Yesterday we launched Mingle, our awesome free iPhone app for New York and San Francisco. The app does a lot of cool stuff with geo-social data aggregation (from Foursquare, Flickr, Instagram and Twitter) and visualisation, and I thought that I’d put some stats together from the data and share it with you.

A quick additional note on the methodology. For the ratio data (most men, women, tourists, locals), I only included venues that had received over 200 checkins during the data collection period. In total, the data consists of about 500,000 checkins for New York and about 250,000 checkins for San Francisco.

Now go check out the Mingle iPhone app, which displays this kind of data in real-time.

Back in March I told you about a couple of projects I’d written in two days over Christmas, and have done nothing with since.

I just happened to be checking our AdSense account for the first time in a while, and noticed that one of the projects – stereotype – was now generating £2.50 a week and gradually increasing each week. It’s not enough to retire (£120 a year… that’s about $200 USD), but it’s better than a kick in the teeth.

The interesting thing about the project is that it’s a wonderful example of self-scaling SEO. I’ve done absolutely zilch in terms of marketing the site or hunting for backlinks, yet the site now attracts 600 pageviews a day, from almost entirely organic search traffic.

The growth of stereoty.pe pageviews since launch, over five months

When I created the site, I invested some time in structuring the page titles to appeal to users browsing Search Engine Results Pages (SERPs), and ensured that the structure of the site always closely linked multiple sections together, so as to create an easily crawlable “mesh” that distributes PageRank evenly across pages.

Most importantly, the site has slowly grown of its own accord. The database was originally populated with a list of 130 countries and 15 topics (women, men, beer, weather). An automated system then searches Twitter for combinations of these phrases (Swedish women, Swedish men, Swedish beer) every day, measures the sentiment, and stores the results.

Pages are only created for combinations that have enough relevant (high sentiment) data. As the data has grown over the last five months, so new quality, non-duplicate pages have slowly been automatically added and cross-linked into the site structure.

I don’t know how many pages have currently been “made live”, but eventually there will be 130 (countries) * 15 (topics) + 130 (country top-level pages) + 15 (topic top-level pages) + 1 (home page) = 2,096 pages. Yes, it could be construed as content-farm like, and yes it faces risk from an ongoing Panda-like update from Google that wiped out my other 2-day project, Fan Ranked. But for the time being I believe it provides some content value that isn’t provided by other websites and is a good example of how to get decent marketing results with little effort.

I made a decision at the start of this year to blog and tweet less. Talk less, do more. Up until now it’s been fairly easy, but I’ve had to interrupt my writing of the last few chapters of the book to write this post.

A couple of days ago Seth Godin wrote about The future of the library. While it is positioned as a love-letter to librarians and the latent potential in a new vision for the library, I see it as a dangerous and (ironically) outdated article.

Throughout our last year of travel, we relied on libraries – from Cardiff to Vancouver to Melbourne. Initially it wasn’t a positive decision to use them, more-so a “I suppose, if we really have to…” decision. But we quickly developed a deep appreciation and respect for libraries and the surprisingly modern services that they provide. So much so, that I now feel it is our personal duty to protect them.

Wherever we went, we could use a library. FOR FREE. We didn’t live there, but we were always welcomed and given warm shelter, free WiFi and a large desk with a convenient power outlet. It seems that many other people were also making use of the service – if you haven’t been to a library for a while and presume that most others haven’t either, you may be surprised.

Libraries are wonderful places to work, and many already meet Godin’s “future vision” of rows of available computer terminals and masses of convenient working areas. Some also have meeting rooms and other facilities that make them even more friendly to the development and exchange of information and ideas.

And yes, they still have books too. I’m currently sitting next to a pile of beautiful library-borrowed books – including wonderfully typeset and illustrated Tufte masterpieces – that are far superior to online equivalents. There’s probably about $250 worth of books in that little pile, and if there’s something specific we need, we can just order it. Isn’t that amazing, when you think about it? We can get any book we want, for free.

Libraries also offer an amazing array of other services, including DVD / Blu-Ray rentals and video game rentals. Mostly for free. As Godin states, I’m sure some librarians resent this “dumbing down” of their establishments, but isn’t this also the most incredible way to get people into the building, together as a community, and slowly inch them towards other activities? Getting people through the door is half the battle, so these services make sense to me. Plus, let me just re-state, they are mostly FREE. You can borrow a $60 video game for FREE.

I’ve also seen libraries with video game consoles inside the library, playgroups, and other non-literature-based services. These have been well used – it was incredible in Melbourne to see kids queuing for the library in the morning so that they could play games that they couldn’t afford to otherwise. And afterwards, they’d often progress to the comics or magazine section that had been strategically placed beside the consoles. Sure, it isn’t Mark Twain, but doesn’t the video game > comic > graphical novel > book progression just make sense to you? It certainly does to me. Well done, libraries.

My main gripe with Godin’s post, however, is that he seems to be almost blaming libraries for a lack of vision. No, Seth, what they’re lacking is CASH. And pointing out that they’re currently crap does not help their fight against cuts. They’re not rolling in money, but instead are facing reduced budgets every year. And you know what you can do with less? Less.

I’d like to return to the start of this post and ask Seth to do something that I’m trying: talk less, do more. It’s fine to point out foibles and wax lyrical about beautiful technical visions. But what libraries really need is support. They ARE heading in the right direction, but what we need to do is encourage people to get libraries back into the zeitgeist and to use them. When people realise how wonderful libraries are we will hopefully see more people willing to demand budget increases rather than cuts, which will help to further evolve their services. But moaning and navel-gazing won’t get us there.

If you haven’t visited a library for a while, I urge you to take 20 minutes out of your weekend and give one a try.

The title says it all, really.

I’m currently running two almost-identical scripts that pull data from the Twitter API every half an hour. The first version of the script runs at :00 and :30 minutes past every hour, and the second version at :05 and :35 minutes past every hour.

The on-the-hour/half-hour script fails (i.e. the API doesn’t respond) about two to three times more often than the 5 minute delayed version. Failed API calls can be identified in the sample screenshot above by 0 values – none of the calls should return a zero value.

So that’s it, really. The most likely reason is that :00 and :30 are peak times for API requests, so you’re more likely to make a request at these times to an overloaded API service. Shift your calls around a bit in the hour, and you’ll get a more robust response. This probably holds true for other APIs too.

I try not to write too much personal stuff on this blog, as I can’t think of more than about three people in the world who actually want to know what I’m doing. Even so, this is going to be one of those posts, albeit a short one. We’re so hectic at the moment that I haven’t had time to update this blog as much as I’d like, so instead I thought I’d explain what I’m up to. Basically, it’s an “excuse blog post“.

We took on a lot of consultancy work last year, especially from August onwards. We got to work with some lovely people and on some excellent projects, but working seven days a week just didn’t leave any time to build our own things. We decided that we would really cut-back on consultancy work after December, and so far we’ve been fairly good at not taking on too much work this year.

At the end of December, I created a couple of 2-day projects (Stereotype, Fan Ranked) to try to get back into the swing of iterative, agile development. Actually, Fan Ranked was relatively successful, given the small investment: it was making about $20 per week in affiliate fees. Unfortunately, it was a little too late. A few weeks after launch, the Google ‘content farm’ update obliterated it from the search rankings (as it should, I suppose), and it now just sits there, gathering about five meagre visits a day. Still, these projects got my development-fu back in motion.

Over the next month, I started to build up a bigger app: Rosella Twitter Games. I made good progress, but soon decided that I should probably finish other commitments before embarking on larger projects. With that in mind, the last month I’ve had my head down with writing Designing Web App Success. I’m about 18 chapters into the 25, and making good progress. I’m REALLY excited about the book – it’s crammed full of good stuff, from business models to personas, typography to testing, development to SEO. I really want it to be The Definitive Manual for web apps. I’m a little fed-up with those books that are 20 chapters of the same thing, full of fluff and inspiration, but no real actionable information. This book is going to be the opposite: about 20 books condensed down into a single ‘need to know’ compendium.

We’re also about to ‘pivot’ the Contentini agency model / website, but I don’t want to say too much about that quite yet. We’ve built some apps under the Contentini brand over the last couple of weeks (e.g. Web Strategy Twitter Trends), so that might give you an idea.

Until next time, whenever it might be.