November 1st, 2004

Filed under:
Technology
2 Comments »

Software That Lasts for 200 Years

ChangeThis continues to do excellent work, publishing polemics on everything from technology to urban planning. The latest, greatest one is from Dan Bricklin, who happens to be the guy who invented the spreadsheet.I haven’t read the whole manifesto yet, but I like what I’ve seen so far:

We need to start thinking about software in a way more like how we think about building bridges, dams, and sewers. What we build must last for generations without total rebuilding. This requires new thinking and new ways of organizing development. This is especially important for governments of all sizes as well as for established, ongoing businesses and institutions.

The ChangeThis folks have also demonstrated tremendous restraint by not forcing me to register to view their manifestos. Well done.

Comments: 2 Responses so far

Chris
November 1st, 2004
1:19 pm

I agree, we need to START building more reliable software, but it’s not as easy as sitting down and saying “okay, I’ll make this software reliable”. People have been building bridges, sewers, and dams for thousands of years, but have been building software for about 50. Also, when people learned how to build bridges, they were reasonably sure that there wasn’t going to be a brand new type of ground that they had to build the bridge on. Currently, hardware is constantly changing, and if quantum computing takes off it will change even more. How expensive would a new ISA modem be right now? Now, imagine how much it would cost to replace a 150 year old piece of hardware with something that’s still compatible. Unless backwards compatibility becomes a major trend, it will be very, very expensive to maintain this software. At some point, it would be cheaper to just rewrite the software to be compatible with modern hardware – and I bet that point is far below the 200 year mark.

Reply

Jonathan
November 2nd, 2004
6:50 am

Maybe he meant 200 internet years? ;-)

Chris has precisely hit the point. Unless we’re writing in some kind of metalanguage that will never have to deal with new protocols, storage methods, etc. it’s impossible to lay down permanent code.

Heck, English won’t be anything like it is now (if it’s still a primary language at all).

Reply

Add a comment