3 ms·

"Pick the language and stack based on your team’s needs and comfort, based on your business’s need and risk tolerance, and based on how easy it will be to produ

"Pick the language and stack based on your team’s needs and comfort, based on your business’s need and risk tolerance, and based on how easy it will be to produce software for your target users in that language or framework. That’s your criteria" Yep especially the part about "how easy it will be to produce software for your target users". Case in point. My company uses PHP/Laravel framework for our SAAS and we didn't just choose it because it was cool but because Laravel comes with a great ecosystem, packages and community and most of my team was good in PHP for a while. I actually tried to go rogue by trying to force Golang on my team (long story) but almost dodged the bullet when my CTO told me to go pound sand just because I wanted to switch to Golang (I am the CEO). I am so glad I have people to tell me that I am wrong when I am. We have not only created a great product but the time to market is 20 times of what could have been if we had switched to Golang just because it was cool (and don't get me wrong, I love Go as an amateur programmer and even wrote an internal tool in it to help our team). But for our core SAAS product, focus should be on time to market, customer needs and our team's strength and I am so glad we stuck to what works for us, PHP and Laravel in this case. I hardly have a customer asking about the Programming language we use (may be 2-3 have asked in last 7 years and those were technical folks from customer's team)

duxup 6y ago
The sheer volume of fretting about what might happen later in a company's life because of day 1 type decisions is bonkers IMO. I participate on some forums where folks are just learning and literally folks will show up "yeah but if you start your project with X what happens later if it takes off and you have to <scale, change schema, whatever>". In reality if some new guy takes off and makes Facebook 2 he'll have so much money that that problem will be probably pretty darned nice to have. IIRC hearing Facebook was using ColdFusion at one point, somehow that didn't stop them...
- cosmodisk 6y ago
  I remember we had some discussions in management meetings about these 'what if' situations and they always end the same: once we have them,it means we now have tons of money and can hire people to do whatever needs to be done or simply throw money on available solutions. Worrying if your blog can handle 1M/day queries when you only have 2 readers doesn't make sense.
  - codegeek 6y ago
    Yea. If I had a nickel for everytime a customer asked "Can your system handle hundreds and thousands of concurrent users" and when asked how many they have now, the answer is "Oh we are just starting out but wanted to plan now", I would be a gazillionaire.
  - Frost1x 6y ago
    The problem is that responsibility and risk in busienss gets passed down to those that do the actual work. You're not really concerned about your blog going from 2 readers to 1M/day, you're concerned that iff your blog volume jumps that much, your job and perhaps livelihood may be on the line because the business doesn't understand why you didn't prepare for this and doesn't understand why you now suddenly need a bunch of more resources to accommodate this change in demand. Most all these long drawn out what-if discussions almost always dance around the real topic everyone's discussing: who is taking the risk and whose neck is on the line should things go south. If businesses were more willing to take reasonably accessible risks back from their employees doing the work, we'd have a lot less of these discussions. Instead they just go on and on. Who has to do what or will get fired and have to deal with explaining to future employers why they were fired. Instead, we end up passing as much risk as possible to customers and innovation suffers.
jpcooper 6y ago
Quite refreshing to read. My first job out of university was at a start-up bank whose first version was, would you believe, built on PHP+MySQL. This disgusted my functional purist build beautiful abstractions wanker self at the time, and I distinctly remember feeling that PHP was below me. They are still running and handling people’s money.

duxup 6y ago

The sheer volume of fretting about what might happen later in a company's life because of day 1 type decisions is bonkers IMO. I participate on some forums where folks are just learning and literally folks will show up "yeah but if you start your project with X what happens later if it takes off and you have to <scale, change schema, whatever>". In reality if some new guy takes off and makes Facebook 2 he'll have so much money that that problem will be probably pretty darned nice to have. IIRC hearing Facebook was using ColdFusion at one point, somehow that didn't stop them...

cosmodisk 6y ago
I remember we had some discussions in management meetings about these 'what if' situations and they always end the same: once we have them,it means we now have tons of money and can hire people to do whatever needs to be done or simply throw money on available solutions. Worrying if your blog can handle 1M/day queries when you only have 2 readers doesn't make sense.
- codegeek 6y ago
  Yea. If I had a nickel for everytime a customer asked "Can your system handle hundreds and thousands of concurrent users" and when asked how many they have now, the answer is "Oh we are just starting out but wanted to plan now", I would be a gazillionaire.
- Frost1x 6y ago
  The problem is that responsibility and risk in busienss gets passed down to those that do the actual work. You're not really concerned about your blog going from 2 readers to 1M/day, you're concerned that iff your blog volume jumps that much, your job and perhaps livelihood may be on the line because the business doesn't understand why you didn't prepare for this and doesn't understand why you now suddenly need a bunch of more resources to accommodate this change in demand. Most all these long drawn out what-if discussions almost always dance around the real topic everyone's discussing: who is taking the risk and whose neck is on the line should things go south. If businesses were more willing to take reasonably accessible risks back from their employees doing the work, we'd have a lot less of these discussions. Instead they just go on and on. Who has to do what or will get fired and have to deal with explaining to future employers why they were fired. Instead, we end up passing as much risk as possible to customers and innovation suffers.

jpcooper 6y ago

Quite refreshing to read. My first job out of university was at a start-up bank whose first version was, would you believe, built on PHP+MySQL. This disgusted my functional purist build beautiful abstractions wanker self at the time, and I distinctly remember feeling that PHP was below me. They are still running and handling people’s money.