User:Geo Swan/opinions/Should the wikimedia software support additional kinds of links?

Should the wikimedia software support additional kinds of links?

Should the wikimedia software support additional kinds of links? There is no question in my mind that this would be a good idea.

Currently we support:

1. regular, old fashioned, uni-directional, links -- various sorts;
2. bi-directional standard, article to article, wikilinks;
3. one-and-a-half-directional wikilinks that (ineffectively) link to section headings within the article;
4. one-and-a-half-directional wikilinks that (ineffectively) link to section headings within other articles;
5. transclusion, as used with navbox templates -- in article space transclusion is only supported for templates;
6. transclusion, in the user name space any page can be transcluded into other user space articles. Transclusions can be several levels deep;
7. category inclusion;
8. redirections;

In my opinion the strongest new kind of link that the wikimedia software allows, not provided in the regular world wide web is the bidirection article to article wikilink. It supports the very powerful "what links here". The world-wide-web has nothing like this. The links are instanteous. It supports redirection. Links don't break when articles are renamed. This is also instantaneous.

I called wikilinks to section headings are one-and-a-half-directional. Wikilinks to subsection heading are far less useful that regular, article to article wikilinks. There is no "what links here" facility that will automatically look up whether a sub-section heading has incoming wikilinks. And any incoming wikilinks to the sub-section heading will break if the subsection heading is changed, even if the change is a trivial change to the spelling, punctuation or capitalization. These are, in my opinion, extremely serious drawbacks to wikilinks to subsection headings. These drawbacks are, in my opinion, so serious that although this kind of link is possible, it should never be used in article space.

I am completely satisfied with the way we transclude templates now. By convention we don't allow other kinds of transclusions in article space. We don't allow, for instance, the transclusion of images, or the transclusion of text. Back in 2006 I experimented with using transclusion to include some images that were used in a large number of articles. This had advantages -- all the instances were the image was included had the same caption. And, in theory, when someone thought it was a good idea to modify the caption, all instances would get the new caption. But, with the current version of the wikimedia software, practically no one but me ended up trying to edit caption, because no one but me could find caption. The wikimedia markup language is amazingly easy to use. Prior to html and gml I learned the old nroff and troff and IBM's script. But while editing wikimedia pages is extremely easy, figuring out how to edit material transcluded into a page isn't.

I'd like to see changes in how both editing and transclusion works. In addition to editing a whole article, or editing a sub-section of an article, by clicking on an edit button, I would like to see the wikimedia software support sweeping the mouse across a block of text, and then clicking an edit selected text button in a drop down menu, that would allow the editor to view and/or edit all the text in the selected material. Usually contributors would just edit the topmost pane. I think this would address the main problem that currently precludes using transclusion of images, or of text that is common to many articles. Ordinary editors could view and/or edit the pages where the transcluded material was found.

The current support of categories is, IMO, problematic. We use categories for pragmatic reasons. They have proven very useful. But they have very serious weaknesses. First categories have no history. Sometimes good-faith contributors, or vandals, decide to strip a category of all its members. There is no way of knowing which articles were listed in the category. And a question that comes up, all the time, is "why is this article in this category". More on this below.

Our current

I think we tell In my opinion the two most