IMO, it is worthwhile to invest in Ext4. The code I have seen in 3 is not nearly as clean as I have seen in 4, even when written by the same developer. It provides a standard way of building an application and separating code. This will make it easier to on-board new developers as well maintain the code base. That alone is worth it for me, but there are other benefits as well. It is much cleaner using ComponentQueries in your controller than sprinkling refs everywhere or using getCmp/find. I definitely prefer the class system in 4. It is far more powerful than using Ext.extend, Ext.reg, and other methods(Looks like there is a basic 'define' method in 3 now, but it just hides a few of these away). It can handle all your class dependencies and namespaces and so much more. Sencha command can then take advantage of that so you don't have to worry about the order they are in your index or minification script. While Ext3 can look the same, it does not do nearly as much as 4 does and I only touched the surface.
Originally Posted by Grolubao
From a business perspective, older versions of the framework will lose official support and have less mind-share sooner as other developers move on. So when new versions of IE come out after support ends and some of your clients upgrade, your application may not work. Your clients won't be on IE 7 forever, it is already 7 years old.
If performance is a concern and you haven't tried 4 then download the trial. If you are already on 3 and it is major undertaking to upgrade, then maybe wait until Ext5, but the longer you wait, the more versions you are jumping over, and the more difficult it is to upgrade.
That may be true, FOR YOU, but the reality is that global use of IE 7 is currently around 1.5%. Surely you can understand that Sencha isn't basing their decisions on the 1.5% of the world that uses a horribly outdated browser. Further, Windows XP will reach its end-of-life support in April, 2014. Since many (most?) of the people using IE 7 are on XP, they're going to HAVE to update anyway in the next few months. Which means very soon, IE 7 usage will plummet below 1%, and probably to nearly 0.
Originally Posted by Grolubao
Our company is primarily on IE 8. Every four years they lease new PCs and do very little to upgrade browser versions in the interim. Thus, our internal users get a new version of IE whenever they get a new computer. Our IT department has progressed by allowing users to request Firefox or Chrome in addition but that's a tedious manual install.
Our external users probably reflect the general browser trend. Our challenge is that our external users were originally subscription starting in 2001 and we only supported IE. This lazily continued over time. However, we're not yet committed to supporting FF/Chrome due to the fact that we need to test/fix over 10 years of web development that was implemented as IE centric. That will change, but not immediately. Unfortunately we are stuck with IE 8 due to past commitments.
I'm guessing that Sencha has done everything possible to make IE 8 and family as quick as possible. EXTJS is quick on Firefox and Chrome. Performance upgrades are always welcome, but I think that making requests for a faster IE should be made to Microsoft and not Sencha.
Just curious, what do you mean with responsive features? The richness of the components is just the reason why I tend to stick with ExtJs.
Originally Posted by xjpmauricio
To all those IE 6, 7, 8 etc.. users
I faced a similar problem as mentioned often in this thread, that the company (hospital in my case) I work for does not support new browsers. However, chrome portable proved an excellent escape. Unless you company does not even copy paste actions or basic file access, this is a very nice workaround.
9 Sep 2013, 12:12 AM
As you point out, global marketshare is something very different from market-specific marketshare. My experience is that the more you deal with enterprise customers, the older the browsers they use and the more likely they are to mandate IE. The reality is that many software vendors, like the company I work for, have to ship to IE7.
Originally Posted by brian428
On the other hand, there's nobody saying that you have to offer all the latest features on IE7. My employer is already saying for some of our newer modules that they need IE8 minimum. It's perfectly reasonable to have an older version that you provide for those customers on IE7, and have the latest version require IE8. Maybe here or there you'll have some unreasonable customer who wants the latest and greatest on a decade-old browser, but those customers should be willing to pay what it costs to provide that (e.g. backporting to an older ExtJS version).
So, probably IE8 as a baseline is quite reasonable for ExtJS 5. I don't know how much work that saves Sencha (probably most of the work is saved in testing) or how much more efficient it could make ExtJS, but it is probably the right call to make at this point, and a good reason to have a major version number increment.
9 Sep 2013, 12:19 AM
As an aside, why does Sencha keep supporting quirks mode? I understand it is useful when you embed ExtJS into a small part of a bigger legacy page, but are there really that many users in that scenario? Quirks mode must be such a handicap on ExtJS, that I can imagine a lot of improvements could be made by dropping it (or not going for a pixel-perfect rendering in quirks mode). Any thinking on this wrt. ExtJS 5?
Originally Posted by Phil Guerrant
9 Sep 2013, 12:40 AM
As I remember, this means a design/layout, that nicely fits itself into any target device.
Originally Posted by halcwb
Perhaps there's something we could learn from other JS libraries.
Every six weeks: A new stable version of Ember.js, just like in Chrome.
Reality is that most enterprise applications are used by large businesses that are still on IE7. Your 1.5% stat has no relevance to a lot of developers here.
Originally Posted by brian428
However, if you look back a few page, Sencha is aware of the performance issues in IE7/IE8 and they have plans for improvement.