The Quest of the Canonical Link
Some thoughts about bookmarking an actual and unique link for a web page in mid-2018.
Managing efficiently bookmarks seems to become a bit more complicated each day right now.
- Naturally, many sites have now switched to HTTPS protocol for securing communication.
- Occasionally, some sites starts to redirect from subdomains to directory paths for performance improvement (as explained for DeviantArt by Dan Leveille for example).
- Sadly, many sites continue to add unwanted kind of anchors, mainly for tracking purpose (such as these gi=, ref=, src=, utm= and so on).
In the near past, it was still easy to determine and retrieve the canonical URL of a page before bookmarking it. An extension for Firefox like Uppity (by Anthony Lieuallen), was enabling for navigating among different levels of a page or site in the manner of file system. Unfortunately, since the huge changes made to the Mozilla add-ons ecosystem (due to the switch to WebExtension tech), these useful options have disappeared (it is now possible to reach up only to the main top level).
Of course, all extensions for Firefox haven’t been affected by this massive switch, and some of them are still useful for bookmarking content. For example, au-revoir-utm (by Anthony Ricaud) continues to removes tracking utm= stuff to render cleaner URL, but, unfortunately, it doesn’t make the job for other types. And, as there is no doubt that some imaginative will not hesitate to create new annoying anchors, it’s not inaccurate to think that it should be useful to get a native mechanism or an extension able to manage dynamic filters (as does some blockers, such as uBlock filtering).
Apart from privacy matter, the main problem for bookmarking these pages which use such additions happens to be that they can create many duplicates. Neither Firefox nor Pocket are currently able to detect this kind of doublets: the same page can therefore be saved several times with multiple URLs. In order to clean these supernumerary bookmarks, using Bookmark Organizer extension (by Sören Hentzschel) can be considered, as it aims to find “no longer working bookmarks, redirects, duplicates”. However, at this time, it seems that duplicates of this sort are not yet detected as such.
In an ideal world, I guess that a modern browser should nowadays give (even if in an optional way) to the user the ability to display only the canonical link in the address bar (provided that it has been correctly specified in the page code). In an era where are growing worries about privacy preserve, there is reason to hope that main actors of the Internet will dig into this question (and not only for bookmarking purpose). By the way, has not Firefox just announced some changes about the approach to anti-tracking (as introduced by Nick Nguyen)? But as some use themselves this kind of anchors for certain of their own sites (such as Mozilla does on its own add-ons repository), it seems still far from over right now.