Serious question. Please hit me with a clue stick.
Regarding 1.9, how can "old-IE compatibility often causes problems of its own" and "simplest way to support older browsers is to use jQuery 1.x on your site, since it works for all browsers." both be true at one and the same time?
This is one of the biggest projects on the web leading the way in throwing shitty old IE under the bus to help push the modern web forward.
They're going to keep big-fixing 1.X so people stuck with IE6/7/8 can keep supporting it, but moving forward with 2.x for modern browsers makes their lives easier in the long run, and provides yet another coffin nail in Microsoft's attempt decade-old attempt to break the web.
They're not so much throwing IE under the bus as throwing developers under the bus that have fortune-500 clients that can only use IE. Some nerd (me) whining about their browser choice is really not going to make their IT director change policies.
Make that Fortune 500s which when jQuery 2.0 is released next year needs to be running a OS which is EOLed by Microsoft with no support or security patches to be affected.
IOW: Cry me a river.
People made these exact arguments that people "require XP and MSIE8" and that they "can't upgrade" years ago, when Vista was released, then when Windows 7 was released and now with Windows 8's release.
Here's news for you: You've had 2 or 3 advance warnings from Microsoft that your IT platform will be obsoleted and unsupported and 2 non-ordinary support-extensions. You've had more than half a decade to plan and execute the migration.
If you haven't completed that migration yet and are still whining, you are incompetent as an IT manager. There is no other way to describe this. If you are still running XP, that is because of incompetence. No excuses.
XP is older than the very first release of Ubuntu. Microsoft has supported XP for longer than Ubuntu's entire existance. As a basis for comparison any version of Ubuntu is supported for 2 years and that's it. XP has been around for 12 years now, and will be supported for yet another year before support is finally dropped. Sometimes enough is enough.
This is a good move by jQuery to streamline their code-base. Sometimes you need to do the spring-cleaning if you want your code to be manageable and have any chance of effectively improving your product, and this is just that spring-cleaning.
Whining about browser choice is never going to accomplish anything. You're speaking the wrong language. Speaking the right language would go something like: “I can increase revenues while decreasing long-term maintenance costs, and I can do it at the current expense level. The only way I can do this for you is if we use this technology.” The person you're pitching couldn't possibly care less about IE/Chrome/Firefox, but they certainly care very much about business metrics.
I think it will. Especially when you can say, "If we can drop IE 6/7/8 support, development costs will go down significantly." Remember, it's not just JS compatibility, think about all the CSS hacks you need to do to get your design just so. HTML 5 is another issue (I'm talking tag support). Requiring flash versions of your custom skinned player. There are so many factors we need to attach to old browsers, that it just makes sense to move forward. Maybe not right this second, but soon. And I think this is a great place to start.
Third parties have been trying to throw shitty old IE under the bus for a long, long time and haven't had much success in doing so. I fear jQuery 2.0 will see very low take-up on mainstream sites, and the resulting fork will be tricky to manage.
>Third parties have been trying to throw shitty old IE under the bus for a long, long time and haven't had much success in doing so.
On the contrary. Lots of big volume mainstream websites (including even Google properties) have stopped supporting IE6 and some even IE7 for years. For IE6 it has been at least 4 years since most sites stopped caring about it.
(Heck, even Microsoft created a site to convince people to drop IE6 it last year).
Mainstream websites will move much faster off of IE8. It's enterprise shops and people making software or websites targeted at them that will take the longest.
True, but it took a LONG time - IEs 6 and 7 were released in 2001 and 2006 respectively, so assuming your '4 years ago' stat is correct, that's still 3 years of waiting for a sweet spot of upgrades, by which time IE6 was 8 YEARS old. IE7 still gives me enough traffic (~5%) to require support, so that's a total of 4 active versions I need to cater for. I really hope adoption of latest IE speeds up, but I've been hoping that for a while now, and I just don't see it. Given that XP introduces a cut-off on IE8, I doubt we'll see the situation changing for a while.
1) Opportunity cost. Supporting them takes time and makes your code more bloated and less able to use cool new stuff without hacks and workarounds (and that cool new stuff can give you a competitive advantage to the other 95%).
2) This 5% is only gonna go down.
3) 5% is too small anyway.
4) There might be 5% of traffic, but how much is quality traffic? Depending on the site, a guy with the latest laptop, OS, browser and everything might be more likely to spend money, compared to some XP using guy that might have gotten there accidentally. This needs more search into your user data to tell.
5) Even if you want to still cater to IE6, you don't have to support it per se. You can just make it generally sure that its user can click around and see the pages, even if they lose a lot of formatting and cool options. I mean, don't even go for "progressive enhancement" stuff. Just let them be able to at least see the page, even if it looks like crap without the necessary css etc support.
IE7 is old enough that I would point users at all of Microsoft's “please upgrade” pages — if you deal with financial or personal information, it's worth doing just for the security benefits.
I still can't wrap my head around this concept that 5% requires support. Tim Ferriss isn't right about everything, but his bit about firing your worst customers is dead on. The Pareto principle suggests that 20% of your customers cause 80% of your workload, and the sensible thing to do is drop those customers. If it's only 5% instead of 20%, it seems like a no-brainer to me.
It really depends what kind of business you're building. If you rely on the network effect, for example, 5% of your user base helps to lock in the rest of your users, and so losing it may have a greater overall impact than just 5%.
Keeping old-IE compatibility prevents some optimizations and makes the code have to do extra checks, but using one file that supports it is simpler than using two files?
Ok, so the problems it causes are pain points for the jQuery devs and not for website devs? Also, the underlying code is more complex and lacks optimizations? Until old drops off the radar completely we'll have to use 1.9 solely or do a conditional check and serve the suitable version - serving 1.9 seems simpler ^^, call me lazy, many have,
Depends what you're doing. If you're building a mobile app (particularly one meant to be distributed via PhoneGap), or a Chrome extension, maybe you don't care about IE support.
Its also a pain for every website serving jQuery because of increased data for every single page load. Cutting out the support for old IE should also help with performance.
old-IE compatibility causes problems with the other non-browser platforms they list, e.g. Windows 8 apps, Firefox OS apps, Chrome OS apps, PhoneGap apps, BB10 apps, node.js, etc. For example Win8 apps using jQuery used to log an error on startup every time, probably because of some oldIE compatibility check which didn't work in the not-quite-a-browser environment.
There's a difference between working as a web page in all browsers and working in all HTML-rendering environments such as the ones listed in the bullet points in the blog. For example, Windows 8 apps or Chrome add-ons have additional security restrictions on .innerHTML.
Regarding 1.9, how can "old-IE compatibility often causes problems of its own" and "simplest way to support older browsers is to use jQuery 1.x on your site, since it works for all browsers." both be true at one and the same time?