Developing at the speed of Apps...
MrChris 2700011K0F Visits (5873)
First things first, as this is my inaugural blog post here on DeveloperWorks it is only fair I give a little background about myself (htt
Having gone through so many roles in my career so far it is only reasonable that I occasionally harken back to earlier times when all I had to worry about was whether the business logic code was correctly routing gas bills to the right print centre; as opposed to today when a typical day involves dealing many different clients from a range of geographies with so many variations in technology that it's hard to keep track of who's doing what.
With the release of Microsoft's Windows 8 operating system and their respective app store I decided the time had come to take on a new personal project to learn both C# and XAML as development languages and also to understand what being in an App Store was really like from a developers perspective. Getting 'in on the ground' to the Windows 8 Store seemed like a great way to do both these things.
And so SmashCast was born as a collaboration with my friend Mike (htt
Having never created an app before we made some mistakes in how we approached the App Store.
For one thing we didn't realise the importance of a name (or SEO in general) in the store, I honestly thought the metadata we put in at submission time would weigh more heavily and was really worried about those 6 tags we could use (podcast, podcasts, vodcasts, crikey I've used half of them!).
We (I) also submitted some bad builds; completely my own fault and something I should have mitigated sooner - see below for how we eventually resolved one of our biggest testing issues - hint, it involved Service Virtualisation.
In fact that's something I never expected, the impact of reviews; I work all day selling big box software to people and obviously helping to make a sale at work is great. But the first tab I open when I sit down at my PC each morning is the app store analytics tab; eager to see if there's a new rating or review, or heaven forfend a new purchase! And every few weeks there's something new, and I'll be nicely chuffed that day.
All in all it's been an interesting experience and something I'm glad to have done (and will continue to do I'm sure).
Now how does all this relate to Integration and Service Virtualisation I hear you ask?
(I'm not sure if these blogs have editors but if they do I'm sure this had a ton of red lines through it by now!).
To begin with I thought about deploying Apache and hosting a feed and writing a script to generate new entries for a test feed etc, but the more I thought about what I'd have to do to get it all set up the more I frankly lost motivation.
After that adding in additional logic so that I could be fed different sample audio/video podcasts for testing the different features of the app was really simple; the hardest part was generating suitable uniqueId's, but I remembered I had a simple number generator in my custom function pack that I could leverage (more on the custom function pack in a future blog post).
With all this in place I was ready to start testing SmashCast at 30,000 feet! Each time I refreshed the feeds from the app I'd get back a feed result with a new entry that was added into the episodes collection and boom I had a simple and effective virtual service that represented a common feed type and made my life significantly easier.
This is just one of the many benefits of service virtualisation and as I publish more entries in this blog we'll cover a range of further topics.
-- Parts of this blog post originally appeared in the following Neowin thread, bear in mind I'm even more ranty in that thread than I am here :)