Geeks With Blogs

News Dave's Mug View David Oliver's profile on LinkedIn Add to Technorati Favorites Blog Directory for Guildford, Surrey
Dave Oliver's Blog Enterprise Technology Thought Leadership in a FTSE 100

The post is a cautionary tail of the latest episode of common sense versus cool.

AJAX is a great technology that allows webpages to have much richer content. Google Maps was the turning point for this technology as it brought the technology to the attention of the Technorati that has enjoyed a superficial level of hype ever since because it is encompassed as the lynch-pin of Web 2.0.

The key part of AJAX is Javascript which is a language that all mainstream browsers, no matter what operating system, have installed by default, no need for additional plug-ins. The issue I have with Javascript is that it isn't a compiled language, it's interpreted so will not work any where as fast as compiled code. Javascript does not multi-thread, so adding more Cores to your machine is not going to make an applications written in Javascript run any faster. A good example of this is that now many webpages are taking a great deal more processor power than they would have say 12 months ago and this is the reason why I do not recommend a low-spec machine for surfing the web these days and perhaps not hardly surprising with the additional burden of video adverts, remember when only high-spec machines could only run video?

I believe that Javascript will re-ignite the GHz arms-race between Intel and AMD and will again become centre stage topic as Javascript doesn't run in multiple threads. Yes, I'm sure that having 2,4,8,12,16,32, etc cores is high on the agenda but the amount of core will eventually out-run it's usefulness. Developers are fully aware that writing applications that use multiple thread isn't child's-play to create so  instances were this is used often require an extra level of justification for the additional time it takes to code them and this fact isn't changing anytime soon. I'm sure that operating systems of the future will even dedicate cores to a particular function such as one for the UI, two to look after security for example, but you can see from this example what is going to be the serious advantage of have 64 over 128? You are just going to have more spare cores idle for most of the time wasting electricity.

Whoever, inside most business today you are running Windows, most likely XP on a machine with a 1+ GHz processor and 512Mb of RAM, am I right? I bet that most of those machines have IE6 and are moving slowly to IE7, I would even hazard a guess that they may even have .Net 1.1 or even .Net 2.0 installed? Perhaps even the latest JVM?

If this is the case then why the hell are businesses worrying about writing stuff in Javascript?

Are businesses likely to migrate to Mac OS or Linux anytime soon? Well what would be the point? It doesn't take a genius to work out that this wouldn't make a great deal of sense right now. Hey, I think I'm safe in not having to do an ROI on this one, it's that obvious, remember the OS change from 98 or even ME to XP and that was just an OS upgrade, can you imagine a platform change? And just like any OS change what would be the justification? Well I know first hand about expensive migration projects, one that was going to save £250,000 ended up costing £1.2 million. Just adds weight to the saying, No free lunches in business. So for my mind, it's got to be a slam-dunk or a twisted arm to make me recommend that kind of change again.

Ok, so if your business is the web or is very important to your business that having the people that use your website have the best possible experience this is of paramount importance but lets be perfectly clear here most businesses do not make their money from the web so making a better investment in what the already have makes far more sense then jumping ship and following the trend. Lets not beat about the bush here, Window Forms applications make rich experiences now, with little development effort, compiled and can happily run quickly on many machines now not some uber-beast of the future. Technologies such as SMS and Smart-Clients have made deploying Windows Forms applications a piece of cake for years over 1000's of desktop all over the world.

The disadvantage is with a Windows Form application that it's first language isn't the web, and the Internet does create many a hurdle, but again lets be realistic here? Exactly how many applications are going to see the cold light of day outside a LAN or WAN? I guess that it's going to be the majority. Software as a service isn't here yet and we are going to have to wait till .Net 3.5 and Orcas before WCF and WWF get all of it's teeth into WS-* that's going to be simple enough for the average Joe developer to string together and that's just the Microsoft world. Java world is going to have to wait for OSOA.

So, being an Architect at a Developers party doesn't mean I'm a kill-joy but it's clear that IT does have to reduce it's run-away costs to regain trust with the business who after all come up with the cash that pays for everything. So be pragmatic and invest in what you have rather than be a cool kid. Lets be honest spending the extra time learning Web 2.0 and AJAX and fighting with the new tools, then spending all the budget building new kit to run it all on and the project is finished sometime in the next few years never over getting the project finished in the next 12 months and the kudos from delivering. Boring? Yes but then 80% of all IT projects fail to deliver on time and to budget and one of the reason is because we in IT just LOVE to chase rainbows that the next big thing brings, the next version will solve all our problems! Can you imagine Isimbard Kingdom Brunel waiting till Bill-Bob's widgets & Co brought out rivet 2.0, remember his involutions came from what he finished. So the moral is, don't wait for BIll, Steve or Eric to come and save you, they have their own agendas and us hankering for Rivet 2.0 is only aiding them, egged on unwittingly by the Technorati who, lets face it, haven't got to stand up in front of your boss and tell them why the projects late. A healthy dose of realism and pragmatism means you sitting on the beach with a pay rise perhaps to sitting in an office wrestling with the next broken promise. Lets just stop beating ourselves up. </rant>

Posted on Wednesday, June 27, 2007 9:50 PM Main , Development Technologies , Technical Architecture | Back to top

Comments on this post: Don't bet the farm on AJAX if you don't have to.

Comments are closed.
Comments have been closed on this topic.
Copyright © Dave Oliver | Powered by: