Why is software so poor? Is open source just as bad?
CIO.com's Ken Harris unleashes a torrent of abuse on the state of software quality in a recent article. While he doesn't bring it up, the problem is compounded by software licenses that disclaim all responsibility for the problems caused by
We call them "bugs" as if somehow they are separate, evil creatures. But they're not. They are problems with the intrinsic quality of our product and the way it is brought to market. With automobiles, we learned that it's not always the driver, or how they drive. Sometimes it's the vehicle, and how it is built. At what point do we challenge software's architecture and design and the quality control process that produces it?
Harris suggests open source as a possible check on shoddy software quality. This may be overly optimistic. Open source by itself does not affect software quality. Knowing that one's code is open for all to see may prevent a developer from taking shortcuts that a proprietary license would hide, and a strong community might root out problems, but neither is a guarantee.
No, I think there is something fundamentally different about software. It is still "magic" in some ways. Just as a mobile phone's convenience trumps the need for a perfect (or even passable) signal, I suspect that we put up with a lot from software because it's still relatively new and drives a tremendous amount of value, even in its buggy state.
Eventually we'll expect more, and we'll get it. For now, I think we're still somewhat giddy by just how much even the worst software can do.
Matt Asay brings a decade of in-the-trenches open-source business and legal experience to The Open Road, with an emphasis on emerging open-source business strategies and opportunities. Matt is vice president of business development at Alfresco, a company that develops open-source software for content management. He is a member of the CNET Blog Network and is not an employee of CNET. Disclosure. You can follow Matt on Twitter @mjasay. 





Back before things went awry, dozens of 32 bit GUI application could run comfortably at once on 512 Meg of RAM and a 7Mhz processor. At least it could under the AmigaOS.
It's amazing the difference a well written Operating System could make to the entire software eco-system. That was back in the mid to late 80's.
Now we just have proof that fiscal, marketting and license leveraging DOES NOT produce "Best Of Breed". Merely the most common lowest denominator.
Mission and life critical software should be treated like buildings and bridges and those who write such software should be licensed and have to follow guidelines.
The software industry has evaded responsibility for too long.
Software is not magic, and it is not that much younger than aircraft and automobiles.
You are making a meaningless distinction.
- by jrepenning August 4, 2008 9:27 AM PDT
- We've been giddy so long as we could spend our Moore's-law winnings on clock speed. Now that that's pooping out, and we're discovering that most software is danged HARD to parallelize at all, let alone robustly, will there be a wave of hard-headed realism? Or just a grudging acceptance of even worse quality?
- Like this Reply to this comment
-
(6 Comments)