3 ms·

This is not really about perfection - which, after all, is impossible - as it's about creating high quality products, no matter how simple or complex they might

This is not really about perfection - which, after all, is impossible - as it's about creating high quality products, no matter how simple or complex they might be. If the wheel was invented with today's manufacturing mindset, it wouldn't actually roll in first couple of years ("available in the 2.0 release!") and probably set half the cave on fire when actually doing so ("You did read the EULA, didn't you?").

corentin 18y ago
Do you actually think the first wheel was perfectly round and that it didn't break into pieces after some time? All those early planes that crashed, those zeppelins that took fire were the works of software engineers? What about the Titanic?
- bmj 18y ago
  Sure, there are examples of spectacular product failures. But the software industry seems to be particularly bad in this regard. We all use software that is buggy, and most of us just shrug our shoulders and say "oh, well" or post on our blogs about it. And, of course, we can tolerate these sorts of things because if my phone drops a signal or an app freezes, it's unlikely it will be as catastrophic as a zeppelin bursting into flames. Product management would tell you that they have no choice but to rush these things to market. And maybe that's true. Also consider that some "important" software came from academic/research settings, where the pressure to bring things to market was much lower (though there is considerable pressure to publish in some cases). As an example: my employer, who writes software for a highly regulated industry, recently hired a new product manager. We have been preparing a minor update release for a core product, and he has been tasked to manage this update. He comes from a company that offered non-critical web applications. When our QC department gave him estimates for testing, he became visibly pale, and pulled me aside to ask how, exactly, they needed that much time to test. I responded that he would understand the first time he is pulled into an audit by one of our clients and is asked to explain the validation process for the software. Quality just isn't the first thing management thinks about--it's all about getting something to market.

corentin 18y ago

Do you actually think the first wheel was perfectly round and that it didn't break into pieces after some time? All those early planes that crashed, those zeppelins that took fire were the works of software engineers? What about the Titanic?

bmj 18y ago
Sure, there are examples of spectacular product failures. But the software industry seems to be particularly bad in this regard. We all use software that is buggy, and most of us just shrug our shoulders and say "oh, well" or post on our blogs about it. And, of course, we can tolerate these sorts of things because if my phone drops a signal or an app freezes, it's unlikely it will be as catastrophic as a zeppelin bursting into flames. Product management would tell you that they have no choice but to rush these things to market. And maybe that's true. Also consider that some "important" software came from academic/research settings, where the pressure to bring things to market was much lower (though there is considerable pressure to publish in some cases). As an example: my employer, who writes software for a highly regulated industry, recently hired a new product manager. We have been preparing a minor update release for a core product, and he has been tasked to manage this update. He comes from a company that offered non-critical web applications. When our QC department gave him estimates for testing, he became visibly pale, and pulled me aside to ask how, exactly, they needed that much time to test. I responded that he would understand the first time he is pulled into an audit by one of our clients and is asked to explain the validation process for the software. Quality just isn't the first thing management thinks about--it's all about getting something to market.