I like Owen's common urls idea elsethread, which would be great for projects that are willing to use those URLs, but that's not something we're likely to require.
Rich Bowen wrote on 6/21/17 4:40 PM: > On Jun 21, 2017 20:12, "Shane Curcuru" <a...@shanecurcuru.org> wrote: > > > If we're doing something new, why not use rel="" values? > > https://developer.mozilla.org/en-US/docs/Web/HTML/Link_types > > We could even abuse the rel= attribute to add some of our own keys if we > really wanted to. This would make link checking super-easy for sites > that use it. > > > Not a clue what you're suggesting. Can you be more specific? Or give an > example? HTML standard(s) allows a rel="" attribute on <a...> tags. One value is rel="license", which "indicates that the hyperlink leads to a document describing the licensing information". Various unofficial microformats also use things like rel="friend" for my homepage link that points to Rich's blog, because he's a friend. We could (ab)use this to add our own set of rel="trademark" values that the Whimsy site checker (either for policy links, or for recommended links like these) could scan for. This is more valuable when some projects want the link *text* to say "source repository" and other projects want it to say "github repo" or the like. But if we're going to propose patches to a bunch of websites, having a set of rel= values that we consistently use might be a really simple way to add structure to the links, even if we're making up our own values. - Shane --------------------------------------------------------------------- To unsubscribe, e-mail: dev-unsubscr...@community.apache.org For additional commands, e-mail: dev-h...@community.apache.org
