6 ms·

Why do we need another generation of DB tools, what is wrong with what we have?

sorenbs 6y ago
The way I think about this is that the old generation of DB tools make it possible for application developers to use databases. With Prisma we want to make it a delight. There are an order of magnitude more application developers today than just 10 years ago. This enable us to invest much more in the tools we all use, resulting in higher quality and a much better experience. So to turn it around - why _not_ build new and better tools?
- kevas 6y ago
  Because someone will build tools off yours and then someone will build tools off theirs. Abstractions on abstractions on abstractions.
  - sorenbs 6y ago
    I believe this is normally called innovation :-P
    - kevas 6y ago
      LPad was innovation too and that broke everything.
      - sorenbs 6y ago
        Was it though?
  - dragonwriter 6y ago
    > Because someone will build tools off yours and then someone will build tools off theirs. Is that supposed to be a reason against? Enabling further progress sounds more like a reason for, not a reason against.
nikolasburk 6y ago
I feel like current database tools don't really cut it! From my own experience, working with databases in Node.js and TypeScript has always been a painful experience – sometimes more, sometimes less, but it was never smooth or enjoyable! With Prisma, we're setting out to build database tools that make developers more _confident_ and _productive_ [1] in their database workflows! We're focusing on type-safety, intuitive data modeling, easy working with relations and overall simpler ways for achieving common tasks with your database. We also wrote at length about what we believe the shortcomings are with databases tools in our documentation [2], would love to hear your feedback about our thoughts here! [1] https://www.prisma.io/blog/announcing-prisma-2-n0v98rzc8br1/ https://www.prisma.io/blog/announcing-prisma-2-n0v98rzc8br1/ [2] https://www.prisma.io/docs/understand-prisma/why-prisma https://www.prisma.io/docs/understand-prisma/why-prisma
- XCSme 6y ago
  I used MongoDB + Mongoose in TypeScript and the experience was ok. Just declare a Schema and it works, you get typings and everything. I think the problem was not the DB itself (MongoDB) but the tooling around it.
- pphysch 6y ago
  > working with databases in Node.js and TypeScript has always been a painful experience Even after reading your (three) paragraphs from the article on why raw SQL and drivers are bad, I'm far from sold. 1. Establish connection 2. Execute SQL 3. Manipulate results 4. More SQL/Commit/close IME a lot of the struggle comes from not understanding how 1. or 2. work and trying to do all the heavy lifting in 3. The amount of times I've refactored dozens of lines of code into <10, with huge performance and readability gains by simply using SQL's potential... Best of luck with the sales.
  - dllthomas 6y ago
    On the other hand, working at a large company I saw a many-line SQL query that had been run monthly for a long while, looked at and edited by quite a few people, which down in the middle of it was comparing (for equality) a city id and a probability. There is absolutely room for improved tooling. I have no idea whether Prisma is it.

The way I think about this is that the old generation of DB tools make it possible for application developers to use databases. With Prisma we want to make it a delight. There are an order of magnitude more application developers today than just 10 years ago. This enable us to invest much more in the tools we all use, resulting in higher quality and a much better experience. So to turn it around - why _not_ build new and better tools?

kevas 6y ago
Because someone will build tools off yours and then someone will build tools off theirs. Abstractions on abstractions on abstractions.
- sorenbs 6y ago
  I believe this is normally called innovation :-P
  - kevas 6y ago
    LPad was innovation too and that broke everything.
    - sorenbs 6y ago
      Was it though?
- dragonwriter 6y ago
  > Because someone will build tools off yours and then someone will build tools off theirs. Is that supposed to be a reason against? Enabling further progress sounds more like a reason for, not a reason against.

nikolasburk 6y ago

I feel like current database tools don't really cut it! From my own experience, working with databases in Node.js and TypeScript has always been a painful experience – sometimes more, sometimes less, but it was never smooth or enjoyable! With Prisma, we're setting out to build database tools that make developers more _confident_ and _productive_ [1] in their database workflows! We're focusing on type-safety, intuitive data modeling, easy working with relations and overall simpler ways for achieving common tasks with your database. We also wrote at length about what we believe the shortcomings are with databases tools in our documentation [2], would love to hear your feedback about our thoughts here! [1] https://www.prisma.io/blog/announcing-prisma-2-n0v98rzc8br1/ https://www.prisma.io/blog/announcing-prisma-2-n0v98rzc8br1/ [2] https://www.prisma.io/docs/understand-prisma/why-prisma https://www.prisma.io/docs/understand-prisma/why-prisma

XCSme 6y ago
I used MongoDB + Mongoose in TypeScript and the experience was ok. Just declare a Schema and it works, you get typings and everything. I think the problem was not the DB itself (MongoDB) but the tooling around it.
pphysch 6y ago
> working with databases in Node.js and TypeScript has always been a painful experience Even after reading your (three) paragraphs from the article on why raw SQL and drivers are bad, I'm far from sold. 1. Establish connection 2. Execute SQL 3. Manipulate results 4. More SQL/Commit/close IME a lot of the struggle comes from not understanding how 1. or 2. work and trying to do all the heavy lifting in 3. The amount of times I've refactored dozens of lines of code into <10, with huge performance and readability gains by simply using SQL's potential... Best of luck with the sales.
- dllthomas 6y ago
  On the other hand, working at a large company I saw a many-line SQL query that had been run monthly for a long while, looked at and edited by quite a few people, which down in the middle of it was comparing (for equality) a city id and a probability. There is absolutely room for improved tooling. I have no idea whether Prisma is it.