The anecdotes from the 80s & 90s when annual PC sales for any given year were a substantial portion of the overall PCs in use at the time are interesting. The realization that software could be written for the newest machine and it would run everywhere soon enough is not directly applicable to the smartphone market. I think that out of collective habit, many apps assume that building for this year's flagship models is the best course. It may be, but it effectively means your market is much smaller, especially for free, network effect apps.
These apps and their ecosystems rise and fall fast. Tablets & smartphones are spreading downmarket with devices being sold today @ 10% of the first androids.
A $50 phone won't run everything, but it will run some things - useful things.
The really fun thing about WhatsApp is that it is essentially a product for the developing world, and to make money doing that you have to be seriously efficient.
Ignoring the fictional statistics, the cheap devices (e.g. Xiaomi) in countries like China still tend to be surprisingly powerful, and you hardly need to design down to them. And it's a rare app, beyond something like video conferencing, that puts any real burden on connectivity.
http://appleinsider.com/articles/13/11/12/idc-data-shows-66-...
And it turns out I misrepresented the fictional statistic. The "real" number is 66%. Though I suppose if you count older mid tier models (this counts sales) still in use, 80% might be pretty close.
IMO you still need to design dow a bit, especially for data poor users. Also smaller screens. In some cases I think craigslist is the thing to emulate.
While I don't normally agree with dismissing something from the source -- in this case Apple Insider -- the way that the article digs to try to generate their manufactured numbers borders on comical. Secondly, the metric that it is "junk" (the article's word) because the ASP is $200+ is asinine, if you will. As is the notion that such makes it a "feature phone" (where ASPs were in the LOW double digits).
A $200 device in many Asian markets is a decidedly premium device. The Xiaomi Redmi, for instance, is a $139 (retail!) decidedly decent mid-range device. In no way does anyone have to design down at all for devices like that, the notion being absurd.
Androids are penetrating markets that were never penetrated by PCs or the web or even residential electricity.
I did, however, say that many of the cheap phones are quite powerful. And they are. And as someone who has several widely deployed games on Android -- one category where there is much more demands than categories like WhatsApp -- it is absolutely no problem at all.
I generally test things on the Nexus One just as a bottom end comparator. Compared to many of the devices from bottom-tier manufacturers now, the Nexus One is grossly underpowered. It is always surprising what it can run with gusto.
Caching 2 pages in RAM also means instant undo when I open the story link.
[1] http://dx.com/s/android%2bphone?PriceSort=up&category=511
Of course it is likely to have other issues, but don't limit yourself to saying what is possible... think about the cost to actually deploy that possibility.
So be very, very careful about trying to 'serve' that demographic.