We're doing *so darn much* with the Web platform these days, from cross-domain access mechanisms to new drawing and graphics tools. But in the end, we still have to deal with different web browsers. This discussion brings the leads from Mozilla (Firefox), Microsoft (IE), Google (Chrome) and Opera (Opera) together for yet another incendiary discussion about the future of the web.
Arun Ranganathan Mozilla
Chris Wilson Web Platform Architect, Microsoft
Brendan Eich CTO, Mozilla Foundation
Charles McCathieNevile Chief Standards Officer, Opera Software
Darin Fisher Software Engineer, Google
20 year anniversary of the web
Apple not participating in the browser panel...
Including Google/Chrome for the first time this year.
Introducing the panel
Asking why Google is getting into the browser game
Darin originally joined Google to focus on making Firefox better and that evolved into a plan to make browsers in general better
Google always wanted to create an open source browser and did not want to create a new rendering engine.
Took a hard look at Gecko and Webkit
Really liked how Webkit had a nice simple codebase with a lot of pluses
Liked that Webkit just focused on being a rendering engine.. not a full platform
Right now Google is working hard to get involved in the Webkit community
Slowly and surely building the Google reputation in the Webkit community
Cheers for decline of IE6
Talking about how the different browsers/platforms can work together b/c at the end of the day it's all about the web as a platform... if browser makers can work together the web wins and in effect everyone wins
Asking IE guy about Silverlight and why not focus on the open web standards that are already similar to Silverlight...
Chris is talking about how he fights for open web standards inside of MSFT. Said he does not work on Silverlight but there are times when Silverlight makes sense...
Challenges with getting everyone to work together but in the long run it's the only way to make things work for the widest opportunity
Silverlight is not taking over b/c of the adoption of web standards and the fact that the open standards are starting to be able to do what Flash and Silverlight can do.
Big challenge emerging is the fact that even with better modern browers there are still multiple versions of the good guys - various Safari versions, various Firefox, etc. Might not have to target at the macro level but still need to be aware of the micro differences
Talking about HTML5 :-)
Talk of making the HTML5 an open source spec... asking what that would mean to MSFT and other closed source corporations
According to MSFT there is not a problem with an open source spec but it depends on the details about contributing back, etc.
Feeling that current licensing schemes are too restrictive
Scary/nervous if people can actually do whatever they want with the license. If someone were to take HTML5 and fork it and go in a totally different direction that would be very challenging....
Opera guy just said you need to do stuff in the open and publicly otherwise you are stupid. Helps you avoid building with a narrow focus that doesn't actually work for the wider world. You shouldn't develop totally closed if you are building wide ranging specs that need to work for a lot of people/systems.
Javascript is the defacto dev language on the web - atleast when talking at the browser level... getting into discussion about Chrome and JS engines
Javascript and performance wars are the new frontier
Fast engines are good and being released now
People are building huge complicated code bases all in JS and the standards body is trying to evolve rapidly to address this trend
Focus on co-operating to make ongoing changes and improvements... it needs to evolve and grow to move forward.
Compared to 2007 the working group is much more active
A world where everyone used the same rendering engine would be back b/c the needs to make the code inter-operable is import and keeps everyone honest. Developers develop to the spec, not the codebase of the browser engine...
MSFT is talking about being focused on JS - looking at holistic performance and not just a javascript heavy benchmark
Generally speaking everyone is focused on Javascript and everyone seems optimistic about the future right now.
Need JS engines to be fast enough in order to support the next level apps that people are trying to build right now. These ultra heavy JS apps will be the norm very soon.
Opera guy talking about how they built a fast JS engine a long time ago and no one cared up until a year or so ago.
Mobile JS engine is really important... if JS is processing heavy it kills batteries and people are not happy.
Moving on to security and privacy
You need to take security seriously and worry about addressing issues for your users as quickly as possible
Always watching for problems and debating whether to wait for the next big version or make incremental improvements. Need to solve issues fast and protect people, even if it means an emergency band-aid and then improve it with next version.
Google talking about security and their sandboxing concept to make the file system, etc. more secure
Second set of coverage if there is a hole in their rendering engine
Cross domain security is tricky and everyone is looking at how best to address it. You need to be doing a lot all over the place but making sure solutions scale and are always being revisited/critiqued/improved
Right now so many things are being discussed in the spec meetings and standards discussions. It is critical that all players are at the table consistently so no one miss out on emerging trends/issues. Lack of interaction was a problem a few years ago when MSFT and others got out of sync but as of recently everyone is collaborating a lot more in order to solve big problems and trying to find good solutions for the web as a whole. Making the effort is important!
The ecosystem and motivations are always changing... easy to say people should have been focused on things like AJAX, etc. years ago but you never know what is next and when things are going to be big. Important to stay plugged in and addressing trends/desires as they arrive.
Talking about mobile web experience. You want to get the user experience of the mobile browser right and still give developers all the tools they need to build apps that work. You don't want to have to build big desktop app and small mobile app. You also don't want/need the full desktop experience if that is the wrong experience for mobile. Right now UI is the really hard part of building mobile browser. Thankfully the iPhone set a precedent and focused on building a real user focused browser.
Mobile web browsers with standards and full support FTW!
Andre
Copyright 2012 Scribble Technologies Inc. All Rights Reserved.
