7 ms·

Why only fools write code first

17 points by kareemm 6y ago

seba_dos1 6y ago
Do only fools write webpages (with nothing but simple text) that are just white empty space when JavaScript is disabled?
- AndrewDucker 6y ago
  Because the number of people who disable JavaScript is so low that they are not worth considering when designing something?
  - throwaway_jones 6y ago
    Even if that were true (it isn't), a site should still consider non-javascript rendering for accessibility purposes.
  - seba_dos1 6y ago
    You have to actively break your site to make it broken this way, and that number may not be so low when you consider particular audiences (which is very likely the case for this post). Not even speaking about making the website unnecessarily fragile from interpreter or network errors.
jasoneckert 6y ago
I can appreciate the angle the author provides in this blog post, because I'm sure it will work for many startups. However, there are many others where it simply won't work because a functional prototype is necessary to procure interest and funding. And the quality and depth of that prototype is key to achieving this as well.
scarface74 6y ago
I once worked for a department in a large at the time F10 company. My manager was the founder of the startup that was acquired. As the founder, he had been scrappy, found a need in market that was a large highly regulated two sided B2B market. His team wrote a web app/mobile app (ruggedized Windows CE devices). By the time his company was acquired, they had 90% of addressable market, but didn’t have two or three of the largest players. After being acquired, the acquiring company made him relocate and most of his former employees didn’t want to relocate. He was under 30 and hired a bunch of “smart people” in their 20s to expand to the other side of the B2B market. This time, with plenty of money and not having to worry about financing, he lost the plot. The “smart people” spent more time being concerned with proper unit testing and code coverage, “good architecture”, “good processes”, etc. two years later, we had a wonderful product no one wanted. But, the three of the five biggest players were interested in the “legacy” product. None of us wanted to be building on a PHP product and raises weren’t forthcoming because our team wasn’t making money. We all saw the writing on the wall and all 15 of us left within the next six months - starting with me. I was the oldest by 10 years and far less idealistic. No one listened to me when I kept asking “how will these conversations about best practices help us in the market? What does the customer want?”
scoot_718 6y ago
Oh yeah, you never want to experiment with code, no, you should write a design doc first.
- lecarore 6y ago
  To be fair, he raises a fair point. The title should be "don't invest heavily in coding from the start, talk to your future customers first, if those are available, if you are trying to make money with your project" but it's less catchy. I've been guilty of building things nobody needed and it's not a nice feeling
- kjeetgill 6y ago
  He's quite clearly talking about getting paying customers before building anything. I don't think this about design docs before prototypes where you're more an employee not an entrepreneur.
tenbino 6y ago
Sometimes it takes writing software and modifying it to even discover what your idea is. It’s like painting.
SahAssar 6y ago
This sounds like it only works for ideas that are easily codeable, otherwise you get stuff like theranos (and IMO hyperloop and wework) where the marketing, hype and customers are there before you even know if it's possible to build sustainably.
anotheryou 6y ago
There are things that need prototypes. You can mock a bit, write about it, but if it works you can sometimes only find out in practice. How would you build a minimal tic-tok competing with insta stories and snapchat? Surely not with youtube playlists or something. Or dropbox: usb-drives in ubers?
wdiamond 6y ago
Suppose you know everything, why would bother to help others? with altruism(a fool?), or for money(a poor?). Suppose you know nothing. Could you sell what you don't know? And if it sells what would you deliver? I believe if you are good enough you won't waste too much money selling.

17 points by kareemm 6y ago

Do only fools write webpages (with nothing but simple text) that are just white empty space when JavaScript is disabled?

AndrewDucker 6y ago
Because the number of people who disable JavaScript is so low that they are not worth considering when designing something?
- throwaway_jones 6y ago
  Even if that were true (it isn't), a site should still consider non-javascript rendering for accessibility purposes.
- seba_dos1 6y ago
  You have to actively break your site to make it broken this way, and that number may not be so low when you consider particular audiences (which is very likely the case for this post). Not even speaking about making the website unnecessarily fragile from interpreter or network errors.

jasoneckert 6y ago

I can appreciate the angle the author provides in this blog post, because I'm sure it will work for many startups. However, there are many others where it simply won't work because a functional prototype is necessary to procure interest and funding. And the quality and depth of that prototype is key to achieving this as well.

scarface74 6y ago

I once worked for a department in a large at the time F10 company. My manager was the founder of the startup that was acquired. As the founder, he had been scrappy, found a need in market that was a large highly regulated two sided B2B market. His team wrote a web app/mobile app (ruggedized Windows CE devices). By the time his company was acquired, they had 90% of addressable market, but didn’t have two or three of the largest players. After being acquired, the acquiring company made him relocate and most of his former employees didn’t want to relocate. He was under 30 and hired a bunch of “smart people” in their 20s to expand to the other side of the B2B market. This time, with plenty of money and not having to worry about financing, he lost the plot. The “smart people” spent more time being concerned with proper unit testing and code coverage, “good architecture”, “good processes”, etc. two years later, we had a wonderful product no one wanted. But, the three of the five biggest players were interested in the “legacy” product. None of us wanted to be building on a PHP product and raises weren’t forthcoming because our team wasn’t making money. We all saw the writing on the wall and all 15 of us left within the next six months - starting with me. I was the oldest by 10 years and far less idealistic. No one listened to me when I kept asking “how will these conversations about best practices help us in the market? What does the customer want?”

scoot_718 6y ago

Oh yeah, you never want to experiment with code, no, you should write a design doc first.

lecarore 6y ago
To be fair, he raises a fair point. The title should be "don't invest heavily in coding from the start, talk to your future customers first, if those are available, if you are trying to make money with your project" but it's less catchy. I've been guilty of building things nobody needed and it's not a nice feeling
kjeetgill 6y ago
He's quite clearly talking about getting paying customers before building anything. I don't think this about design docs before prototypes where you're more an employee not an entrepreneur.

tenbino 6y ago

Sometimes it takes writing software and modifying it to even discover what your idea is. It’s like painting.

SahAssar 6y ago

This sounds like it only works for ideas that are easily codeable, otherwise you get stuff like theranos (and IMO hyperloop and wework) where the marketing, hype and customers are there before you even know if it's possible to build sustainably.

anotheryou 6y ago

There are things that need prototypes. You can mock a bit, write about it, but if it works you can sometimes only find out in practice. How would you build a minimal tic-tok competing with insta stories and snapchat? Surely not with youtube playlists or something. Or dropbox: usb-drives in ubers?

wdiamond 6y ago

Suppose you know everything, why would bother to help others? with altruism(a fool?), or for money(a poor?). Suppose you know nothing. Could you sell what you don't know? And if it sells what would you deliver? I believe if you are good enough you won't waste too much money selling.