ADF application browser support strategy

840 views
Skip to first unread message

Jan Vervecken

unread,
Nov 13, 2013, 2:58:10 PM11/13/13
to adf-met...@googlegroups.com
hi

Because I wanted to avoid hijacking the recent IE11 discussion topic [1] I created a separate more general but related discussion topic.

Consider you are creating an ADF application for end-users that use a spectrum of modern browsers of which you don't have control.

Consider ADF Faces as a user interface component framework providing a level of abstraction.
ADF Faces, also intended to abstract the application developer from browser-specific issues.
(Correct me if I am wrong, I could not find documentation confirming this.)

What strategy will allow you to sufficiently anticipate browser evolution for your ADF application?

- [1] "ADF 11g and WIN8.1 using IE11"
at https://groups.google.com/forum/#!topic/adf-methodology/NP4g98AvrCw

many thanks
Jan Vervecken

SHAY SHMELTZER

unread,
Nov 13, 2013, 4:16:47 PM11/13/13
to adf-met...@googlegroups.com
Just a note that this would probably be a challenge for any other framework/architecture you use for your application.
So this question is maybe a bit more generic than just an ADF question.

For example, IE8 also broke Google functionality and MS's own outlook web interface (both non-ADF apps :-) )
http://www.theregister.co.uk/2013/10/21/internet_explorer_11_breaks_google_outlook_web_access/

ADF provides quite a good protection from these changes usually, and we do fix issues and certify new browser versions as they come out - so you don't need to change your app code. But when a browser vendor creates a version that differ so much, there is going to be some delay in adopting it.

Shay
--
--
You received this message because you are subscribed to the ADF Enterprise Methodology Group (http://groups.google.com/group/adf-methodology). To unsubscribe send email to adf-methodolo...@googlegroups.com
 
All content to the ADF EMG lies under the Creative Commons Attribution 3.0 Unported License (http://creativecommons.org/licenses/by/3.0/). Any content sourced must be attributed back to the ADF EMG with a link to the Google Group (http://groups.google.com/group/adf-methodology).
 
---
You received this message because you are subscribed to the Google Groups "ADF Enterprise Methodology Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to adf-methodolo...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

shay_shmeltzer.vcf

Jan Vervecken

unread,
Nov 14, 2013, 11:31:44 AM11/14/13
to adf-met...@googlegroups.com
Thanks for your reply Shay.

Sure, it is not only a challenge for ADF applications.
But my question is about approaching this challenge for ADF applications.

One aspect of my question is the "delay in adopting" that you mention. Such delay can be considered twofold, one delay for Oracle to certify new browsers, another delay for ADF applications to migrate to JDeveloper versions that the browsers are certified for.

For the first delay, what does Oracle suggest that ADF application developers do to "bridge the gap" between uncertified and certified?
For the second delay, how can ADF application developers know which browser version will be certified (or will NOT be certified) for specific versions of JDeveloper?

many thanks
Jan Vervecken

John Flack

unread,
Nov 14, 2013, 12:14:20 PM11/14/13
to adf-met...@googlegroups.com

Don’t talk to me about IE8 – I had a .png image on a page and it looked great in IE7, IE9, IE10, not to mention Firefox, Chrome, Safari and Opera (which isn’t even officially supported).  So I deployed for user acceptance, and immediately heard that they couldn’t see the image.  Turns out the image had a transparent background to let the underlying background color show through and the users used IE8 which doesn’t show .pngs with transparent backgrounds.

SHAY SHMELTZER

unread,
Nov 14, 2013, 1:16:30 PM11/14/13
to adf-met...@googlegroups.com
Certification means that we ran a set of tests. Supported means that you can file bugs and we'll aim to fix them.
So far when we released a new version of ADF we certify with the latest available browser at that time - I think it is safe to assume we'll be continuing this practice.

If you look at the certification/support doc for browser you'll note that for most browser we have a + sign next to the version.
This means that we tested the numbered version, but we support higher versions as well.
This is because most browsers have been quite good at doing upgrades that don't break older apps.
This is something we can't say for IE though - this is why you don't see a + sign there.

If you think there is something more that Oracle can do on our side it would be good to hear.

In terms of what developers can do - test your application on the new browser that came out and let Oracle support know if you are running into any issues.
This will allow us to fix those in the upcoming version/patches.

Shay
shay_shmeltzer.vcf

Sten Vesterli

unread,
Nov 15, 2013, 8:56:14 AM11/15/13
to adf-met...@googlegroups.com
What we can do as ADF developers: Create a set of automated regression tests so we are able to move easily to later JDev versions. I see organizations running two or three different ADF versions concurrently, because they don't have the resources to perform manual re-testing after a version upgrade.

If you invest in building a regression test suite that can run automatically, you can easily recompile your application in the latest ADF version in order to get the browser support you need.

Best regards

Sten Vesterli

Mark Robinson

unread,
Nov 15, 2013, 2:40:01 PM11/15/13
to ADF Methodology
I think another challenge of upgrading ADF is that the ADF version is tied very tightly to JDeveloper.  This JDeveloper version is then tied to other things like the supported version of WebLogic and SOA Suite.

So it's not too unreasonable for a large company to get caught in a really nasty web where they have to upgrade everything all at once to upgrade anything.

Mark


Jan Vervecken

unread,
Nov 17, 2013, 10:22:20 AM11/17/13
to adf-met...@googlegroups.com
Thanks for your reply Shay.

So, does the following assumption, match the assumption that you suggest we can make?
There will be a next version of 11g R1 and 11g R2 and 12c that will certify with the latest available browser at that time.

Maybe the "missing + sign" (currently for IE) is what is causing a "gap", between uncertified/unsupported and certified/supported, which is a reality for ADF applications, but at the same time something that Oracle doesn't seem particularly concerned about (other than getting a browser certified a.s.a.p., which is good).
If you ask what more Oracle can do, maybe it could try to help ADF application developers bridge that "gap" (with features in the framework, temporary workarounds, good information of which version will be certified, ...).

regards
Jan Vervecken

Jan Vervecken

unread,
Nov 17, 2013, 10:24:22 AM11/17/13
to adf-met...@googlegroups.com
Thanks for your reply Sten.

Although easing migration is an important aspect of a good "browser support strategy", the idea of automated regression tests is itself a more general topic.

As it is also an interesting subject, I created a separate topic, "automated regression tests for ADF applications"
at https://groups.google.com/d/topic/adf-methodology/mbaWcoY7ww8/discussion

regards
Jan Vervecken

Jan Vervecken

unread,
Nov 17, 2013, 10:24:59 AM11/17/13
to adf-met...@googlegroups.com
Thanks for your reply Mark.

Thank you for pointing out important dependencies that make the idea of "bring your ADF application to the latest version of JDeveloper for up to date brower support" sound a lot less lightweight.

regards
Jan Vervecken

Timo Hahn

unread,
Nov 17, 2013, 12:20:33 PM11/17/13
to adf-met...@googlegroups.com
Shay, 
opening a SR is a good approach, in theory. However, we opened a SR on this matter and all we got from support was the simple 'IE11 is not supported' message.This we already knew. 
The intention of the SR was to get some kind of statement from Oracle giving us the feeling of getting help at some point in the future. Nobody make Oracle responsible for this mess.
I know we don't get actual dates when such a problem gets fixed, but as Frank answered on the other thread I opened on the ADF EMG, it's good to know Oracle is aware of the problem and working on some kind of fix.
We may control our development environment, but controlling the internet users environment is not feasible. 

Timo

Jan Vervecken

unread,
Dec 14, 2013, 5:59:43 AM12/14/13
to adf-met...@googlegroups.com
fyi

In Oracle service request 3-8091598761 we got this feedback from support :
"... After a discussion with product management I doubt that there will get any change on the way it is being done now. ..."

Also, I hope Oracle understands that things like this can change the perspective from Oracle customers on Oracle ADF.

regards
Jan Vervecken

raman nanda

unread,
Dec 14, 2013, 6:02:25 AM12/14/13
to adf-met...@googlegroups.com
Hi,

This happens on a lot of service requests btw :-)

Regards,
Ramandeep
Reply all
Reply to author
Forward
0 new messages