New iPods and SVG in Chrome

09.09.2008

I know it’s a mixed post. I don’t mean for these – two very separate subjects – to be confused and even connected, except by one fact: that I don’t want to write two individual posts.

I have just discovered that my 6-month-old iPod Touch is now outdated my an ever-so-slightly better model. As far as I can tell, all that the new iPod touch’s sport is a built-in speaker (“for casual listening”) and volume buttons on the side, as well as a slight curvature of the spine. So I’m not really that bothered. Still it’s a shame that mine is now practically redundant…
On an even sadder note I find that Google Chrome doesn’t yet have even basic support for Scalable Vector Graphics. I find this a shame and certainly a downer on Google’s claim that they wanted wanted to build a modern browser for today’s Internet. I know SVG isn’t quite in the public domain yet, but I feel that’s mainly because of lack of support from the major browsers (yeah… Micro-soft!).
If anyone from Google reads this, please think about SVG support in Chrome, please. And also, can somebody escalate my ticket for the AJAX APIs in Google Code as that would be a real bonus to some projects I’m working on. Thanks.
G’night all.

PHP, Zend Studio for Eclipse, Subversion… hmmm

13.08.2008

Since one month ago I’ve been working for a PHP-oriented software house. We’re growing. Some good stuff is being churned out, but things are becoming a little unstable as the workload piles up and current project management methodologies start to buckle.

So tomorrow we look forward to a full day of meeting that will tackle this issue and provide us with a clear path of action towards all round developer joy.
I have been given the unholy task of managing our internal systems development. This means looking for ways to improve. But for me to give an accurate idea of the possibilities I’ve had to get stuck in to some areas that up to now I’ve been fairly cautious of: Eclipse and Subversion.
Eclipse is something I don’t go more than two days without hearing something about. I’m told it’s a developers IDE. It’s extensible, scalable, customisable, pliable, malleable, something-able, able-able. The list goes on.
The depth of it scared me. What’s all this perspectives and views? So many options, so little documentation. Plugins or standalone? My research led me down many paths – of which a greater proportion had dead ends… or rather large unsuspecting precipices with no foreseeable bottom.
I came to conclude the following: if you already use Eclipse and wish to add supreme PHP functionality into an existing installation, go here: http://static.zend.com/topics/Installing-Studio-v601-Plugins.pdf. If on the other hand you are happy to use a separate installation (or don’t have/need a full Eclipse install) download Zend Studio for Eclipse for your platform and install the standalone version.
Then came my wrangles with Subversion. This is something I’ve been itching to get my head around for a while now. I finally managed it with the help of a brilliant guide - http://svnbook.red-bean.com/. I set up Subversion as a daemon process and created my first repository based on my long-running project.
Then I put the two together. Couldn’t be simpler. So my plan is hatched: move some existing/ongoing projects to Subversion (SVN) and get the team hooked up with Zend Studio (ZS) (for those who want to use it), probably even Eclipse (as we’re doing a bit of Flex work now too!).
The next problem came though when I found that projects created in ZS based on a checked-out version of an SVN repository suffered from a distinct lack of code assist. How silly. I tried the basics. No joy. What would be the use of this? Well further research revealed the cause and the solution. So for all of you writhing in agony over this: don’t go elsewhere, just fix it!
The fix is published here for your benefit. I don’t know if this will ever be addressed in a future build of ZS but for now this will do.
Ah that was cathartic.
I wanted to do my links for this week but I haven’t been doing any web-hopping (or whopping as I will call it) for a while, so I have little to share. Except perhaps for a couple of sites which ignited my passion for randomness: http://xkcd.com/ || http://creativescrape.com/
Enjoy!

Ideas That Will Never Make Me Rich

25.07.2008

I got obsessed with the idea that I could come up with something huge. I felt I could be the one to produce the next big thing. But as time has passed and brain-ache has set in, I’ve come to the conclusion that I simply don’t have the necessary time, resources, and inspiration. I got close though.

A good friend of mine told me that the best way to find inspiration for a new and unthought of idea is to take two very unique and disparate concepts and form a path between them. In essence, this path would be the genius idea that has never been thought of before… and what will ultimately make you rich… supposedly.

Therefore I’ve sort of given up. I found that the last idea I formulated had already been realised (if only a matter of weeks before). I was annoyed because I thought “noone will have done this yet,” but at the same time I figured that I only missed the boat by a few weeks… which is not bad for a first attempt!

I’m not just rambling on about my failed attempts to fly… I do have a point. I think it’s this: as a result of all this thought and conceding failure, I’ve realised that it’s best to start by improving. Now that I think about it, some of the best inventions are really only improvements on previous inventions; not entirely new concepts.

I’ve recently taken on a new role with a local web design and development studio. They have some interesting projects on the go and quite a few upcoming. I’m already involved in two or three. I’m really enjoying being part of a team of young developers. It’s quite refreshing. I’m looking forward to helping them bring some new offerings to market. Whilst my position is as a PHP Developer, m main aims are to improve processes, increase productivity, and diversify the projects we work on (even though they are already quite diverse!)

I guess my biggest aim is to increase productivity. At first glance this may seem like an easy step as there are many obvious areas that can be tackled. From my [developers] point of view, I am finding that poor coding and a lack of standards (even agreed standards in the team) are a major hinderance to productivity.

This is what my main project is focussing on. It’s taken me a little over 18 months to work out what the purpose is. My memorised tech spec has ballooned out of all proportion in between times. I’ve considered rewrites and porting to frameworks, but in the end I figure I can keep going with what I have.

Simply put I am aiming to build the definitive software production package for web design & development teams. Is it an application, a framework, a toolset? No. I think it’s really a platform. It’s a complete application on it’s own. It has a simple roll-out like many modern web-based applications. It’s highly customisable and configurable. It’s very modular. Beautifully written code (if I dare to say so). You might just think it’s another content management system, but in actual fact it is so much more.

My greatest vision for it is that it can be used by a team of designers and developers to develop new bespoke applications with little to no effort. For increased enhancements, a deep API coupled with a simple installation method will provide all that is needed for deep use of the core system. For the designer, a 99% code-free skinning system will allow easy installation and manipulation of designs. A full CSS-friendly codebase supported by cascade hints will allow for instant debugging of design flaws.

There’s a long way to go with this at the moment as I am working on it alone in what little spare time I have. I’m keen to get a 1.0 version live as soon as possible. I feel like this is a game-changer. It could speed up development cycles, make a lot of people’s jobs easier, improve client happiness and even improve the end-user experience.

I just want to plug it in to everywhere! I know that if the core is written well, the harder stuff will be easier and others will plug in what they need.

This weeks links:

www.pimpthatsnack.com // www.wilkesdesign.com // www.csscookie.com

RadScheduler and .NET 3.5

09.04.2008

Dear Susan,

At the moment I’m working heavily on a project that is using Telerik’s superb RadScheduler. I know it’s still only in BETA, but it came at just the perfect time for us and our client so we jumped on board.

This meant a 1.1 – 2.0 shift (yikes!) and AJAX thrown in too. The main issue though came up a couple of weeks ago, when – here in the UK – we moved into DST. What a nightmare!

The problem is that one of the specifications of the project requires that it be ready to accept users from any country and map all times local to them! So I spent two months rebuilding all the time functions, saving all times to UTC etc etc

Without going into all of the details of DST in different countries and how we use the data in other areas of the site, suffice to say there is no manageable way with either the Scheduler or .NET 2.0 to map DST for each and every rule save creating them yourself.

And then I stumbled upon TimeZoneInfo in the System.Core of .NET 3.5! What a God-send. Thanks to some of the simple methods in this beautiful class we have succesfully implemented a method for displaying all appointments in the Scheduler to the correct local time.

This class encapsulates Windows’ built-in timezone database in the registry. I will be posting the code for adapting the Scheduler later. Basically: FORGET THE SCHEDULER’S BUILT-IN TIMEZONEOFFSET PROPERTY (i.e. set it to ‘00:00:00′ and don’t change it)