Showing posts with label Google. Show all posts
Showing posts with label Google. Show all posts

Thursday, December 11, 2008

Google Native Client

OR Changing the web one press release at a time

Google have released a new experimental browser plugin called Native Client that allows embedded application to interact with the operating system. Traditionally browser to OS interaction has been severely constrained for security reasons. Plugins that have attempting to provide more options for interaction have proven to be highly insecure. ActiveX attempted to do this by allowing the user to decide what plugins it trusted but users are generally poorly positioned to make these sorts of decisions. Google's Native Client plugin aims to provide secure interaction by identifying trusted modules for interfacing with the OS. I'm guessing that Google will be making the decisions about what modules can and can't be trusted. The up side is that you can develop an application and know that it will be as secure as the trusted module that is doing the work for you.

A nice feature of this system is that existing applications can apparently be quickly compiled to work with a minimal number of changes:

"As a test, Google ported an H.264 encoder that is about 11,000 lines of conventional C code. To get the encoder to compile and run in Native Client, they only had to add approximately 20 lines of code and modify the Makefile. The resulting binaries are completely portable and can run without requiring recompilation on any operating system supported by the Native Client. "
Safer than ActiveX: a look at Google's Native Client plugin

It's not hard to see that this has the potential to transform the internet (not that old chestnut again). Especially when it's combined with Google's Chrome browser and the growing popularity of Google apps.

Wednesday, September 10, 2008

Google Homepage Word Count and SEO

Marissa Mayer is one of the people responsible for the Google homepage and I just discovered a fascinating post from her on the Official Google Blog. It seems that a series of emails lead Marissa to become very aware of the number of words that Google uses on it's homepage. This lead them to adopt a 28 word policy. That means they use no more than 28 words on the basic Google homepage. Therefore when they decided to add a link to their Privacy page then '©2008 Google' dropped the Google to become '©2008'.

One of the reasons for Google's success is the simplicity of it's homepage and so you might wonder why more people don't copy this approach. It's more than a little ironic that the answer is SEO (Search Engine Optimisation). Google determines a pages page rank by 'reading' the content of the page and looking at what pages link to (and from) that page. Therefore having more relevant text and links on your homepage has the potential to increase your page rank and therefore make your site easier to find on Google. So Google has less words on it's home page so more people will use it to find your site and you must have more words on your home page so Google will let people find it.

Tuesday, July 15, 2008

Google and Radiohead's House of Cards

Radiohead and Google have released a clip of "House of Cards" that uses no cameras or lights. The clip consists entirely of 3D data created using Geometric Informatics and Velodyne LIDAR. "Geometric Informatics scanning systems produce structured light to capture 3D images at close proximity, while a Velodyne Lidar system that uses multiple lasers is used to capture large environments such as landscapes. In this video, 64 lasers rotating and shooting in a 360degree radius 900 times per minute produced all the exterior scenes." They have also released an interactive version of the clip that uses the Flash player to allows you to pan and zoom the clips data. The technology was managed by Aaron Koblin. It's worth a visit to his site to see art works he created using the flight patterns of birds.


Tuesday, November 27, 2007

Googlebot : accessibility in SEO clothing

You might think this an odd title.  I'm currently doing a makeover for a clients website.  At the end of four hours of CSS tinkering I was very proud to say the site looks exactly the same.  That is, of course, until you disable the CSS and see a lovely clear page of sequential text and links.  The original site looked pretty much the same with the CSS on or off and that was the problem.  This change isn't the consequence of my clients new born interest in accessibility.  It would be unfair to say my client doesn't care about accessibility but it would be fair to say he would find it hard to justify spending money to make the site more accessibile.  What my client has become very interested in is Search Engine Optimisation (SEO).  This is where you focus on changes to your website that will provide a better page ranking in Google.  The client thought SEO was important enough to engage a company that specialises in this dark art.  My changes to the site are the result of the SEO's report.  What I found fascinating is that most of the issues raised in the report are exactly the same issues that I would raise from an accessibility perspective.  That is they are about semantically correct HTML.  That seemed a little odd until I realised that Google's official site auditor is blind.  The Googlebot can't see your site.  It can only read your site and interpret what it thinks the site is about.  The sad fact is that most of the worlds successful websites only care about one blind user : the Googlebot.  Therefore if Google did nothing else to promote site accessibility than provide the world with a search engine then it is doing more for accessibility than any other company in the world.  For the rest of us we can stop trying to get our clients interested in accessibility.  Instead we can work on getting them excited about SEO knowing the outcome will be exactly the same.

Blogged with Flock

Thursday, October 25, 2007

Google v Microsoft

Quote of the day :

"Mozilla’s financial statement really puts the browser battle into perspective. It’s not Firefox vs. IE as much as it is Google vs. Microsoft."

Mozilla banks Google bucks; Builds a portfolio

Saturday, June 16, 2007

Gears, Pipes and Mashups

I'll be doing a presentation this Thursday evening at the Victorian MUV. The presentation will be an overview of the Mashup landscape combined with emerging trends as evident in Google Gears and Yahoo Pipes. We were talking about this a bit at Thursday's talk at the Gordon TAFE. I was pretty surprised that many of the people there were unfamiliar with these developments. Which is good news for my talk as it means you may find it all very new and exciting.

Jody Fenn will also be talking about UI Prototyping with Acrobat. This is a technique we've been using more and more as the applications we develop grow in complexity. But it's a great way to develop, test and get feedback on relatively straightforward websites. I can strongly recommend Jody's talk to anyone involved in interface design or development.

Tuesday, June 12, 2007

Google Street View : Ed Ruscha Revisited

Thinking about Top Street Views on the trip home I suddenly remembered how old an idea this actually is. In 1966 the American artist Ed Ruscha created a fold out book showing "Every Building On Sunset Strip". The technique was an automated camera attached to the side of a car. By driving down the street at a fixed speed he was able to photograph every building. Sound familiar.

Every building wasn't Ruscha's only photographic book. Similarly prescient was 1967's "Thirtyfour Parking Lots" photographed from a helicopter. Inspired by this book and the easy access of Google maps the blogger at AMP decided to create a Flickr set of parking lots.

Ruscha eventually went back to his painting. But it may turnout that his most durable contribution was his trail blazing photography.

Google Street Views : Product Placements

I guess most of us would have had a look at Google Street View (read more ) by now. In case you haven't it's basically a series of 360 panoramas of a city. You can use Google maps to look around the city streets that have been mapped. Where they've mapped a city they've done large chunks of the city. The way they make these is to drive around the city in a van with a special camera mounted on the roof. One unexpected outcome is all the odd details that are hidden in the details. It hasn't taken long for Top Street Views to take a Digg approach to rating these hidden moments. For example is this really a house breaker.

Looking at this shot of the Street View team it seemed to me it wouldn't take long for Google to start selling product placements. Imagine two girls holding a giant logo waiting in a street at a certain time for the Google van to take their picture. Or an actor from a current TV series doing something in character that would provide a hint to an upcoming "special event". There could even be artistic interventions. Perhaps lonelygirl15 could make an appearance. The opportunities for viral promotion escapades seem endless.

Tuesday, January 23, 2007

Dream Reader

Flock browser RSS Reader

A wish has formed in my mind. An idea about my RSS reading future. It will look like Flock's RSS reader. But there will be one major difference. My feed list will be stored online. I realised a while ago that I wanted my bookmarks online in one central repository. That way they could be accessed and updated whereever I go. But for some reason I haven't extended this idea to my RSS feeds. But now I understand. Anything that is accessed online should be managed online. Therefore, RSS feeds should be managed online.


Of course, it's easy to manage your feeds online using Google Reader and in Firefox you can quickly add a pages feed by telling it to use Google Reader. It's a slightly awkward process because Google keeps asking where to store the feed. But it's not too painful. The thing is I don't want to read my feeds in Google Reader. It's nowhere near as friendly as the Flock Reader. I guess what I'm talking about is a feature request for Flock. I want the Flock Reader to source my feeds from Google Reader.



technorati tags: , ,