: Is it necessary to browser QA different versions of non-IE Browsers? For example, is there a chance that Chrome 25 will render an element differently than Chrome 26? If the answer is "no",
For example, is there a chance that Chrome 25 will render an element differently than Chrome 26? If the answer is "no", a link to a resource that explains why there are no differences in redering between versions would be ideal.
Edit
Thanks for the feedback. Everyone seems to indicate that, yes, it's possible that something could change in the redering engine. In my original question, I asked for a resource if the answer was "no". Now that the answer seems to be "yes" can anybody point to a specific redering difference between to versions of non-IE browsers?
For example, Chrome 23 renders X as inline, while Chrome 24 reders x as block.
More posts by @Holmes151
3 Comments
Sorted by latest first Latest Oldest Best
Yes there is but there is no way for you to be exhaustive. The best bet is to check the last few versions of each browser with an emphasis on what your visitors use. Even one version of a browser can render slightly differently on different platforms. I have seen Firefox and Safari recently behave differently on Windows and Mac and Chrome differently between Linux and Windows.
Chrome is constantly changing and, while not every version makes it on user's machines, there are literally hundreds of versions out there. The only way to know the differences would be to look at the Change Lists but this is a monumental task.
Chrome has just changed its rendering engine to Blink which is a fork of WebKit. It will take time before most people transition to it but from this point on you have to expect that differences in rendering between browsers will increase as Safari and Chrome will no longer share a rendering engine. Opera will shift from WebKit to Blink later but that will only change the mix and not add variety.
Plenty of Linux distributions build their own Chromium which have a different release schedule which makes for plenty more potential versions. The good news is that actual differences are usually small, so it is better to check various major versions than several continuous minor ones.
Testing the most popular major versions of each browser is a good policy to follow:
Example:
Internet Explorer: 7.xx, 9.xx, etc
Firefox: 3.xx, 4.xx, 13-19.xxx
Chrome: 22.xx, 23.xx, 24.xx, etc
Look at your site's statistics and make sure to test the major versions of each browser that as a whole makes up at least 80-90% of your total traffic. Using your actual user usage data can narrow down the list of required tests to the ones that matter.
Yes, there could be reasons.
Two simple reasons are that not everyone leaves on auto-updating, and secondly there can be and have been bugs introduced in newer versions that broke the functionality of older versions.
Terms of Use Create Support ticket Your support tickets Stock Market News! © vmapp.org2024 All Rights reserved.