Context
Hello!
When working on the implementation of the new accessibility preferences, it became apparent that the font-size increase that we supported previously in the Extra accessibility
user preference was not possible to update without a lot of changes to the codebase.
Vote object
Drop the font size increase preference from xwiki standard flavor.
Clarifications
Here are a few pros and cons I identified related to this vote object:
(+) The former implementation did not consider the color theme font-size into account. If the wiki admin already set a large font-size-base, the users with accessibility extra turned ON could have smaller text than the regular user.
(+) The former implementation is nearly impossible to extend. It relied on loading a new CSS depending on the user preference. This could end up quite a large cost if multiple preferences with multiple options take this architecture. This and over details in our font-size system brought a lot of technical debt. See the discussions on the PR for XWIKI-21492, especially this one, as a proof of this technical debt.
(+) Most users nowadays are already used to playing with the zoom level of the page (Ctrl
+ +
) to fit their preferences. This feature can be useful (understand “better than nothing”) for some users but I don’t see many cases where it would be critical for the user experience.
(+) The extra accessibility feature was outdated and we did not receive many user reports regarding it (if any at all?). In my opinion this is a hint that the feature was not used a lot. Removing a part of this feature would not impact many users (and as said above, not in a critical way, there’s easy ways around this feature regression).
(+) The object of this vote influences only the display, I don’t see it breaking any extension or custom code.
(-) It’s a regression on feature.
(-) It can impact the user experience of any regular user of XWiki (not only admins or devs).
Opinion
I think the pros far outweight the cons, and for the sake of updating the extra accessibility feature without introducing even more tech debt I think this feature should be dropped.
Note that adding this feature back could be an improvement we consider later down the line, but :
- it’s not a critical one
- it could overlap with modern browser features (need to check if it’s actually relevant or an alternative solution would fit better)
- we need to define and apply new best practices in order to avoid it being a large technical debt.
For what it’s worth, here’s my +1
Conclusion
Did you find out additional pros and cons related to this proposal?
Do you agree with the object of this vote?
Thanks in advance for your inputs!
Lucas C.