What the title says. I'm building my own little thing with Swift and Vapor. Does anyone here have experience with it? What hosting provider do you recommend? I was thinking Heroku, but I don't like the idea of being trapped with them
Use server based solutions. I would suggest go with shared hosting which starts with $5 or so, something like Digital Ocean.
Yeah, I was thinking DO would be nice.
I built WordTester with .NET Core and Blazor
I actually started developing my current project with Blazor, which is really cool, but I was worried about the latency, considering how Blazor works. Did you find any issue with that?
I haven't noticed any issues with it
Blazor WASM?
I am currently building in Next.js + Chakra UI + Strapi + Cloudinary + Render + Vercel which can quite quickly get me to a fast and scalable solution.
I use Next.js/React, Supabase for backend, and Vercel for hosting.
The big advantage is that you don't have to pay anything for hosting costs when you're small, and everything regarding deployment/hosting is basically taken care of for you, so you never have to worry about it. This lets you focus exclusively on your app.
I've used Heroku before, and while it's also relatively simple, you still need to manage various things like your dynos, configuring how deployment works, making sure your app doesn't go down, etc. It's just extra overhead that distracts you from the real work.
Python, Django, jQuery, mySQL, DigitalOcean. That's my stack.
I always recommend using what you know. Your customers couldn't care less about your stack. You also are unlikely to face any scaling issues as an indiehacker. Don't go flavor of the month, go without whatever you can build fastest with.
That’s why I picked Swift. I develop for Apple platforms and while I know other languages, the one I use every day is Swift, so it made sense to stick with it.
the best language to use for a SaaS is the one you know, but if you want to make a proper business out of it you need to think about how you going to find other people to help with the coding, how much that will cost and how fast and reliable you can build with the technologies you have selected and how much the support and bug fixing will cost.
Thank you for your response. This is one of the many times I’ve read the same thing, to use what you already know. Us developers usually want to experiment with different languages and technologies and building a business out of a stack we barely know is not a good business advice, unless that stack gives really extra powers. I thought about Haskell for the backend, but as I’m not too familiar with it, I know I would be wasting a lot of time building even the simplest thing.
for serious projects you have 3 options
There are some specific (machine learning) cases you can involve Python or GO when necessary (Never for page rendering)
there are some use cases for frameworks (when single developer) but most are just lose of resource when scaling up is needed.
Flutter is shaping up as a good front end alternative to Typescript and .NET but difficult to quickly build 2-3 teams with 4-5 devs each, unless you jump 100% remote that have it's challenges (depending on the project)
I personally will try not to get involve in the mess that modern PHP(Laravel) and Node development has become, spend to much time fixing the problems that comes from them, but if that is what you know that s what you use.
Same can be said for React, it is very good for business as companies pay top dollar for someone to keep the mess running but it is loosing ground fast.
unless you are using one of the 3 mentioned above you will have problems scaling up, not that big of a deal if you plan for it.
You said 2. Node with Typescript and later say to avoid Node. Can you please expand?
NodeJS is an option you can use and I do not recommend using it.
In software development you can use almost any language to build almost anything. Many people (especially beginners) make the mistake of using language and/or framework just because it is the only one they know or they heard can do this kind of project and in most cases the selected language and/or framework is not the best choice even if it is technically capable of achieving the specific task.
Creating a system is not just putting together a bunch of code to execute a task (except when it is a very, very simple one), it is a balance between time, system requirements, development and financial resources to build, support and maintain the system.
Thanks, I've never use Node on backend, but I did use Django. So I have just one more question. You say python, but not for page rendering. Why do you think that?
I’m already set on using Swift for backend, tailwind + htmx + hyperscript for frontend. So totally and completely different than most other are using. But I don’t think the most used languages/frameworks are really good at all.
I also have no issue with the stack not being common if I ever have to hire people. Engineers should be able to pick up a new framework easily, right? ;)
Hasura (with Postgres) + Elixir + React on render.com which has been very productive and performant.
If I were starting again today I'd probably use Supabase + React or Elixir's Phoenix LiveView
I'm a Firebase user and I think Supabase looks really promising and I hope it succeeds, but I think I would wait until it was more mature and battle tested.
I'm building https://sraksha.com/ with Clutch.io + Supabase.io (still under the wrap).
I use Laravel and React (NextJS), I also have a SaaS using Serverless(python) on AWS.
I love both but I prefer Laravel <3
Flutter, Express.js, Mongo, EC2 on Aws.
Ik, very complicated stuff 🙃.
I find that the simpler the solution is, the better. Our first priority as funders should be to solve the main and only problem of our customers. Knowing what docker is won't get me too much closer to that