Category Archives: Industry

Mobile stats for January 2014

Naga IT Services
Filed under: Industry, Responsive

As part of research for something I’m working on, I plucked out some statistics for mobile vs desktop for the world, and for South Africa. The numbers below are a guide to the state of play as at January 2014. The mobile / desktop split is a bit of a false dichotomy, but it’s a useful shorthand for small things vs big things, new things vs old things.

Worldwide web usage is about 24% mobile vs 76% desktop. In South Africa, it’s more like 63% mobile vs 37% desktop. The tipping point (when it was 50 / 50) was around October 2013. It’s important to remember that many users in SA are mobile-only: they don’t have access to a desktop computer with an internet connection.

In terms of the split of operating systems, the largest is BlackBerry with about 34%. Android is at about 20%, as is Series 40 (Nokia’s featurephone operating system). Symbian (Nokia’s newer OS) is about 6%, and Apple’s iOS is at about 4%.

Blackberry and Nokia are the handset leaders, but Samsung is also popular, taking around 15% of the handsets in SA.

Some sources:

Responsive Frameworks

Naga IT Services
Filed under: Industry, Responsive

When I saw the headline for Jen Kramer’s article “Responsive Design Frameworks: Just Because You Can, Should You?” on Smashing Magazine, my first response was worry. I thought that she was going to say how great frameworks are and that we should use them all the time. I quite strongly believe that the answer to the headline question is no (and not just because of Betteridge).

That’s not what she says, though. Kramer’s article is well thought-out, well reasoned, and sensibly pragmatic. “It depends” is the most common answer to software questions for good reason.

Am I saying we should ditch customized solutions for frameworks? No, of course not. A fully customized solution has its place in the Web development world, just as canned CMS themes have their place. If your client has the time and money to achieve perfection and a fully customized solution is a sound approach for the project, then why not?

The price concern is very valid, but even when I was freelancing I wouldn’t use off-the-shelf templates unless I could help it.

Browser support

Another thread of her argument is around not needing perfection for every client, and “Can we get something that’s “pretty good” instead?” (by using a framework). This bit is complicated for me. I completely agree that we can’t achieve perfection, but for different reasons: browser support. It’s interesting (and a little scary) looking at Boostrap’s support table and Foundation’s compatibility table. Bootstrap is actually not so bad: it goes into a bit of detail and explains that some of the features (e.g. CSS3 things) are more granularly supported.

Bootstrap is built to work best in the latest desktop and mobile browsers, meaning older browsers might display differently styled, though fully functional, renderings of certain components.

Of course, it’s all a bit complicated. Some of the support / compatibility table data is about what has been tested and really works, rather than what will be kind of okay. Even having a “supported browsers” list feels a little like it’s sending the wrong message, though.

As builders of things for the web, we have a duty to support every browser, and to build things in a future friendly way. Our sites might not looking as amazing on the stock browser on a Nokia Asha as they do on the latest Chrome on a high-end MacBook (dwsntbeetsieb refers), but it should work on both (and everything in between). We can then use stats and experience to guide us in where we then optimise the experience.


Over the past few weeks, I’ve kept coming back to John Allsop’s classic A Dao of Web Design.

The journey begins by letting go of control, and becoming flexible.

It’s difficult to accept that lack of control, to accept the ebb and flow and move into thinking about proportions rather than pixels, but it’s essential for building sites that will live in our multi-device world.


Naga IT Services
Filed under: Industry, Talk notes

Last week I was at RubyFuza. Hats off to Marc for another excellent conference. The talks, and the people there, gave me a lot to think about. I got lots of ideas for things to do at, or improvements for, RailsBridge and for things I want to do at work.


I can’t remember exactly which talk (or talks) it was, but I was reminded of the importance of network connectivity, especially for South Africa. This mostly make me think of performance concerns: data is money. When people view your site on their phone, they are paying to do so. They’re paying for every extra 1mb png of someone smiling at a salad that you add to your site.

Other people

Another theme I picked up on was “other people.” Pair programming came up in a couple of the talks, and I was reminded of how much I like pairing. I want to investigate how I can bring it into my current job, especially into the design and UX side of things. Where can two people working together on one thing, even for a short burst, add value for a client and for the designers?

Another part of it was about code quality. Writing and maintaining modular code is not just about the reusability, but about sharing it. Other developers should be able to pick up your code and continue where you left off. They shouldn’t have to do that by reading through a huge trail of comments: methods should be well-named and short enough that they can understood quickly.


I like attending conferences. Although I love my job, time away from the office can be useful. Just like when I’m on holiday, I find that my brain unhooks, de-clutches a bit, and I can think about problems from a different angle, or I come up with new ideas that I wouldn’t have otherwise.

It’s also great to see some familiar faces in a different environment, and to meet new people. I have to admit that I was a bit anti-social at this one: I stayed around the people I knew, and I should really have made more of an effort to talk to new people. I hope I’ll do better at the next thing I go to. ScaleConf, anyone? I bought my ticket today.

Thoughts on Ouya

Naga IT Services
Filed under: Industry, Projects

A few weeks ago, I got myself an Ouya (trigger warning: newsletter signup doorslam): it’s a $100 games console, powered by Android. I bought it because I’m an avid gamer, and the market and interface tingles my professional interests.

The console itself is beautiful: it’s tiny, and feels solid and nicely weighted. The controllers are disappointingly crappy by comparison: hollow plastic and cheap-feeling. The buttons are particularly bad, but the sticks are actually quite nice and solid.

User Interface

The dashboard set up is interesting and different, but is a little clunky and immature compared to other systems. I’m sure that this will improve over time as they tweak bits of it. The on-boarding sequence is also kinda cool and not too painful.

The messages and notifications try to be more friendly, less formal, than other consoles, and it mostly works. It makes you feel more a part of a community than the other big name consoles: like people wrote the text rather than the marketing department or a back room developer.

The market

The marketplace is the most interesting thing about Ouya for me. Their USP (Unique Selling Point) is that every game is free to try. That’s quite something. Trials can be downloaded in one click from the dashboard, and the biggest, brightest, call to action on every game’s page is the “Download for free” button.

Most of the games that I’ve checked out are small in size (less than 100MB), so it’s not a problem or a worry to download lots. It seems odd to download from the dashboard, though: do people often want to download a game just based on the thumbnail image and the name?


I have a few problems around the pricing of the games. The first, and biggest, is that you have to dig to find the prices. They are only displayed in a pop up window from the game’s individual page: they aren’t displayed in listings anywhere, or on the Ouya site.

This makes the “Download for free” button feel a little disingenuous (“Download trial for free” or “Try for free” would feel more truthful). It also explains why sales stats don’t seem to be very good: Ouya makes it very easy to download trials, but almost hides away the buy call to action.

The second problem, which is mitigated a little but the first (but not in a good way), is that prices vary wildly: from $1 to $35. Some developers are pricing like the App Store or Google Play, some are pricing like console and PC games. In a way this makes sense: many of the games are crossovers from other platforms, and the prices on Ouya match the ones there.

However, this ties in with the problem of hidden prices and could result in disgruntled users. They may download the trial, come back to buy the game and find the pricing is more like Steam, when they were expecting more like App Store.

Other thoughts

For game developers, I think the market is great. There’s a small (compared to the other platforms) audience of passionate and interested gamers, and interesting things are happening here. I think that there’s a real choice to have your voice heard.

I think that the Ouya is fun and could become a home for awesome indie things, but it needs some changes.