I suppose I could start appending footnotes to my ranty emails. Johnathan is definitely right regarding jQuery while I was generalizing. Yes, jq1.8 will be supported - but, if you wanted to, you could still run a site using jq1.4. The jQuery team is moving beyond LT IE9 because losing the bloat is certainly more performant, especially as the web scurries further away from high-speed connections. Even now, many of us are supporting old IE by pulling in additional stylesheets or scripts. The practice doesn't change if on detection you choose to load jq1.8 instead of 2+. As the web moves forward, the experience for old browser users will increasingly suck - polyfills bust performance budget.
Google Apps / Play pulled support for IE8 on November 15. Link to Techcrunch below. It's not fatal, but it's the same premise - IE8 users will get the "you should consider upgrading" message. It's the beginning of the trend, but it's definitely a trend. I just browserstacked my dusty G+ profile and there is a polite message. I didn't see it on Calendar or Gmail. It more in-your-face on Play. It's there and it's not. It works, it's gradual, but it's goading.
John's right, too, when he makes the point that the decision has a lot to do with the difficulty / expense to support. The question to me is if a library website is built by taxes and tuition, is there a point where the redundant work for backward compatibility becomes a disservice?
Michael // ns4lib.com
From: Jonathan Rochkind [mailto:[log in to unmask]]
Sent: Tuesday, February 19, 2013 11:57 AM
To: Code for Libraries
Cc: Michael Schofield
Subject: Re: [CODE4LIB] A Responsibility to Encourage Better Browsers ( ? )
On 2/19/2013 10:22 AM, Michael Schofield wrote:
> Now that Google, jQuery, and others will soon drop support for IE8 -
> its time to politely join-in and make luddite patrons aware. IMHO,
I would like a cite for this. I think you are mis-informed. It is a misconception that JQuery is dropping support for IE8 anytime soon. And I'm not sure what you mean about 'Google' dropping support for IE8.
[The mis-conception comes from the fact that JQuery 1.9 will not support IE <9, HOWEVER, JQuery 1.8 will be supported indefinitely as feature-complete-compatible with JQuery 1.9, and supporting IE <9.
JQuery 1.9 is just an alternate smaller JQuery without IE 8 support, yeah, but JQuery 1.8 has no EOL and will be supported indefinitey feature-complete with 1.9].
Anyway, I think it's clear that the web developer with our level of resources can not afford to support every browser that may possibly exist.
We have to decide on our list of browsers we will actually spend time ensuring work with our code. (You can also, like JQuery-mobile, have a list that's supported as 'first class', and another list that is supported with graceful degredation -- and then others which you don't look at at all, and may fail miserably/unusably).
That decision is generally based on a combination of popularity of browsers among your users as well as difficulty (expense) to support.
If you can politically get away with no longer supporting IE8 even though it's popular among your users, I guess that could be legit. It depends on your 'business needs', right?
Once you've decided to stop supporting a browser, especially one that may be popular anyway, a secondary question is whether to let it just silently potentially fail (you generally aren't spending time analyzing whether it will in fact fail, work as intended, or degrade gracefully -- that's part of the point), or actually sniff user agents and give the user some sort of warning that your site may not work with your browser.
If you are going to give a warning, I'd recommend it be a relatively unobtrusive warning that still lets them proceed to use your site anyway if they want to ignore your warning, rather than one that locks them out.