Smartphones segmentation

Today I ran into rather “happy” problem: needed to determine which devices should I buy to test on, if I want to cover most of the smartphone market.

Android

The valuable resources for android became Android Developer portal – mostly dashboard. However, it overviews only Android OS versions. The other resource, OpenSignal research, covers specific devices. To sum up, to test hardware features and to determine usability on Android, you should have:

  1. Samsung Galaxy Ace S5830
  2. Samsung Galaxy S II
  3. Samsung Galaxy Note 10.1 2014 (or any other with 4.3 – Jelly Bean. I just prefer this one)
  4. Samsung Galaxy Note 3 (or any other with 4.4 – KitKat. I just like the size and usability of this one).

iOS

There I had to draw the line: old (pre iPhone 4) devices does not support iOS 7.1 update (according to extensive wikipedia page). Frankly, it would be enough to buy:

  1. iPhone 4 (oldest smartphone with iOS 7.1 capability)
  2. iPhone 5S (newest today)
  3. iPad Air (personally prefer this one in front of all those mini’s and old thick iPads)

Windows Phone 8

As Windows Phone is the 3rd popular platform for smart phone users (and I own one :)) it’s coverage also is a must. We will orientate ourselves only into Windows Phone 8 platform cause 7.5 really sucked. IMHO (as I was unable to find any resource), it is enough to buy mediocre Nokia Lumia 920 to cover the major of the market. This shouldn’t be a problem since all GUI in Windows Phone 8 is pretty basic and shouldn’t be a problem unless you use phone camera or other specific thing in your app.

 

Additions To Web Pages You’ll Like

It has been a long time since I wrote last. This time I will look over some interesting bootstrap plugins and other interesting tools to produce high quality web page.

  1. JsTree. Trees made simple in Javascript. Quite usefull stuff;
  2. Bootstrap Modal2. Fixes almost everything what is wrong with default witter bootstrap modal;
  3. Select2. Adds some nice things / touch / wtw you call it to your selects like: item grouping, multiple value choosing, autocompletion and etc;
  4. List of Twitter Bootstrap resources. If bootstrap has it, it will porobably be there;
  5. NVD3 javascript library. Nice and painless svg charts for your page;
  6. probably will be added later…

P. S. Just recently, my contribution to TinyMCE autocompletion plugin was pulled in to the main repository. Cheers, everyone! ;)

Estimation

Today came across an interesting post on agile blog about estimating time and cost of the system which is being build. The idea is that no matter how good your team and you are, you will always make mistakes in estimating. Especially complex systems. If your complex system is on date then there is a probability that the quality will really suck and the system itself might be not usable at all.

 

Interesting Articles Today

The first one is this (Losing faith in the startup where I’m employee #1) and the second one is this (First employee of startup? You are probably getting screwed). Interesting reads for today.

Quote of the day (by mzarate06):

During meetings, I always feel like a killjoy by attempting to be realistic … about the brand new idea they come up with every other week and that I need to get started on right when the meeting ends.

I tire from that very thing, very quickly. I’ve worked with too many “idea” guys that always want to see 10 ideas started, vs. 1 completed and executed really well. Maybe they’re trying to find that 1 great idea by trying out many, but a lot of times I see startups start one thing without finishing another. In the end, after XYZ time frame, the company really doesn’t have 1 solid feature to show for it, only multiple half ass ones.

What’s worse, at least from what I’ve experienced, is that a lot of those ideas or features are secondary, or tertiary, with regard to what the core product should really support

Windows and Malware

There is an interesting post / comment in redit. The guy claims that he now owns a botnet of ~10k bots and you can ask him anything about it. One interesting idea he mentioned was that all antiviruses suck and they are made this way for  a purpose. It is better for antivirus vendors to charge users on monthly basis for “updates” than to set up some defense mechanism to prevent virus infection at all (then there wouldn’t be any monthly charge). Also, he compiles his malware code for every machine personally. So such malware is not detected by heuristic antivirus software (of cause, he does have perl scripts which randomize code automatically).

In his opinion, it is possible to create such a tool that would prevent malware execution at it’s roots. On example of such tool – disk drives with “read – only” option. This way you setup your system (which you’ve got from vendor directly and triple checked hashes and so on) and after that, you put your disk to “read-only” mode. This way, any of the system files, MBR and etc could not be affected. All files and other useful / necessary stuff should be stored on a network drive / server (which would not allow any type of executable files). This is simple and effective however not so user – friendly: think, you want to update your PDF reader? Remove read – only lock, do the update, and then set the lock again. This would be pretty awful to use from user perspective.

Another approach is simply to monitor all system files, MBR and etc. For this kinda work he recommends GMER(been there, tried that, works as expected). However, this approach too is a no – go for an average windows user.

One more interesting thought: use only open source software (especially for encryption). If you don’t – you’ll never be sure you use the software which does what it advertises (for example, in US there are no such thing as 100% privacy).

On average day he does 40$ on bitcoin alone. That’s a day. It takes around 30 minutes of work a day. Pretty good job, right? Also, as he does not use stolen data (think – credit card numbers), but only sells that to a third party so he’s in so called “grey” area.

In the end, an interesting thought on malware market share :) (citation from one of throwaway236236 posts):

  • Linux: low market share, educated users -> hopeless
  • OS X: low market share, slowly rising, uneducated users -> some day maybe, but I never coded on OSX and am too lazy to learn to