Linked by Thom Holwerda on Wed 3rd Apr 2013 22:27 UTC
Google It's apparently browser engine day today. After Mozilla and Samsung announcing Servo, Google has just announced it's forking WebKit into Blink. Like WebKit, Blink will be open source, and it will also be used by other browser makers - most prominently, Opera has already announced it's not using WebKit, but Blink. Update: Courtesy of MacRumors, this graph illustrates how just how much Google contributed to WebKit. Much more than I thought. Also, Chrome developer Alex Russell: "To make a better platform faster, you must be able to iterate faster. Steps away from that are steps away from a better platform. Today's WebKit defeats that imperative in ways large and small. It's not anybody's fault, but it does need to change. And changing it will allow us to iterate faster, working through the annealing process that takes a good idea from drawing board to API to refined feature."
Thread beginning with comment 557612
To view parent comment, click here.
To read all comments associated with this story, please click here.
RE: Is Blink Open Source?
by Neolander on Thu 4th Apr 2013 15:14 UTC in reply to "Is Blink Open Source?"
Neolander
Member since:
2010-03-08

It's always amazing how much NetApplications and StatCounter stats can differ...

http://www.netmarketshare.com/browser-market-share.aspx?qprid=0&qpc...
vs
http://gs.statcounter.com/#mobile_browser-ww-monthly-201203-201303

The interesting thing about StatCounter, though, is that they did publish a detailed explanation of why it is the case in their opinion. And if what they say is true, NA's metrics do indeed suffer from some major methodology issues.
http://gs.statcounter.com/press/open-letter-ms

Edited 2013-04-04 15:15 UTC

Reply Parent Score: 3

RE[2]: Is Blink Open Source?
by Tony Swash on Thu 4th Apr 2013 16:50 in reply to "RE: Is Blink Open Source?"
Tony Swash Member since:
2009-08-22

It's certainly hard to find a way to objectively and accurately measure mobile web usage by platform and so it is hardly surprising that NetApplications and StatCounter stats differ. One way to try to work out which of these estimates is more accurate and which is the outlier is to look at other sources and methods of measuring web usage by platform. I am thinking of these sorts of data sources,

http://allthingsd.com/20130311/watching-a-video-on-your-phone-youre...

http://www.fortune3.com/blog/2012/11/cyber-monday-2012-ecommerce-st...

http://blog.gogoair.com/?p=193


I would say that the balance of the reports of mobile OS web use using different measuring systems is that iOS over performs in relation to device numbers and Android under performs in relation to device numbers.

That in itself is not particularly surprising when taken in the context of the mass of separate independent data sources that show that iOS over performs as a general computing platform and Android under performs as a general computing platform.

One key facto in all this is of course the iPad, which is the mobile device most used on a per capita basis for mobile internet usage, and which continues to sell in very high numbers compared to other tablets. Android tablet sales continue to be weakish even if you include forked Android platform such as the Kindle.

Reply Parent Score: -1

RE[3]: Is Blink Open Source?
by Neolander on Thu 4th Apr 2013 18:25 in reply to "RE[2]: Is Blink Open Source?"
Neolander Member since:
2010-03-08

I'd say that you do have a remarkable talent for picking examples that mainly target the core iOS demographics. That's perhaps not so surprising, though, since as an Apple user you probably belong to one or several of those demographics yourself.

Gogoair only provide services to large, and thus highly expensive air companies. This means that their data will have a bias towards people which have lots of money to spend in plane tickets. Perhaps an extra $20000 of average household income in the US could have an influence in this area? http://techcrunch.com/2012/09/26/forrester-iphone-app-users-young-a...

Another thing worth noting about Gogoair is that it solely provides service to northern american companies, which may tend to artificially inflate the specificities of the american population. Since a picture is better than a thousand words... http://gs.statcounter.com/#mobile_os-ww-monthly-201303-201303-map

The Oolaya example chosen by AllThingsD is also interesting, since that company does not take care of any major video website like Youtube, Dailymotion, or Vimeo, instead providing smaller "inline video" services to various online newspapers. Depending on how said services are implemented, it may well happen that they do not get the full share of Android video views because those are displayed a Flash video player without ever seeing Ooyala tech (remember, the Flash Player plugin is still installed on a large number of Android devices).

Also, this selectivity towards use cases where video is a "bonus content" rather than the main thing that people want to see may again bias towards the iOS user base, but experiment would be needed to check if and why this is the case. All that I say for now in this area is that specific examples may come with specific results.

And then there's Fortune3, which is nothing less than an e-commerce company. Bias towards e-commerce was one of the main criticisms of StatCounter regarding the methodology of NetApplications, and in the specific case of mobile ecosystems (where, again, iOS users tend to have much more money to spend), it's easy to see why it would be a problem.

Just as an aside, I don't think that no matter what each of us says, we are ever going to agree on this matter. You have an Apple agenda that you feel an unhealthy need to push no matter how ridiculous it gets, and myself I find that company remarkably irritating for various reasons ranging from moral opinions to design philosophies. Because of this, and because we manipulate data that is as hard to measure, and thus far away from hard facts, as usage statistics, it appears to me that this discussion cannot go anywhere. So perhaps we should just switch to some other conversation subject which we can agree on, such having a good laugh at Microsoft's attempts at going touchscreen.

Edited 2013-04-04 18:26 UTC

Reply Parent Score: 5

RE[2]: Is Blink Open Source?
by Lennie on Thu 4th Apr 2013 23:30 in reply to "RE: Is Blink Open Source?"
Lennie Member since:
2007-09-22

There are a few things you need to do keep in mind when comparing hitslink and statcounter:

- I always get the impression Hitslink has more business-like websites as their customer, statcounter has less of that

- statcounter measures by pagevisits, hitslink by sessions. So if Chrome users are heavy Internet users by visiting many, many pages, they will show up higher

- Hitslink changes their statistics based on the ratio of users in a certain country. So if they measure on their US-customer websites that a lot of users from China use IE. They will proclaim a lot of users worldwide use IE.

Reply Parent Score: 4

RE[2]: Is Blink Open Source?
by PresentIt on Sat 6th Apr 2013 17:21 in reply to "RE: Is Blink Open Source?"
PresentIt Member since:
2010-02-10

The interesting thing about StatCounter, though, is that they did publish a detailed explanation of why it is the case in their opinion. And if what they say is true, NA's metrics do indeed suffer from some major methodology issues.


Actually, both sites have sucky stats. StatCounter doesn't weigh statistics from different countries, so a hit from someone in the US would count as, say, 50 hits in China.

StatCounter even admitted that they used flawed methodology, as did NetMarketShare a while ago.

Don't trust either of them.

Reply Parent Score: 2

RE[3]: Is Blink Open Source?
by Neolander on Sun 7th Apr 2013 06:39 in reply to "RE[2]: Is Blink Open Source?"
Neolander Member since:
2010-03-08

I don't understand this geo weighting thing, actually: surely, these services would know where each hit that they receive on a website comes from, right?

Edited 2013-04-07 06:47 UTC

Reply Parent Score: 1