Having previously added tablet support to the Android version of the educational flag recognition game Flagpole, we’ve now brought the iOS version up to speed with proper iPad support (including support for landscape orientations).  This new version (version 1.3) is available on the App Store now.

The release of the Kindle Fire just over a week ago has breathed new life into Amazon’s Appstore. One of the consequences of this is that Rectangular Software’s flag recognition app, Flagpole, has sold several copies to Kindle Fire purchasers in the last few days having previously barely registered on the Amazon store.

Until now Flagpole has only been optimised for phone-sized screens. Today we’ve released an update that improves the experience on the Kindle Fire and other larger devices such as the 10.1″ Samsung Galaxy Tab. As part of this update the app now also works in landscape orientation as well as portrait mode.

In addition, version 3.4 includes the updated flag for Malawi that didn’t make it into the previous version.  The new version is available from the Android Market right now and has been submitted to the Amazon Appstore where it should appear some time in the next week.

When considering building a smartphone app for your business, it is important to understand the nature of your target audience. The landscape changes rapidly and differs from country to country. There are multiple mostly incompatible smartphone platforms battling for supremacy, constantly gaining or losing market share. Yesterday’s top dog might be tomorrow’s has been.

A glut of surveys are published throughout the year that attempt to provide an overview of the current state of the smartphone world but often they just end up adding to the confusion. I lost count of how many separate occasions Android was reported to have overtaken the iPhone in the last year. The problem is that the surveys often measure different things in different ways. Some count only new activations while others monitor total device ownership. Sometimes they can’t even agree on the definition of a smartphone and, even if they can, the data presented in one survey is often from a different territory to the data in another survey.

Any surveys that you consult have to be interpreted carefully if you are going to use them as the basis for important decisions such as which platform(s) to prioritise. The first question to consider is where are your target audience? If your business is constrained by geography then you’re only really interested in local statistics, for instance those for the UK. Alternatively, if you aim to reach the entire English-speaking world then the data for the much bigger US market is more significant. And if you intend to engage in multiple languages across the world then you should be looking at global data.

Secondly, while surveys based on the number of new activations provide a good indication of where the market is heading, they don’t tell you how many users each platform has right now. Many users are tied into 18-month or 24-month contracts with mobile operators. They are still potential users of your app but they don’t show up in the new activations surveys because they are not buying new devices.

With all this in mind, I wanted to highlight a couple of surveys that have been published in the past week that provide a useful snapshot of where things stand right now (as of the end of the third quarter of 2011). The first is from Kantar Worldpanel ComTech and covers Great Britain. It shows that 43.8% of the population now owns a smartphone (and this is increasing significantly with 69.1% of new phones sold being smartphones). Half (49.9%) of British smartphones run Android, a phenomenal increase over the last 18 months. RIM’s Blackberries (22.5%) are in second place ahead of Apple’s iPhones (18.5%), suggesting that the oft-predicted demise of the Canadian manufacturer is yet to materialise in the UK. Microsoft’s Windows phones have been almost completely ignored by British buyers so far.

The second survey, from Nielsen, provides an equivalent view of the US market. The ratio of smartphones to non-smartphones (43% vs. 57%) is almost identical. Android is number one in America too but not quite as dominant with a stronger showing from Apple pushing RIM into third place. Microsoft has at least registered on the other side of the Atlantic but is still a distant fourth.

The broad similarity of the UK and US figures might lead you to assume that the picture is the same across the world. That’s not the case and if you plan to target other countries you should strive to determine the local situation. As an example, in parts of Asia the popularity of Android is even higher, driven in part by loyalty to home-grown brands. For instance, figures from the previous quarter show 85% Android penetration in South Korea, home to Samsung and LG, and 71% in Taiwan where HTC is based.

Several Beep Test for Android users have suggested an enhancement to the app to make it easier to keep track of what level you are on while running, without having to look at the screen. Specifically, Android’s speech capabilities mean that we can do better than the three short beeps that indicate the start of a new level. Starting with version 2.2 (available today from Android Market and Amazon Appstore), the Beep Test app now announces the level number in addition to playing the beeps, which means you no longer have to keep count for yourself.

This version also improves the experience on Honeycomb tablets. There is more that we can do in this respect (a landscape layout for instance), but these updates are a step in the right direction.

Rectangular Video Poker was originally designed for Android phones and, as such, while it was usable on tablets, it didn’t really integrate with the Honeycomb experience. That has changed with version 2.4 released today and available now on the Android Market. This version improves the layout for large screen devices and takes advantage of the Honeycomb action bar to provide a tablet-optimised experience.

Following on from adding Google accounts authentication in Appmonger 2.1, version 2.2 moves on to the next most requested feature – the ability to alias and/or group apps for reporting purposes.

There are two use cases that this new feature addresses. Firstly, when you translate your app into different languages the translated names show up in Appmonger as separate apps. This is because the data from Google Checkout does not include the app’s package name so there is no way for Appmonger to reliably infer which different app names refer to the same app. You can now assign the same alias to each of these translated names and they will be grouped together as one app.

Similarly, you can now organise your apps into groups. By assigning the same alias to different apps you can treat them as a single entity for reporting purposes. For example, some developers release several different variations of the same basic app and might want their reports to show the combined figures rather than figures for each individual variant. This is now possible.

To assign aliases to apps, simply select the “Groups/Aliases” option from the main Appmonger menu and then tap on each app that you want to enter an alias for.

Appmonger 2.2 is available now on Android Market.

In recent weeks we’ve been seeing higher incidences of cancelled orders on Android Market than usual. It seems to be due to a change in payment processing that Google has yet to explain to developers.

One way that Google has always differentiated Android Market from Apple’s App Store is by offering refunds to users. Initially there was a very generous 24-hour refund window that was subsequently reduced to a barely useful 15 minutes. When a user requests a refund, that order is assigned the CANCELLED status (it shows up in Appmonger‘s list view as red/pink) and the Google Checkout order history will show the message “Cancellation requested from phone” (or some non-English translation thereof) as below:

The other scenario in which orders could become cancelled was when the customer’s payment card was declined. Google gives the customer 7 days to provide alternative payment and then automatically cancels the order if payment is still outstanding. In this case the order is assigned the CANCELLED_BY_GOOGLE status (it shows up in Appmonger as orange) and the Google Checkout order history contains the text “The customer did not provide valid credit card information in repsonse to the email Google sent. This order was automatically cancelled”:

In the last couple of weeks we’ve been seeing a third kind of cancelled order. These are also assigned the status CANCELLED and therefore, like refunds, show in Appmonger as red/pink. But these cancellations are not due to customers requesting refunds. These cancellations occur several hours after the orders are made, well outside of the refund window. Nor are they a result of declined credit cards since Google Checkout shows that the authorisation succeeded. The salient text in the order history is “We could not complete your order in a timely fashion. Your order has been cancelled automatically. Please retry your purchase”:

Trying to get an explanation out of Google is like trying to get any other information out of Google – i.e. not a very rewarding exercise. There are some suggestions that these cancellations are related to the recently-introduced carrier billing but I don’t think that’s the case. The most benign theory is that Google is simply cancelling orders where the user fails to download the app. However, the wording of the e-mail sent to the customer implies that there is a payment processing problem on Google’s side. If that’s true then developers are losing money from lost sales.

Whatever the real reason for the cancellations, Google could easily clear up the confusion by communicating better with Android merchants.

There are many important milestones in the early life of a new country. The declaration of independence, diplomatic recognition from other states, and membership of the United Nations to name just a few.

South Sudan is the world’s newest country. It gained independence on July 9th this year and joined the United Nations five days later, but until now it has lacked the prestige of an entry in Rectangular Software’s Flagpole mobile app.

South Sudan becomes the 234th flag to feature in our educational game  for Android and iPhone. The updated Android app is available now on the Android Market. The iPhone update has been submitted to the App Store and should be available soon. Also included in these versions is the change of Libya’s flag back to the pre-Gaddafi red, black and green design.

Tech conferences are great when somebody else is footing the bill. A couple of days away from the office, visit a new city, meet new people, have a few beers, maybe even learn something if you can find the time. When you’re running your own company you tend to look at these events differently. How much will it cost and what benefit will I derive?

For a UK company, such considerations tend to rule out the big US events such as JavaOne, Google I/O and Apple’s WWDC. The time and expense involved is simply not justifiable. Realistically, from Rectangular Software’s base in darkest Kent, it’s only the London conferences that tick enough boxes (though the company finances might stretch to a Eurostar ticket if I can find a suitable excuse).

Being largely concerned with Android, last year’s Droidcon UK was one of the few conferences that looked like it might be worthwhile. Unfortunately, I was busy with contract work for a client at that time so I had to give it a miss. So this year I made sure to sign-up early. Droidcon 2011 begins four weeks from today (I believe tickets are still available). The programme features such Android luminaries as Mark Murphy (a.k.a Commons Guy, a.k.a. that bloke that answers your Android questions on StackOverflow when you get stuck) and Nick Butcher and Richard Hyndman of Google.

This will probably be the only conference I attend this year. If you see me, come and say hello (I’ll be the one looking considerably greyer than the picture on his company’s website).

One of the major limitations of Appmonger has been that it requires you to use your merchant ID and merchant key to authenticate with Google Checkout to download Android Market orders. At first glance this doesn’t appear to be a limitation at all – it is simply how Google requires clients to connect to the Order Report API. But for reasons that are not at all clear, Google does not provide merchant keys for all Checkout merchants. It seems that all US and UK accounts have merchant keys. In other countries some accounts do have this basic integration option but many don’t. As a result, the number one most requested feature for Appmonger has been for some alternative way to connect.

Finally we have a solution. Appmonger 2.1 provides the option to use your phone’s Google Account to authenticate with Google Checkout. Appmonger gets an authentication token from the Android AccountManager and then mimics what your browser does when you download CSV data manually from the Google Checkout website.

To use this new mechanism, select the “Google Account” radio button on the Checkout credentials screen. If there are no accounts listed (only accounts associated with Google Checkout are shown), you will need to add one.  This can be done from Appmonger (select “Add Google Account” from the app menu) or from the device’s settings (under “Accounts & Sync”).

You will need to grant Appmonger permission to use your Google account.

Users with a merchant key should probably continue to use that as it is Google’s officially supported way of integrating with Google Checkout and as such is less likely to break if Google changes something.