10 Jun
2014
10 Jun
'14
12:04 p.m.
On Tue, Jun 10, 2014 at 11:02 AM, Denis Gervalle <dgl(a)softec.lu> wrote:
Hi Vincent and all,
On Tue, Jun 10, 2014 at 9:08 AM, vincent(a)massol.net <vincent(a)massol.net>
wrote:
Hi Andrea/all,
On 10 Jun 2014 at 08:46:45, Andreea Popescu (andreea.popescu(a)xwiki.com
(mailto:andreea.popescu@xwiki.com)) wrote:
We see here that Marius have done the most for our UI, and he know how
costly it is to support different browser. I have my own experience in that
field with another product, so I really understand your point Marius.
Hello all,
Following the "Start supporting IE10 and IE11" thread I’d like to propose
to vote if we will support IE10 and IE11.
FTR the thread in question is at
http://markmail.org/message/zjlbcfkl5cqnrwfk
The most important reasons for supporting these
versions are:
- Internet Explorer 8 and Internet Explorer 9 are still widely used by
users, but many of them are currently choosing to update to IE10 and
IE11.
- according to netmarketshare.com (Browsers ->
Desktop Share by Version)
from a total of 100%, IE10 is used by 6.85% users and IE11 by 16.61%
users.
It’s true that IE10 is not as popular as the
older versions (IE8 - 20.85%
and IE9 - 8.89%), but its usage percentage is still significant.
Here’s my +1.
On the above-mentioned thread Marius had proposed to support only the
latest version of each browser, which is basically what we’re doing except
for IE. The rationale is that it takes time to support all browsers and the
community can only do so much. Companies offering paid services on top of
XWiki could support the other browser versions for their customers.
While I understand that, my POV is a bit
different: I’d like that we
support the top browsers/versions used by the majority of people, because I
feel XWiki will be successful if the majority of people can use it without
problems.
I fully agree.
Looking at the graph at
http://dev.xwiki.org/xwiki/bin/view/Community/BrowserSupportStrategy if
we were to support at least the top 6 browser/versions we would cover 70%
of the user base. For IE this means supporting IE8-11. Note that strangely
IE8 is the top browser in market shares, way above most of the others… Not
supporting it ATM would mean “loosing" 20%+ of users. If we have problems
supporting, say the top 10 browsers, then we can drop to top 6, if that’s
an issue, then to top 5, etc. ATM we’re supporting 5 browsers and with
IE10-11, this would mean raising to 7 browsers. I propose that we try it
with the additional 2 browsers (IE10-11) and see if it’s a burden for us or
it’s manageable.
However, this is not so black and white this should be part of the
discussion. While we may increase our luck of supporting many browser by
simply using an appropriate javascript and CSS framework (Prototype and now
JQuery, Bootstrap), testing and ensuring that a lot of browser version is
fully compatible could be a tremendous task that I sure we probably do not
fully achieve now.
IMO, we should ensure the most essential features of XWiki degrade nicely
in any browser by having a not too convoluted UI for them. Aside of it, we
should be able to benefit of the latest HTML5 improvement to provide really
appealing interface, and supporting only the latest browser is best for
allowing that. To achieve that goal, IMO we should declare that we only
support the latest browser, and that essential feature are supported on any
browser in degraded form. We should carefully test latest browser, and only
test essential features on a chosen subset of old browser (the most
popular, like IE8).
If you ask, what are the essential feature, I would reply by a very clear
example, to edit a document, having a fully WYSIWYG editor is not
essential. If you want it really and you decide that it is worth the
effort, you may upgrade your browser. Of course, we already have the actual
WYSIWYG editor almost working properly, we should not completely break it
now. But we may clearly state and accept that it have some limits, and put
effort on new features instead of trying to fix old features in IE8 for
example, event if this limit the IE8 compatibility of the editor (if it is
still basically usable of course). Overtime, we may accept more issue with
IE8 and a such non essential feature, when the market share continue to
drop.
If have the now the ability to sustain a better effort in browser support,
the browser I would like to see better supported are the mobile ones. We
have not put enough effort on the mobility aspect, and we should not
consider Safari, Chrome or Firefox as single browsers, their mobile counter
part also require some attention.
So there are 2 questions in here:
- Starting the support for IE10-11: I’m +1 for that
+1 as well
+1
- Deciding if we want to support only one version
for each browser or
instead support the highest browsers in marketshare. I’m +1 for the
marketshare approach.
+1 to only support essentiel feature on any browser,
testing the one or two
that have the highest marketshare for confirmation.
+1 for Denis' suggestion to support only basic features in older IE
versions. Note that in practice this is what we're doing since we
don't hurry to fix issues with IE8 that are not critical, i.e. that
don't affect important features.
Thanks,
Marius
Thanks,
Thanks
-Vincent
_______________________________________________
devs mailing list
devs(a)xwiki.org
http://lists.xwiki.org/mailman/listinfo/devs
--
Denis Gervalle
SOFTEC sa - CEO
_______________________________________________
devs mailing list
devs(a)xwiki.org
http://lists.xwiki.org/mailman/listinfo/devs