Innovation in Software

Actually, Vagueware Ltd technically hasn’t. Whilst the registered office remains the same, and we remain working in Manchester, our correspondence address has had a tweak to:

13 Crossland Road
Manchester
M21 9DU

Please send any correspondence to us at this new address from now on. Sending correspondence to the old address means it may not be received after the end of this month.

Vagueware is a Rails shop. I have written less than 1000 lines of code in languages other than Ruby (assuming you don’t count SQL as a programming language) in the last 3 years. That’s going to change sooner than expected (hello C my old nemesis), but obviously I have an interest in Rails and Merb and want to see what is happening. Because of workload, I’ve not been able to keep up to date as much as I’d liked, which I’m now redress.

This evening on Twitter Will Jessop pointed me to the “What to Expect in Rails 3.0″ at O’Reilly:

It’s not for the newbie – you need some knowledge of the inner workings of Rails to be able to understand what some of Yehuda Katz is talking about here, but in short, simple terms:

• In case you didn’t know, Merb and Rails are merging with what seems to be a goal of “best of both Worlds”
• Agnosticism in relation to several components, so if you don’t like ActiveRecord and you love jQuery than that becomes simpler
• Rack is the future to give all sorts of goodness you can read about over at the official website or even better this introduction blog article
• Lots of refactoring and performance increases (I didn’t realise quite how big an effect callbacks had on performance)
• What seems to me to be a much cleaner way of handling JavaScript in various parts of your app
• Better Extension support through an API

All good stuff, and the launch later in the year should result in some really interesting changes to some applications being possible.

There’s a reason why less than 10% of readers of this blog use Internet Explorer: it is not innovative, flexible or stable enough for Alpha Geeks. And this blog, if it’s for anybody, is for those of us in society who are constantly looking around the corner snatching a view of the Next Big Thing in software.

The trend-setter in the browser industry over the last few years has been Firefox, with some notable thinking coming out of Safari and Opera. Firefox though is where the really clever stuff seems to happen first, and it’s gaining pace. Everybody is obviously raving about Firefox 3.5 that was released yesterday, but yesterday’s release is of no interest to us.

I was rather piqued instead by Firefox.next (aka “Namaroka” or possibly Firefox 3.6), that Mashable.com pointed out to me. From the wiki:

Namoroka will focus on the following areas:

Performance

Observable improvements in user-perceptible performance metrics such as startup, time to open a new tab, and responsiveness when interacting with the user interface. Common user tasks should feel faster and more responsive.

Personalization & Customization

Simplify the development, discovery, installation and management of browser customization and functional extension. Where possible, provide a custom fit user experience based on a user’s interaction history. Act in the user’s interests, leveraging existing knowledge about their identity and browsing habits.

Task Based Navigation

Allow users to organize their tabs, history, downloaded files, and other resources according to the task they were attempting to accomplish. Provide support for executing common web-based tasks, mash-up style, without having to visit a website.

Web Application Support

Blur the distinction between web and desktop applications, providing web developers with the tools required to create rich application experiences for a user who is connected or disconnected from the Internet. Act as the intermediary between web applications and the user’s OS desktop.

System Integration

Integrate with the look and feel of the host operating system, including data-level interactions with existing system services such as dictionaries.

That’s a pretty lofty set of goals for what is meant to be an incremental release. I’m liking the focus on performance with some pretty solid goals set out in the priority list:

• [P1] achieve dramatic, human-perceivable (>50ms) speed increases on common user tasks
  • opening a new tab
  • loading a bookmarked page
  • autocompleting a location in the Awesomebar
  • play rich media content

Wow. To be honest, I hadn’t noticed it takes more than 50ms to do those things right now but a quick test shows it does take a little longer than I’d thought. Firefox.next is going to be quick for most people then.

What most people are really going to notice though is the UI changes in relation to personalisation, customisation (how cool does “provide a custom fit user experience based on a user’s interaction history” sound?), and the task-based navigation. The ability to see what a user is doing in software and respond to it is something I have spent a lot of time tinkering with in the Vagueware labs, and I’m loving the look of their jumping off point, the new “about:me” page:

Pretty. Of course, this means that for real utility to occur, managing what you delete out of a browser history is going to be important – many users might not want certain sites showing up in here, but will not want to nuke their entire activity if it means the browser has to learn from scratch how they work, and where they work.

Lastly, of course 2010 is going to be the year web applications become mainstream. You thought they were already? Nowhere near. Very few people in B2C or B2B environments right now are developing desktop software, everybody from CIOs to users want the application to live on the web. The drawback through is that not everybody is connected all of the time, so however Firefox.next deals with that issue it needs to do it better than good: it needs to fly.

So what about 4.0? Nobody seems to be talking about that yet, but if you have ideas, that comment form below just loves your attention…