Traversing the road to success.

Once every few years, a new sic-fi technology gets to the point where it can become massively adopted and change the technological landscape forever. Microsoft released the Kinect barely one year ago, but setting a new standard for interaction in the gaming industry and breaking the Guiness Record for becoming the fastest selling gadget ever was just the tip of the iceberg for this revolutionary device.

Microsoft left the USB connection to the device opened by design, and the development community rejoiced. Several Kinect hacks appeared during this year and some development communities are starting to gain momentum: Open Kinect, OpenNI (supported by Prime Sense, who holds the patent on the Kinect’s chip) and even Microsoft itself is funding an accelerator program for startups that use Kinect for other commercial uses.

Far away from Silicon Valley, in a garage in Argentina, three techies saw the upcoming wave and dared to dream. Why not using the Kinect for analyzing people and giving brick and mortar stores something like Google Analytics? A new concept was born, gathering automatic data on the behavior of real flesh and blood visitors. This concept was named Shopperception.

Shopperception allows to gather information that was impossible to know up to now. Visits, average times, locations, grabbing products from the shelf (conversions), putting them back (returns), hot and cold shelf zones, just to name a few. Shopperception is the first objective way a real life store can perform an A/B test to understand the impact of a change in a shelf before deploying it into hundreds of shops.

Shopperception is to the Retail Industry what Google Analytics was to the web.

But, why are analytics and A/B tests so important? The web is what it is thanks to the analytics and A/B tests, that’s why. If you don’t know what I am talking about, then you never asked yourself what happened with those bouncy icon infested sites where you could never find what you where searching for, or do what you wanted to do. The web today is more natural. Buying a book in Amazon is a simple experience, and so it is looking at your friend’s birthday pictures in Facebook. But, how did they do it? Great websites are not oracles of user behavior, they gather data about their users all the time, and use that data to improve what they offer. The result? More users, more sales and a better usage experience for you and me.

But, should we start worrying about our privacy? No. Shopperception is anonymous, as all website statistics are today. Agile Route takes your privacy very seriously. Shopperception does not allow the shop owner to know what *you* bought, but it does allow him to know how many people grabbed the chips, put them back and then reached for the oats. It allows him to serve you better, perhaps by displaying the healthy food closer to you and provide you with more variety so you get to choose among better options in the first place.

Listen to this interview the Wall Street Journal made to Holly Finn about Agile Route’s Shopperception, and you’ll see what I’m talking about.

The overwhelming answer from the media and the press on Shopperception worldwide, shows that this first effort in joining the offline and the online worlds is but the first tiny stone in a huge avalanche that is coming.

The only question that still remains to be answered is: can three people and a $200 device change a trillion dollar industry forever? I guess we’ll find soon enough.


Goodbye Steve

I never had the chance to meet Steve Jobs in person, but the sudden realization that I never will just struck me down with the most indescribable sadness. His life was an inspiration to us all, a lesson on chasing our dreams (and catching them), a fire on the shadows of conformism, the proof that true innovation comes from following your own path.

Half the globe away, I found the foothold I needed in his words on Standford to stop wasting my time and start doing what I really wanted: build my own company.

We are still struggling, since dreams are like adventures, and have uncertain destinations. I have been developing for both Mac and iOS for the last few years, out of business and pure pleasure. I hope we can eventually succeed, but if we don’t, at least it was worth it. Sometimes the journey alone is worth it, and Steve sure made the Apple road a sight worth seeing.

My condolences to all others, who just like me, have just lost their most beloved flesh and blood hero.

Raúl I. Verano (aka Rula)
Mac & iOS developer and President – Agile Route
Buenos Aires, Argentina

Tired of the default iOS buttons? Yes, we too. Don’t want to spend hours of graphical design just to put a new button style? Nah, we neither.

About a year ago, Erica Sadun posted how to use the undocumented UIGlassButtons from iOS to generate a shinny button, and then copy it to be used as an image in your real buttons. The post is still accessible here:

But, what if you could avoid the pain of generating each button also and just select the color of the button on your code? We thought it wouldn’t be too hard to do it, and Juani (our star programmer) did it.

I just compiled his code in a project that will demonstrate how to use these buttons. The CrystalButton class is just a subclass of UIButton so all you have to do is use them like normal buttons (even in the Interface Builder) but be aware that now you have a new property to set: the tint color of the button.

@property (nonatomic, retain) UIColor * tintColor;

Without further delay, here is the CrystalButton sample:

Play with it. Enjoy it. Tell us what you think.

– Rula.

We launched a few apps with the Mac App Store’s opening. One of them, Social Clipboard, was clipboard history app which also allowed you to post your clippings to FaceBook, Twitter/Twitpic or DropBox. We sent press releases, and contacted every possible blog and review site (we already knew the value of this, see our post: “What happens when the media pays attention“). The combination of the lack of diffusion we got, Apple’s poor ranking system (which gives everything to the top and nothing to the rest) and perhaps our inability to explain a complex value from one icon and a name (we tried several icons, as a matter of fact) gave it very poor sales results.

We still believed in the idea, so we cut a lot of functionality and made a simpler application from that one removing the social/networking components: Clipboard History. The message was now clear, and the functionality too. After the 2+ weeks Apple takes to review apps, it was launched at $0.99, the lowest possible price.

We got a few days of dubious results (26 downloads the first day, 21 the second, 33 the third and 18 the fourth), and starting to loose the only boost the App Store gives (the visibility of being a new app just because they are ordered by date) we decided that it was time to try something drastic. So we went free (at least for a while). That was two days ago.

The first day we got 1775 downloads (100 times the downloads we had the day before at $0.99). The second day we got 5358 downloads, and Clipboard History climbing to Top 1 in the charts around the world (in some countries to number 1 in Productivity as I write or even as overall number 1 Free App).

Where does this path lead to? Without in-app purchases to be able to charge for some extra functionality, or iAds to make some ad money, we still don’t know. But we have a few ideas. We have some time to figure this out. We will share with you our findings in this blog.

At least now we have crossed for the first time the hardest gap in the ecosystem Apple created: putting our software in the hands of the people.


So, you are building a 3D game. Everything looks somehow fine while things are moving, but then you take a screenshot to upload it to iTunes and: the horror! You realize the awful truth. Some texture borders seems crappy and pixelated.

Oh, you forgot to turn on the anti-aliasing. That’s easy to solve. Except that… wait, there is no anti-aliasing in OpenGL ES 1.1!  Damn!

Well, we’ve been there too. While we were porting Cube X Words to the iPad, the 3D dice that looked awesome in Mac OS, looked just like a BMP drawn in MS Paint in the iPad. So we did what every self-respecting developer does first: we googled for an answer, but alas!, we found none…

How to solve that? The hard way: mess with everything you’ve done in order to draw a 4X image and then scale it down. But, but… stretching a texture to 4X and then down looks bad, I have to re-do all my textures!  It sometimes works, it sometimes doesn’t, it surely hurts the performance and may end up looking like crap again…

By accident, we found an easier way. It involves a bit of modification to the textures and another bit to the code.

We found out that if you leave a couple of transparent pixels outside of your texture, when the projection is blended, it kind of anti-aliases the texture borders (since whenever OpenGL calculates the pixel’s color value, it mixes the color with the transparent part). So, voila! We just had to give our textures a bigger canvas, and make the sprites bigger accordingly. Just like this:

It worked like a charm. You can check it out in this screenshot. See that the rotated E’s border looks just fine.

Well, I hope this helps you too.

– Rula.

So, you’ve coded, tested, crafted an icon and submitted your app to the App Store. Now what?

Most people think they have a lottery ticket, and now it is time to sit and hope to be noticed by Apple, be featured, get a gazillion downloads and be the next Angry Birds. Dudes: with 1 chance in 350.000, it doesn’t work like that. Sure, that guy from Germany just did it with Tiny Wings, but so did the guy that won the lottery yesterday in your city. The difference between “that could be you” and “that will be you” is simply huge, but we delude ourselves and tend to forget that.

We knew that from the iTunes Store and the time we launched Kolonists, so we took our experience to the Mac App Store. You have to work hard. Make a presentation on who you are, a package with information on the App that could easily be blogged (and not just a copy&paste of the app’s description from the App Store), some screenshots and hi-res logos. Then make a list of every god-damn blog and review site out there and send your package to them, now crossing your fingers again with a much better chance that at least one of them will take the time to have a look at what you’ve done.

We got a review for gNotifier at MacWord. Twitter went crazy. It took us in one day to Top 9 Paid Productivity (with 57 downloads, yes, Mac App Store sales volumes are very disappointing). Then, tough luck: Sparrow came out, the press attention shifted 100% to them and the darted to the top in no time. The result: we got gradually lost in the way to a current rate of 10-15 downloads a day.

But there is something to learn here: since there is absolutely nothing you can do in the App Store to get noticed, your only bet is outside. So, start writing and taking screenshots, and find a creative way to call the attention of the real people that can make you noticed: the reviewers.


Tag Cloud