The zenith of bad architecture. For people confident enough to use a helper (ai, framework, code generator) but not confident enough to independently plan large original applications it will always be cheaper to start over every time than maintain software at scale, but that doesn’t make it cheap.
Can not understand what original designer(s) intent / thought process was unless there's documenation (physical and/or comments). As software complexity increases, using educated guesses / testing becomes more difficult to do in way that doesnt' introduce other issues. (vs. relevant documenation / explaination). Lot of times easier to start from scratch than do guestimate thinking/approaches.
NASA has a long term/historical projects with public documenation about supporting long term project issues.
Shrinking large moving mechanical devices down to something that isn't going to chop off / cause physical body harm certainly would be in the 'less dangerous' department. (newer tech, high electrical voltages not withstanding) But not having to think about how something works / purpose can certainly lead to dangerous abuses as much as new/creative uses.
IMHO, the pitfalls are in not understanding/being aware of the limitations/edge cases. aka Not unplugging 70's CRT TV meant TV still drew power vs. unplugging meant took longer for CRT to warm up before being able to display image. Note: For areas with frequent electical storms, power surge due to lighting strike could destroy plugged in 70's TV. Today's battery powered displays, dropping from height & destroying display more relevant risk than worrying about lighting strike causing power surge.
Unlike most mechanical hardware where have to design upfront to minimize costs/ensure things fit/work together; with software one can very easily intentionally/unintentionally 'sweep things under the carpet" in quest to 'get it done'. see [1] vs. [3]; [2] vs. history of [4];
Quote by Arthur C. Clarke : "Any sufficiently advanced technology is indistinguishable from magic."[5]
mechanical digital camera[2]; 1800's Fax machine[4]
niplo disk [1] vs. flexible dot-matrix display manufacture by offset lithography [3]
Related "Ask HN: Is human technology advancing faster than human wisdom sustainable?"[6]
The zenith of bad architecture. For people confident enough to use a helper (ai, framework, code generator) but not confident enough to independently plan large original applications it will always be cheaper to start over every time than maintain software at scale, but that doesn’t make it cheap.
Who says is easier to rebuild than to fix?
The rebuild vs fix problem depends a lot on the state of the codebase and on the inherent complexity of what the system actually does.
> Who says is easier to rebuild than to fix?
People who don't understand the codebase
Can not understand what original designer(s) intent / thought process was unless there's documenation (physical and/or comments). As software complexity increases, using educated guesses / testing becomes more difficult to do in way that doesnt' introduce other issues. (vs. relevant documenation / explaination). Lot of times easier to start from scratch than do guestimate thinking/approaches.
NASA has a long term/historical projects with public documenation about supporting long term project issues.
TX company that still uses punch cards (2013): https://www.chron.com/news/article/Conroe-company-still-usin...
Can you give an example or two?
Shrinking large moving mechanical devices down to something that isn't going to chop off / cause physical body harm certainly would be in the 'less dangerous' department. (newer tech, high electrical voltages not withstanding) But not having to think about how something works / purpose can certainly lead to dangerous abuses as much as new/creative uses.
IMHO, the pitfalls are in not understanding/being aware of the limitations/edge cases. aka Not unplugging 70's CRT TV meant TV still drew power vs. unplugging meant took longer for CRT to warm up before being able to display image. Note: For areas with frequent electical storms, power surge due to lighting strike could destroy plugged in 70's TV. Today's battery powered displays, dropping from height & destroying display more relevant risk than worrying about lighting strike causing power surge.
Unlike most mechanical hardware where have to design upfront to minimize costs/ensure things fit/work together; with software one can very easily intentionally/unintentionally 'sweep things under the carpet" in quest to 'get it done'. see [1] vs. [3]; [2] vs. history of [4];
Quote by Arthur C. Clarke : "Any sufficiently advanced technology is indistinguishable from magic."[5]
mechanical digital camera[2]; 1800's Fax machine[4]
niplo disk [1] vs. flexible dot-matrix display manufacture by offset lithography [3]
Related "Ask HN: Is human technology advancing faster than human wisdom sustainable?"[6]
-----------------------------------------------------------------------------------------
[1] niplo disk : https://web.archive.org/web/20091026194656/http://geocities....
[2] mechanical digital camera : https://www.youtube.com/watch?v=KCzMLl-cB94 [3] : (behind paywall) flexible dot-matrix display manufacture by offset lithography : https://repository.lboro.ac.uk/articles/journal_contribution...[4] : (electro-mechanical image) : The history of faxing: from revolutionary to digital evolution in 2024 : https://novatech.net/blog/the-history-of-faxing-from-revolut...
[5] Aurthur C. Clarke quote disucssion : https://lab.cccb.org/en/arthur-c-clarke-any-sufficiently-adv...
[6] Ask HN: Is human technology advancing faster than human wisdom sustainable? : https://news.ycombinator.com/item?id=43357677
Oh, I thought perhaps you were referring to the country.