jQuery 2.0 Will Drop Support For IE 6, 7, 8

benfrog writes "The developers of jQuery recently announced in a blog entry that jQuery 2.0 will drop support for legacy versions of Internet Explorer. The release will come in parallel with version 1.9, however, which will include support for older versions of IE. The versions will offer full API compatibility, but 2.0 will 'benefit from a faster implementation that doesn't have to rely on legacy compatibility hacks.'"

Comments

Like

[Krojack]

I like this however I'm guessing anyone that starts using this will have a boat load of complains thus 2.0 won't be used for several years. Sadly. Damn IE slowing down even non-IE users!

Re:Like

[Anonymous Coward]

I agree that a lot of situations will not be able to move to 2.0, but I bet you will see it implemented a lot more than you would expect.

Re:Like

[Kate6]

You see the bit where it says "full API compatibility"? That means with a few lines of PHP (or some equivalent back-end language) you can have a look at the user's browser USERAGENT string, figure out if it's an MSIE browser or not and serve the preferable version of jQuery... 1.9 for legacy IE, 2.0 for IE9 and for standards compliant browsers. And the "full API compatibility" means the rest of your code will play nice with either.

Re:Like

[Anonymous Coward]

I agree. You see more and more websites that are dropping compatibility with older versions of IE, and direct users to Firefox or Chrome. It's a good thing too. It's annoying to dumb down on features for website, just because a small portion of them refuse to upgrade to better browsers.

Corporate websites, it's not going to matter. They'll just continue using an old version of jQuery on Firefox 3.6 or something.

Re:Like

[Anonymous Coward]

User agent sniffing is for tards. Just like PHP. Are you a tard? Much safer and less tarded is to use IE's conditional comments.

Re:Like

[cheesecake23]

Sadly. Damn IE slowing down even non-IE users!

Even more sadly, this is not news. IE has been effectively sabotaging the non-IE community for years. The need to specifically code for IE6 and IE7 due to their crappy compliance with web standards has swallowed immense amounts of global developer effort, which could otherwise have been invested in improving the interwebs for everyone.

Just die already.

Re:Like

[man_of_mr_e]

Or use Modernizr's browser identification.

Re:Like

[man_of_mr_e]

This is fine for 1.9/2.0 but what happens when 2.1 and 2.2 come out? The API's will diverge.

I hope 1.9 stays api compatible, but I would doubt it.

Re:Like

[TheSpoom]

While normally useragent-based feature detection is a bad thing, it seems like one could selectively insert a script tag for either jQuery 1.9 or jQuery 2.0 (they are supposed to be API-compatible, yes?) depending on whether or not the user is using a legacy IE, without any performance hiccup.

Re:Like

[Kate6]

There's already a fair bit of divergence between what you can do in IE and what you can do in other browsers.  IE (including IE9) doesn't implement as much of HTML5 as other browsers.  IE's SVG support is in its infancy compared to other browsers', and completely nonexistent before IE9.

So we might end up with some jQuery functionality silently being disabled in legacy versions of IE too.  Big whoop.

Re:Like

[dingen]

Once a version of jQuery 2.x comes out with an API which isn't compatible with 1.9, IE8 better be long forgotten.

Re:Like

[Billly Gates]

Sadly. Damn IE slowing down even non-IE users!

Even more sadly, this is not news. IE has been effectively sabotaging the non-IE community for years. The need to specifically code for IE6 and IE7 due to their crappy compliance with web standards has swallowed immense amounts of global developer effort, which could otherwise have been invested in improving the interwebs for everyone.

Just die already.

You Cheesecake23 have been misled by a vocal minority!

Re:Like

[Krojack]

What about some jQuery plugin that requires 1.9 or 2.0?

Sorry but I don't like having to write a website 6 times over. One for each IE 6, 7, 8, 9, 10 then Firefox and Chrome. Writing a site now and once I'm done go back and add CSS tweaks for special IE is already a pain in the ass.

Re:Like

[cyber-vandal]

This is a far better way since anything non-IE pretending to be IE will just ignore these. Web designers have been using these for years to load special CSS workarounds for all the bugs in IE.

Re:Like

[Kate6]

If, as the article suggest, there will be "full API compatibility" then any plugins that require one of them should also work with the other.

Don't get me wrong, full API compatibility is one heck of a claim to make and I guess we'll have to see just how close they actually get to it indeed being full.  But assuming they largely make good on their claim, this just isn't really major news.  They're creating a more optimized version of their library for more modern browsers.

Big whoop.

Re:Like

[man_of_mr_e]

jQuery comes out with new API's all the time.. Remember live was deprecated by delegate, then by on... this stuff seems to happen *EVERY* single version.

Re:Like

[dingen]

But .delegate(), .on() and .live() all still work with recent versions of jQuery, so your old code still runs with newer versions of the library. New features being introduced is not bad. The pain starts when old features are dropped.

Re:Like

[madprof]

Holy moley. I might be saying what others say but please, for the love of everything good, use conditional comments.
They've been around since IE5. You will love them.

Re:Like

[Kate6]

I don't love anything about IE.  IE is that extra browser I have to load VMware to test stuff in.

Re:Like

[nedlohs]

Given most (not all) web development makes things worse, that seems like a good thing.

Re:Like

[TheSpoom]

I hear the people saying that, but what would you do to make the page load either one or the other jQuery, but not both? Loading two jQuery instances causes all sorts of problems if not handled well.

Having done web development for many years, believe me, I'm very familiar with all variety of IE-handling hacks. At the moment I can't think of a way to conditionally load jQuery 2.0 if and only if not using legacy IE, and load jQuery 1.9 if and only if using non-legacy IE or non-IE, using IE conditional comments. Besides, using server-side handling keeps the HTML output cleaner since all the user gets is the single, correct script tag.

Re:Like

[shutdown -p now]

XP (and therefore IE6/7/8) is not going to be with us forever, either.

Re:Like

[madprof]

IE9 is better, albeit rapidly falling behind the curve.

Re:Like

[Kate6]

I don't disagree.

Re:Like

[Serious Callers Only]

Another possibility is to use jquery 2 for all other browsers and jquery 1.9 for ie 10 via conditional comments.

Re:Like

[kawika]

What are you specifically worried about? Dell is still using jQuery 1.2 on their site from 2007, and it seems to work. So in 2014, something will break in the (supported) 1.9 version that the jQuery team refuses to fix, but you can't upgrade to 2.1? Can you say what that something might be?

Re:Like

[kawika]

Conditional comments do exactly that. There is an example in the jQuery blog post.

Re:Like

[Art3x]

with a few lines of PHP (or some equivalent back-end language) you can have a look at the user's browser USERAGENT string, figure out if it's an MSIE browser or not and serve the preferable version of jQuery... 1.9 for legacy IE, 2.0 for IE9 and for standards compliant browsers.

Or do it with HTML comments (this actually works):

<!--[if lt IE 9]>
<script src="jquery-1.9.0.js"></script>
<![endif]-->
<!--[if gte IE 9]><!-->
<script src="jquery-2.0.0.js"><</script>
<!--<![endif]-->

Less code. Also you've offloaded the CPU processing from your server to the user's browser.

This was in the article by the way. . . . Oh . . . nevermind.

Re:Like

[Kate6]

Does this also load 2.0.0 for browsers that aren't IE at all, or would you have to do something extra for that part?

Re:Like

[leenks]

I'm forced to use IE8 at the office (thin client crap for internet access) and already there are some sites I cannot use, and others that degrade to basic versions. The sooner this happens the better IMO - the more sites that do not work in IE8 the faster my IT department will be forced to realise that they have to upgrade from Win2003 server and/or install a different browser. Oh, and that really, Internet Explorer isn't "secure" just because it offers an illusion of security lockdown...

Re:Like

[Pieroxy]

And this is the reason I don't use jQuery. When something breaks down on a newer browser and you have to manually patch your JS as fast as possible, you've better patch your code. Patching jQuery might prove to be quite complicated depending on the fix.

Not mentioning that support for legacy browsers is exactly what got jQuery where it is in the first place.

Methinks some people are shooting themselves in the foot.

Re:Like

[omfgnosis]

If you look at a set of the available conditional comments in an editor with proper syntax highlighting, it'll become obvious immediately how they work: everything that is targeted at IE (or specific versions thereof) is technically inside an HTML comment and ignored by other browsers; IE parses these comments and, in the presence of conditional directives it determines (by version) whether to treat them as if outside an HTML comment. Everything that targets non-IE browsers is outside of a comment, but IE will treat it as a comment.

This is a comment, unless loaded in IE versions 8 and below:
<!--[if lte IE 8]><script src="/path/to/jquery19.js"><![endif]-->

This is a comment, unless loaded in IE versions 9 and above:
<!--[if gte IE 9]><script src="/path/to/jquery20.js"><![endif]-->

This is not a comment, unless loaded in IE:
<!--[if !IE]><!--><script src="/path/to/jquery20.js"><!-- <![endif]-->

Re:Like

[jmerlin]

User Agent hacks are incompatible with caching. Bad idea. Use conditional comments so IE gets its familiar old diaper and every other browser gets a shiny new car. Either that or propose an If-User-Agent header so proxies and caches can figure out whether your request is the same as a previous one.

Re:Like

[Kate6]

Thank you -- finally someone who's provided a coherent reason for why conditional comments are preferable.

That being said, isn't any page containing dynamic content of any sort already incompatible with caching anyway?

Re:Like

[Alex Zepeda]

Plenty of tards lead kick ass lives.

Re:Like

[Sephwrath]

Wow that is the most complex troll I've seen in a while. Congratulations sir, I award you one internet... pity that you won't be able to view all of it correctly.

Re:Like

[hairyfeet]

Really? Because i thought the move to mobile was quickly making IE obsolete. After all we are seeing Chrome and iOS everywhere, and the market has been focused on consumers over corporate for a few years now so i figured the days of IE dictating squat were finally behind us, thank the FSM. Was i wrong?

Re:Like

[SolitaryMan]

You see the bit where it says "full API compatibility"? ...

This is the same unicorn as "platform independent". Many people talk about, but nobody really ever saw it.

Re:Like

[jmerlin]

No. The plugins for Apache etc. for PHP tend to respond to requests like this:

REQUEST:
If-Modified-Since:Sat, 14 Jul 2012 03:43:11 GMT
RESPONSE:
200
Last-Modified:Sat, 14 Jul 2012 03:43:11 GMT

To prevent caching, but there's no reason any dynamic page shouldn't be cached unless it's critical that it not be stale. For instance, your front page, even with a listing of your most recent blog postings, is unlikely to change even once per day, so why re-query the database on every GET? One step is to use a backend cache like memcached to prevent repeated SELECTs, another is to set the proper headers to allow browser and proxy caches to cache the output. It can be required that conditional GETs go back to the server so you can determine if a new post has been made since the time of cache, too, allowing you to reply with a 302 to save yourself the need to re-send the entire content again. Out of the box, PHP on Apache might never respect caching, but any site with any amount of hits without an absurd budget for bandwidth and load balancing should be taking advantage of it.

Re:Like

[toejam13]

You can just do this on a Layer 7 switch like a load-balancer.

rule old_browser_rule {
  when HTTP_REQUEST {
    switch -glob [string tolower [HTTP::header User-Agent]] {
      "*msie 5*" -
      "*msie 6*" -
      "*msie 7*" -
      "*msie 8*" {
        pool jquery1_pool
      } else {
        pool jquery2_pool
      }
    }
  }
}

virtual www_80 {
  destination 1.1.1.1:80
  rules old_browser_rule
  profiles http tcp
}

Re:Like

[u64]

saveie6.com: We want to achieve the following:
    Get everyone to use IE6.

Remember that Microsoft was the moron that actually *wanted* the whole world to use IE6.

Re:Like

[man_of_mr_e]

You don't appear to be thinking clearly about this.

If my site uses jQuery 2, and I server 1.9 to IE users, then I want to upgrade to jQuery 2.1, or 2.2, i'm stuck because jQuery 1.9 may not be compatible with 2.1 or 2.2 or whatever later version i'm using. I'm then forced to continue to use jQuery 2 because it's compatible with 1.9.

Re:Like

[jcfandino]

Ok, but doesn't jQuery 2.0 have some new feature not present on 1.9? Maybe my application needs that feature, maybe I can avoid using it, but then, what's the point of using the new version of the library?

Re:Like

[Jane Q. Public]

No, XP is not going to be around forever. However, last I checked (which has admittedly been a while), something like 30% of PC users were still running Windows XP.

I feel it is a big mistake to abandon such a huge user base, even if Microsoft does want to force people to move on. The PC market is not the same as it was 8 or 10 years ago, and lots of people are satisfied with what they have.

Re:Like

[shutdown -p now]

XP just went down below 30% in June (which is also the first month when Win7 is above 50%).

As always, it's a matter of cost/benefit ratio. Keeping support for IE versions below 9 is relatively expensive, because they do way too many things differently from other common browsers, so you need two very different code path for the same things. At some point it does make sense to just split the two, and move new feature development to the one that doesn't has to bother with old IE, keeping the old one on "life support". That's pretty much what jQuery did with 1.9 & 2.0.

Re:Like

[yuhong]

BTW Server 2003 ends support on July 14, 2015, slightly more than a year after XP:
http://support.microsoft.com/lifecycle/search/default.aspx?alpha=Windows+Server+2003

Re:Like

[Jane Q. Public]

"As always, it's a matter of cost/benefit ratio. Keeping support for IE versions below 9 is relatively expensive, because they do way too many things differently from other common browsers, so you need two very different code path for the same things."

I don't disagree with you, but I also don't think you got the entire point. It's not just about browsers, it's about Operating Systems.

It isn't that XP doesn't support IE 8 or 9... it's that IE 8 and 9 don't run on XP.

Many systems that run XP just fine won't run Windows 7 or 8 worth a damn. And a lot of people don't much care... they're happy with XP. So they don't see the need to upgrade both their hardware and software, just to run a new browser. It's probably a bad business decision, in a lot of cases.

So they'll stick with XP as long as they can, and run Firefox or something instead. And when that doesn't work anymore, they'll go with Linux and keep using most of their old tools, rather than shelling out hundreds per machine for newer Windows.

Re:Like

[bpkiwi]

Re:Like

[datavirtue]

Uh...I think his suggestion is easier to maintain. Let's see, change in one place or two hundred? Gotta think about this.....

Re:Like

[makomk]

Yes, it does. If you look closely the first script tag is in a comment as far as non-IE browsers are concerned whereas the second script tag is outside the comments and gets parsed.

Re:Like

[omfgnosis]

Ok, but doesn't jQuery 2.0 have some new feature not present on 1.9?

Nope. That's the point, at least for now. 1.9 and 2.0 have the same API features, but 2.0 loses the cruft necessary to support oldIE, reducing jQuery's footprint and preparing for a time when that cruft is superfluous. Surely 2.x will eventually diverge from 1.9, but probably not for a while.

That said, even if that were already the case, another point of 2.0 is modularity; presumably, as new features for 2.x are introduced, it will be possible to develop modules for 1.9 that support oldIE, allowing 2.x to progress and allowing developers with greater need for backward-compatible support to progress as well.

Re:Like

[jbolden]

No they won't. Lets break this into groups:

Small business -- This segment is cheap. But they are the ones that benefit the most from SaS. They will upgrade to the system that allows them to use SB apps.
Home -- Home has entertainment and rich content needs. They should be moving over already and that's happening with iOS / Android.
Enterprise -- Here the switching costs can be enormous but the problem is software not hardware. Hardware costs are trivial. Windows 7 offers Windows XP compatibility mode. These people can be easily moved.

Re:Like

[omfgnosis]

Three short lines of HTML hardly compares to the dozens of lines of jQuery devoted to old IE support. I'll probably have a snippet in my editor and won't write a thing.

Re:Like

[Ash Vince]

There's already a fair bit of divergence between what you can do in IE and what you can do in other browsers. IE (including IE9) doesn't implement as much of HTML5 as other browsers. IE's SVG support is in its infancy compared to other browsers', and completely nonexistent before IE9.

So we might end up with some jQuery functionality silently being disabled in legacy versions of IE too. Big whoop.

I found some interesting "divergence" on friday afternoon just before I left work.

I wanted to grab an element of html using jquery, then detach it from the page and then append it to the body in a new window I had just opened. I also wanted to do it in a pretty seamless way so that code that wrote text into this pre tag (it is a debug window) would all do so using the same variable. This all worked fine in firefox, but if I tried it in IE8 I got a message back saying "unknown interface" or some such crap. Then as I have the IE developer tools I got asked if I wanted to debug, well of course I did so I clicked yes. That locked my machine up, no ctrl-alt-del, no nothing, just a frozen mouse and me going for the reset switch.

I tried again after having done a slight change and got the same thing. It seems that sending a detached jquery html element between windows in IE8 is a great way to cause a complete OS lockup in XP. There is definitely something seriously wrong there!

Re:Like

[FrangoAssado]

Yes. For any non-IE browser,

<!--anything>

starts a comment, and

<anything-->

ends a comment (if inside a comment). This means that the first script tag (the one loading "jquery-1.9.0.js") is considered a comment in any non-IE browser.

The second script tag is preceded by

<!--anything><!-->

which opens and then immediately closes a comment (in non-IE browsers). So the second script tag will not be inside a comment.

Re:Like

[oamasood]

If you're using a single layout file for your s, then you'd only need to change the jQuery include in one place.

Re:Like

[Jane Q. Public]

Besides, like the other responder, I still don't think you're getting the point.

Many people won't (already don't) want to upgrade their operating system just to run a new browser.

Gaming is less of an issue with each passing month, with companies supporting Linux and OS X more and more.

And as for productivity software: sure, there are enterprise applications that are Windows-only, but those are also becoming more of a commodity. With most of the functionality of Microsoft Office (plus some features of their own), Open Office and Libre Office have made office software not just a commodity, but a free one.

I haven't been in IT for years, but if I were, unless I had proprietary software that I could not run natively, or under Wine or a VM, I would not use Windows at all.

Re:Like

[jbolden]

First off you aren't really presenting any evidence that people would go to those sorts of extraordinary lengths to avoid buying new machines. You are just reasserting that they don't really need software.

Gaming is less of an issue with each passing month, with companies supporting Linux and OS X more and more.

I've been on OSX for a decade now and Linux for 18 years. I don't game but for OSX its pretty much the same, games come to OSX a couple years after they come out for PC. There is much less selection and anyone who cares deeply about PC gaming owns a windows box. I think the real change as far as the gaming market goes is that gaming has mostly moved off of Windows and its less of a reason to upgrade the computer. The casual gaming market thriving though on phones and tablets and this might start driving an upgrades but we are still years from that.

And as for productivity software: sure, there are enterprise applications that are Windows-only, but those are also becoming more of a commodity. With most of the functionality of Microsoft Office (plus some features of their own), Open Office and Libre Office have made office software not just a commodity, but a free one.

Just go to the forum where people are discussing Outlook. People who are heavy Word users don't tend to post on /. but no OO does not support most advanced Word features. I think its great that OO and Google Docs are making a basic suite available to people cheaply. I've installed NeoOffice (a Mac only OO fork) on several family member's computers to meet their needs. But anyone who is going to be using these programs regularly and heavily still likely needs things that OO just doesn't have. The low end office suite is a commodity, the high end is still a pricey piece of software people are willing to pay for.

Re:Like

[TheSpoom]

Two things. One, asking how to do something somehow makes me a bad web developer? Two, according to your siblings your code does not work. Perhaps you should not be so quick to judge next time.

Re:Like

[TheSpoom]

Thanks, I didn't know you could do conditional comments that would flow through unless in IE. That's fairly useful and I'll probably use it now. Still, parsing the user agent string on the backend doesn't use a huge amount of CPU power so I mostly see this as a style thing.

Re:Like

[omfgnosis]

It's not a style thing. UA string is not reliable, there is a long history of vendors spoofing other browsers' UA strings. Many, if not most, end users are able to alter their browser's UA string. There aren't any non-Trident browsers (that I'm aware of) that parse conditional comments, and if they do they would not mimic IE's conditional paths. Conditional comments are the most reliable way to do this, and that's why they exist.

Re:Like

[shutdown -p now]

Anyone not keeping up to date with Vista can be told to keep up to date, and WU will bug you to update IE in any case. The problem with XP is that there's no way for people using it to get IE9+, and it may be hard to convince them to use some other browser.

Re:Like

[Jane Q. Public]

"People who are heavy Word users don't tend to post on /. but no OO does not support most advanced Word features."

Yes, it does support "most" advanced Word features, but I would be the first to admit not all. They are not in the same places in the menus, but most of them are indeed there.

But that's kind of beside the point. MOST people don't need or use most of Word's "advanced" features.

I've installed NeoOffice (a Mac only OO fork) on several family member's computers to meet their needs.

That's nice. And it shows that you are at least a couple of years behind the curve. Open Office and Libre Office have had native OS X versions out for quite a while now. NeoOffice hasn't been necessary for a year or two.

"But anyone who is going to be using these programs regularly and heavily still likely needs things that OO just doesn't have."

I very much disagree. I have been using Open Office (or Libre Office) as my primary office suite for many years now. I regularly correspond with people who use Word, Excel, etc. And I have very seldom had any problems at all.

"... the high end is still a pricey piece of software people are willing to pay for."

Of course. But most of the people under discussion here aren't "high end" at all.

Re:Like

[jbolden]

Yes, it does support "most" advanced Word features, but I would be the first to admit not all. They are not in the same places in the menus, but most of them are indeed there.

Completely configurable line sizes and colours.
Page Style sheets that are linked to each other.
Advanced integrated bibliography handling
Related content management i.e. tying multiple documents together on different axis

etc... Those things aren't at different places they just aren't there.

But that's kind of beside the point. MOST people don't need or use most of Word's "advanced" features.

Not true. As I said above the /. crowd tends to be ignorant of these sorts of advanced features because they aren't heavy office productivity users. Most users do use a few of these advanced features. The problem for porting is they all use different ones. So anytime you do attempt to port people over to OpenOffice they do in-fact experience a loss of features they were using. The cost may justify this loss, but it is really there. Look at the OO forums they are filled with people asking "how do I do X in OO that I used to do in Word". And similarly for Excel and Powerpoint.

That's nice. And it shows that you are at least a couple of years behind the curve. Open Office and Libre Office have had native OS X versions out for quite a while now. NeoOffice hasn't been necessary for a year or two.

I'm aware of their native versions. They are inferior to NeoOffice. For example 18 months ago I was getting frequent crashes with OO's version of Base (medium sized database, 7m records) that I wasn't having with NeoOffice. In OO and LO float windows (ex style bars) float like they should in MS Windows not OSX, a partial port of the interface. And the inclusion of services is important. I can't think of any good reason to use OO/LO over Neo on a Mac.

Of course. But most of the people under discussion here aren't "high end" at all.

Why do you think that back when people got Microsoft Works often for free they still bought Office?

Re:Like

[TheSpoom]

But why would a non-Trident browser spoof an IE user agent unless it was intentional?

Re:Like

[Jane Q. Public]

"Completely configurable line sizes and colours."

This option is there, it is just not easy to find. It is also a little funky in the way it operates... but it IS present, and it DOES work. You can define any text color you want. Go to (in OS X), Preferences | LibreOffice | Colors, then Add, Modify, or Edit.

"Page Style sheets that are linked to each other."

This is in the Format | Page dialogue of Libre Office, in the "Organizer" tab. (I am using 3.5.0, which is not the most recent version.)

Advanced integrated bibliography handling.

Your use of the term "advanced" is ambiguous. It does do bibliographies. But without knowing what you mean by "advanced integrated", I have no way to really address your objection.

Related content management i.e. tying multiple documents together on different axis

Again, I'm not sure what you mean by "tying... together on different axis" means. Are you referring to Pivot Tables for spreadsheets? That feature is most definitely there. If that's what you meant, I don't understand why you think it's not. It has been part of the suite for quite a while.

Those things aren't at different places they just aren't there.

Yes, they are, as I have demonstrated. Except perhaps for the last item, because I am not sure what you meant by it. But if you meant what are commonly called Pivot Tables, yes that is there too.

As I said above the /. crowd tends to be ignorant of these sorts of advanced features because they aren't heavy office productivity users.

But the features you mentioned aren't missing. They are there.

I'm aware of their native versions. They are inferior to NeoOffice.

Maybe on your computer. I can't speak to your own personal difficulties. But when Open Office came out with their own OS X version, I immediately switched from NeoOffice, and I have had no trouble at all since. As mentioned, I am now running LibreOffice (because I was not about to give Oracle any support in that direction) and I have had exactly no troubles with that, either.

"In OO and LO float windows (ex style bars) float like they should in MS Windows not OSX, a partial port of the interface."

I have no idea what you are talking about. They "float" just fine for me, and dock when they are supposed to dock. Dockable windows is a per-program interface decision, and is not unique to either Windows or OS X. I just opened my Style bar and it behaves like pretty much any other floating OS X window that I have used. (The docking part being an option, as I mentioned. You don't have to use it.)

"Why do you think that back when people got Microsoft Works often for free they still bought Office?"

Jeez, dude, you're talking about like 10-12 years ago. Where have you been?

In all honestly, I don't know what you're objecting to. The features you claim are "missing" are right there... just, as I said, maybe not where you are expecting to find them (if I understood what you meant in your descriptions). The glitches you complain about, I have not had. So... our personal experiences have been different, I can accept that. But I do not agree with your "feature" arguments.

Re:Like

[Jane Q. Public]

I almost forgot to add:

Line spacing is customizable, down to hundredths of an inch. If you need finer control than that, I admit that perhaps you need a different program.

The text sizing does leave a bit to be desired, seemingly limited to adjustments of only 0.1 points. But that may be something I haven't found yet, either.

Re:Like

[jbolden]

Your use of the term "advanced" is ambiguous. It does do bibliographies. [libreoffice.org] But without knowing what you mean by "advanced integrated", I have no way to really address your objection.

Here is a link http://office.microsoft.com/en-us/word-help/apa-mla-chicago-automatically-format-bibliographies-HA102435469.aspx

Word will extract any reference citations in a document into a digital form which allows for import / export into bibliography managers. The purpose of this is that many people have a bibliographic library you want an application (extension) to manage that library allowing for attachments... The Word processing application needs to maintain that data though at the document level so that bibliography information stays sinced between different managers.
In addition in terms of styles there are 4300 commonly used bibliographic databases each that has slightly different formatting and information. That's 4300 import scripts that need to be written (custom styles). If you cite from non-Latin based sources the number is slightly over 5000. The Word community has done that leg work, the OO community hasn't.

Related content management i.e. tying multiple documents together on different axis

No that's not a pivot table, I agree OO/LO got those. What I'm talking about is what Office calls tagging documents by "Content Types" (note this is not how OO/LO use this term, they use it to mean essentially MIME type). Its a CRM concept that a document carry metadata automatically as it gets passed around and versioned. So for example if you include an image of a graph you can also include the data that built the graph inside the metadata for the image. For CRMs certain information about a document can be maintained, and forced (see http://office.microsoft.com/en-us/sharepoint-server-help/introduction-to-content-types-HA010149551.aspx). The majority of the user interaction for the CRM stuff would be at the CRM level and OO doesn't have an equivalent of SharePoint, but the real problem is that OO lacks this sort of CRM support at the document level. This allows me to search for "all documents generated by project 2013478" or "all documents generated originally (i.e. first version) out of the California division prior to 4/23/2011"

In terms of lines and colors, you are right that is in Libre office. That story of how it didn't get into OO is nuts.

Anyway good refutation on most of these which shows the degree of progress in the last few years. It seems like LO/OO is progressing at a fast clip even without Sun which is wonderful to see. I have the rMBP and Office looks terrible. If its going to be a year before Microsoft releases a version that supports retina, Libre Office supported retina out of the box and Neo is working on a partial solution with the intention of waiting until OO does the work of switching hundreds of icons to vector so it might be a chance for me to take a look at Libre Office.

The crashing on Base on OO was infuriating. Base is one of the primary reasons I use Neo/OO since there is no Access for Mac. I used write the most, since I use it all the time with people who don't own Office. But If I want to do something I would do in Access on a PC I use that, so for me at least its the primary application in the suite. I've never really regretted being on Neo, even when they lagged behind OO (which was the reason I was trying OO in the first place).

I have no idea what you are talking about. They "float" just fine for me, and dock when they are supposed to dock. Dockable windows is a per-program interface decision, and is not unique to either Windows or OS X.

In windows floating toolbars are supposed to act like document windows. In OSX floating toolbars a

Re:Like

[robsku]

Obvious solution could be server side code to identify old IE versions, provide 1.9 for them and 2.0 for others - and not using any 2.0 specific features, only taking advantage of having faster jQuery lib than 1.9.

Re:Like

[robsku]

Are you a turd?

Re:Like

[robsku]

Never used conditional comments, am a known IE basher yet I know it's possible to do just that, load one or another using conditional comments - I would use server side code still though, if for nothing else at least for cleaner HTML, as you say... and I would not worry about browsers faking to be IE - no browser is set to do so by default, so it's end users own issue anyway.

They seem to be missing the point...

[Anonymous Coward]

I thought a large part of the appeal of jQuery was that it allowed you to avoid writing the crufty legacy browser code that you would normally have to re-implement for every application. If they no longer offer that then what is the point of using it?

IE Version Code Breakdown?

[lavaforge]

Does anyone here have enough knowledge of the JQuery codebase to say how the IE-specific cruft breaks down by version?

IE6 is a monstrosity; that's pretty much a given, and IE7 isn't great either. I could see dropping support for both of those being a big win in terms of cleaning up the codebase. That said, how much do they gain by dropping IE8 as well? It was only released 3 years ago.

Re:IE Version Code Breakdown?

[tobiasly]

Does anyone here have enough knowledge of the JQuery codebase to say how the IE-specific cruft breaks down by version?

IE6 is a monstrosity; that's pretty much a given, and IE7 isn't great either. I could see dropping support for both of those being a big win in terms of cleaning up the codebase. That said, how much do they gain by dropping IE8 as well? It was only released 3 years ago.

The "promoted reader comment" in the linked Ars article actually answers exactly that:

Drizzt321 wrote:
Wow, removing support for IE8? That's a really bold move. I can see IE6 & 7, those are rather old and should be deprecated and people should be really encouraged to move to newer versions.

John Resig has said in interviews that most of the IE6 and IE7 code is needed for IE8 as well. While IE8 has far better layout and CSS engine evidently it still has pretty bad DOM API. Resig pointed out that dropping support for IE6 and IE7 would have very little impact on the size and complexity of the library unless they drop IE8 as well

Re:IE Version Code Breakdown?

[kawika]

As far as bugs and quirks go, a lot of the ones in IE6 and IE7 are also present in IE8. In specific, things like eating HTML5 tags, the lack of true opacity support and the proprietary IE event model that used attachEvent instead of addEventListener. There's quite a bit of code in jQuery to deal with IE event issues, for example the lack of a bubbling change event. IE8 managed to plug some memory leaks and wasn't as bizarre with the "attroperties" issue as IE7 was, but they still have a lot of sins in common.

Re:IE Version Code Breakdown?

[gsnedders]

IE8's big change was all the brand-new layout engine (thus the fairly rapid development in IE9/10, coming from nowhere to being right up there with CSS Level 3 module support). It changed relatively little elsewhere.

Re:IE Version Code Breakdown?

[yuhong]

Yea, IE7 changed very little about the DOM but IE8 did change the DOM to be a bit more Level 1 compliant.

Re:IE Version Code Breakdown?

[yuhong]

In particular, XHTML and DOM Level 2 are 10 years old! Though to support DOM Level 2 in IE9 required major changes to the DOM.

So it's like Python 3

[tepples]

Developers tend not to write programs for Python 3 because most end users who have Python installed have Python 2 installed, and some Python environments (such as Windows) can't easily have multiple versions installed side-by-side. Likewise, developers will continue to use the jQuery 1.9 branch because until April 2014, when Windows XP reaches its end of life, a lot of people will still be using Internet Explorer 8.

Re:So it's like Python 3

[dririan]

some Python environments (such as Windows) can't easily have multiple versions installed side-by-side

The Python installers on Windows install to (by default) C:\Python$MAJOR$MINOR (Python 2.7 installs to C:\Python27, Python 3.2 installs to C:\Python32). For scripts that require a certain version of Python, you may have to open it with the proper version yourself, because (obviously) .py files can only be associated with one program at a time. But for scripts made to be run by end-users, (assuming they don't ship their own Python runtime in the first place) a simple EXE wrapper can use the right interpreter. What part of this isn't easy?

Also, many developers make things compatible with both, which is made much easier with 2to3. Quite a few setup.py scripts use 2to3 automatically even, to be compatible with both using the normal python setup.py install.

Re:So it's like Python 3

[the_demiurge]

No, adoption of Python 3 is limited by library support, not install base. Libraries need to be rewritten for Python 3 and some major ones (e.g. Django) are not yet ready. In this case, jQuery 1.9 and 2.0 will have the same API, but 2.0 will not work on older versions of IE in order to optimize for speed. Your comment about Windows XP is apt; this is akin to software dropping support for older OS versions.

Re:So it's like Python 3

[IICV]

Yeah, but like the article says, jQuery 2.0 is API compatible with jQuery 1.9. This means that regardless of which one you're dishing up, your code remains the same.

It's like if Python 3 and Python 2 had exactly the same features, and you just picked which one worked better on your computer when running a Python program.

Re:So it's like Python 3

[ultranova]

The Python installers on Windows install to (by default) C:\Python$MAJOR$MINOR (Python 2.7 installs to C:\Python27, Python 3.2 installs to C:\Python32). For scripts that require a certain version of Python, you may have to open it with the proper version yourself, because (obviously) .py files can only be associated with one program at a time. But for scripts made to be run by end-users, (assuming they don't ship their own Python runtime in the first place) a simple EXE wrapper can use the right interpreter. What part of this isn't easy?

The part where you need an EXE wrapper to do something the Python executable could itself do: get the version by parsing the first line of the script and delegate to another version if it's requested.

Re:So it's like Python 3

[dririan]

That would be nice, but Python unfortunately doesn't do that. As I posted above, you could still do the same from Python, though. You could check sys.version_info to see if it's the desired version, and if not use Python's built-in Windows registry library to find the right version.

Sidenote: if you wanted it to work with Python 2 and 3 without using 2to3, use this: (sorry, it doesn't want to indent properly, so you'll need to add indentation as needed)

try:
import _winreg as winreg
except:
import winreg

You can then get the install path for the version of Python you want (winreg.QueryValue(winreg.HKEY_LOCAL_MACHINE, 'SOFTWARE\\Python\\PythonCore\\2.7\\InstallPath'), replacing 2.7 with whatever version you want), and run it with your script (sys.argv[0]) as the argument.

Too soon

[knetcomp]

I hate old IE versions as much as every other web developer, but I don't think this is the right way to go yet. One of the main reasons most developers love jQuery is because it allows them to forget about IE quirks and lack of compliance, and just write code. I think it would be better if they continued to support IE in their main branch, but also offer a "lite" version without IE support.

Re:Too soon

[Billly Gates]

I hate old IE versions as much as every other web developer, but I don't think this is the right way to go yet. One of the main reasons most developers love jQuery is because it allows them to forget about IE quirks and lack of compliance, and just write code. I think it would be better if they continued to support IE in their main branch, but also offer a "lite" version without IE support.

Knet, why don't you see 16 bit Windows 3.x apps anymore? Why did windows95 not fail? Why did XP not fail?

The answer is developers said enough is enough and stopped porting to ancient standards. XP should have died years ago and IE 6 will keep staying as long as people think their platforms are fine. Look it runs everything!

Corporations and cheapskates are setting the standard locked forever and IE 8 is 3 years old. But it is what IE 6 should have been and was still far behind. It is time to move on and I am not saying this for the sake of change for something that works.

Mobile devices are demanding HTML 5 and CSS 3 with JIT javascript engines. This legacy shit is a pain in the ass and moves the external costs to the developers. Maybe if internet growth was still desktop oriented but its not. It is a pain to write 2 different websites.Worse Windows 7 comes with IE 8 so these users who do leave XP by 2014 will probably lock their corporate desktops to it until 2020 while we STILL WAIT for iphone 2007 functionality. That is ridiculous.

Re:Too soon

[shutdown -p now]

I think it would be better if they continued to support IE in their main branch, but also offer a "lite" version without IE support.

They did just that. The "lite" version is called 1.9, the full one is called 2.0.

Hate

[Anonymous Coward]

This is horrible - not because i like these browsers, but because so many people use jQ for just that - make web sites work in as many browsers as possible.

Missing the point

[bobetov]

All, one of the big reasons for this move is to provide a cleaner, smaller, faster version of jQuery to apps dedicated to a given platform, such as iOS apps running as HTML5 shells. These applications know their "browser" environment, and thus can benefit from lack of IE support with no cost in terms of audience size.

No one who has a general-public website should use 2.0 for years to come, but if you're developing a quickie Android app, or one of the dozen or so javascript-based desktop shells, etc, then this move is for you.

Re:Missing the point

[frednofr]

For webkit (used by Android and iOS), all you need is Zepto http://zeptojs.com/

IE6 and 7?

[Anonymous Coward]

Find, finally kill the bastards. But 8?! This is the last IE available for XP, which is still widely used in companys....

Good thing

[mpol]

I think it's a good thing. It might even be better if first IE6 and IE7 support was dropped, and IE8 support later, but that might be a lot of work.

I axpect that for the first few years jQuery 1.9 will be the most used version, but you can afcourse load jQuery 2.0 on newer browsers, and only load 1.9 with conditional comments.

I'm wondering how the big CMSs will handle this. Will Wordpress come with 2 versions for a few years?

Re:Good thing

[man_of_mr_e]

More than likely, it's the same code for IE6/7/8 that they're dropping. If you leave it for 8, you may as well leave 6 and 7, because you don't gain much.

Re:Good thing

[Billly Gates]

Too bad I am working on a corporate oriented site. I can imagine IE 6 and 7 will be over 50% usage by the biggest customers who spend the most, even if they are a small minority overall.

I should not have to be reading books on IE 6 hacks in 2012.

Re:Good thing

[Billly Gates]

Issue I have with conditional comments is the bloated javascript libraries are still loaded and now they are bloated by 2x, even if they are ignored by the browser. IE 6 crashes with too much javascript.

For example in IE 6 I can load msnbc.com fine and it still is compatible. But it will freeze up randomly when these annoying social media scripts start up. I have to alt tab and kill it and use a custom hosts file for it to even load half the time.

The costs add up with ISPs now putting in caps and raising bandwidth rates as their competitors die off. 2 megs a view is very expensive and not friendly to those still on dial up vs half that without jquery ... especially 2 jqueries loading up.

So in essence I am in favor of 2.0 but will not use it for awhile for these reasons. If old IE finally dies out by 2014 then I will upgrade. I have a feeling the corporations are terrified of new browsers after IE 6 hell and will stay with 8 until 2020 until developers stop supporting IE 8. It is already becoming the new IE 6 of this decade and MS now has IE on an annual release trek so IE 18 will be out by then.

Re:Good thing

[Billly Gates]

I have to support a lot for a gov site. I don't know what your corporate policy may be, but clients start getting /really/ flustered when I show them the pricing structure for IE support...

I'm willing to go back to IE5, but nobody's ever taken me up on what that would cost. IE7 is an extra 20% on /all/ development costs (plus features being removed or replaced with crippleware) and IE6 is 50% beyond that.

Some of those numbers are to keep up with training on new technologies while having to support relics.

Now, there's a lot of infrastructure and systems in place to try to make this work -- but it's all stuff that takes real time for me to learn, write up, troubleshoot. It slows down debugging, makes the libraries slower, and basically results in horribly 'forked' applications. Or flash.

I realize a boss is a boss is a boss and can say anything they want. But have you tried giving yours an itemized timesheet where they see that "IE6 support constitutes 35% of the total cost of the project" ? Or that "making this little circle draw right here in IE7 actually slows the application by 5000%, and while you can fix it, the resulting fork has tripled your code base and made your 5 minute test suite take an hour?

At a certain point, your /boss/ will start billing IT for refusing to install firefox or chrome, even if it's just for your web app.

At an old job our web app actually got people off of IE5 when the CEO of a fortune 500 dragged their CIO into their office and told them to find a way to upgrade, get firefox installed for an entire department, or resign. Now, that company didn't bill extra for old support (they should have). We just didn't have the resources to get our application working in IE5 (or I wasn't good enough, same thing).

The thing is, your time is probably cheaper than those desktop upgrades. Because as an employee, they think of you as a black-box that gets-the-job-done-per-month. You need to present IE6 (and 7, and 8, and probably 9 for the lack of webgl) as an unecessary operational expense. Just like some apocryphal mainframes in a modern bank where 100 meg hard drives cost thousands to replace, but they still need that old COBOL reporting system from the 60's... IE6 is a liability and a threat to your mission success.

I am creating the startup and I am my own boss. :-). I need vendors and manufacturer support. My plan is to make it simple and probably not bother with IE 6 at all since most places use it for intranet apps. I can not find any good links or books on IE 7 hacks so I use IE 6 ones (though IE 7 is at least aware of the correct box model) so I might as well support it. Sigh

Re:Good thing

[man_of_mr_e]

FYI, anyone that thinks IE6 doesn't support the correct box model is incompetent.

IE6 fully supports the correct box model, and if it's not doing that for you, it's because you're doing something to get quirks mode rendering. If you don't know why that is, you are incompetent to be doing your job. Seriously.

Re:Good thing

[man_of_mr_e]

Lucky you. Not everyone works in that kind of environment. Many of us sell software to clients, and they won't even consider us if it doesn't support IE6. You can't even get to the point of arguing against it, it's in the RFP. You don't meet it, you don't get the job. You don't even get to argue your case.

Your argument is naÃve, and simplistic. Most people who bring their potential client a bill that's 20-50% more will simply get fired and someone that will do it for the regular price will get hired.

Congrats on having an audience that lets' you dictate the terms to them. You have no idea how lucky you are.

Re:Good thing

[yuhong]

Yea, in fact IE6 was the first version of IE where MS focused on standard compliance instead of adding new features, partly thanks to complaints from the WaSP and also thanks to DOCTYPE switching.

Use 1.9 until 2014-04-08

[tepples]

The alternative is sticking with jQuery 1.9 until April 8, 2014, and then flipping to jQuery 2.0.

Re:Use 1.9 until 2014-04-08

[pak9rabid]

The alternative is sticking with jQuery 1.9 until April 8, 2014 [microsoft.com], and then flipping to jQuery 2.0.

Not necessarily. Just because MS doesn't officially support it anymore, doesn't mean it's going to just go away. Unfortunately, in our market (web-based MLS software), we're going to have to support these legacy versions long after MS ditches them.

Re:Use 1.9 until 2014-04-08

[dingen]

No, the alternative is just to keep on moving forward and forget about old versions of IE. Simply don't support them. When enough websites do this, the IE-users will figure out the internet is broken with their browser and move to something which does work.

Re:Use 1.9 until 2014-04-08

[JDG1980]

Not necessarily. Just because MS doesn't officially support it anymore, doesn't mean it's going to just go away. Unfortunately, in our market (web-based MLS software), we're going to have to support these legacy versions long after MS ditches them.

Agreed. In fact, I think there's at least a 25%-50% chance that Windows XP will never go away. What I think will happen is that a substantial number of large corporations and state/local/national governments will still be on WinXP in 2014, and unable to migrate up for a variety of reasons (logistics, legacy compatibility, IE6 ActiveX crap, etc.) As a result, I think there will be heavy pressure on Microsoft to lease the source code for WinXP to trusted third parties for continued maintenance and service. Expect a carrot-and-stick approach from the affected organizations: lots of money offered in contracts if WinXP continues to be somehow supportable, combined with threats that if they have to redo everything to leave XP, they might as well switch to MacOS or Linux instead of upgrading to the newest version of Windows. And, of course, governments have even more tools than this at their disposal if they don't like what MS is doing.

There's a very real possibility that we will see a de facto fork of Windows within the next five years.

Re:Use 1.9 until 2014-04-08

[yuhong]

Yea, MS already has the expensive Custom Support Agreements, and already have Shared Source too that could be extended for this purpose.

Re:Use 1.9 until 2014-04-08

[jbolden]

Windows 7 offers an XP compatibility mode. Microsoft has already handled this problem.

Dojo Toolkit

[MatrixCubed]

Another reason why Dojo Toolkit is more attractive than jQuery. Clients don't care that the JS executes however many milliseconds faster, and they also don't care that the developers have an easier time not supporting older browsers.

What they do care about is stuff that "just works", and being able to add new features at the speed of *click*. Like any tool, if it hinders you from delivering either of these, it's fit for the trip out behind the woodshed.

I'd like to hear comments from other toolkit devs (Sencha, YUI, etc).

Re:Dojo Toolkit

[Anonymous Coward]

That is how I feel about programming C and openssl, libjpeg, X11, tcl, libogg, etc.. For some reason, though, I never seem to be able to get much done.

Re:Dojo Toolkit

[cyber-vandal]

Yawn. Why is it that nerds who generally hate typing are so hung up on reinventing the wheel all the time just to meet some meaningless code purity standard. You might not have deadlines but I do and anything that speeds my work up is a good thing. Shit code in pure Javascript abounds; JQuery has nothing to do with the quality of the developer.

Re:Dojo Toolkit

[steveb3210]

Riight..

Because looking at code littered with var e = document.getElementById('stuff'); is just so much more elegant and non-shit-looking than $('#stuff')..

Re:Dojo Toolkit

[colinrichardday]

And javascript itself isn't a crutch for people who can't be bothered to learn other languages?

Re:Dojo Toolkit

[colinrichardday]

And why do the popular mobile platforms support javascript only?

Re:Dojo Toolkit

[RobNich]

It's a one-time download of the library, and jQuery is not an interpreted language, it's an object containing methods that are used by the developer in writing code.

In any development project, there's a trade-off made between fast development, fast execution, and code maintenance. jQuery handles the first two, and a good Javascript developer can keep the third from slipping.

Re:Dojo Toolkit

[cyber-vandal]

He is actually because I produce high quality code to a deadline. I don't sit there dicking about reinventing the wheel just because it fits some meaningless notion of developer purity. By the way 40KB downloads virtually instantaneously on a corporate network and then sits in the disk cache.

IE8 = "latest" version for many

[tverbeek]

I can see dropping IE6 & IE7, because there's no sound reason for anyone to still be using them. But IE8 is the terminal version of IE for Windows XP, which remains one of the most widely-used operating systems on the planet. It's not going to go away just because someone doesn't want to support it any more.

Re:IE8 = "latest" version for many

[dingen]

The problem is that IE8 handles Javascript in much the same non-standard way als IE6 & IE7. If a library such as jQuery includes support for IE8, it's supporting IE6 & IE7 as well.

Re:IE8 = "latest" version for many

[cyber-vandal]

You can use conditional comments to load the older version of JQuery for IE6, 7 and 8 so it's not as bad as it seems. Web designers have been doing this for years to hack around Microsoft's incompetence,

Re:IE8 = "latest" version for many

[jellomizer]

However my organization just Upgraded to IE 8.

There isn't any plan to be going to 9 in the near future as there are too apps that have issues with following Open Standards.

Re:IE8 = "latest" version for many

[dingen]

That's why the jQuery team is going to provide security fixes for 1.9 for a while.

Re:IE8 = "latest" version for many

[denis-The-menace]

Chrome + IEtab = solved

configure IETab to use IE when chrome hits one of your ass-backwards web apps.

Re:IE8 = "latest" version for many

[steveb3210]

Why do organizations target specific versions of browsers - I really don't understand why its so hard for people to write cross-browser web applications - I do it every day at my job..

Re:IE8 = "latest" version for many

[Billly Gates]

Not the statistics I see on statcounter for the US. China is an outliner that gives a false impression on netmarketshare/appliances etc because they only run pirated software over there. Unless you make Mandrin websites that shouldn't matter.

Windows 7 in the US and Europe already hit 50% and XP is dying fast. There were barely be double digit marketshare by the end of next year. It is time to move on and start supporting newer standards. Look at this way? The iphone came out in 2007 with css and html 5 support for its apps. That was 5 years ago and we still have these issues because it is so far behind and it is common that we should be implementing css 3 and html 5 and use JIT javascript engines to not leave these users behind, but the corps will make us use IE 8 until 2020 if they can.

Browsers should be upgraded frequently like Windows Update. If developers stop supporting like they did with 16 bit Win 3.x apps, dos, netscape, the corps move on. If not we still would be using Windows 3.1 with Netscape 4.7 today. Why? It works fine!

IE 8 in the US already drops down to 11% markethsare on the weekends according to www.gstatcounter.com! It is on its way out and it is time people moved iwth the times.

Re:IE8 = "latest" version for many

[colinrichardday]

What if a developer needs a (standard) feature that a browser doesn't support?

Re:IE8 = "latest" version for many

[dingen]

Then that feature doesn't work in that browser. Doesn't mean the rest of your web application has to fall apart.

Re:IE8 = "latest" version for many

[leenks]

How exactly does one install that on a terminal server session where everything is locked down? And yes, I'm a developer not a member of the admin staff.

Re:IE8 = "latest" version for many

[leenks]

Website? Missing the point. The problems in business are typically "web applications" not websites - either internal custom applications that they have lost the source for, the developers left, or they use technologies nobody has heard of, or enterprise apps that were purchased long ago and upgrade paths are too expensive or disruptive.

Hundreds of staff where I work were forced to stick with IE7 until earlier this year because a single app used by a handful of users in HR wouldn't work on IE8. On thin clients. Why on earth they couldn't provide a separate IE7 TS instance for those users is beyond me. But this is the insanity of corporate IT departments.

Re:IE8 = "latest" version for many

[multicoregeneral]

It's unreasonable to expect the entire industry to stand still because the granny contingent refuses to buy a new pc or tablet. Every other piece of the technology industry is about rapid change. Web development has brief periods of rapid change, and then we stand still for YEARS, waiting for everyone to catch up. It's a huge mistake, and it's holding up progress. The standard needs to be a reasonable four years of compatibility. No more, no less.

There are already issues with Jquery and IE 6 and 7. Take fore example, slideUp and slideDown functions will throw your content aimlessly across the screen (certainly in 6, not sure about 7). Show and hide don't work well either when you're dealing with timing events. 8 has always been my bear minimum standard for Jquery.

Here's the thing though: IE 8 will go away... if every website in the world stopped supporting it. How many millions of sites use Jquery-latest from google?

Re:IE8 = "latest" version for many

[Anonymous Coward]

Except that support for conditional comments has been removed from IE10, which means that you need to include the IE10-supporting-version outside of a conditional comment, which means IE6, 7, and 8, will also try to load that one.

Re:IE8 = "latest" version for many

[marcosdumay]

It's unreasonable to expect the entire industry to stand still because the granny contingent refuses to buy a new pc or tablet.

Those grannies are paying your salary (well, paying the typical web developer's salary, may not apply personaly to you), are you sure you want them to go away?

Re:IE8 = "latest" version for many

[Ragingguppy]

Windows XP is now 10 years old. It was released in 2002. How long do developers have to keep supporting it? Its time we moved on. I say we drop flash too in favour of HTML5 canvas with webGL.

Re:IE8 = "latest" version for many

[Anonymous Coward]

Chrome installs without needing admin privileges (I think - you could always just try it and see), and I assume IETab is an extension for Chrome (only on Windows) that runs IE in a tab in Chrome.

Granted, you might get in trouble for having non-company-approved software installed, but that's a risk only you have the power and information to weigh. You could also talk to the IT department about it.

Re:IE8 = "latest" version for many

[TheCycoONE]

The first part of your comment is true, 10 is dropping them. The second part isn't - conditional comments can be written either to exclude or include all browsers that don't support them (with comment syntax or tag syntax). When using comment syntax IE6-9 can execute a piece of code that all other browsers see as a comment.

Re:IE8 = "latest" version for many

[tverbeek]

"IE 8 will go away... if every website in the world stopped supporting it."

Clearly false. As you noted (earlier in your exercise in circular illogic), there's a contingent of people who will not upgrade just because you want them to. They will (rightly or wrongly) just blame your crappy webapp for not working properly, and keep using the same software.

Re:IE8 = "latest" version for many

[tverbeek]

The question of when Windows XP 1.0 was released is irrelevant, because it was the "current" product for many years after that. The more important landmark is when a replacement (Vista) became available, and more to the point when an attractive replacement (Win7) became available. That's much more recent.

Re:IE8 = "latest" version for many

[genik76]

If you work in a big corporation and have to implement a feature for web site and have to support only a certain browser, for example IE 8 and get something done until 5 p.m in the allocated time., do you think that

a) what about the the users who have installed the other browsers? Should I spend the next four hours doing unpaid work to support those as well?
b) go home and spend time with my family/watch a movie/drink a beer/go to a bar

I think the answer is pretty clear.

Re:IE8 = "latest" version for many

[BeanThere]

I know I know! Because a lot of software developers are, in fact, idiots.

Re:IE8 = "latest" version for many

[BeanThere]

Actually, in most cases those 'grannies' are not the ones paying developer salaries. Which, thankfully, is why such ancient tech is actually being relatively rapidly phased out now.

Re:IE8 = "latest" version for many

[multicoregeneral]

False? Hardly. How many people do you know of in North America that still use IE 6? Not many. The last stragglers are the corporate users that haven't figured out how to use VMware or Virtualbox yet. IE 7 is going down the same path. The reason for all the upgrading? Nothing works in IE 6 anymore. Nothing. Not even sites by Google or Microsoft. Catering to them only makes the problem worse.

Re:IE8 = "latest" version for many

[thsths]

> Windows XP is now 10 years old. It was released in 2002.

That is not the point. The point is that Windows XP was still being sold last year on netbook. To be honest, Windows 7 basic just does not make a lot of sense on netbooks, especially if you have only 1 GB of RAM. So there is no upgrade path, and in any case it would not be worth it.

People are not going to replace brand new hardware just because you tell them to.

Re:IE8 = "latest" version for many

[datavirtue]

Yeah, if they're using IE6 they'll get malware and be shutdown by a drive-by before even getting to your site, so nothing really lost there. IE 7 and 8 will be dead quicker than you think.

Re:IE8 = "latest" version for many

[datavirtue]

Wrong. It is not a problem with open standards it is a problem with IE9 not being configured properly. Hardly anyone in IT these days knows anything about IE browsers. They are basically ignored. Default config for IE9 doesn't render a lot of sites properly. Hell, I've had numerous problems using it on Microsoft's site! The built in backward compatibility so you didn't have to languish with shitty old browsers.

Re:IE8 = "latest" version for many

[datavirtue]

No they will not. They will ask their tech support/tech friend who will tell them to download Firefox or Chrome. They will download it in 1 minute and be on their way.

Re:IE8 = "latest" version for many

[ultranova]

That's why the jQuery team is going to provide security fixes for 1.9 for a while.

Wait, what? JQuery is a Javascript library, which means that neither the client (because it's loaded from a remote machine) nor the server (because it runs in the client machine) can trust it. So how can it possibly have any security to fix?

Re:IE8 = "latest" version for many

[Eskarel]

XP is also an outdated insecure pile, which needs to go as badly as IE 6.

Re:IE8 = "latest" version for many

[julesh]

Because the javascript application itself has to be able to trust it, as it may be used for data transformation of data from third parties (e.g. inserting text provided by a chat service into a web page) that could enable vulnerabilities (e.g. cookie stealing, XSS) if it worked incorrectly. OK, it sounds unlikely that such a problem will surface at this point, but it isn't impossible,.

Re:IE8 = "latest" version for many

[Billly Gates]

XP is also an outdated insecure pile, which needs to go as badly as IE 6.

Agreed. I view this not as an IE problem but an XP one. Which OS runs these dinosaurs? While the dinosaur OS which is XP. Get rid of that and old IE goes away. However, IE 8 wont sadly as it comes with Windows 7 and many corps still believe in locking browser versions down are now locked to IE 8 as they were with IE 6.

Re:IE8 = "latest" version for many

[Eskarel]

It's not as bad though. 7 was a jump, 8 was is a major jump, but once you're working with 8, you're past the "checking for browser not functionality" javascript headache and the upgrade path isn't too bad. 9 to 10 will be a bit of a headache as, however cool Windows 8 may turn out to be, it doesn't seem to be entirely ready for anything that doesn't have touch, and I don't really see an awful lot of enterprises upgrading to it period.

Re:IE8 = "latest" version for many

[shpoffo]

"640K of memory should be enough for anybody"

This won't really affect anything.

[Kate6]

"Full API compatibility" means the same identical code should be able to work with the slower-performing version 1.9 on legacy IE browsers *and* with the more optimized version 2 on IE9 and standards compliant browsers.

All you need is some back-end code to examine the user's browser's "useragent" string and figure out which version of jQuery to serve.

<?php
preg_match( '/MSIE ([0-9\.]+)/', $_SERVER[ 'HTTP_USER_AGENT' ], $matches );
if ( ( count( $matches ) == 2 ) && ( floatval( $matches[ 1 ] ) < 9.0 ) )
     echo "<script type='text/javascript' src='jQuery-1.9.min.js'></script>";
else
     echo "<script type='text/javascript' src='jQuery-2.0.min.js'></script>";
?>

Re:This won't really affect anything.

[flux]

What's the point in using 2.0 in the first place if 1.9 will do? Obviously 2.0 will have some functionality or fixes 1.9 won't. If you really want to do what you suggest, you would still have two code paths in your system, one for 2.0 and one for 1.9 (or possibly one for 2.0 and then gracefully degrade when there is no 2.0).

Re:This won't really affect anything.

[Kate6]

RTFA.

2.0 will implement the same features as 1.9 but in more optimized ways, by relying on browser features not available in legacy versions of IE.  The whole point of having a standardized API is that you can provide different implementations of that API.

Re:This won't really affect anything.

[dingen]

This is the web we're talking about. Especially with so many people on mobiles these days, bandwidth is limited so size matters. Even is 2.0 offers exactly the same features as 1.9, 2.0 will be a heck of a lot smaller, meaning faster downloading, faster loading, faster parsing. Especially on phones, that's not just a nice thing to have, it's a necessity.

Re:This won't really affect anything.

[Kate6]

It's not just that 2.0 won't need to check which hacks are needed to support all the different weirdness of different legacy versions of IE...  It's also that HTML5 has a lot of features that can make much of jQuery's functionality much simpler, and HTML5 isn't really IN IE8 and older.

Re:This won't really affect anything.

[cyber-vandal]

Speed.

Re:This won't really affect anything.

[steveb3210]

echo + html tags is a huge fail...

Re:This won't really affect anything.

[Kate6]

Oh who the heck cares?  My point is that jQuery isn't actually dropping support for legacy versions of IE -- they're just creating a more optimal version of their library for newer browsers.  And I was trying to keep my code example reasonably clear and simple.

By all means, though, go ahead and start a whole sub-thread about what the most optimal way to do the browser detection would be.  There seems to be a lot of commenters who feel it ought to be done using IE conditional comments.

Re:This won't really affect anything.

[furbearntrout]

IIRC, the code path for 1.9 will be longer and slower, but it won't any additional code in your application.

Re:This won't really affect anything.

[FutureDomain]

What's the point in using 2.0 in the first place if 1.9 will do?

If you'll read TFA, they're doing it so that jQuery will be much faster and more modular. If you just use 1.9, you're punishing everyone who has a sane browser. Their code will run slower so that it won't break for the drones using ancient versions of IE. By offering both and choosing server side, you have the speed and support of the new versions and still support grandpa's Windows XP computer. Eventually Windows XP will die a long deserved death and version 1.9 can go with it.

Re:This won't really affect anything.

[dingen]

If you don't use a library, you have to create your own selectors, event handlers, ajax calls etc. It just moves code from the library to your own script, it doesn't reduce the overall size of the code unless you don't really use the library that much.

Re:This won't really affect anything.

[flux]

Indeed the article says 1.9 and 2.0 will be compatible. Yet there will eventually be 2.1, 2.2, and now you're stuck with 1.9 and 2.0, unless those new versions also don't introduce anything that 1.9 doesn't support or unless you're willing to consider the differences and support both in some way. Will 2.x series involve pure optimizations only? A gap will be introduced, and it won't likely take that long either.

Dropping IE8 support at this time is unacceptable

[JDG1980]

I don't have any problem with dropping support for IE6 and IE7. These should have been phased out years ago, and their only reasonable present-day use is not as web browsers, but as legacy applications platforms for badly coded ActiveX apps.

But IE8? For users with Windows XP, this is as high as you can go. IE9 is not supported. And WinXP continues to be supported by Microsoft through the end of 2014, and is still in very widespread use among businesses, and to a lesser but still significant extent among home users.

I wish IE8 would go away (or more specifically, that Microsoft would take the time to backport IE9 to WinXP). Its continued persistence means a lot of css3pie hacks in the website I maintain, which are necessary to get rounded corners and drop shadows in this outdated browser. But it won't go away just because Google wants it to. And saying to use Firefox or Chrome instead of IE is not going to fly in most workplaces, which rely heavily on Internet Explorer's integration with Active Directory and its support for Group Policy. As far behind as IE falls in other categories, no third-party browser matches its abilities on these vital business fronts.

Version numbers causing confusion

[dnwheeler]

I think the primary cause of the confusion is that there are two version numbers. It sounds like the "new" version is dropping IE6-8 support when, in fact, both 1.9 and 2.0 are "new" versions. It would have been better if they shared a version number with some sort of modifier:

2.0 - 2.0 without legacy support
2.0L - 2.0 with legacy support (IE6-8)

Simply changing the version number makes the difference clear while mitigating the worry/panic.

Dewey

[michaelmalak]

In the news from 2013: jQuery adopts the Dewey-decimal numbering system for its versions. The next version will be 1.9.3.2.

Re:Dewey

[Anonymous Coward]

In the news from 2013: jQuery adopts the Dewey-decimal numbering system for its versions. The next version will be 1.9.3.2.

Thats ok.. Long ago Microsoft adopted the Screwy-decimal system for internet explorer and as such started the browser one up version war with Nutscrape. End result, by the time they got to version 6 nothing worked worth a shit in either browser and you had to run both browsers if you wanted to use the net effectively at all.

No doubt Microsoft will screw with something so again the browser wars will piss off internet application coders. You can almost bet it will have to do with multimedia application UIs in silverlight .net coding. Core intranet apps that only work with IE are the bread an butter of Microsoft, html5 cross compatibility with IOS or Android is not in their best interest, otherwise they would have dumped .net and activex years ago!

I know they are touting complete compatibility with html5 but sorry I just do not trust the jerks, we have been down this road before and to my way of thinking it is only a matter of time before they deliberately break things again. Silverlight, activex and .net are not dead yet and you can bet that there is going to be a huge compatibly war with Apple and Google. Don't for one minute think that the assholes in Redmond are going to to give up on trying to dominate the net!

Legacy Branching?

[kiehlster]

Why don't we manage these APIs in a manner like version controls? Maintain a 'core' branch that's compatible with the latest browsers and then have additional 'legacy' branches that break off whenever a browser version is deprecated. I know your first instinct is to argue, "Just install multiple versions of the API," but why not make the API a repository in and of itself? And for those people who want every ounce of speed out of their API, you can just offer a 'core' version of the API that only supports the latest browsers and doesn't include the legacy branches.

jQuery native support

[JDG1980]

This is a bit off of the main topic... but any idea why none of the browser vendors have implemented native support for jQuery in their JavaScript engines? It seems to me that this would be an easy way to improve performance, considering how many websites use it, and considering that implementing native support for its API would remove the need to go through 2 layers of interpreters.

Viewing any web site with IE 8 is unsafe

[tepples]

Just because MS doesn't officially support it anymore, doesn't mean it's going to just go away.

Even if Windows XP itself doesn't go away, old IE will if web sites start putting up wording like this: "As of April of 2014, Internet Explorer for Windows XP is no longer supported by Microsoft. It has known security defects that Microsoft will never fix, through which someone could break into your computer. Viewing any web site with Internet Explorer for Windows XP is potentially not safe. For your safety, we recommend using Firefox, SeaMonkey, Chrome, Safari, or Opera software to browse the web."

Re:Viewing any web site with IE 8 is unsafe

[shutdown -p now]

Most sites don't care about governments and large organizations, and target individual customers directly.

It's a prisoner's dilemma

[tepples]

When enough websites do this, the IE-users will figure out the internet is broken with their browser

But if not enough web sites do this, IE users will defect from web sites that no longer support IE to web sites that advertise that they still support IE. It's a prisoner's dilemma.

Re:It's a prisoner's dilemma

[Billly Gates]

Google was instrumental in stopping IE 6 and 7 in late 2009. Hell 1/5 of the webs traffic was IE 6 in the US (not China) until Youtube, gmail, and others put notices that we wont support your browser anymore.

It worked very well. That and advertisements for Chrome helped too for clueless users who did not know where to get a better browser as statistics show the majority of new Chrome users are former IE users and not FF.

Business will upgrade when people stop supporting it. They left DOS behind, Win 16 apps behind, and Netscape intranet software behind too. If enough people stop supporting it they will leave. It is the opposite of software being sold RIGHT NOW in 2012 that still requires XP and even IE 8 keeps these guys locked in. Not everyone is big like google and yes, they lost some Google Doc accounts to Office 365 for their refusal of IE 7 support, but it works.

People are ignorant and if something works fine and developers take external costs and time to themselves for free they will never leave. Have the cool CSS 3 sites only work in IE 9 or later and people will eventually upgrade.

Re:jQuery freezes at 1.9

[dingen]

Windows XP doesn't have to die for IE8 to go away. Every other browsers' latest version runs great on XP.

Conditionally include 1.9/2.0

[kawika]

True, Walmart.com will have IE 6/7/8 visitors for years to come, but there are plenty of places where someone might be using jQuery but not need IE 6/7/8 support. It could be a cutting-edge public web site, a desktop app using an embedded HTML rendering engine, or a mobile app where oldIE doesn't matter. You can either use jQuery 1.9 for now, or use graceful degradation to give those people a really basic experience and a nudge to upgrade.

EXE wrappers and third-party libraries

[tepples]

But for scripts made to be run by end-users, (assuming they don't ship their own Python runtime in the first place) a simple EXE wrapper can use the right interpreter. What part of this isn't easy?

That depends. Is there an automated tool to create and deploy such EXE wrappers yet?

Also, many developers make things compatible with both, which is made much easier with 2to3

And harder when your program relies on third-party libraries that have not yet been ported to 3.

Re:EXE wrappers and third-party libraries

[dririan]

That depends. Is there an automated tool to create and deploy such EXE wrappers yet?

Not that I'm aware of, probably because it's about five lines of C#. It would also be doable fairly easily from Python as well, but it seems kludgy to make a Python script to find the proper Python runtime then call it on another Python script. Read HKLM\SOFTWARE\Python\PythonCore\$MAJOR.$MINOR\InstallPath (or HKLM\SOFTWARE\Wow6432Node\Python\PythonCore\$MAJOR.$MINOR\InstallPath on 64-bit Windows if you want 32-bit Python), if the key doesn't exist then display an error (preferably with a link to download the proper version), otherwise run $InstallPath\python.exe $SCRIPT.py.

That being said, again, most Python programs will ship with their own Python runtime (with py2exe or similar) if they depend on a specific version like that. Additionally, if you have users with enough smarts to be able to figure out installing Python on their own and don't expect them to change the installation location (or this script is just for yourself), you can obviously use a one-liner BAT file, but I wouldn't expect to see that outside of closed deployments.

I'll agree it's not as simple as it is on Linux distros (although some use python27, and some use python2.7, for instance), but it's certainly not difficult.

And harder when your program relies on third-party libraries that have not yet been ported to 3.

This I fully agree with, and is unfortunately true. However, most libraries on PyPI have been ported to Py3k these days (at least ones with ongoing development). A list of PyPI packages that haven't been ported to Py3k yet can be found here. You could (most likely) easily run 2to3 on other libraries yourself, and fix whatever minor problems arise. It would be harder in libraries with C extensions, but it's almost always trivial for pure Python libraries.

Re:I don't use jQuery, here's why

[steveb3210]

If by 250k you mean 80k, then sure..

-rw-r--r-- 1 xxxxx staff 85260 Apr 30 10:28 jquery-1.5.1.min.js

If you're building any sort of complicated website these days, you will use alot of jQuery.. Do you really want to go write a custom ajax handler? why bother..

Plus, using jQuery allows you to leverage tons of other pre-written libraries which were all written on top of jQuery and will avoid alot of cross-browser bugs and repeated code..

hmm

[buddyglass]

Maybe they were inspired by Google, which declared Chrome 20 would only support only OS X 10.6+, i.e. 80-85% of Mac users.

Re:Should Firefox below 10.0 as well.

[furbearntrout]

IE 8 still uses the legacy JavaScript API and DOM. Cutting 6 and 7 wouldn't do much.

one word

[Johann Lau]

Good.

High-profile computer intrusions

[tepples]

If governments and other organizations aren't willing to move beyond XP

Then their IT departments are acting as if they want proverbial egg on their face. The first post-EOL exploit for XP will result in high-profile computer intrusions at such organizations.

Re:High-profile computer intrusions

[yuhong]

AFAIK Stuxnet targeted Win2000 and was created before Win2000 end of support but was discovered just after it ended support, which means patches to close the holes it exploited are not available without a Custom Support agreement.

modernizr?

[jason777]

I wonder if those older browsers can still be supported with Jquery 2.0 by using Modernizr?

Conditional comments work both ways

[vgerclover]

They are called Downlevel-revealed Conditional Comments, and they are described in the linked article.

<![if !IE]>Seen by everything but IE.<![endif]><!--[if IE]>Seen only by IE.<![endif]-->

You're welcome.

Why?

[WaffleMonster]

I'm having trouble understanding the "speed" argument. Couldn't you provide different code paths for different browser versions and still maintain full performance?

I'm sure this means the size of the library is increased some but so what? aint that what cache is for?

Seems a little odd considering one of the major highlights/selling points was browser compatibility. Write once without worries of platform specific hacks. If you adopt the position of getting rid of old browsers much of those same historical compatibility issues solve themselves.

Re:Why?

[bussdriver]

Gone is the big reason for jQuery. So all the added bloat keeps it afloat while we should be bitching about official support for document.getElementBySelector().

I always wondered why jQuery (last I read the source) didn't implement multiple functions and then dynamically reference the proper function after a test? That caused a minor speed gain for my own compatibility code for quite a long time now.

Of course their other features that are kind of fun do not help matters.

Re:Why?

[dingen]

I'm sure this means the size of the library is increased some but so what? aint that what cache is for?

Mobile is the main reason. Obviously when bandwidth is tight, size matters a lot. And on top of that, phone browsers typically are a lot worse at effectively caching scripts than their desktop counterparts.

Re:Why?

[equex]

I think the codepath selection code is what is getting trimmed here.

Good

[diego.viola]

Good, people need to stop supporting crap if we want the crap to go away.

Bad move jquery

[detain]

You cant run IE9 on older versions of windows can you ? Isnt one of the big reasons people used jquery in the first place was because it works well on most all browsers that are actually used? For any site with a fair amount of traffic you cant afford to just cut out so many of your users. Its true that you will be able to add in of If IE9 load 1.9 otherwise load 2.0 but that seems kinda lame and evnetually theres going to be a larger gap between 1.9 and the current 2.0+ version. I think they should still include support for IE9 and just have it automaticaly handle switchign between the faster and slower code. Either way its way way too early to drop IE8 and maybe even IE6-7

I guess I will be dropping jQuery

[elabs]

Cross-browser compatibility is the only reason I even use jQuery.

When all browsers are equal, why use jQuery?

[elabs]

Browsers are converging, there is no question about it. I don't doubt that as browsers get closer and closer to standards that jQuery gets slimmer and it's bac-end code gets simpler. So... why will we even need it anymore when regular JS and HTML5 work equally well?

Re:When all browsers are equal, why use jQuery?

[Shados]

The functional (ish) programming model, the better handling of events, the abstraction of bug and workaround...

and the plain fact that as much as browsers are converging, they still simply aren't the same. They use different javascript engine, so stuff like Safari and Chrome have quite a few differences.

Jquery's not my favorite javascript library, but it beats vanilla javascript anytime (and thats speaking as someone who spent quite a bit of time in the last few years submitting major patches to more than one of the major javascript libraries out there.)

Re:When all browsers are equal, why use jQuery?

[dingen]

Because the DOM's API is so bad, libraries such as jQuery will always be a blessing from the heavens.

Re:Dropping IE8 support at this time is unacceptab

[yuhong]

Not till the end of 2014, till *April* 2014.

Re:Vary: User-Agent

[jmerlin]

Vary doesn't support grouping of values, so the likely outcome is that you get N caches for each of N unique user-agents. Without nonstandard "smart" behavior, proxies wont be much help, and with it, they become a barrier to change. The first problem that needs to be addressed is that PHP never 302s by default for conditional GETs, the second is transparent proxies. It seems overall to be much simpler to just use conditional comments, as most web servers and proxies do a good job of caching for static files.

Re:Dropping IE8 support at this time is unacceptab

[jbolden]

Why do you think businesses are still on XP? Because people have been tolerant of slow business upgrades. As XP becomes more painful businesses will be forced to migrate. That's a good thing. Vista was released January 30, 2007. Windows 7 ships with an XP compatibility mode to handle legacy. There is no good reason that any business computer at all anywhere should be using XP. One of the ways of making that happen is for stuff to start breaking in XP.

Re:jQuery freezes at 1.9

[dingen]

One of the reasons why Internet Explorer is always the odd one out is that Microsoft only updates their browser with major versions. Every time a new major version of IE comes out it's sort of okay, but it doesn't take long for them to fall awfully behind again as the web just keeps on going forward. Other browsers continually include new functionality, but Microsoft just sits around and does nothing for a few years until they've conceived a new major version. They catch up briefly and start falling behind again.

Re:Dropping IE8 support at this time is unacceptab

[Billly Gates]

Workplaces should actually be the least problematic. You can dictate what version of IE people are using, and at this point any responsible IT department REALLY should have a migration plan in place for getting rid of any legacy XP boxes.

Heck, if a workplace hasn't already moved to 7 by now - they've got bigger issues than worrying about jQuery.

Many are not. Even if they did switch to Windows 7 it comes with IE 8 and corporations paranoid and scared from the IE 6 experience will also lock that browser from being upgraded too and refuse to upgrade until 2019.