Alan Coleman

I’m still trying to get my head round Microsoft’s crisitism of our comapny website this week. Yes, Microsoft. The comapny that revolutionised the web with Internet Explorer, the Firebrands that changed the face of search with Bing and bossed development with Silverlight (Remember that?).

Don’t get me wrong, there’s loads about Microsoft that I can’t get enough of. Bill Gates, the PC, Windows, .NET, Windows Server and T-SQL. Love it all. However I can’t help feeling that Microsoft are still playing catch up with the Web and Internet, it’s like a loveable friend with too much money that gets into the same sport as you. All the gear, no idea!

It’s slightly diffrent with Google of course. All they’ve got to do is sneeze and we all start clapping and coding with a sense of blind panic like the worlds going to end, which in a way it could!

That said, the loveable friend with no idea is still a friend and needs to be humoured like everybody else, we wouldn’t want him going home and taking his ball with him would we?

I came across this issue after kicking off an automatic upgrade to WordPress 3.3. The visual tool-bar disappears and it’s no longer possible to swap to code view. After a little searching about I found this solution.

You’ll probably need to install “use Google Libraries” plugin.

I’d only ever played around with JavaScript as a way of achieving something within a development environment. I’d given little thought to object orientation and code reuse- ability, focusing more on write once, use once and move on as soon as the thing was working.

Then I read DOM Scripting: Web Design with JavaScript and the Document Object Model which changed everything for me. Off the back of that I moved onto Advanced DOM Scripting: Dynamic Web Design Techniques. Just as good although significantly more complex. I did feel at least that I had a good grasp of JavaScript and its client side implications.

Things are different now, everything about how we write code and deploy it is unrecognisable from a decade ago. Time and expense are an issue with everything we do and with applification gaining momentum none of us can hang around.  Now we’re using frameworks, in my case the jQuery library, for our client side scripting. Development can be fast, although the environment on which the framework is being laid can often present issues surrounding access, especially if you’re dealing with legacy code.

For myself there are two drawbacks to using a framework. The first is the Framework itself which must be downloaded to the client browser, along with any plugins, which can be weighty and as with any JavaScript can cause havoc in an untested environment. The second is less to do with the framework itself and more to do with the developer, who no longer writes his own code, but rather gives instructions to someone else’s. Not the same level of code satisfaction but just the results are just as much fun.

So I’ve gone from reading  about JavaScript using the Document Object Model to reading about frameworks that all but dispense with the considerations of the former.

Happy days.

I’ve been looking at content management today. We need to upgrade our current CMS as the front end isn’t giving us the flexibility we need. There are a million CMS products available, all offering more simplicity than the next, the big list of corporate clients, the 30 day trial and sandbox etc.

Most products centre content management around the web pages, using in-page editing for ease of use and simplicity. I installed a copy of DotNetNuke 6 to take a look at, works really well with its flawless AJAX editing, plug ins and extensions. The thing is, we already have a database schema in place. when we set about this project a few years ago we always thought that the front end would be replaceable and it was the back end that we were going to get right. At the moment our front end is a simple table editor, and it’s this we’re looking to improve.

This is where I’m having the problem, we don’t need everything that comes with an off the shelf CMS, forums, galleries and alike. We’re just looking for a flexible table editor that gives us some WYSIWYG functionality for the editors. The .NET MVC process looks like it would do the job, but I don’t want to have to rebuild everything just to give us a WYSIWYG editor, there’s got to be more result for the work involved.

What to do about a CMS.

So now I can update my blog by using the fantastic WordPress Android app. Brilliant, why has this taken so long? WordPress, Android and HTC, what a fantastic combination.

Last month Joyent ceased to support their Connector suite of applications, the combined email, contacts and calendar app, next month they’ll be stopping the service completely. The Ruby on Rails app is several years old now and can’t really keep up with other offerings on the market, besides Joyent have moved their focus elsewhere, outside of basic hosting. Fair enough, I can see why they’re making the move, the product is no longer central to their business aims and as time creeps on the app will drain resources as it becomes harder to maintain.

Not everyone at Joyent used Connector, I did although not a daily basis. It was a useful, secure and reasonably stable place to keep my contacts and important snippets of information. Now that the service is coming to an end there’s one small question I have to ask. What now?

I have to keep this information somewhere and realistically it has to be online in the form of some sort of application. Irritatingly there’s no way I can move all my contacts from Joyent Connector in some sort of import export process, so I’m going to have to move them one by one. Mmmmm.

So now I’m looking at either Highrise from 37 Signals or Contacts that form part of Google’s Gmail. There’s pros and cons of both. I’m already using Gmail and it’s free but on the other hand I think I’d like to use something that’s not Google or Yahoo, I use those apps all day every day and sometimes fear Goo-hoo overload. However, I’m already using Google calendar and that seems to link seamlessly with Gmail, so maybe using contacts is just an extension of that. Google apps are free too.

I’d rather stamp on my own fingers that have to do this again in a hurry!

We run a dynamic website that retrieves information from a relational database. There’s lots of content involved across loads of domains and in several languages, this makes the site fairly heavy. Performance is as important as ever and it’s only going to become more so as users patience decreases inversely as bandwidth increases. Mmm.

One solution is to cache content that doesn’t need to be refreshed that regularly. The time between refreshes can be anything, a few minutes, days, weeks or never. It’s a fairly simple process that lends itself to any scripting language that reads and writes to a database. The best part is that if you’re already reading content from database tables and displaying it, you only have to adjust your script to check another table for the cached content date and act accordingly.

Here’s a really simple example of how it might work, this ones in pseudo VB:


'Query your cache table for content
objRSNavBuild.Open "SELECT * FROM tblNavBuild WHERE BuildType = 2 AND CountryID=2 AND LanguageID=1"
'Set refresh variable
navBuildInterval = 720 '24 hours
'Set timestamp from query
lastBuilt = objRSNavBuild.Fields("timeStamp")

'Otherwise write out your cached content
else
Response.write objRSNavBuild.Fields("navigationBuild")
end if


One of the most attractive aspects for WordPress has to be the method of upgrade. A simple message appears on the dashboard in the manner of desktop software and you’re one click away from all the latest goodies. For me, this is the point where on-line apps take over from applications confined to the desktop. All the processes are seamless, and most importantly are tested thoroughly before release – to the point that end users can action a single command and watch the magic.

It does recommend a backup prior to upgrade but as my install isn’t mission critical I didn’t bother, the upgrade took about 20 seconds. The beauty of this blog entry is that I don’t even know what’s new in 3.1.1! Does it matter? Not really, just as long as it’s a simple process.

I’m considering moving some of our CMS editing at work over to the WordPress platform, the WYSIWYG editor works so well, especially the more recent versions, and entering content has been a stumbling block for our users for years. At the moment they’re writing the HTML tags themselves when inputting content, the upside is that the mark-up is quite clean, but the downside is that content people don’t want to get involved with mark-up (fair enough). If I can get the WordPress install working on IIS then we could have an awesome solution.

The 5oth anniversary of Yuri Gagarin’s journey into space seemed like an appropriate time to talk about Russian graphic design

I’ve always liked the simplicity of Russian design. The heroics of the people, the sweeping blocks of colour and style.

I find it amazing that a society that was so isolated from western culture managed to produce graphic design that was so similar to that produced in the west. Of course it was different in terms of the subject matter, which couldn’t have been much further away from knocking out Coke and lifestyle products.

But it’s kind of like all graphic designers did the same course at the same college before going back to their respective states. The use of space, texture and typeface of Russian graphics all comply to the basic ‘rules’ of design. Those underpinning concepts of aesthetics that appealed to us in the west seemed to come as natural to those in the east too.

You couldn’t say the same about music.

Check out this poster, the colours, simple text and feeling of speed all point to ‘just enough’ and ‘just in the right place’. It’s awesome. Reckon I’ll be bring some Russian graphic style to the pages of this website over the next few months, Komrade.

I’ve been working on a project over the last week that involves fairly heavy use of Stored Procedures. It’s not a new technology and we’re not that new to it either, it’s just that during the development cycle the logic of Stored Procedures seems to have fallen off the radar.

Stored Procedures are perfect for us on a number of fronts. We run our entire website on a handful of files so the code base tends to be huge, moving some of that code out to a more appropriate home makes complete sense. There’s less possibility of an accident in the few thousand rows of code, and it also allows us to extend our code separation policy to new levels.

The creation of an abstraction layer between content and code also creates some interesting possibilities. Applications other than the website can execute the procedure and make use of the results, also developers can test queries without having to fish the SQL out of the code and replace all the variables, this often goes unmentioned but for me is one of the more useful aspects of the technology.

CREATE PROCEDURE sp_getproduct
(@UnitPrice money,
@UnitsInStock SMALLINT = '10')
AS
BEGIN
SET NOCOUNT ON;

SELECT [ProductName] FROM Products
WHERE ([UnitPrice] < @UnitPrice) AND ([UnitsInStock] > @UnitsInStock)
END
GO

I’ve been working with cookies for the last couple of days, also known as a web cookie, browser cookie, and HTTP cookie (got that from Wikipedia). For some reason I always end up getting confused about what cookie is written where and how I get my code to respond to it, it’s simply never as easy as writing code when cookies are involved.  Chris Pederick’s awesome Firefox Developer Tools go a long way to eradicating the hit and miss of deleting cookies from a session or domain, doing the same in IE or Chrome is nothing short of a nightmare. Firefox to the rescue, again (Just seen that Web Developer tools are now available for Chrome).

Whatever the working issues, cookies do allow us to manipulate our site to respond to users in different ways, which can be very powerful. Couple this with Google Analytics and we have a blinding set of tools for testing what works, with who and where. Again, very powerful. Also from a development perspective, the idea of making changes to a page and those changes staying with the user (without the URL changing) throughout their visit is quite satisfying, although not smug.

I’ve been using WordPress to blog in various places for a number of years now, but it’s only recently that it’s started to live up to what it should have been. WordPress seems to be converging at point where all web apps  have aimed, almost like it’s stepping up to the mark.

I was fairly slack with upgrading and ended up using an ancient version that didn’t even support the one click upgrade that comes with this one, so I ended up reinstalling the whole thing. No trouble at all, just the usual fiddling around with phpMyAdmin (Needs are a must with that one!).

The changes where nothing short of incredible, in the space of a couple of years the entire package had gone from a database update tool to a proper content publishing package. All the frustrations I had with the old versions and their issues with formatting content had been resolved and improved beyond recognition.

It’s awesome, and I’m a huge fan.

Amid much fanfare Firefox 4 was released today. I watched and listened to the event live on the web, and heard the download count increase in its hundreds of thousands to whooping and applause. It sounds boring, a load of geeks sat around talking about their latest browser and how the various plugin will or wont work. But for some reason it made for compelling viewing. Programmers talking about rebuilding the JavaScript engine, marketing people talking about the release schedule and developers talking about the new website, I sort of wished I was there – glugging coffee and twittering continuously on a silver Mac.

The browser itself is awesome. The first thing to notice is the blisteringly fast page rendering speed, almost certainly down to the work the programmers and developers put into the JavaScript engine, the same thing that Google did about 18 months ago when they released their browser. I’ve only being playing around for a couple of hours so have yet to fully grasp the capabilities of this release.

The upgrade was smooth, reinstalling the latest and most compatible version of Fire Bug and even reopening the tabs that where open on the old version. The little things.

So where next? Pretty much all the real estate that is possible to be displayed on the screen is now given over to content and it’s difficult to see a point in the future when connection speed will come close to matching page render. But does it matter?

Right now it’s just great to indulge in something so slick and usable. It’s been an emotional day.

So we get broadband installed with our new Sky package and it’s advertised at a fairly respectable 20 Mbps. Start the machine up and everything connects fine, the wireless password works and the router is doing the business, wicked.

Or is it?

A speed check revealed a paltry 3Mbps, I couldn’t hide my disapointment.

Is there any point ringing customer services? Probably not, the 20 Mbps advertised is probably a best case scenario for a user living at the exchange browsing at 3am, for everyone else, which means most customers, the actual speed still only provides for an adequate browsing experience.

The capability of the Web is flying forward at brake neck pace whilst the media we are consuming is getting heavier by the day. It’s at this point that I make a profound comment about the direction of the web and the infrastructure that supports it, but there’s not really much point.

Sky are no different from HSBC, they know we don’t have the time to complain, especially when they know that we know that nothing will change. No?

I’ve been using Google Analytics a lot recently and at work it’s become an invaluable tool that guides all our online activity. The wealth and sheer depth of information available is staggering, as is the functionality afforded the user courtesy of their front end developers. This is what can be achieved when you have a budget like NASA and an equally sizable resource of talent! That’s not to take anything away from the application though, from a business perspective it’s difficult to imagine how we ever managed without this kind of feedback before. From a developers perspective I found the information at hand especially useful when rewriting URLs in a friendly fashion and anticipating any potential damage.

I think this kind of application is what Web 2.0 is really about, and more importantly it represents a turning point in the history of the web and how we develop it.

So is Google Analytics better than The Poseidon Adventure? The only reason I’m asking that question is because it’s on the box behind the one I’m typing this blog entry on, and it’s a brilliant film. A badly acted 70s seascape epic about scantily clad women trying to escape from an upturned liner. Whilst Google has certainly had an impact on my life professionally, it’s not yet at the point where it’s going to overtake the effect that Susan Shelby had on a 11 year boy from suburban Essex.

As the spring sun shone through our office window this morning I came across a couple of great things online that made the already great day even better.

The first is an online code editing app from Mozilla Labs called Bespin (Bespin video). Although only in Alpha at the time of going to press its potential is already mind blowing, a fully functioning browser based coding environment that keeps your file in the cloud. Fantastic. There’s more, included is a dashboard style file manager and command line functionality. The whole thing is put together with JavaScript using the HTML 5 Canvas Element and makes for an intuitive work area that I’m sure will at some stage offer some sort of CVS.

The second item from today is the discovery of Personas, which is also from Mozilla Labs. Essentially it’s a plug-in for the Firefox browser that allows users to skin their browser in a variety of different patterns and themes. It’s all in real time so as you navigate down the list the whole browser changes before your very eyes. Okay, it offers no additional functionality, but it does look good, great in fact.

On the same lovely morning someone sent me a link regarding an extension to Microsoft’s Expression web environment that checks your web layout in IE 6,7 and 8. So is this an admission from Microsoft that their browsers are incompatible with each other? Or just an acknowledgement that their browsing capability is a badly organised, over funded river of shit that simply does not work properly.

Redmond, you are a long way behind.

I have a couple of great additions to my world of web development that are worthy of a mention.

The first is the discovery of WebDevRadio, a site based around a set of regularly updated podcasts on web development issues. The podcasts themselves are hosted by Michael Kimsal, an genuinely friendly bloke who discusses anything from the weather in his part of the world to the finer points of his favourite PHP class of the moment. It’s all very listenable and down to earth, in fact he comes across like he’s talking it over with a few mates down the pub. On the back of listening to the latest podcast whilst doing my ironing (2 shirts, that was it) I tried to find some UK based podcasts and came across Fresh Egg, a design and marketing agency who I think are based out of Reading in Berkshire (shit county). If it’s not Reading then I apologise, but not to the people of Berkshire (shit county). Their podcasts stand in complete contrast to those of Michael Kimsal, both in terms of delivery and content. For a start, their overwhelming self importance gives listeners the feeling of being spoken down to, not good. Then there’s the intensely annoying accents, that middle England sixth form slur copied from daytime Radio 1. The content isn’t much better, instead of insightful discussion the podcasts contain a series know it all sound bites designed to fool potential clients into thinking that ‘These guys really know their stuff’. A truly awful listening experience.

Americans? Best music, best literature and best web development podcasts.

The second point I’d like to talk about is my new Whiteboard. It sits on the wall next to my desk and came complete with a set of coloured pens and accessories. It’s a really good way of visualising concepts, brainstorming ideas or even just a large area to design schemas or normalise database tables. Big, long sweeping coloured arrows make for a great environment to refer back to at the turn of the head. It’s a bit like being at school but without getting laughed at in the showers after cross country. The thing is, all of the above is just a way of avoiding what I’m stuck on at work, ISAPI_rewrite. Oh joy.

I’ve been working on the same project for several months now so haven’t really had the opportunity to mess around with new web page layout ideas. After looking around at what’s going on in the more visual fields of web design and development I was amazed at how much things have changed in such a short period of time.

All of a sudden, the web is looking great. It’s almost as if the whole thing has had a makeover following the initial hype of Web 2.0. Or maybe this is Web 2.0, it’s just that it took a little longer to happen than we at first thought it would.

One area of development that is particularly interesting at the moment is JavaScript libraries. I’ve made some fairly dramatic structural changes to our site at work using JavaScript working in conjunction with the Document Object Model, but these are scripts that I’ve written myself for a particular purpose and as such are fairly bullet proof. A library takes it a step further by offering the kind of functionality that I’d never have the time to develop.

My initial impressions of the Mootools framework was excitement tinged with the usual IE disappointment, and because of its Object Orientated nature I found debugging to be a much more drawn out process. Although it has to be said, that in itself was a superb lesson in Object Orientation.

Although we’re yet to go live with any of the functionality I’ve been playing with, it is great fun, there are simply hours of pleasure to be had from playing around with the framework and snippets of code. The whole thing reminded me of unwrapping my first metal toolbox as a child (Blue Hamerite finish).

Mootools

This is a great website too http://css-tricks.com/

2008s project at work will be the next step in our re branding exercise, the development of an in house application to manage our content, otherwise known as a CMS. The technology will come courtesy of Microsoft’s .NET framework, that being a logical progression from the classic ASP (Visual Basic) set up used to create the web site and Intranet.

It is an exiting, if terrifying prospect, one that threatens a complete immersion in a whole new area of uncharted technology. However there is hope, already the human logic of those that developed this framework has started to show itself as little different from those from the open source sector. For instance, I felt almost at home with the concept of Microsoft’s Data Tier Components, and begrudgingly acknowledged that someone has probably put a great deal of thought into what I’ve always known as, Layer Separation.

This is only the beginning of a momentous project, but what is clear already is that the .NET Framework is not quite the bloated WYSIWYG editor that people like me used to say it was.

What’s this, Something positive about Microsoft?

Maybe, only time will tell.

When I first built my website things where different, the web was still very much a media to look at and navigate, rather than to interact with like it is now. Initially I hand built HTML pages to show my photos and organised the various albums into a file structure, to save on server space and download times I compressed all the images down to 30k. Laughable when you think about what we can do now, but it served a purpose. They were my photos on my website and I thought that was a million miles away from having them sit in dusty Truprint envelopes under the bed.

From flat pages I moved to an open source CMS, Plogger, that I hosted and adapted to my own needs, and whilst it wasn’t great it made the process of publishing photos less painfull. But as with most things in life, the novelty wears off. As much as one would like to, there isn’t time to get stuck in and develop functionality and improved scripting because there’s too much of that to do at work. When this happens things don’t get used and stagnate, it becomes a corner of the web that sits redundant. I don’t update my photos because there’s no technological incentive, and I still have to work everything through Photoshop. More importantly, it’s not a scalable solution both in terms of time and space.

Also, there’s privacy to consider and this is going to be a big deal over the next few years. My brother has always said that he doesn’t want pictures of himself on the web, a little paranoid maybe, but its also his right and I should respect that. My other brother has kids and pictures of them is a whole different ball game. I don’t have the time to develop permissions on my own site so that only family or friends can see them, so why not let flickr do it for me?

I’d always thought of flickr as a site for people who didn’t have their own site, which is of course, ridiculous. For a few quid a month I get to upload an unlimited amount of images at whatever resolution comes off my camera, that’s really good. Not only that, the application interface works better than a desktop operating system, who could have predicted that we would be at this point so soon?

To a certain extent I’m admitting defeat, my photo folder cant possibly keep up with what the people at Yahoo can offer, and to not embrace it would be as pointless as it is unpractical. I know this isn’t in keeping with the pioneering spirit of open source development, but that spirit is based on ideal scenarios of intelligence and time, both of which I don’t have an endless supply of.

So what if the photos aren’t on my site, the web is changing, and that was always going to happen.