At least that's the plan according to Intel who've released details on their new Larrabee graphics chipset. Remembering the days of the old software rendering it was both a blessing and a curse. The performance was severely limited compared to the hardware 3D accelerators, but it did mean that you could run games without one, which was great for laptops.

Nowadays even the budget laptops are thrown a basic Intel graphics chipset with a 3D accelerator, so why the return to the software, over hardware, based architecture. Well as CustomPC reports :

This means that Larrabee will not only be completely compatible with DirectX and OpenGL, but that it can also be easily opened up to other new APIs with its x86-based architecture. On top of this, it could also theoretically work with future versions of DirectX via a software update, without having to upgrade your hardware.

Intel also suggest that the architecture will scale well with multiple-cores and even hinted that they may be supporting multiple card configurations much like nVidia's SLI technology.

Could Larrabee be Intel's break from budget, to premium graphics card vendor or will the software based architecture fail to perform compared to the optmized hardware based architecture of their competitors? Larrabee won't be released until 2009-2010, so there's plenty of time for us all to speculate until we are able to find out.

If you've heard anything about Windows 2008 you'll know one of the big selling points is Microsoft's new built-in virtualization toolkit called 'Hyper-V'. What I wasn't aware of is why Hyper-V was such a big deal over Virtual Server and just exactly what goes on behind the scenes with virtualization products.

Thankfully Ars Technica have written up a handy guide to Virtualization, the techniques employed and just how it all works. I look forward to the future parts to the article.

I stumbled across this article "8 Web Design Mistakes That Developers Make" a while back and as I made my way through the list none of my colleagues (or myself) sprang to mind as being guilty of such mistakes on more than a handful of occasions.

Instead, I began to think about the kinds of mistakes I know we all make again and again and so I put forth my only list of 8 design mistakes that developers make. I'd love some feedback especially if you spot anything not on the list or have tales of that 'friend of a friend' you know who had a few mistakes of their own to add.

8. Know your limits

Before you start on your design project, determine if you're upto the task. If you've never designed a site before and you're tasked with the revival of a hip fashion conscious urban brand's website, now may not be the time to try your hand at a little Photoshop but instead hire a design creative. If you're designing a small extension to an existing site with an existing style, then it will be much easier to dip your toe into the creative waters when you already have the image files and you are following someone's existing styling and guidelines.

It is much the same principle when learning a new language, you don't try to design an enterprise application out of the box you start with 'hello world', the creative world is no different.

7. Believe good design can be learned

I've often heard developers moan that design and art are creative skills that one simply posesses or does not, that there isn't some manual for how to be a good designer and so there isn't much room to improve their design skills.

This is simply not the truth. While being a good designer does indeed involve a large slice of inate skill and an artistic eye for detail that few have, you can certainly improve your design skills a great deal by learning about spacing, color theory and typography. It probably won't lead to a position as art director any time soon, but it will improve your designs greatly.

6. Look and I mean Look at design around you.

You know how you copy and paste that cool bit of code you saw somewhere else? Here's a secret, designers are thieves too. The world is full of design inspiration, in nature as well as existing man made designs. Take a magazine and actually take the time to look at the layout, how the photographs and images are placed, the fonts used, etc. Next time you hit a website, study the graphics and maybe take a peek at how they did it. Look at the interface for your phone, your kettle, even the menu on your DVR. There's design all around you, some of it good, some of it bad. The more you notice it, the more you can begin to identify what works, what doesn't and how to improve on others ideas. You can then use this as inspiration on the next UI you design.

5. Be inspired by other designs, but only a little

By this, I mean don't turn into the 'bevel guy', the person who continually designs their buttons in the same way, overusing the same designs again and again. While copying individual ideas from somewhere else is generally accepted, there's a rather distinct line between taking one or two design elements and incorporating it into your work and outright plagerism. If you like someone's design work, hire them, don't rip them off. If you're doing your own work, then do your own work and make your own unique design don't copy another work completely. Doing so isn't just disrespectful and unprofessional, it is also hindering your the development of your own design skills as you rely on someone else to do the work for you.

4. There's a reason for branding, use it

Many companies have branding guidelines for their design work. Such guidelines usually entail detailing precisely which shade of blue to use, which fonts and images, etc. This helps to ensure a uniform look and gives a professional feel. Using random fonts that you like, clipart and picking alternative color schemes or merely getting colors 'close enough' makes will give your work an obvious amateur feel. Such conduct should be entirely limited to PTA newsletters.

3. Apply new technology only where appropriate

Designers are just as guilty of this as us developers (Flash, anyone?) but nothing is likely to turn a user of quicker than inappropriate use of technology. While I'm sure you can't wait to turn the billing system over to Silverlight that doesn't mean you should. Think carefully about your justification when using a new technology, if it boils down to 'this is going to look sweet!', then you are more than likely setting yourself up for failure. AJAX is a particularly overused technology that springs to mind, if you want a fantastic example of the perfect use of it, head over to any of the news stories on MarketWatch.com as during the stock exchange hours, the tickers for companies listed in the stories will regularly update with the latest prices and information.

2. Use the right tool

Related to number 3, but not the same point. I've often seen developers using MS Paint to create graphical elements for their projects. While I know where they live and shall be administering a severe talking to shortly, they could have avoided this by simply using Paint .NET. If your design work is difficult or turns out looking poor (and you've followed the above steps) then the problem might not be you and could be the software you are using, try to find a creative to ask what they'd use in your situation. There's a staggering number of design applications and most serve specific functions. While Photoshop might let you design in CMYK for print work, that doesn't meant it is the best tool for the job.

1. Relate the design to your user, not the data

Without doubt the single biggest mistake a developer makes when tasked with design work, is by coming at the problem from the 'well goodness, how do I fit all this data onto the screen neatly' angle. Wrong, wrong and I say again, wrong. Good design comes from asking what your users need to see. Maybe you need to display a list of contacts and while you could just throw on a datagrid and display all their information such as address, email, phone number and such, it may be more appropriate for a user to see only their names and hover over or click on a user's name in order to see a contact card with their information on (much like MSN Messenger).

Always remember to keep in mind the task your user is trying to acomplish and provide them with all the information and tools they might need, without bombarding them with too much data and too many options.

Been a while since I updated, this is what happens when life gets in the way of geeking. However, my return to posting is sparked by an exciting new project I'm working on, which for the moment is named Jigsaw and has a five minute logo that looks like :

This'll get updated once I get the time/inclination to work on the design as it is something that at the moment looks clearly designed by a developer ;)

So what is this Jigsaw thing?

Jigsaw is borne out of a frustration of mine when working with current CMS/Blogging offerings to cobble together community sites. It was never simple with a lot of CMS packages offering great blog style experiences with limited forum functionality. In the case of Joomla you're looking at user management extensions, a seperate forum application and a bridge between the forum and the CMS. This kind of thing isn't unique to Joomla and while there are a few CMS that do offer reasonable forum support out of the box, it regularly breaks with new versions of your CMS or Forum software, has skining issues when trying to make the sites look the same or as is the case with PHP Nuke are so hideously complex that all but the most determined are scared away.

Jigsaw will be my attempt at fixing this particular problem of mine, taking the lessons learned from the admin/CMS toolkit that I bolted onto my old FurtherDown site and creating a more robust, flexible and secure system.

The basic idea

At the heart of it, Jigsaw takes the idea that with a blog and a forum you essentially have the same data, a main post with zero or many 'replies'. These replies can be to the original post, or to other replies. The only difference between what we commonly identify as a forum and a blog is how this data is displayed. Often times with a CMS system you want to combine the two ideas so that forum posts make it to the 'front page' of your site, where the information is structured more like a blog.

What I want to do with Jigsaw is have a base structure where a post objects belong to some sort of post structure, this will in turn have identified categories and possibly subcategories which in the case of a forum would constitute the different forums. By choosing a display style for this post stucture, you can determine if it is shown as a blog or as a forum. There will also be the option to create aggregate structures that will pull posts from existing post structures based on a defined criteria.

Lost? Think of a site like Gizmodo, where you have a number of approved contributors adding content to the site. Each contributor could have their own blog, these could possibly be private if you want the users to see the site only in the agregated state or public if you'd like to give them the option of breaking the site down by author. Now, the main site could be an aggregate structure that will only pull posts from the blogs of those in an approved user group, say 'contributors'. You could also then allow any user to start up their own related blog on the site, so that along with your 'approved' content, the site is also generating a lot of related user created content.

So then what? well why not throw on an aggregate structure that grabs the best of the user generated content as well? How do you tell what the best user content is, add a digg style plugin to the user blogs, where other users can rate or vote for content to make it to the 'best of' page, then you just stick on a value for the threshold of good content and rather quickly you have an involved interactive and user contributed community site with almost no administration.

That's not the only neat trick Jigsaw has planned up its sleeve, but I'll reveal more of those once I'm closer to a working project, for the moment I have two rather large blank whiteboards, a basic Visual Studio project and some cool ideas. I'll be updating as my progess continues.

The nanotechnology revolution can't get here soon enough as far as I'm concerned and after seeing this concept video from Nokia of their thoughts on future technology, well I'm desperate to do a Buck Rogers and freeze myself for a few decades until the technology exists.

For now, I'll just have to settle for escaping into a game of Deus Ex until it arrives.

Full credit to the AskTheCSSGuy on this one, for creating a detailed and beautiful how-to on creating a table with dynamically highlighted columns. I've then tacked on my own quick 5-minute hack to make the highlighting more fluid by removing the button clicking and moving the highlight to mouse over, which you can see by clicking here or on the image.

While most of the US was glued to the TV this weekend for the Superbowl, it seems the guys at Microsoft have been rather busy, releasing both Windows Server 2008 and Vista Service Pack 1 RTM.

Hyper-V is still in Beta though and won't RTM for another '180 days'. Ouch.

Google have posted a response to Microsoft's offer to buy Yahoo for $44.6bn on their offical company blog. Naturally, given their history with Microsoft and the rivalry over issues such as desktop search software, they are rather critical of Microsoft.

“Microsoft's hostile bid for Yahoo! raises troubling questions. This is about more than simply a financial transaction, one company taking over another. It's about preserving the underlying principles of the Internet: openness and innovation.
...
Could Microsoft now attempt to exert the same sort of inappropriate and illegal influence over the Internet that it did with the PC?”

It seems rather hypocritical of Mr Drummond to be levelling such accusations at Microsoft, given his company happily provided a censored version of their site for Chinese users and rewarded the 'openness' of a CNET reporter who detailed private information (found by using google) on Google's CEO Eric Schmidt by blacklisting CNET reporters.

Perhaps their mantra should be changed from “Do no Evil” to “It isn't Evil when we do it”.

Interestingly, as well as the Yahoo search engine, it'd give Microsoft control of Yahoo's search marketing tools, which I could easily see being reshaped into a more credible competitor to Adwords.

In honour of WKRC Cincinnati's Good Morning show, tomorrow has been named 'Dance Party Friday'.

 Work should be this much fun for everyone.

Marketwatch.com has a fantastic use of AJAX and web services on their site. Like most financial news sites, they include the stock symbols and information on the companies involved in the news story. The page takes on a new life when the stock market is open, with the symbol information automatically updating as you read the story. So instead of just reading about the wonderful success of MyCorp's stock rising through the roof, you actually see it happening.

Genius.