betterton.net

Finally, the best of both worlds - offline web apps

I started out as a developer with client server applications (Oracle Forms apps, to be precise). When I moved to the world of web applications like many I never looked back, thinking all the frustrations of client side apps were solved.

They weren't (at least not straight away) but as time has gone by web apps have got better and better from a usability point of view - give me Gmail over Outlook any day. So that old usability argument for client apps is now pretty much dead in the water.

But web apps have always been limited by their need for internet connectivity to function. When you spend a lot of time on trains with intermittent connectivity - or you want to quickly add something to a list without waiting for the server to refresh - that's a seriously big problem.

I'm currently reading HTML5 and CSS3 by Brian Hogan (http://www.amazon.co.uk/gp/aw/d/1934356689/ref=redir_mdp_mobile) and with WebSQL and Offline support, we may finally have bridged web and client apps. I can now develop a web app that has its own local database and assets, and will work just fine disconnected from the server.

This probably won't be news to those who have been looking at HTML5 before, but to me this is truly game changing. Cross platform disconnected web applications are now finally a reality, and I can see why for organisations like the Financial Times they are a serious alternative to native applications.

As web developers we may finally be approaching the best of both worlds and I'm personally really excited about it.

Where there's muck there's brass and Dropbox

Back in 2007 Joel on Software wrote a great blog post on how solving gnarly problems is what makes a business successful. As the yorkshire saying goes, where there's muck there's brass - nobody in our industry got rich solving easy problems.

It's a saying that has stuck with me for a long time. In the social solutions group at Demand we solve three gnarly problems - building scalable community solutions, live blogging events, and moderating community content - and that's what makes us successful. Google solve search, Facebook solve social, Apple solve great design, the list is endless.

But to my mind probably the best example of a company that is making A LOT of brass for solving a gnarly problem is Dropbox. I still remember when I first got a Palm Pilot Personal in 1996 and started encountering the problem of How To Sync Stuff Reliably. The Palm Pilot was good if you wanted to sync with one PC, but syncing to your home and work PCs? Not so much.

Then it was keeping local files synced with files on a Windows file share so I could work disconnected from the network. Windows 2000 supposedly had a solution, but it never worked reliably. Various other solutions have been tried and discarded over my career in IT, but none of them worked - until Dropbox.

It works great as a file storage and sharing system both within our London office and in giving us access to our colleagues files in Austin, and their new Dropbox for Teams offering is a thoroughly convincing solution for small groups. I also use it to keep my Photoshop Lightroom catalog and photo archive backed up and accessible on different computers. It "just works" and everyone I've recommended it to has loved it.

By solving probably the gnarliest problem I've encountered as a computer user, Dropbox throughly deserves the $4bn valuation they recently received. It's also a great lesson that all businesses and projects need to ask themselves - what gnarly problem am I solving?

Why Steve Yegge was spot on about Google+

Recently a Google engineer, Steve Yegge, posted an epic rant / manifesto / resignation letter about what he sees as Google's biggest weakness. He says that with a few exceptions, people there think of Google as a product company. Because they've had such a phenomenal success with some products (search, gmail) and because they're all so clever, they assume they can consistently design the perfect product.

He contrasts this with Facebook and Amazon, who think of themselves as platform companies. They provide platforms and let other people figure out how to deliver products on that platform.

He points to Google+ as a prime example of this. There is still no full API available, so the Seismics and Hootsuites of the world can't add features, we're reliant on Google for Android and iPhone apps, and if you're on another platform like Blackberry you're totally out of luck except for a very average mobile optimised site. (One of my first thoughts when I started playing with Google+ was "wow, I bet the API for this is really cool." My next thought was "oh.")

I think Yegge has hit the nail on the head. Too often in software we think we know our users and audience so well that we can design and build exactly what they want. We're often wrong, which is why successful software projects tend to be ones that start simple and grow based on feedback from their users. It's an approach we reccomend to Pluck customers - start building your community with our out of the box functionality, then learn what your community wants before you spend loads of time building custom features for them. After all, we've learnt from our customers and incorporated what _they_ want in our functionality, so it's normally an excellent starting point. I do hope the powers that be at Google take on board what Yegge says and make the kind of radical transformation into a services based organisation that delivers platforms Yegge advocates. I'd hate to see Google+ fail - our industry needs a serious competitor to Facebook and Twitter - and I fear for Google in the long term if they continue on their current path.