“ became clear to me that learning how to write a specification was going to be my hardest learn-on-the-job experience yet. Why? Because I was actually having an unlearn-on-the-job experience.”
—> concept of unlearning to progress is really interesting, it seems underrated as a mechanism for how on some becomes a better engineer..
“Back during school, and when I first started working, I used to think that writing specs was a useless exercise. What even was the point? Just design your system in your head, talk it through with others if you need to, and go build it. How could you even know all about your system before you start building it anyway? Any spec you could write would just be incomplete and not good. Writing specs was an unnecessary, bureaucratic task best to be avoided. Time spent on doing “paperwork” would be better used by actually building your product.
The sentiment above is still common in the technology industry, which I think is a shame. After three years in the industry, I have changed my mind, and now believe that a having good specs is critical for making your project a success.”
—> I like this take, it’s a good example of “tradition is a lot smarter than you think”
.. more takes of where tradition in engineering is actually underrated would be really interesting
“ Imagine now that your specification gets magically air-dropped into the Huaqiangbei. Only your spec - not your product. If a month later, successful clones of your product emerge in the Huaqiangbei, then you have succeeded in writing a good spec.”
Cool litmus test.
At the end of the day a spec is a communication tool to tell other humans how to pull off the original vision. It’s more efficient than words and talking.
2 comments
[ 2.9 ms ] story [ 14.6 ms ] thread—> concept of unlearning to progress is really interesting, it seems underrated as a mechanism for how on some becomes a better engineer..
“Back during school, and when I first started working, I used to think that writing specs was a useless exercise. What even was the point? Just design your system in your head, talk it through with others if you need to, and go build it. How could you even know all about your system before you start building it anyway? Any spec you could write would just be incomplete and not good. Writing specs was an unnecessary, bureaucratic task best to be avoided. Time spent on doing “paperwork” would be better used by actually building your product.
The sentiment above is still common in the technology industry, which I think is a shame. After three years in the industry, I have changed my mind, and now believe that a having good specs is critical for making your project a success.”
—> I like this take, it’s a good example of “tradition is a lot smarter than you think”
.. more takes of where tradition in engineering is actually underrated would be really interesting
Cool litmus test.
At the end of the day a spec is a communication tool to tell other humans how to pull off the original vision. It’s more efficient than words and talking.
It’s a medium.