Timeline for How can developers be assured that their software is fault free?
Current License: CC BY-SA 4.0
15 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Dec 4, 2022 at 4:28 | history | edited | Dewi Morgan | CC BY-SA 4.0 |
added 8 characters in body
|
| Jun 9, 2017 at 18:11 | history | bounty ended | alecxe | ||
| Mar 18, 2016 at 13:40 | comment | added | Dewi Morgan | @DaveBoltman Kinda depends what your spec is. Unsure there's much call for "a program that does one NOP then terminates". But that level of "correctness" is also trivially true for all other non-defective CPU instructions. But if you can calculate 0.00 faults per 100klocs of NOPs, then you have 10M NOPs. Even if what you wanted to do was "nothing for 10M clock cycles", such a busy-wait would fail code review: only case I can think off offhand where that code could be considered "not defective" is a NOP-slide. | |
| Mar 18, 2016 at 11:17 | comment | added | Reversed Engineer | No, I'd say NOP is the most bug-free code ever written. It just wasted once CPU cycle, and (ignoring hardware bugs), has 0.00 fault per 100000 lines of code. | |
| Mar 18, 2016 at 8:48 | comment | added | AdrianHHH | @HorusKol redundancy also protects from hardware problems. The Space shuttle had 5-way redundancy of the main flight control computers (see 2nd paragraph in en.wikipedia.org/wiki/Space_Shuttle#Flight_systems ). | |
| Mar 17, 2016 at 21:57 | comment | added | HorusKol | It should be noted that even with this massive effort to zero-defect code, NASA also uses redundancy to protect from software bugs. | |
| Mar 17, 2016 at 12:40 | comment | added | user | @Luaan Forget a billion dollars in equipment, how about half a dozen lives in a very high publicity accident? | |
| Mar 16, 2016 at 17:07 | comment | added | Luaan | @DewiMorgan And it's still absurdly expensive. Probably worth it dealing with something where a single error can mean losing a billion dollars in equipment, but oh well :) | |
| Mar 16, 2016 at 17:03 | history | edited | Dewi Morgan | CC BY-SA 3.0 |
I'm wrong! Yay!
|
| Mar 16, 2016 at 16:58 | comment | added | Dewi Morgan | @Benjaminssp - Looks to me like either the program was reopened, or they started anew, and achieved even better results... and that I am way out of date in my above post. Editing, thank you! | |
| Mar 16, 2016 at 15:44 | comment | added | Benjaminssp | Here’s a source, though it doesn't say anything about the program's discontinuation: fastcompany.com/28121/they-write-right-stuff | |
| Mar 16, 2016 at 15:27 | comment | added | rdans | do you have a reference for this? | |
| Mar 16, 2016 at 13:54 | comment | added | Niels van Reijmersdal | +1 For tests also have faults if you check them in large. People just make mistakes :) | |
| Mar 16, 2016 at 13:49 | review | First posts | |||
| Mar 16, 2016 at 15:49 | |||||
| Mar 16, 2016 at 13:45 | history | answered | Dewi Morgan | CC BY-SA 3.0 |