"And remember – you are building an MVP (Minimum Viable Product), and will be in the process of creating a POC (Proof of Concept). So get your product out as soon as possible."
but also create cron tasks, jobs, queues, log monitoring, application monitoring, docker support and ofcourse CI
I guess most people stuck on projects because they(we) try to implement too many things at once...
Yeah, this is really two posts: (1) build a PoC, and MVP; (2) build a future-proofed architecture/platform.
You shouldn't start (2) until you have found product market fit and have enough paying customers that it would be shame to lose them and start over finding new ones.
That said, I often do (2) before then but I do it full well knowing that I'm doing it for my own edification and no other goal.
This seems to be a general set of guidelines for developing web apps/APIs and it somewhat overlaps with the recent HN post “Tackling Webdev as a Bioinformatician: why is it so hard?” by Jessime Kirk:
I don’t find this set of guidelines useful; it seems too Python/Django/12-Factor/BizModel specific. Web, Mobile, and Cloud has changed the world but every set of guidelines seems to start with a toolchain/language and then solves the same problems within that scope. For instance, a more generic set of choices might be:
1. static site vs dynamic
2. custom DNS domain vs 3rd party
3. github.io vs CDN vs PaaS vs IaaS
4. APIs: auth, payments, email, notifications
5. Front end framework
6. Backend architecture
7. Storage: BLOBs, files, rows, columns
8. Client platform: web, mobile, library
etc.
4 comments
[ 3.7 ms ] story [ 21.9 ms ] threadbut also create cron tasks, jobs, queues, log monitoring, application monitoring, docker support and ofcourse CI
I guess most people stuck on projects because they(we) try to implement too many things at once...
You shouldn't start (2) until you have found product market fit and have enough paying customers that it would be shame to lose them and start over finding new ones.
That said, I often do (2) before then but I do it full well knowing that I'm doing it for my own edification and no other goal.
https://news.ycombinator.com/item?id=23100002
I don’t find this set of guidelines useful; it seems too Python/Django/12-Factor/BizModel specific. Web, Mobile, and Cloud has changed the world but every set of guidelines seems to start with a toolchain/language and then solves the same problems within that scope. For instance, a more generic set of choices might be:
1. static site vs dynamic 2. custom DNS domain vs 3rd party 3. github.io vs CDN vs PaaS vs IaaS 4. APIs: auth, payments, email, notifications 5. Front end framework 6. Backend architecture 7. Storage: BLOBs, files, rows, columns 8. Client platform: web, mobile, library etc.