Earlier in the year I was approached by the nice people at Apress, who asked me if I was interested in writing a book. I’m still not entirely sure why they approached me – are all the people who know what they’re talking about tied up with more interesting stuff? – but flattery gets you a long way with me, so I said yes.

Provisionally titled “Pro iOS Table Views“, the book’s going to cover the UITableView and associated gubbins to a level of detail that most iOS titles can’t do. The table view is to iOS apps what reinforced concrete is to architecture – it’s virtually omnipresent, suffers from a perception that it’s somewhat dull, and yet is capable of doing amazing things with a bit of love and attention.

While this probably won’t make me the next J K Rowling, hopefully it might contribute a tiny amount to the sum total of human happiness – if only by reducing the amount of swearing that occurs when an iOS developer meets some of the more – ahem – interesting quirks of the SDK.

The book’s a deep-dive into how the table view works and what you can do with it. The core Apple documentation is good, but only scratches the surface – and it also makes massive assumptions about prior knowledge of design patterns like delegation and data sources.

Huge amounts of really clever stuff is strewn liberally across the blogosphere and fora like Stack Overflow, as well – so the intent for the second part of the book is to pull some of that together to demonstrate Cool Things That Can Be Done (thus addressing the perception problem…)

The process so far is giving me a new respect for technical authors, because it’s bloody difficult to hit the sweet spot between blinding you reader with science, and insulting their intelligence with stuff that’s humiliatingly Janet-and-John. Then there’s the rampant paranoia that somewhere in your code lurks egregious howlers that will cause the entire industry to split their sides with mirth at your pathetic attempt at preventing memory leaks.

At the moment it’s slated for publication in the New Year during MacWorld. Which will explain my increasingly frantic expression in the run up to my deadline before Christmas.

* The post’s title is a P G Wodehouse quote, as well as being a fairly accurate description of my technique…

Tumblr is a fascinating example of how to get user generated and community sites right.   Fundamentally it’s a microblogging platform, but I think this sells it very short of the whole picture.  It’s succeeding by taking the interaction patterns that are part of a whole variety of other services, and creating something which is more than the sum of the parts.

Not only is it a compelling user experience, but it’s also become the basis of some very interesting virtual communities.   Amongst a certain subsection of teenagers, Tumblr has gone viral and is getting a significant amount of their online time – in fact, for heavy users, I’d say it’s overtaken Facebook as their “platform of choice”.

Stripping the platform down, I think you can identify six key reasons why Tumblr’s become so “sticky”.

Ease of use

As a microblogging service, it’s trivially easy to post content via the web interface.   There’s a simple rich text interface for text, as well as single-click post options for rich content like photos, audio and video.   There’s also some Tumblr-specific functionality like quoting existing posts, and embedding chat-like Q&A functions – but all of this is done through a very simple dashboard.   There’s the ability to personalise the appearance of your blog through themes, as well.   The content that gets created seems to lean towards the creative end of the spectrum from what I’ve seen – there’s a lot of striking photography being posted alongside the more angsty outpourings of stereotypical blogs posts.

Community

Although each Tumblr blog is a separate property – each one has a unique URL – they’ve very cleverly baked in some community functionality.   You can “follow” other posters Twitter-style so that their posts show up on your dashboard, as well as interacting with their content either by commenting or “liking”.  Liking seems an incredibly simple but powerful function to me – it doesn’t require the amount of effort that writing and posting a comment takes, but it allows you to flag to someone else that you’ve read their content.  Facebook have made this a central part of their user experience as well, and I think that it works powerfully in both directions.   As the author, I know that my postings are being read and assessed; as the liker, I’ve got a quick way of interacting with people without the overhead of a full comment.   There seems to be a trend going on here of enabling lighter-and-lighter touch interactions – responding to an email takes a large amount of effort compared with posting a quick comment, which is more effort than a simple “like” or “digg” or “vote”.

Recycling content

Reblogging is another facet where Tumblr overcomes the silo-like nature of most blogging platforms.   Reblogging content from someone else’s Tumblr is trivially easy, and is bidirectional – the reblogging shows up on the original post as well as on the blog of the reblogger.   By also bringing through comments that rebloggers make – and threading the responses – it turns a basic blogging service into something more akin to a forum; but without the topic-centricity that forums have.   This is something that has been tried many times through aggregated comment networks that attempt to connect many disparate blogs – Tumblr has an advantage in owning the core platform, but nevertheless it’s still a neat implementation of something that never really seemed to work too well previously.

Competition

Being able to create and publish your own content is incentive enough for a significant number of users, but Tumblr has taken a leaf from the book that Flickr wrote when they introduced tumblarity.   It’s a mystical calculation of popularity based on a “secret” algorithm – half of Tumblr’s users take it very seriously and spend time trying to reverse-engineer the calculations in an attempt to game it, while the other half profess to be above such sordid considerations and ignore it.

Whether you love or loathe this kind of ranking, to many it’s a compelling feature – particularly for the teenage Tumblr population.  Having an external metric of your “popularity” is something that seems to appeal to teenagers – and it’s made a richer process by the way in which tumblarity is also measured across different groupings of blogs – UK blogs, male bloggers, music bloggers and so on.

Reinforcement

Even the most elegant user interface and richest feature set is of little use if use is irregular.   A phrase I hear often from my own tame sample set is “keeping up with my dashboard”.  The flow of content from other bloggers, and the flow of their likes and comments, means that there’s a constant reason to check back and see what’s going on – and then post new content while you’re there.   It makes the site very sticky, to the point of overtaking other service like Facebook and Bebo for some users.

Bridges to the real world

The final piece of the usability jigsaw is the real-world aspect.   The sterotypical picture of bloggers are sad, pasty individuals who hide in bedrooms and don’t interact with the real world.  No doubt there are plenty of those on Tumblr, but there’s also a significant portion of users who DO interact offline, to the extent of meeting in real life.   Tumblr facilitate this by listing real-world meetups and providing schwag – stickers, name tags, badges and so on – that they provide free of charge.   Quite what the real world makes of gangs of teenagers descending on public spaces to meet face-to-face for the first time I’m not sure, but it’s a fascinating process to watch being organised by people who’ve never met other than online.

Having said all this, there are the standard questions of how Tumblr’s business model will fare in the long-term.  It’s a free service, it doesn’t carry ads, and there are no premium options that can be purchased.  It can’t be cheap to operate, either – bandwidth and hosting bills must be significant.  Ultimately this will probably be its downfall, assuming that it can manage to ride out changes in fashion – while Tumblr users are undoubtedly passionate about it, they’re also likely to be fairly fickle over the long-term.  But nevertheless it’s a fascinating case study for the kind of interaction patterns that make a service attractive to those kind passionate users.

After two and a half years, I’ll be leaving Headshift in the New Year.   Put like that, it doesn’t seem all that long, but then again thinking about everything that I’ve worked in that period it feels like a lot longer.   I’ve had the opportunity to work with some incredibly talented people on some utterly unique and fascinating projects, so the decision to go has not been an easy one to make.

What am I going to do?  I’ve got nothing specific planned in terms of jobs at the moment, other than taking a break to pick up on some personal projects that I’ve been playing with for a while.   There are fascinating times ahead – a lot of factors technological, political and societal all seem to be converging together at the moment, so it’s difficult not to get just a little bit excited about some of the possibilities out there.

In the short term, I’ve got three things that I want have a crack at.   The way mobile technology has changed in the last few years is fascinating – the iPhone that I’ve got in my pocket would be almost indistinguishable from magic from the perspective of just 5 years ago.   Actually building apps for the iPhone has a learning curve that’s less steep than vertical – but making the device actually *do* something is incredibly satisfying.  Getting it to do something useful is harder, but at the moment I’m attempting to build an iPhone app for Wordr.  This is a fantastic quirky little site built by Rattle – think Twitter but quicker – and it’s also the perfect use case for a mobile interface.  It also means getting to grips with lots of different facets of Objective-C and the iPhone SDK, so it’s enormous fun.

Then there’s also the plumbing behind the scenes.  I’ve spent a fair amount of time over the last couple of years dabbling with Ruby on Rails, and have probably go to the point where I know enough to be dangerous.  It would be nice to know enough to be safe, so I’m planning to take some time to level up on some of those skills – particularly the way that you can follow a direct path from high-level user specs to low-level code testing with the speccing and testing frameworks that Rails provides.   One of my hobby horses over the last few years has been how bad *most* software development is at delivering something that end-users actually *want* – and the way that Rails can be used to work seamlessly with user experience design is one of the best ways I’ve found of improving the chances of getting it right first time.

And finally, the last few months have seen a real change of attitude around public data.  It used to be that the profit motive was uppermost, and data was treated as some kind of unrefined gold that could be somehow transformed by quasi-private bodies like the Ordnance Survey into something of value.  What actually happened is that these organisations act as a brake on innovation.  Now we’re about to be hit by a tsunami of raw Government data – the people responsible *really* get it.  Raw data is all very well, but it’s visualisation which adds the value – and tools like Processing and the capabilities of cloud computing offer the possibility of doing interesting things with it.   It’s also an excuse to indulge my strange predilection for mutilating innocent cuddly toys – I think perhaps we’re getting to the point of running out of new things to do with a screen and a keyboard, so making data and interaction visible and tangible strikes me as where the real innovation is likely to be.

In some respects, it might seem a bit mad to be doing something like this when the economy is upside down.   Eventually I’m going to have to earn some money again, so obviously I’ll be keeping my eye open for interesting things to do with interesting people.   But if there’s one thing I’ve learnt over the last few years, it’s that these people pop up from the most unexpected places, so I’m not even going to try and guess where or when I might find them.