Bootstrap: What would be the advantages of dropping IE10?

[Herst]

Thinking aloud, probably just relevant for post-4.0 (right, @mdo?):

• #23770 could be reverted, possibly improving JS minimization and be a micro-optimization for undefined checks
• Notes about missing pointer-events (as added in e.g. #23826) could be removed.
• The pointer-events workaround for #23222 might be worth checking out again.
• SVG images fix not needed any more (#23476)
• Some of the notes about disabled <a> buttons could be possibly removed
• #22493?
• Doing flex-basis: 100% instead of flex-grow: 1 might not be needed any more (#23652).
• No need for -ms- prefix for flexbox properties (as noted by @alecpl, thanks!), dropping -ms- prefix for touch-action

_Can I use_ comparison: http://caniuse.com/#compare=ie+10,ie+11
IMHO, most noteworthy from there:

• pointer-events
• hidden attribute
• const and let (not whole ES6 though, no idea whether Babel can make use of this)

Some other possibly relevant things:

• Various touch improvements, sadly most of them have to do with the pointer events API, but also various touch related (but mostly IE/Edge specific) CSS properties:
  
  
  
  • https://msdn.microsoft.com/library/bg182636.aspx
  
  
  • https://blogs.msdn.microsoft.com/ie/2013/07/31/ie11-touch-browsing-for-todays-web-and-beyond/
  
  

* https://blogs.msdn.microsoft.com/ie/2013/10/22/enabling-new-interoperable-panning-experiences-through-the-css-scrolling-snap-points-specification/

(Concerning desktop browser usage, it is below the 0.27% of IE8 and 0.29% of IE9 according to http://gs.statcounter.com/browser-version-partially-combined-market-share/desktop/worldwide/#monthly-201811-201811-bar, with no significant numbers in any continent. It's also considered unsupported by Microsoft on any desktop end-user operating system, see https://support.microsoft.com/help/17454, under _"What is the Lifecycle Policy for Internet Explorer?"_; Other UI libs which don't support IE10: [1], [2], [3], [4])

---

Dump

• #25587

[alecpl]

No need for -ms- prefix for flexbox properties - smaller css output.

[mdo]

Appreciate the discussion, but it's sticking around for v4.

[tarkhov]

I have now rewritten a few components on a clean javascript and will end in the next month - https://tarkhov.github.io/postboot/

[kukuleta]

, Z

24 Nis 2018 Sal 05:46 tarihinde Alexander Tarkhov [e-mail]
şunu yazdı:

I have now rewritten a few components on a clean javascript and will end
in the next month - https://tarkhov.github.io/postboot/

—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub
https://github.com/twbs/bootstrap/issues/25100#issuecomment-383785579,
or mute the thread
https://github.com/notifications/unsubscribe-auth/AdKINlnS_zdUYuBPcx9TaLbNpJRt0iqrks5trpITgaJpZM4ROPiE
.

[zoldello]

In version 5, if IE11 and lower support is dropped, bootstrap can iterate much faster, will not be dragged down by dinosaurs and can fully embrace new CSS and HTML feature (even right after they are released). Version 4 can be the version those still needing to dry-nurse IE11 and lower, can use.

[coliff]

@zoldello IE11 support won't be going away anytime soon. It is still very popular (has more desktop users than all versions of Safari combined - and even Microsoft's own Edge browser!). It's the only preinstalled browser on Windows 7 and Windows 8.1 which are supported by Microsoft for many years to come.

Of course I agree that IE10 can be dropped in the next major version of Bootstrap. That is unsupported by Microsoft and has a tiny marketshare (0.1%)

[Kout]

@coliff The decline of IE might be quicker than expected. Last week one of Microsoft officials called Internet Explorer “a compatibility solution” (eg. not a browser :grin:).

Our part of developers can be giving the IE less care. I know all the arguments of how much money can bring a 0.1 % of users visiting our e-commerce website. But I also think, we are helping to keep this dinosaur alive with too much care and conservativness.

[coliff]

@Kout Let's not go off-topic too much here...

• IE10 support should be dropped from Bootstrap as usage is now 0.07% globally.
• IE11 is still very popular (2.37% globally), still more used than all versions of Safari for MacOS combined and still more used than all versions of Edge combined.
  Reference: https://caniuse.com/usage-table

[XhmikosR]

BTW I agree with @coliff's answer. That being said, I'm not sure what the verdict will be for v5. Because dropping IE11 makes JS cleaner. We could suggest a specific snippet for any needed polyfills from https://polyfill.io/ thus making the transition smoother. CSS is another issue, but we'll need to assess that when time comes.

[Kout]

@coliff You are aboslutely right. In ideal world it would be nice to separate Bootstrap version supporting IE11 from the non-IE11. Eg. having for the first one extra stylesheets, polyfills etc. With rise of HTTP/2 having more requests is not big deal anymore (and rather “punish” IE11 users than the majority with modern browsers). In plain words the strategy of jQuery and their slim/full versions. But I don’t know if it won’t be a hell for maintenance.

[dvmuccillo]

IE11 isn't recommended anymore by microsoft itself.

Can't we afford the price and remove its support?
When v5 release come maybe someone will still be developing to support a software marked as dead or that will touch a very old app and try to upgrade it.

But aren't there no more positive points than cons as to the removal of the support?

[Herst]

@dvmuccillo

But aren't there no more positive points than cons as to the removal of the support?

Feel free to create a new thread discussing the possible upsides of dropping IE11 support (this thread is about dropping IE10). I am sure the list of advantages would be much longer than the one here, since much more changed between IE11 and Edge than between IE10 and IE11. I guess this would apply even more so if only rather recent versions of Edge would be supported.

[dvmuccillo]

@Herst my bad!
Sorry and thanks for the tip.