It wasn't particularly accurate for me, pointing to London and I'm a good hour's train journey from it.

You'll be glad to know you have to opt in on each website and there are good uses for it. Just like at the moment various apps on the iPhone and Android can tell you the nearest restaurants, events, weather, news, etc. The internet will soon be able to do so as well, if you want it to.

The reason for the -webkit, -msie, -moz and -o prefixes is that CSS3 isn't finalised yet. For example, the format of parameters for background gradients is still being worked on and both Webkit and Mozilla have different ideas for how to implement it. For Example:

background: -webkit-gradient(linear, top, bottom, from(rgb(0,0,0)), to(rgb(255,255,255)));

background: -moz-linear-gradient(top, rgb(0,0,0) 0, rgb(255,255,255) 100%);

The above would generate the same linear gradient, but as the format of gradients isn't finalised, there is no standard to the parameters

It should also be noted that Webkit currently processes anti-aliasing before it processes any CSS transformations making things like text-shadow and CSS rotations look awful

Chrome (and I presume therefore Android's browser) does actually contain Webkit and Safari in it's User Agent string. In fact the developer tools in Chrome still use Safari UI elements.

The Geolocation spec says it has to be opt in. The reason for it's existence is the recent increase in internet use on mobiles. With Geolocation, your mobile could tell you where the nearest restaurant is, what's on at your local cinema, give you the phone numbers of local taxi companies and tell you what the weather is like without having to have multiple apps on your phone.

But this is not something completely under the control of the designer at this time. All the major web design companies and major companies with inhouse web designers design for different browser groups, not resolutions. If you take the BBC as an example, their website serves up a fixed width output based on which browser you access it with.

The point behind picking 960 as a width is that fluid layouts don't work the vast majority of the time across browsers and, without multi-column support as is coming in CSS3, content is hard to read beyond a certain line length.

Also, pixels are used in the vast majority of designs still because relative sizes, such as em, are unweildy as each size is relative to it's parent container and has noting to do with the client resolution.

I see a large number of problems with the belief that HTML5 will defeat Flash. It will become another competing format, just like Silverlight, but I highly doubt it will defeat Flash in any real meaningful way (and even if it did, it wont remove the things people find a problem with Flash, other than possibly the searchability issue).

Firstly, most web developers these days will write javascript for a framework such as Prototype, MooTools or JQuery because there are so many individual nuances to the javascript notation for each browser that it would be too frustrating to write it without such a framework. This problem will still apply to HTML5 so unless these supposed IDEs pick 1 framework and run with it, we're going to end up with a complete mess of features, frameworks and IDEs. By the time one of them becomes as usable as Visual Studio or Flash CS5, Silverlight and Flash would have both moved on in leaps and bounds attempting to stay well ahead of HTML5.

Secondly, by the time HTML5 is ready for real use (i.e. supported by the vast majority of browsers currently in use), W3C will likely be back around the table discussing HTML6 so who knows what Flash and Silverlight will be capable of by then and what uses we will have for them.

Thirdly, at the moment, HTML5 is far more complicated to write for than either Flash or Silverlight, in most cases it has a significantly larger file size, uses more resources for most of Flash and Silverlight's current usages and it is a lot harder to obscure code or source material, meaning sites like Hulu just wont use it.

There is a place for HTML5 (I'd be happy just for an implementation of canvas in IE so we don't have do all our chart generation through PHP) but it is not, and is never really likely to be (as HTML5 and not a future variant) the death of Flash and Silverlight.

Lastly, I'm sure most people here are happy to use Flashblock to stop any obnoxious Flash site but what will you do when canvas,SVG and CSS (no need for javascript) can reproduce all those annoying flashy adds with little to no way to block them all?

Yes, I have to manage 3 large insurance websites with no CMS tools... Thanks

They've already started doing it: http://www.9to5mac.com/Flash-html5-canvas-35409730

I presume you mean the full width. If so, you're going against the grain of usability on this one I'm afraid. The majority of designers with good knowledge of usability and CSS will build the design to a width of 960px wide as this allows for a usable design inside the max width of the majority of their visitors. If you start pushing wider, the text of the main body becomes too long in line length and therefore hard to follow on longer articles. Unfortunately, although CSS has a min-width and max-width property, they are not very reliable and are usually used only to fix problems in IE. Hopefully, once CSS3 and HTML5 are properly supported we'll have the flexibility to provide designs that adjust properly to a browsers width.

In fact, they'd be worse. You can block Flash by either not installing the plugin or using one of the many Flashblock plugins. You wont find it so easy with HTML5.

You mean the HTML that doesn't do vector graphics? Canvas adds support for it but you are then either writing SVG for it or dynamically generating it through Javascript, either way, drawing it in Flash is much easier. Also, these days, if a Flash file is created properly, it too is searchable.

I hate Flash as much as the next guy, but I have to agree that from a designers point of view, Photoshop and Flash will still be the tools to use for a good few years yet. I'm just glad that you can at least block Flash if it gets obnoxious... That wont be as simple with HTML 5.

Maybe you should have considered my whole statement. If I installed SC2 at any point when Bnet2 is running I can upgrade anything on the computer and still play. Even if you buy a new PC you can just copy the folder from one PC to another and it will still work.

Personally, I'd rather this over the old CD protection because over the course of Diablo 2's existence, I've had to buy 3 copies of it due to damage. So I have no idea how some of you think that this possibly breaking in 10 years (and no one managing to make a fake server app in that time) is any worse than the possibility your CD will break.

Starcraft 2 is pretty much idientical to this, except the single player which can be played in offline mode.

You wouldn't need to. The single activation they are talking about is connecting your CD-Key to your Battle.NET account. Once that is done you can just use that login to play, no matter who's actual installation you are playing on. Also, as long as you have logged in at least once on the installation with the internet connected, the game will allow you to log in while not connected (On beta, it logs in but then kicks you back out because it can't load any of the screen you have access to (multiplayer ladder, achievements, news, etc.))

I think this is probably down to the fact the beta is designed for performance testing and Battle.NET is only used as the set up medium (Blizzard have stated once a game is started it runs P2P), so it could be dropping you out of games because it's failing to update performance logs, etc. online

While I've been playing the beta I've noticed a peak in connection at regular intervals throughout games (both vs CPU and another player), even to the point of having to wait a moment for the game to sort itself out which suggests some kind of data transfer happening.

My suggestion would be to wait and see if the same problem occurs with the release version (obviously from someone else).