Most, if not all of those have patches in browser.js. Google have long since proven they aren't able/willing to test in Opera (they are only a small start-up after all :P )

Sony removed OtherOS from the Slim first. Then Geohot started poking around looking for ways to re-enable it.

The massive irony being that Opera's built-in mail client was using labels instead of folders before Gmail had even launched.

I can death-grip my Nokia N96 (generally noted among the weaker quality hardware Nokia has put out of late) right around the antenna and get no drop — in bars or db — whatsoever. I can take off the battery cover so I can get an even closer death-grip and still effect no visible change upon my reception.

Is there any particular reason why iTunes on Windows cannot be written to function correctly?
Does Windows perhaps randomly corrupt data on USB ports messing up the connection? Does it randomly mess with the file system causing iTunes to lose track of files and corrupt their metadata? And if so, how can I edit tags and manage my music in Winamp, Foobar, MP3Tag, WMP, or a host of other programs, as well as reliably sync my music and data with Nokia's PC Suite without ever seeing issues such as iTunes is being alleged (and to which you do not appear to be disputing) to experience?

Seriously; what UI changes wouldn't do wonders for the GIMP.

The article implies that this event was some time ago now (they mention Opera Mini growth, and Opera link launching, which puts it between 2006/2007), but that they were predicting huge growth; presumably back when Opera Mini was still in its infancy. The article also mentions "millions of NOK" which is *at the very least* hundreds of thousands of dollars. Opera Mini usage has been growing at a phenominal rate (data usage has also trebled since the last point on that chart) and I doubt very much that Opera is a small contract even when this event occured but it definitly isn't now. Especially with Opera Turbo, and to a lesser extent Unite also needing server farms.

Remember that company almost certainly hasn't lost just one contract as a result as this, they've probably lost several more contracts since then too.

Have you filed bug reports?

That said, I REALLY dislike having the tab close button on the tab itself, and dislike the new "locked tab" icon's transparency. I normally have Way Too Many tabs open (38, right now), and the tabs are pretty much exactly as wide as their favicon (normally displayed on the tab). Unfortunately, I am still accustomed to being able to click the active tab to put it in the background -- but now, doing so closes the tab. Phoey.

You can of course configure Opera not to have close buttons on the tabs.

That's not how it works in reality though. Every browser ever made has bugs, sites will have to work around those bugs, and depending on the developers in question you might get one, two, or as many browsers as the developer is willing to test in, sets of workarounds.

Not to mention so much of the web is still a horrible mess of tag-soup where each browser vendor has had to make up support, and each other vendor has had to guess or reverse engineer what its competitors are doing.

Further to that Opera has, on at least two occasions that I remember, encountered examples of broken code because they support too many standards. Most notably in the case of when they introduced WebForms 2 support and several sites using previously non-standard values (the spec said all unknown values should be treated as "text") for <input type=""> suddenly were using values that meant very different things. There were also many problems caused because Opera was I believe the only browser to ever correctly support the third parameter of addEventListener(). Opera has since had to break its support because Mozilla concluded that too many sites would break if they implemented fully-correct support and ultimately I believe the spec has, or will be superceded with a version that reflects that end result.

And as a final point in my not-so-subtle suggestion that the biggest problem with site compatibility is actually web developers not knowing their job properly: Opera 10's user-agent is actually "Opera/9.80 (Windows NT 5.1; U; en) Presto/2.2.15 Version/10.00" because several high-profile sites assumed that no browser would ever reach a double digit version number (Flash has also seen this problem in a few places since version 10 of the plugin arrived.)

Yes, this was continually explained to people asking for https support in Opera Mini, but still it kept being requested. In the end implementing it seemed like the better choice, though they still explicitly recommend against using Opera Mini for secure transactions on their Opera Mini FAQ.