[tog] Tog Mobile App

Sean Nicholls seannicholls at gmail.com
Sat Mar 3 01:42:08 GMT 2012


Note: since this reply is annoyingly long, I want to point out that to anyone reading that I'll be at this weekends Dublin Science Hack Day at DCU, working on the Tog App if you're there and want to team up. My twitter handle is @sean_nicholls.

Hi James,

I hope this email ends up displaying on the Tog mailing list properly; its really doing my head-in, a Google groups would be much easier, but I digress.

Take a deep breath, it's a long reply:
>Hi Sean,
> I've spent the last few years working in mobile
> development (various platforms, mostly native Android and J2ME),
> and was thinking of a doing something like this myself, but I don't have the
> time at the moment (new baby and house :)
> I have some feedback below,
> but like the others said, if you're pony-ing up the man-hours, you make
> the calls!
> 
I think you can probably predict my initial response... If you would like a Tog app on the other platforms, I'll gladly do what I can to help, but it'd be on your shoulders to make it happen.

I will outright say that I plainly do not know the demographic of potential or existing Tog members, nor am I particularly inclined to run any sort of market surveys on such, but I think it would be a fair assumption that of those that do own a smart phone and make it a point of downloading apps of interest on a regular basis, iOS, Android and Blackberry would probably be the top of the list.

I am fully aware that Windows Phone 7 is not yet even on the radar for most people, but I firmly believe that it is a platform that will eventually rival iOS and Android. Microsoft are putting a big effort behind it, and frankly with Android's splintering, and disenfranchisement of Apple for various reasons (App review delays, rejections, ejections, mac costs etc.) developers are looking for viable alternative. Android will never be that whilst the hardware and OS is so fragmented.

In any case, it's just a nice side-effect of using C# that you can easily port to Windows Phone 7; from my point of view the main targets are iOS and Android.

And again, I'll go back to my previous comments. I would rather produce something of quality, using tech I have some experience with, than learn a whole new toolchain and produce an app that I would not be confident about.

>As unfashionable is it may sound,
>there are still plenty of people with Symbian and J2ME devices out there
>(and Blackberry etc...).
>I know that's a lot of platforms to support,
>but if there is the option to use something cross platform that ticks
>all the boxes,
>it is worth considering. These are the best numbers I
>could find on short notice (doesn't show J2ME)...

Just to drive the point home a little further, I've worked on Blackberry before. its a HORRENDOUS platform. You literally could not pay me to work on it, the toolchain is slow and buggy, the documentation sucks, oh and I think RIM have removed them, but they used to have some really laughably amateur videos online by the the developers themselves of the OS/API which made me lose respect for them so much. 

That's to speak nothing of the variability in UX/UI and so on and so on...

No, if you want to do Blackberry, God speed and good luck, but holy crap keep that dirty stuff way from me!

> Regarding google maps, I would be a lot happier to see OSM
> in there :)
I would like that too, but it's another problem of cross platform capabilities and time/effort involved. Open Street Map does not, to the best of my recollection have any sort of developer libraries for mobile platforms. Sure the API is there, but from what I've read in the past, in order to make use of it in a mobile app at any decent frequency, you would have to make use of a proxy/caching server of your own to not overly burden the OSM servers.

So, at least for version 1, it's not on the cards. But it is definitely something I would like to do and will probably be looking into doing for some other projects I've got on the cards, but not for the immediate future.

>Why no app store presence? What kind of
>performance do you need that HTML can't meet?
>Many web app platforms
>allow the content to be bundled or cached locally, and most have hooks
>into device functionality.

Web apps, unless bound into a native app that shows it via web view, cannot be listed on App stores - hence, no presence, and I am not yet convinced that Apple will like the idea of a web app wrapper. I know the last time I was bored out of my skull and read the full TOS for App Developers, it explicitly prevented them. Android is probably a different matter.

There are hooks into "most" device features, but not all, and frequently using such features causes performance issues.

I'm too tired right now to go find all the extra links, but there's also significant data on the subject that shows that people really dislike using apps that are not native over-all.

Web Apps also present an issue when it comes to native UI features. Yes, there is emulation, but often it is poor emulation and you lose a lot of "snappiness" of the UI, which mobile users have come to expect. It's really a matter of User Experience and over all design.

If others would like to remake the Tog app using HTML5 or other technology, then please go ahead, and I'll try to share what I can to help that process, but as a personal thing, I would like to go on with this using C# and Mono for now. If it turns out being a poor decision, then on my head it'll be.

>I have no problem with you choosing what you are most
>comfortable with,
>but I wouldn't even consider using anything that needs
>to be paid for
>and I _would_ seriously consider PhoneGap, it being a
>common solution in this space.

I think thats a pretty idealistic view of things. You cannot access software without a computer, which costs money. Up until recently you could not develop at all for iOS without completely investing in the Mac ecosystem, including OS, developer tools etc. all of which have associated costs.

But the way I look at it, it comes down to this:  MonoTouch and MonoDroid are completely free for "evaluation purposes". You just cannot make a commercial app and put it on an App Store.

People will still be able to build the apps, test them and modify the code. They just wont be able to put them in a commercial context without further licensing. This is in fact, true of many Open Source projects and I think fits well within the context of a hackerspace.

Regards,

Sean










-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.tog.ie/pipermail/tog/attachments/20120303/cb7a5f93/attachment.html>


More information about the tog mailing list