Thankfully this functionality was eventually removed from the UI specifications after concerns from myself and others were raised. Just to make this perfectly clear; I think this is a bad idea, and here is why:

• Users are very wary when it comes entering their credit card details. Spaces magically appearing in their credit card number could freak a few of them out.
• Most e-commerce website I’ve ever used ask for the credit card number as one long string of numbers with no spaces. Auto-spacing goes against this convention, so a user may actually go back and try to remove the spaces (I know I would).
• Validation scripts for credit card numbers usually ask for the input to have no spaces, or will remove any spaces on submit.
• If you do want to space out the credit card number, why not use different input boxes for each set of numbers? It looks cleaner and is simpler to implement. Just have your JavaScript automatically jump between boxes as the user types.

Even though this functionality was removed, I decided to build a prototype of it in action because… well because I can and I was bored. :) Please, (please!), whatever you do don’t use this code in production. I created it as an experiment and should be treated that way. It’s very simple to get around the auto-spacing and then all hell breaks loose! (not really, but it is easy to break).

The spacing is implemented by counting the length of the string inside the input field, but you could also monitor the current position of the cursor. Anyone wishing to do this see this excellent set of answers on Stackoverflow.

You can see the prototype here. Remember, don’t use it. It’s not big and it’s not clever :)

After a little head scratching to work out what was failing, I worked out it was because I was using HTML5 elements such as ‘article’, ‘header’ and ‘footer’. There was no problem displaying them on the page, as I’d used Remy Sharps excellent HTML5 Shiv script. It only failed when trying to pull in these ‘new’ elements via Ajax.

Note: Before you read on you’ll be happy to hear that this issue no longer occurs in jQuery 1.7.0. Horray!

If you can’t upgrade, for whatever reason, I hashed together a little work-around for browsers IE8 and below, so read on. I admit the work around isn’t pretty, but it works. I tried for a few hours to get IE8- to recognise the ‘new’ elements after an Ajax request, but to no avail. Eventually I had to wrap the HTML5 tags in a div using conditionals:

The div is added for browsers IE8 and below, you can then use it as a hook to pull in the elements inside the div. I was hoping that wrapping the HTML5 elements in a standard div would be the end of it, I’m afraid not. The .load() method still didn’t work, IE just ignored the elements it didn’t recognise. I used jQuerys much more customisable .ajax() method to fix the issue in IE6, 7 and 8:

This method works in all browsers I’ve tested in as I’ve added a fall back (or should that be forward?) for newer browsers. The solution isn’t ideal (I really hate the conditional comments), but older versions of IE aren’t going away any time soon, and it works. Maybe there’s a more obvious solution that I missed? Leave a comment if there is, I’d love to know!

You can see a working example here. The page I’m loading from is here and the JavaScript in full here.

The real part of any non-trivial zero of the Riemann zeta function is 1/2.

I’d love to say that after reading the book I understand the Maths behind it but alas I do not. That’s not surprising though as it’s a problem that has yet to be solved, 159 years after Riemann paper on “On the Number of Primes Less Than a Given Magnitude”. It doesn’t look like it’s going to be solved any time soon either! If you’re interested it’s one of the Clay Mathematics Institutes Millennium Prize Problems in which you can earn yourself US$1,000,000.

While reading Prime Obsession it mentioned an interesting method for finding prime numbers called “Sieve of Eratosthenes” (pronounced air-a-TAWS-the-nEEss). Eratosthenes was a librarian at the great library of Alexandria around 230 B.C.E. It’s quite obvious how the method works once you know it, and easy to see why it’s called a sieve. The method goes like this:

1. Write a list of numbers in which you wish to identify the primes, say 2 – 100 (exclude 1 as it isn’t required)
2. Take the first number 2. Leaving 2 intact, remove every second number.
3. Move onto the next number 3. Leave 3 intact and remove every third number
4. Take the next number 5. Leave 5 intact and take away every fifth number
5. Repeat until you get to the end of your list

The end result is the set of prime numbers up to your given number. Essentially all you are doing is removing any number that is divisible by the 1st, 2nd… Nth number. Clever stuff!

As a little side project I decided to implement this method using JavaScript, displaying the result in the browser. You can view the demo here. For fun I created two implementations, one relying heavily on the DOM, the other only using arrays. As you’d expect the DOM method is slow; looping through large tables multiple times doesn’t make for a quick execution time. The second method using a single JavaScript array is much quicker. I’ve added a stats panel to the right hand side of the page so it’s possible to compare results from both methods over different input values.

I’ve added a limit to the number you can enter into the input box (5000), I don’t want anyone blaming me for crashing their browser by entering 10 million! :) Feel free to copy the demo if you want to see what happens at higher values.

The sieve in action showing prime numbers up to 500.

Thanks goes to John Resig for his very handy Array Remove method.It made removing non-primes from the numbers array very simple indeed.

Prime Obsession is a superb book but a word of warning, it contains a lot of higher level Mathematics, so may not be to everyone’s taste. It certainly made my brain hurt at times; not always what you need on that 8am commute to work! View the demo.

I’d attached the jQuery change event to the output elements like so:

So the plan was, once an input value had been changed using the jQuery().val() method the change event would fire and it would quickly colour in and out the element (using highlightFade) to notify the user that it’s value had changed. Unfortunately the change event wasn’t firing, so I had to find another way to monitor the output elements.

What I ended up doing was using an array to hold the previous values of the input elements then checking to see if they had changed in the calculation. If changed then fire the event:

It’s quite hard to explain exactly what’s happening so I’ve put together a small demo. Try changing one of the top input fields then tab (or click out) of the input element to recalculate. Is there an easier (or better) way to do this? Comment below.

It really is that simple! You may notice the “x-webkit-speech” attribute so will have guessed it currently only works in Webkit; and as far as I’m aware only in very recent versions of Google Chrome (note: “The Input Speech API will now be available by default on Chrome’s dev-channel”). The Speech API seems very new and is still a draft, so it will be a while before it’s fully adopted by other browser vendors.

A quick demonstration of the Speech API in action.

To demonstrate the speech input I’ve put together a very simple festive demo that queries the Flickr API looking for images related to a value entered. So if you have a microphone and the dev channel of Chrome, why not give it a try. Chrome records the input of the microphone before sending it off to an external server to be processed, where a value is then sent back to the browser. At first I thought it was all processed inside the browser but I guess that’s a little to much to expect! It isn’t 100% accurate and the demo accepts Googles first guess so you may get a few strange results. The server returns multiple guesses which a user can select from, but this hasn’t been added (to my demo at least).

Since the API is very new it took a little trial and error to work out how to detect if the Speech API was available in the browser; but with a little help from Paul Irish I managed to crack it:

This function will return true if the API is available. It’s worth noting that since it’s very new this function is likely to break in the future as the properties always seem to be changing. It works at the moment, so get thinking about what this brilliant new API can be used for! It’s a big plus for accessibility if used correctly and I can also see some cool applications in web gaming too. View the demo.

Mr Doob and a little kitten sitting in a 3D tree.

The File API allows a user to upload a file (an image in this experiment) to a web application simply by dragging it onto a designated “droparea”. Once dropped, you can manipulate the data any way you like. The API is able to handle multiple files natively, so just select all the files and drop them onto the area. No need for Flash uploaders or multiple browse boxes. Of course not all browsers support it (IE) and there are a few inconsistencies in the specification; but they will work themselves out over time.

So what does the experiment actually do? Well it allows a user to drag an image onto their browser window which is then “uploaded”; the data is then passed to a canvas element and then this data is passed to Three.js to be plotted in 3D space. The X and Z-axis is plotted in the usual manner to represent the default 2D image using particles; the fun bit comes when you add the Y-axis. This is calculated by taking an average of the RGB values (greyscale) and plotting it accordingly. When using an image of a human face it’s amazing how accurate the 3D plot looks.

Due to the amount of data involved it is recommended that you use small thumbnail images when you try the demo, as larger images equal a larger number of particles. This is an initial version so I’m going to work on a way of dealing with larger images. This will probably mean ignoring a certain number of pixels when looping through the image data. Looking forward, since it can be done with a static image I see no reason why it can’t be done with a video, but that’s a project for another day!

Update: I’ve updated the demo slightly; as Mr.doob pointed out people may have tried to drag directly onto the render area which didn’t work. So now it does :)

Want to see what samples were used in the new Girl Talk album?

The first application I really like is Mashupbreakdown by developer Benjamin Rahn. Mashupbreakdown is a visual representation of the samples used in the latest album by Girl Talk called “All Day”. That in itself is very cool, but where does it get the sample information from? Well that’s the really clever part; it gets it from Wikipedia.

Wikipedia has a page for the “All Day” album that contains all the sample information used in each track. Mashupbreakdown scrapes this page and and visualises it. Since Wikipedia can be edited by anyone, if you spot a mistake / new sample you can easily update the page information and the application with change accordingly!

At the moment it only shows one album but there’s no reason the application can’t be adapted to handle more albums. Benjamin has a post on his blog appealing for more breakdowns so I’ll be checking back with the site now and then to see if any other breakdowns have appeared. Great work Benjamin.

When chemistry and HTML combine you get the periodic table of HTML elements!

The second application that really caught by eye is the Periodic Table of Elements by developer Josh Duck. When I find a site I like, I immediately do a “view source” (don’t all Web Develepers?) to see how it’s been built. The Periodic Table of Elements takes any URL you pass it and tells you what HTML elements were used on the page. It’s basically a prettified / simplified version of view source! The application may not be that useful but kudos to Josh for taking the time to build it. Definitely one of those ideas that I wish I’d thought of!

Unfortunately a slight over-site on my part didn’t make deployment easy. I followed the instructions to the letter (more than once!) but received the same error on commit:

At first I thought this was a permissions issue but that turned out not to be the case. The solution in the end was because I was saving the do_update.cgi script in Windows format in Notepad++! Doh! Windows and UNIX new lines just don’t mix!

To save a file in UNIX compatibility mode in Notepad++ click “Edit” then “EOL Conversion”. Make sure UNIX Format is greyed out (why greyed out? no idea) when you save your CGI script. Upload to the server and you’re done. Won’t be forgetting to do that again any time soon!

It just so happens that I’ve had a request for a customisable header image on an future project, so decided to have a play and see how difficult it is to implement. Well it isn’t difficult at all, very simple in fact.

First you will need to edit your functions.php located in your theme directory. Go ahead and create one if it isn’t in the directory. Now copy and paste the code below into the file. I’ve commented the code but it’s all quite self explanatory. Change the height and width of the image and change the directories where needed.

By adding the code above you will now see a new option under “Appearance” in the admin panel called “Header”. From there you should be able to see all the images and options we just set in the functions.php file. The only thing left to do now is add the header image to the theme.

Where you place the following code will be depend on your theme setup; I’ve decided to place it in the header.php file as I want the image to appear on every page. You may only want the image to appear on specific templates e.g. pages, archive, category (you could also use WordPress conditional tags to do this).

And there you have it, one header image that you can customise via the WordPress admin panel. An ID is attached to the image for CSS styling where needed.

The example above is quite a simplified version of the header image functionality. It is possible to place text over the image and even have a different image per blog post by using the custom thumbnail functionality, but these are beyond the scope of this blog post. Copy & paste and enjoy!

What Aloha actually does is make the content editable right on the page; no need to jump into the CMS, find the page, make the changes then preview the result. Just click on the text you wish to edit and type away. Done! It’s hard to believe it actually works until you try it. Why not give it a go on their demo page. It incorporates a few HTML5 technologies which could cause issues with older IE versions, but the demo page seems to work (with a few quirks).

Quickly and easily edit the demo webpage.

The core has been written to be very small and streamlined, with additional functionality being added by the way of user contributed plug-ins. There’s already an extensive API for developers to play with so I’m sure many of the missing features will be plugged very soon.

There are alternatives about such as TinyMCE and CKEditor, but they’re integrated directly into a CMS’ administration pages rather than as an inline editor like Aloha. Now before you get too excited there are a few cons:

• It isn’t as simple as just dropping the scripts into a page, it needs to be integrated into a backend system for the changes to be saved.
• Still in early development so could be quite buggy.
• Doesn’t work with Opera Browser at the moment but this is being worked on.
• Some features aren’t available yet, like image / media insertion. But I’m sure these will be developed as it matures.

So there are still a few issues to iron out but it all looks very promising! There were even murmurs on Twitter today about a WordPress plug-in and a Drupal v7 module (not 6 though I presume) in development. So it won’t be long before we’ll be able to try it out on a live site; hopefully it’s also simple enough for clients to use too!