backtop

Print 17 comment(s) - last by erple2.. on Dec 4 at 8:44 AM
The report may seem startling, but there is more to the story than meets the eye

A recently released report indicates a growing trend among Google Android app developers who are concerned with a number of different issues, though the numbers don't initially tell the entire story.

The report indicates "82 percent of those surveyed feel that the design of the Android Marketplace makes it difficult for apps to be noticed. Often, apps will be released and quickly buried by other apps, and difficult to find again."

57% of Android developers are not pleased with profits; 52% of developers have seen their apps downloaded less than 5,000 times; 68% said they were somewhat or not as likely to work on future Android apps.

Despite the concern by the Skyhook Wireless report, there is at least one significant point worth noting before anyone begins to believe the Android ecosystem is doomed for failure. For starters, the company only interviewed 30 developers out of at least 15,000 registered App developers, which is a minute sample compared to the pool of overall developers.

The full report compiled by Skyhook Wireless can be viewed here (PDF).

The Apple iPhone has significantly more developers and apps available, but Android, which is new to the market, is expected to gain momentum as multiple phone carriers across all four major platforms begin to use the OS.


Comments     Threshold


This article is over a month old, voting and posting comments is disabled
Hmmm..
By spathotan on 12/2/2009 1:30:36 PM , Rating: 2
As an Android user (MyTouch) I can say the marketplace is a little...linear I guess. However its good enough for me. And with many new Android devices coming out, Motorola Cliq and Droid here recently, as well as the Samsung Galaxy and HTC Hero, their is basically an Android phone under every provider now.(Hell, ntellos has the HTC Hero @ $100) My phone is every bit of good if not better at basic functions than an iPhone 3G.




RE: Hmmm..
By Bender 123 on 12/2/2009 1:47:44 PM , Rating: 3
It works...sorta...

I have the Hero and have found all the apps I need, but the marketplace system is horrible for anybody else. Would it be so bad to have a top 25 list? Let me see what my peers are getting and what tools are working best for others.


RE: Hmmm..
By mmeytin on 12/2/2009 2:10:57 PM , Rating: 3
Android Market (on 1.6 and higher) lists top Paid and top Free apps for the market as a whole, as well as for each individual category (Communication, Entertainment, Finance, etc.) Does your Hero run an older version of Android and/or Market? The Market has improved a lot with 1.6 release, but I'm not sure if every device got a new version of it already.


RE: Hmmm..
By Bender 123 on 12/2/2009 6:32:59 PM , Rating: 2
Hero has 1.5...but it appears we are skipping straight to 2.1 shortly.

What is 2.1 called, anyway? I know 2.0 was E'Claire, so I am hoping for some Android Flapjack/Fritter/Fudge Brownie/Fruit Cake.


RE: Hmmm..
By sandytheguy on 12/3/2009 6:34:24 AM , Rating: 2
I believe it will be called Flan.


RE: Hmmm..
By michael2k on 12/2/2009 2:04:13 PM , Rating: 2
What's a basic function?

Spec wise it sounds like an excellent phone, if a bit underpowered in the CPU department.


RE: Hmmm..
By spathotan on 12/2/2009 2:15:44 PM , Rating: 2
Sorry, I ment navigation through the menus/apps for example. On the iPhone basically everything looks the same which made it harder for me to find whatever. The iPhone is designed for fragile minds and baseline intelligence but that dosent necessarily reflect all users, just to clarify.

Also the only time I experience any sort of system "lag" is if im exiting out of a demanding game/app. Normal use is smooth.


Nice sample size
By krotchy on 12/2/2009 1:29:56 PM , Rating: 4
In other news, I asked the 5 software engineers at my company what they thought of the Android marketplace's design and they all looked at me confused.

Seriously, is a sample size of 30/15000 even worth bringing up?




RE: Nice sample size
By spathotan on 12/2/2009 1:33:47 PM , Rating: 3
No. Its false bad publicity to spike iPhone sales for christmas.


RE: Nice sample size
By sprockkets on 12/2/2009 10:33:49 PM , Rating: 2
Just a question/clarification: You made two points, and your first sentence implies that nobody in your company even knows what Android is, right?


RE: Nice sample size
By FITCamaro on 12/3/09, Rating: 0
RE: Nice sample size
By Gondorff on 12/3/2009 5:01:41 AM , Rating: 2
This.

I thought this was an interesting article until I saw the sample size. That this article was even published is abhorrent, as well as is the fact that this kind of shoddy study would get any media attention at all.

I know it's your job to report all the tech news, but you've got to draw the line somewhere. Polling 30 people is clearly on a bad side of that line.


RE: Nice sample size
By erple2 on 12/4/2009 8:44:53 AM , Rating: 2
I don't know what the methods were for the sample size generation. Did they look at 500 developers, then pair down the sample size to 30 using proper survey techniques, or did they cherry pick those 30 developers to prove the point they were after?

It's not infeasible that those 30 developers are a representative sample size, and could therefore accurately reflect what the developer attitudes are.

Other proper studies can be done with as low a "market share" and still get accurate results. 30/15000 is about 1/500th of the market (or 0.2%). If that were related to, say, cars, that would represent approximately 2000 car owners in the USA (assuming that 1 million cars were sold). That's not necessarily a bad representation.

Gallup polls and political polls usually have fewer responders than 0.2% of the voting population, but they can show accurate trending data.

However, you'd have to read the original .pdf to determine whether the survey was done correctly.


Accurate Sampling
By Admiral Sauce on 12/2/2009 2:22:50 PM , Rating: 5
I'm a political science student. We spend a great deal of time verifying accurate barometers in polling. With a population size of 15,000 developers, seeking a 95% accuracy with 5% variance, the sample size would need to be 375. A sample less than one tenth the necessary minimum would be thought of as a punchline to a joke, not a means of arriving at even generally indicative statistics.




RE: Accurate Sampling
By heulenwolf on 12/2/2009 5:10:01 PM , Rating: 2
I'm with you, Skyhook's "report" seems like marketing. They must get some royalty from Apple for every iPod Touch and iPhone sold, since they each use Skyhook's service with no charge to the user. I'm sure they see the Google Streetview Vans as competitors, as well.

From an engineering precision standpoint, the use of percentages is equally misleading to the lack of necessary sample size, statistically.

I'm curious to see where Android goes. We shouldn't let press or marketing outlets pronounce it dead before its even started walking.


RE: Accurate Sampling
By FITCamaro on 12/3/09, Rating: -1
Why the marketplace?
By GruntboyX on 12/2/2009 7:00:17 PM , Rating: 2
It would seem that it would be revolutionary to allow mobile users to search the internet and download and run programs advertised throughout the web.

I guess everyone sees this as a way to keep malware and virus to a minimum. How about external links to the app in the marketplace. I read a blog about a cool app. There is a link and it takes me to the marketplace to download?




"Mac OS X is like living in a farmhouse in the country with no locks, and Windows is living in a house with bars on the windows in the bad part of town." -- Charlie Miller









botimage
Copyright 2012 DailyTech LLC. - RSS Feed | Advertise | About Us | Ethics | FAQ | Terms, Conditions & Privacy Information | Kristopher Kubicki