They Write the Right Stuff (1996) • Hacker News Board

Re: They Write the Right Stuff (1996)

#31

Post by kats » Wed, Jun 17, 2020, 1:32 AM UTC

sizzzzlerz wrote: post #13

I certainly do not begrudge the efforts that go into building the shuttle software the process they must follow to ensure their code remains at such a high level of quality. When you have essentially an unlimited budget you can achieve a lot with competent managers, staff, and a process that is well-defined and understood and, most importantly, enforced. I don't know the details here but I'd be willing to bet that th…

"An analysis accomplished after the Challenger accident showed that the IBM-developed PASS software had a latent defect rate of just 0.11 errors per 1,000 lines of code—for all intents and purposes, it was considered error-free. But this remarkable achievement did not come easily or cheap. In an industry where the average line of code cost the government (at the time of the report) approximately $50 (written, documented, and tested), the Primary Avionics System Software cost NASA slightly over $1,000 per line."

https://history.nasa.gov/sts1/pages/computer.html

I should really link to this article as well:

http://www.ganssle.com/tem/tem400.html#article2

Re: They Write the Right Stuff (1996)

#32

Post by yummypaint » Wed, Jun 17, 2020, 8:08 AM UTC

beyondcompute wrote: post #24

One of the most influential and interesting articles to me that I share with all the developers with whom I think I can have an interesting conversation (and it influences some quite a lot)! I remember there were some comments: > Are you prepared to pay 10x for a text editor, say? My answer: Hell yeah, if I am earning thousands using this editor, why not pay couple hundred bucks for a really well-made program with go…

I think there is demand for those things, but commercial entities are poorly equipped to provide them. The entire growth mindset is antithetical to providing a stable and consistent product. Product tiers get introduced to extract more money though upselling, creating an inherently adversarial relationship between the company and end users who end up with an intentionally crippled product. As the company gets larger, the incentive structure surrounding middle management takes over and completely wrecks the product as individuals compete to game internal metrics. Unnecessary and regressive changes make their way into updates because "product launches" look good on people's resumes. Nobody cares what changes were actually made because they arent incentivised to. Add innevitable attempts at vendor lock-in / weaponization of the sunk cost falacy against customers and it makes perfect sense that most non-crap software (that also remains non-crap as time goes on) seems to be open source.

Maybe we need a patreon style system for providing long term stable funding to important open source projects. I would sleep easier knowing that FOSS developers for packages i use will have stable income and be able to continue their work.

Re: They Write the Right Stuff (1996)

#33

Post by beyondcompute » Thu, Jun 18, 2020, 1:40 PM UTC

yummypaint wrote: post #32

beyondcompute wrote: post #24

One of the most influential and interesting articles to me that I share with all the developers with whom I think I can have an interesting conversation (and it influences some quite a lot)! I remember there were some comments: > Are you prepared to pay 10x for a text editor, say? My answer: Hell yeah, if I am earning thousands using this editor, why not pay couple hundred bucks for a really well-made program with go…

I think there is demand for those things, but commercial entities are poorly equipped to provide them. The entire growth mindset is antithetical to providing a stable and consistent product. Product tiers get introduced to extract more money though upselling, creating an inherently adversarial relationship between the company and end users who end up with an intentionally crippled product. As the company gets larger,…

Nicely said! I myself prefer the small-businesses’ products. Like several-person organizations that provide “underdo the competition” or “do one thing and do it well” type of software. But I agree on how “bureaucracy” and political games inside larger organizations lead to introduction of not-always-sensical features or re-designs.