The Elegance of Simplicity and the Future of IT

[Jonathon Watkins]

<div class='os_post_top_link'><a href='http://www.economist.com/displaystory.cfm?story_id=3307363' target='_blank'>http://www.economist.com/displaysto...tory_id=3307363</a><br /><br /></div>If you are looking for clear, insightful and thoughtful analysis, then it's hard to beat the UK's Economist Magazine. Recently they published a 15 page survey/series of articles on the future of the IT industry, focusing on the conquest of complexity. It's a cracking read and it's well worth setting the time aside to go through the various articles: <br /><br /><i>""It is time for us to rise up with a profound demand," declared the late Michael Dertouzos in his 2001 book, "The Unfinished Revolution": "Make our computers simpler to use!" Donald Norman, a long-standing advocate of design simplicity, concurs. "Today's technology is intrusive and overbearing. It leaves us with no moments of silence, with less time to ourselves, with a sense of diminished control over our lives," he writes in his book, "The Invisible Computer". "People are analogue, not digital; biological, not mechanical. It is time for human-centred technology, a humane technology."</i><br /><br />The essential argument is that simplicity is the 'Next Big Thing' for the industry. Gartner reckons that IT complexity will cost firms worldwide some US$750Bn and that 70-80% of IT spending goes on fixing things rather than buying new systems. Chris Capossela, the head of Microsoft's desktop applications says that technology has perversely ended up making life more complex. Intel's abandonment of the 4Ghz Pentium 4 also points toward the new 'better, not faster' school of thought. Most modern PCs (and to a lesser extent, Pocket PCs) have an immense amount of horsepower under the bonnet. The question is, can this power be used for good, or for awesome? ;-) To be truly successful does complex technology needs to disappear? Marc Benioff, the boss of Salesforce.com reckons that you need to push all the complexity to the back end in order to make the front end very simple. What's your take? Can computing complexity be tamed and can IT deliver a kinder, simpler life?

[dunos]

The KISS principle of software development proves itself yet again. Shame that so many developers fail to keep it simple, stupid.

Dan

[Sven Johannsen]

I think Marc Benioff has nailed a significant point. To make the outward appearance simple, the guts are going to be very complex. Much like the wizards that MS is so fond of.

I'm afraid the complexity we have now is driven by user demands and expectations. Take Word for example. It has been noted that the average Word user uses less than 10% of the program's capability, yet we still demand more. Pocket Word is actually a more effective communication tool in my mind. It allows you to consider what is said, rather than how it looks :wink:

[Felix Torres]

What's your take? Can computing complexity be tamed and can IT deliver a kinder, simpler life?

Depends: There are two approaches to this particular task depending on whether you think complexity is necessary or not:

One is easy to implement but a dead-end road, judging by the fate of the companies and products that have used it, the other much harder and leads to imperfect products that evolve over time and seem to have "the legs" to go on and on.

The latter is the layered approach that *hides* complexity but builds in full functionality and exposes it through a discoverable interface that offers multiple ways of getting to the destination and is, above all, extensible and expandable. It never integrates software too closely with the hardware and tends to virtualize the hardware wherever possible so that migration to newer tech can be simplified. Above all, it never tries to optimize the product for any given point in time because it recognizes that technology is a river and that today's state-of-the-art is tomorrow's paperweight.

The former approach is the craftman's approach; it creates a carefully thought-out, fully optimized product for the present, regardless of the consequences down the road. The future can take of itself. Features that cannot be implemented are deemed unnecessary. Simplicity is implemented not by hiding but by pruning the feature set. The intent is to provide a polished experience for a single point in time, a finished product.

Needless to say, we're talking PocketPC vs PalmPilot. :-D

Other comparable pairings are left as an exercise for the student.

(Oh, of course, there is a third school of thought that says complexity need not be hidden; if you can't deal with complexity you should stay away from the tech. ;-) )

[bvkeen]

I think the desire for simplicity is high among consumers, but does IT Really want build simpler software? I wonder if it has evolved to where it is because complexity breeds job security. I this is the case, I don't see it changing.

[bothari]

Personally i don't see this even beginning to happen until we have some measure of "fake" ai in software (think Pohl's Heechee saga more than HAL in 2001). Computers are exact in their communication and need "exactness" in their input (eg. Delete them all... what's "all"? who are "them", etc)

Implicitness is dificult to interpret...

Intel's abandonment of the 4Ghz Pentium 4 also points toward the new 'better, not faster' school of thought.

This is a bad example. Intel gave up the 4ghz due to massive problems in getting it up to production and the fact that it would still get it's behind tanned by the athlon64/Opteron from AMD. Intel is loosing ground both in mind space and commecial success, and it must stop just following the new leader...