CZ:The Article Checklist

Basically, the Article Checklist is (or will be, if we start using it) a set of standardized notes about every article in the Citizendium. The proposal on the table is to place this checklist on the "Talk" page of every article. The checklist tracks data that we can use to organize article improvement projects and to compile new statistics meaningful to humans. It does this by automatically compiling useful categories, linked from Category:Checklisted Articles.

Testing starts March 8: please see The Big Cleanup. You can also discuss this proposal.

How it works

The checklist is a template (if you're interested, here is an introduction to Mediawiki templates). Basically, you fill out the checklist simply by typing workgroup category names, code numbers, and yeses and nos. The template takes your input and spits out two things: (1) your answers, prettily formatted in a table, and (2) new and very useful categories such as "Developed Articles," "Stub Articles," "Philosophy Orphans," and many others.

An example of the Article Checklist in use can be found at Talk:John Doherty (fiddler). Note not only the checklist, but all of the "categories" listed at the bottom of the page: all of those categories are automatically generated by the checklist. If you're interested, feel free to play with the parameters of the template on Talk:John Doherty (fiddler), to see how they change the category list; but you might want to see below for help doing this, particularly for filling out the 'status' field, which uses numbers 1-4.

A complete list of articles that make use of the Article Checklist can be found at Category:Checklisted Articles. Checklisted articles make use of a variety of useful categories, including but not limited to:

1. Developed Articles
2. Developing Articles
3. Stub Articles
4. External Articles

• Internal Articles = 1-3
• Nonstub Articles = 1-2
• Underlinked Articles

The technically-minded may be interested to see the template found at Template:Checklist, which is the "engine" behind the checklist.

How we could use this to organize massively distributed work

So, how could the Article Checklist be used?

First consider that we have done remarkably well with The Big Speedydelete. This shows wonderfully what a big crowd of smart people can do if given definite, useful, self-directed tasks, clear instructions, and a way to get credit. Suppose, then, we start another "Big" project, a Big Cleanup, which does all the relatively simple stuff that needs to be done, or checked, in articles. For instance:

• Is the title bolded?
• Are the correct workgroup tags used?
• Are the old Wikipedia interwiki links removed, if the article came from Wikipedia?

And so on.

Well, when people do The Big Cleanup, they can put the Article Checklist on the talk page of every article. One of the items on the checklist is: "Basic cleanup done?" If you say "yes" to that, that means you've gone through the "basic cleanup" list, which can be found below, and you've done everything on that list. If you mark "no," then the article is added to "General Cleanup" and (supposing it's a Physics Workgroup article) the "Physics Cleanup" categories.

If we use the Article Checklist as part of The Big Cleanup, not only would we track the information that the article had been cleaned up, but for relatively little work, many useful categories that would result. Perhaps the most important would be the "status" information about the articles in our database: this would allow us to determine what percentage of articles are "developed," "developing and past stub stage," "stubs," and "sourced from elsewhere and undeveloped," and how many articles there are in each category. This sort of data would be really useful for workgroups to get an idea of where they are at, what their needs are, and so forth.

But this would be only the beginning. These dynamically generated categories would make new projects possible. We could tackle orphans by workgroup; we could upload images and templates by workgroup (or alphabetically); we could expand stubs alphabetically, or by workgroup; we could make and execute some hard decisions about relatively unedited articles from Wikipedia, Britannica, and other sources.

Then, if we expand the checklist and sets of categories, we might keep track of any of dozens of different attributes: need copyediting; needs rewrite for style; intro needs to be rewritten; bias; education level inappropriate; appears to be original research; and so forth.

The blank template

Here's a copy of the blank template that you could (but don't yet!) copy and paste onto the talk page of an article.

{{checklist
|                 abc = 
|                cat1 = 
|                cat2 = 
|                cat3 = 
|           cat_check = 
|              status = 
|         underlinked = 
|             cleanup = 
|                  by = 
}}

The template fields explained

Here are explanations of each field on the checklist. Bear in mind that we can, especially now, add and remove fields.

The 'abc' field

We would be putting the checklist on the "Talk" pages of articles. In category lists, then, articles will all appear under "T" for "Talk," unless we specify otherwise. Therefore, we should give the name of the article here for alphabetization purposes. For example: Russia; United States of America, The; Doherty, John.

Workgroup category or categories

Should be identical to the workgroup list found on the article itself. This is necessary in order to generate certain categories automatically.

Fill in 'cat1' before 'cat2' or 'cat3'. Template:Checklist permits only three categories, 'cat1', 'cat2', and 'cat3'.

Check categories?

The small-font parenthetical note is generated by 'cat_check'. This is used to indicate whether, in the opinion of the person filling out the checklist, the category list is OK or, instead, needs review by editors. Simply write 'y' or 'n' (or variations on these) for yes or no.

This generates handy "category check" categories, such as "Category:Philosophy Category Check" or, if no categories are specified, "Category:General Category Check".

Article status

There are four options for 'status' (examples to be given later):

1. Developed article: complete or nearly so
2. Developing article: beyond a stub, but incomplete
3. Stub: no more than a few sentences
4. External article: from another source, with little change

Specify them by adding the appropriate number to the 'status' field. This populates categories that could be very useful indeed for us, including categories corresponding to each of the four options, as well as "Internal Articles" (combination of 1-3) and "Nonstub Articles" (combination of 1-2)--and for each of the categories listed. Thus, for example, if we have the Article Checklist on all of our articles, then we can produce a complete list of "Philosophy Stub Articles" or "Music Developed Articles" or "Biology Internal Articles".

Underlinked article?

An "underlinked article" is that has none of the main expected links to it. For example, if "tree" is not linked from "plant" or "biology" or "botany" or any other such expected "parent" topic, then it is an "underlinked article." Note, all orphans are underlinked, but some underlinked articles are not orphans. Yes/no as before. When in doubt, mark as underlinked, or get advice.

Basic cleanup done?

This variable, 'cleanup', is just yes or no (so, 'y', 'Yes', 'no', etc.). What it indicates is that the article has gone through a "basic cleanup," which means

• Bold the article title, if necessary.
• Remove all unused (red) templates, images, category tags, and interwiki links.
• Add workgroup category tag(s).
• Add (or remove) CZ Live tag as appropriate.
• Check the "Content is from Wikipedia?" box if the article is sourced from Wikipedia.
• Complete the Article Checklist.

Checklist last edited by?

Simply type ~~~~ (four, not three, so the date is given).

The extensibility of the system

If this checklist can be found atop the talk page of all new articles, clearly, we would like the data to be kept up-to-date, and the system extensible.

There is no reason that we cannot keep this data up-to-date. This is not a foregone conclusion, but it seems entirely possible that people will develop a strong interest in keeping data about, for example, the "status" of the articles they work on--and thus, the other data as well--up-to-date.

The system is extensible in that it will be easy enough to add new fields to the checklist, as long as they are optional fields. We might produce variables that allow us to keep track of other sorts of data we might find useful, for example, how much copyediting an article needs, whether the article has any number of specific problems, such as improper use of footnotes, not enough footnotes, stylistic problems, etc. Any or all of these variables could then be used to track the overall maturity of an article, and (by producing appropriate categories) to organize large-scale efforts to tackle particular problems.