WebKit browsers tested for compatibility 17
Quirksmode measured the CSS and JavaScript compatibility of 19 different WebKit based browsers. The conclusion? All 10 mobile WebKits the study's author has identified are subtly or wildly different, meaning there's no singular "WebKit on Mobile". Quirksmode sums that thought up, saying "This is not consistency; it’s thinly veiled chaos."
Indeed. So how does Palm's WebKit-based browser compare to the others?
Please note while taking in the following data that the testing was done for compatibility. Other factors, notably user interface and performance, have been left out of the equation.
Here's the criterion for inclusion of the tests used in the study:
My main criterion for test inclusion was that a certain method, property, or declaration must be supported by at least two WebKits, but not by all. I made an exception for geolocation, since it’s so totally crucial for the mobile platform. (Geolocation is currently supported only by iPhone 3.1)
Here's how the Pre's browser measures up against some of its biggest WebKit based competitors:
iPhone 3.1
iPhone OS 3.1 does a bang up job as far as standards compliance is concerned, certainly leaving some space for improvement for webOS. iPhone OS3.1 even scores 100 on the Acid3 test, a test that WebKit on webOS is incapable of even taking.
G2 (Android 1.5)
It's a similar story for the G2.
G1 (Android 1.0)
And it seems that the Pre's browser just edges out the G1.
Here's how all of the WebKit based browsers tested compare:
It looks like Palm certainly has some work to do in terms of web standards compatibility on its flavor of WebKit, but it looks like it gets many of the important ones right.
And to reiterate, lack of compliance doesn't necessarily correlate to a lesser user experience, as Quirksmode points out:
S60v3 WebKit, which is by far the worst-scoring one, at least has a workable user interface, something that much-higher scoring WebKits such as Iris don’t.
For more on how the different variations of webkit based browsers stack up, check out the full study.
Thanks James McP
17 Comments
too bad it doesn't explain what each of them do
Well you can take a guess at some and Google search the others.
Like the location one. For example, going to the m.flickr page & searching for nearby doesn't do anything on the Pre, but on mobile Safari I assume it would work.
It's basically just what CSS capabilities it supports (Cascading Style Sheets) which is used to layout the elements on a page and also how you can select those elements via pseudo selectors. http://kimblim.dk/css-tests/selectors/
Sorry but that's incorrect. It's testing DOM compliance, not CSS compliance.
Someone should update this to explain it. I agree. I dont want to waste time looking around the internet for it. Also it would be awesome to see the Pre start updating these capabilities!!! Keep us update Precentral!!!! Thanks!
I respectfully disagree and think that this (and other) articles that provoke a bit of thought and act as a launching pad for more self directed search are exactly what I want from my news media. I don't want to be spoon fed analysis from this or any other news source. This is a good place to start.
Now that said. I have NO idea what half of this means :-) best get's to Googling...
nb
I disagree with your disagreement. It's appropriate to consider one's audience. In this case, much of the audience is likely to find many of the terms unfamiliar.
It's fine to expect the audience to look up one or two unfamiliar words, but in an article of this length there shouldn't be quite so many uncommon, unexplained, technical terms.
What it means, in general, is that the perception that being made with WebKit means a browser will be standards-compliant is false. There are versions of WebKit that do very poorly in testing.
What it means for most people is nothing. The most common browsers (IE and Firefox) are generally less compliant than the desktop WebKit variants (Chrome and Safari). If most browsers are ignoring the standards anyway, then those that follow them can actually be at a disadvantage (like when a site designed for IE won't display properly in a WebKit browser).
However, what it means to a Pre owner is that Palm needs to improve it's browser. There are lots of mobile sites that are built to work in the iPhone browser. By not supporting all of the standards and features that the iPhone does, you may have troubles accessing certain mobile sites. For example, if you've ever used Google Reader on the Pre, you know that it doesn't behave like it should.
Does Webkit include a Javascript engine? Or does each implementation roll its own? If the former, then these results are pretty surprising. If the latter, then saying both browsers are Webkit-based doesn't mean anything if you're comparing Javascript engines.
I do believe it has a core js engine but others can use their own js engine such as Chrome uses V8 for js and Safari uses SquirrelFish(though i "think" SquirrelFish is webKits core js engine).
After much work I've "produced" a page that explains all about the quirksmode tests.
http://www.quirksmode.org/blog/archives/2009/10/there_is_no_web.html
(IOW, if you RTFA as provided above and click - *once* - to see the actual comparison, every one of the test cases is explained in detail, including a description of the desired behavior, some common failure modes, and sample code.)
Took HOURS of searching to find that...
WebOS's browser still doesn't support anchors (Go to top of page) etc. And that's kind of silly, because it should.
I cannot believe the timeliness of this article. In the past two days we have been testing HTML5 and other forms of local storage on various mobile devices for a new project. We were shocked to learn that, while a significant level of HTML5 local storage is supported in webOS itself, only cookies were supported in the pr?
I agree that palm has some serious work to do in this area, and amzed taht they did not implement the same technology that they used in their OS in their own browser.
BUT lets not forget WebOS is till in 1.2 update and coming from Palmwhich has been outstretching thier resources from the day they have visualized WebOS. Lets give it more time. I can only assume that with their resources that they are working on priortized issues and i am happy this is not on the top of their list. it is still above Saf 3.0 and androind 1.0 and just below IphoneOS2.2
One thing I'd like to know - were they testing the WebOS, or just the browser app in the Palm Pre? They're both webkit based (I believe) but may not be exactly the same in how they implement it.
I'd like to see a test of WebOS to see if it is fully compliant and if it also scores a 1 out of 100 for Acid3. In other words, the HTML, CSS, javascript environment for scenes, stages, etc.
I may get flammed for this but for mobile devices web browsing should be basically eccentric with taking their time with the flare to make sure it is right. Any more just bogs the device down and drains the battery more.
I would like to see the geolocation put to uses. This would be good to have when on Flickr.