Let's not start a pissing match

This reminds me of a discussion I once had with an editor for two fairly large "sister" websites (that shall remain nameless), one for Java and one for .NET. Whenever an article was posted on the Java site it would attract 50-100 comments whereas articles on the .NET site would be lucky to get 10 comments on a good day. I questioned the editor about this, asking if the .NET site was attracting less visitors. He assured me that the traffic was in fact comparable but, for whatever reason, the .NET community was just far less active in the comments.

Of course whether this was due to the open source nature of Java attracting a more 'community' feel due to a host of other reasons is open to debate. But assuming the editor was honest about the numbers though (I have no reason to doubt him) it does demonstrate that trying to measure the effectiveness of a website shouldn't be solely based on the participation rates.

Personally I never post anything to twitter, I only use it as an information source by following various companies and people of interest. I find this very useful and I'm sure there's millions more who think likewise.

Agreed there's little in it for the vendor to provide upgrades, however as I said they seem to be doing a reasonable (certainly not perfect) job of providing upgrades all the same which is quite a positive sign. I'm not aware of any vendors "lying" or reneging on specific upgrade promises, I only know of a couple of cases where there have been slight delays. Do you have any examples?

Of course for those who really want to, there's always 3rd party ROMs like CyanogenMOD which add an absolutely insane number of new features even to old phones like the original HTC Dream, though that's getting firmly into "enthusiast" territory and clearly isn't suitable for the average end user.

Well, I'm comparing Apples to Androids actually ;) Where was my 'apples to oranges' argument? I'm talking about the difference of OS versions in the wild (which seems to be what really matters, not how things got into that state). I'd agree with you if that chart was showing versions of Android *currently being shipped*. It's not, it's comparing versions *in the wild*, same as the iOS figures, so I think it's fair to compare them. I agree that there is a difference in how the two situations came about. Some vendors are still shipping with older Android versions installed (nothing worse than 2.1 though AFAIK), and that clearly has an impact on the chart. Since the Android updates go out over-the-air though, the uptake of these releases is clearly far higher than upgrades being applied manually to old iOS devices.

Bottom line is, there's a bunch of old Android phones running 1.5 and 1.6 that likely will never have their OS upgraded either, same as the iOS 2.x situation you described. Who's "fault" the fragmentation is (vendor vs user) doesn't really matter so much given that if you're an app developer, you'd need to be compatible with at OS versions from at least the past year or so regardless.

As an aside, I've got two colleagues at work here with the 3G. One upgraded and has recently rolled back to 3.x, the other refused to upgrade after he saw the grief the first guy had. That was what motivated me to post about the issue in the first place. Your experience was clearly different so I guess it's a bit of a mixed bag.

Unfortunately with 6 month or less shelf lives of phones, there's not a lot of motivation for manufacturers to upgrade old handsets, unless there are glaring support problems that are costing them money. Having said that, most android phones do seem to be getting an encouraging level of upgrades, even if it takes the vendors a bit longer to release the upgrades than many people might like.

What I don't understand is why is no one complaining about the state of fragmentation of iOS?

Given Apple are a single manufacturer with a very small (iOS) product range, they seem to have done a pretty good job of messing things up, arguably worse than Android even with the far greater diversity of companies and products involved.

Maybe you should see on-demand Flash in action, rather than make up FUD about russian roulette and browser crashes without any facts to back your statements up.

I've been running Flash on my phone since June, viewing the Flash content I choose to on a daily basis. It's seamless enough that I hardly think about it. I've never had a browser crash from playing Flash content, there's no "roulette" involved. It does in fact "just work" (though I wouldn't go quite so far as to call it brilliant). And as a bonus, if I want to emulate an iPhone, I'm always free to uninstall Flash completely.

I've been using Flash on my Nexus One for a couple of months now, and I find it pretty useful and flash video perfectly watchable for the most part. I also follow the tech/mobile press pretty closely and I'm not sure what the "most reports" you're referring to are - my impression is that the response in the press has been broadly in agreement with my experience.

Here's an interesting followup to the article:

http://newteevee.com/2010/09/02/is-flash-on-android-shockingly-bad-or-shockingly-great/?utm_source=newteevee&utm_medium=recent-posts

Open the browser on your phone then select: Menu -> More -> Settings -> Enable Plugins -> On Demand.

That means Flash is disabled by default and a placeholder will be displayed instead, but you just need to touch the green arrow to load and play the flash content if you want to see it. Works a treat, performance is fine, and if you really do want the content it's there with a single press.

Having said that, I find Flash performance to be fairly acceptable for the most part on my Nexus One anyway, and having it on demand like this is much, much, much better than being told you can't have it at all.

By that logic, if I leave my front door open year round yet don't get burgled, my home must be burglar proof!