Forum:Out-dated gradient code phaseout

So due to gradients being only standardised fairly recently the (few) pages that use them have been using 5 different versions of the code for each gradient. But I think we can finally start considering removing the different vendor codes and using only the standardised one, or at least keeping track of statistics to determine when we should remove them.


 * Firefox 16+ supports the standard code, as of December 2013, 11.2% of Firefox users use v. 16 or under. Given a large amount of those were likely v. 16 and since then some of those people have very likely ungraded to a newer version we can assume removing the -moz- vendor code will have minimal impact, if any.
 * Opera 12.1+ supports the standard code, as of December 2013, 41.7% of (non-mobile) Opera users still use v. 12 or under. Removing the -o- vendor code will likely heavily impact Opera users for now. However Opera users constitute less than 2% of all browser users, so removing support would impact 1% of all browser users, so we may consider dropping support for it.
 * Both Chrome and Safari use the -webkit- vendor code, but Safari does not yet support the standard code and continue to require its vendor code. However Chrome 26+ supports standard code, as of December 2013, 3% of Chrome users use v. 25 of under. So Chrome users would be unaffected by dropping the -webkit- code, but Safari users would. However Safari users are only 4% of all browser users, so removing support would have as much impact as removing the Firefox code which I already supported removing.
 * Internet Explorer 10 previous required the -ms- vendor code, however it and IE 11 now both support the standard code, the -ms- vendor code can be safely removed.

So that is the summary, we can drop the -ms- codes, and if we decide we no longer wish to support stragglers or Safari users we can drop support for the others, but I think we need to see what the community thinks first. Because we may decide to remain supportive but check the statistics every few months to see if the change is percentage over time is enough to finally drop support. - The Light6 (talk) 11:13, February 14, 2014 (UTC)