My reply on @rastin
blog post: http://www.rastinmehr.com/2012/09/10/many-mobile-apps-are-actually-services/
Nice post Rastin,
I just bring my experience in this direction, when started with ohanah we've tested and experienced using a mobile js fw (for that occasion I'did use sencha touch, altough I wont use it again...). This was pratically an empty app that relies on the logic of com_ohanah running on a joomla sintallation and its exposed rest apis (thanks to nooku :). It was a bit of a growing experience, first we created a working solution then we tried to see what we could do with its api and architecture, more then a finished product was a way to show our clients the way, so that they could start thinking what they could do with it, in fact you could immagine a scenario where using the joomla installation just as a backend structure to offer an event service only on mobile, just to name an option...
On our actual project we decided to completely decouple the backend from the frontend, so we went (for few reasons i won't explain in here) using node.js (with express.js as fw) and key value db like redis.io on the backend to build the logic, and the api server, while being our self the first to consume our own apis on the frontend where we are using ember.js as a js mvc fw to build the client, even if for the moment our client will be just a desktop client that run in a browser (an html5 app).
Building a mobile app to connect to service is not just about building the client and doing the coding homework, there is much more to be thought as the mobile experience is difference, probably people will be on the go, different needs, and different features have then to be thought to make it appealing and not just replicating everything you find on the desktop on a small screen device. That's why i'm personally not that much in favour of building a service and with just some responsiveness bring it to the mobile, if mobile is so important just apply a mobilefirst approach and build for it without caring about the desktop experience in the first place, or viceversa.
In conclusion I agree with you, people shouldn't think about wich tecnology to use in the first place, they should focus on what they services offer, wich problem it solve and to whom, tech is just a consequence choice for the service you 'r building
I think this will be a great topic for a next anahita hangout, in case count me in as i would love to enjoy the convo.(that's why i posted it here and not on your blog directly :)