Google announced in May that they would remove IE6/IE7 support in GWT 2.6 (coming December this year) and remove IE8 support in GWT 3.0.
I understand the reasons of this choice, I support it, and I'm not here to argue about it. People that make the effort to stay up-to-date don't want to have a slow GWT because of people who don't.
The main reason we chose GWT technology is because it handles browser quirks for us. The company I'm working at need to have our app running on IE down to IE7. We had to fight hard to have them drop IE6. Last months, employees just migrated their browsers to IE8.
I'm afraid we'll be stuck with the 2.5.1 forever.
IE migration is a broad topic but narrow answers (yes/no) are good enough for me! I probably won't understand details about the inner mechanisms of GWT anyway.
First question: is the word 'remove' appropriate ? Meaning widget won't work with those browsers, or should I say 'stop', meaning that old widgets will still work but patched and new functionalities won't cope with older versions of IE ?
Second & third question:
If this is not possible: fourth question. If you had to start a new application that should run on IE7/8 would you use GWT 2.5.1 ?
We have a medium size project based on GWT in our company; It's a mature software, with more than 100,000 users and has performed well so far. However, GWT technology seems to become obsolete and I personally see no bright future for it, in competition with brand-new client-side rivals such as Angular.
Officially, support is dropped for running the GWT compiler or server-side tooling on Java 7. The GWT distribution is still compiled to run on Java 7 for this release, but no guarantees are made about whether or not this will work. Future versions will compile bytecode for Java 8+.
We finally decided to disable the ie6
permutation in 2.6 by default, and only remove it in 3.0. That means you'll be able to re-enable support for quirks mode in IE, IE6 and IE7 in GWT 2.6. In GWT 3.0, support will be removed though, i.e. classes that support the ie6
permutation will be deleted, and we'll start doing things that won't work in those browsers with no hook to workaround them.
Starting with 3.0, it'll become harder and harder to keep supporting those antique browsers, because nothing will be made to make it easy (nothing will be made to make it hard either, it's just that "we'll no longer care about them"). To keep things simple, I'd say it'll be impossible to support IE6/7 in GWT 3.0, because basic support (DOM-level) will have been removed and other things refactored (if the ie6
permutation was the only one with a specific implementation, it's likely that the hook where that implementation is plugged in will be removed, to make the code simpler and easier to read and maintain).
I don't know yet how we'll handle the IE8 case; probably the same as IE6/7: disable by default in 3.0 then remove completely in the next release.
No.
I'd use 2.6 then (then start looking for another job ;-) )
If you love us? You can donate to us via Paypal or buy me a coffee so we can maintain and grow! Thank you!
Donate Us With