it so a "yes" or a would give just the text, no date amd everything else would be interpreted as a date, or I could create 2 separate params, one to indicate "yes" (any other string would also work) and one just for the date (if just the daye was given it would also indicate yes). I think something like the second would be better, like {{articleissues|npov=yes|cleanupdate=March 2007|unreferenceddate=January 2006}} Mr.Z-mantalk¢Review! 04:42, 1 April 2007 (UTC)

I think this is the best idea I have heard in a month. It would greatly reduce the visual distraction of multiple cleanup templates. CMummert · talk 03:24, 1 April 2007 (UTC)

This is brilliant. Since it can apparently be coded to add the article into the appropriate categories, can it also be (or is it already) coded to automatically link to the relevant policies? Natalie 04:28, 1 April 2007 (UTC)

Yes, of course. The template above is just a proof of concept. The real one would have to support different dates for different problems, and the text can be made to link to any desired location. CMummert · talk 04:43, 1 April 2007 (UTC)

• Clearly a good idea, please publicise it when it's ready for beta-testing. Guy (Help!) 11:17, 1 April 2007 (UTC)

Okay, the new version now. I've updated it to allow for easier addition of messages and to put different dates for different issues. If there are no objections, I will also move it into template namespace soon, probably Template:Articleissues. Now it works as follows. {{articleissues|cleanup_date=March 2006|NPOV=yes|restructure_date=January 2007}}. To associate a date with an issue, just add "_date" after the issue name. To just activate an issue, without the date, simply use the issue name with any string of text. If an issue name is not used, it will not be displayed (saying something like "NPOV=no' is not necessary and will actually activate the NPOV issue). Using the above example:

{{multiple issues|cleanup_date=March 2006|restructure_date=January 2007}}

This makes adding new issues easier and controls the size. If anyone has any tips on how to get the icon to stay in the top left corner, I would like to know how. Right now, it keeps moving down. Mr.Z-mantalk¢Review! 18:02, 1 April 2007 (UTC)

I've moved it into template space and simplified it some more. If anyone wants to add more issues, the general format is as follows:

-->{{#if:{{{issue_date|}}}{{{issue|}}}|<br/>'''the text to be displayed'''{{#if:{{{issue_date|}}}|<br/><small>'''Since {{{issue_date}}}.'''</small>{{{category|[[Category:relevant category from {{{issue_date}}}]]}}}|<includeonly>{{{category|[[Category:relevant category]]}}}</includeonly>}}}}<!--

If adding the new issue to the bottom of the list, the final <!-- is not necessary but should be added to the issue previously on the bottom. It also needs some usage notes. Mr.Z-mantalk¢Review! 03:50, 2 April 2007 (UTC)

Made some formatting changes. The code for adding an issue has changed as well, so let's finalize the format before adding more issues. I think the icon looks better mid-left than top-left (convention?). –Pomte 12:04, 2 April 2007 (UTC)

I contacted the owner of SmackBot about this template. SmackBot adds dates to maintenance templates; it would be desirable to have dates added to this as well, but it is technically somewhat harder to do. I think that the bot operators should have a chance to comment before anything is finalized here, because a small change in syntax might make their work a lot easier, and the bots are vital for managing these maintenance tags. CMummert · talk 12:20, 2 April 2007 (UTC)

Thanks for contacting me. Dating this as described is not too hard. My main concern is that it may be a solution in search of a problem. Typing {{cu}} {{unref}} {{wfy}} is easier than {{Articleissues|wikify=yes|unreferenced=yes|cleanup=yes}} . Previous "multipurpose" templates have fallen by the wayside, I realise this is slightly different. Rich Farmbrough, 12:43 2 April 2007 (GMT).

This will also affect certain bots that looks for these tags, so they will need to be modified (some bots see {{wfy}} and add the data param). Also, we could create a bot that if it sees multiple cleanup tags, it can combine them into this new template.↔NMajdan•talk 13:32, 2 April 2007 (UTC)

Well, tech problems may be solved and readability may improve somewhat (although I see no big problem with individual templates), but one thing I feel is that it will get more complicated for us behind-the-scenes gnommies used to simple and easy-to-memorize tags like {{sources}}, {{wikify}} and the such. I mean, no harm in learning new things, but still... Hoverfish Talk 22:09, 2 April 2007 (UTC)

I've eliminated the _date parameters in {{articleissues}} so that {{Articleissues|wikify=any text|cleanup=Apr 2007}} also works (see {{checkdate}} for details, and tell me if it's overkill for our purposes), though I don't foresee anyone setting such unconventional values to the parameters, in case they happen to disrupt bot activity. –Pomte 00:14, 22 April 2007 (UTC)

"Clinical trials"

Now I have put the new template {{Articleissues}} for trial in articles below for trial:

• Parapsychology Wooyi^{Talk, Editor review} 22:51, 7 April 2007 (UTC)
• Please add more as you use the template.

readability

I think the main readability issue is that articles are too technical for general audiences. Can you understand this article? 69.140.152.55 (talk) 20:37, 1 October 2008 (UTC)