Show HN: Canonic: A low-code platform to build APIs in minutes | Hacker News Reader

Show HN: Canonic: A low-code platform to build APIs in minutes

canonic.dev

54 pointsprathama34 comments

I feel like this could become a go-to service for experimenting and prototyping APIs for me, given a reasonable pricing model.

A few oddities caught my eye, though, e.g. duplicates of the edited field were created when leaving the field panel.

Is there a plan to add a feature to export the API in a standard like OpenAPI?

Thank you so much for going through the platform. We've just launched and sadly some bugs sneaked through. We are actively working on the fixes and they should be out asap.

Lock-ins have been our biggest concern here at Canonic. We wanted to ensure that if the user wanted to move/integrate with different services, that experience be as seamless as possible. We are exploring both OpenAPI and GraphQL schema export options. Definitely on the roadmap!

> Lock-ins have been our biggest concern here at Canonic. We wanted to ensure that if the user wanted to move/integrate with different services, that experience be as seamless as possible...

Now THAT is worth me paying for! I'm now really looking forward to learning about your pricing, as I'm quite interested in using your platform. All the best!

Thank you! Coming soon :)

Canonic Team, Congrats shipping this. UI appears slick & menus snappy.

Faced an issue creating column in a new model[1] - Can't click on any of the button in image [1] (I'm on Mac & Firefox if that helps)

We are in similar space & open source : we automatically generate REST & GraphQL APIs automatically on any Microsoft SQL Server, Postgres & MySQL databases. Do give us a try. https://github.com/xgenecloud/xgenecloud

[1] : https://imgur.com/efPfFRx

Thanks for trying it out!

We have been experiencing some issues on Firefox, should get sorted in the next 24 hours.

XgeneCloud looks slick! Will definitely give it a go, thank you for sharing it across

So I've spent a bit of time playing around with this (self taught dev, mostly front-end) - really impressed so far.

The app is a bit buggy and features are relatively sparse, but it is intuitive to use and I can see the promise.

Within 20 minutes, I had an API structure fairly well plotted out, along with REST endpoints, a GraphQL schema and playground, documentation, and places to add webhooks (although no way to programmatically add webhooks easily, as far as I could see).

I may be over-rating this simply because it comes at a good time for me - I recently spent hours reading through the JSON:API specification [0], and trying to figure out how it is different to the OpenAPI specification [1], and how to actually get into implementation, and generally getting bogged down in levels of detail that I never knew existed.

The only fear I have with this service is that the pricing structure will be too high. I would be keen on a usage-based pricing model based on amount of data stored, with a free tier for personal projects. This would allow me to play around with it myself for free, while internal business tools with higher usage would be paid.

Interested in talking about what pricing may turn out to be - if you want to get in touch, my email address is on a feature request I submitted about adding a datetime field.

(edit to remove extraneous character, in the process adding lots more extraneous characters)

[0] https://jsonapi.org/format/ [1] https://swagger.io/specification

Thank you for your kind words.

I've personally gone down that rabbit hole as well, especially when you're looking to get started quickly.

As for the pricing, our plans are to make it usage based. The goal is to have an unlimited time free-tier (with reasonable usage restrictions) along with paid individual/team/enterprise plans.

Unfortunately, our system was having some issues, so we missed out on your feedback, I'd be really grateful if you could drop us an email at hello@canonic.dev and we can take the conversation forward!

Thank you

Looks nice conceptually! I am getting a lot of jittery glitches with the graph dashboard and especially the CMS (unusable.) On FF 81.0.

I like Airtable for rapid prototyping and small data stores. It seems like this would be more comprehensive for that use case? Unclear on how to approach user login and permissions (of the project and not the canonic platform)

We are actively working on the FF bugs, the experience is quite non-ideal on it. Hoping to get it sorted asap.

Airtable is amazing and ️ fast for smaller projects, however like you mentioned, we want to be more comprehensive and scalable. One big differentiator there is that we support adding custom endpoints and APIs that can include custom business logic as well.

We realise the documentation is a bit sparse around auth login and permissions. I'll make sure the documentation is updated soon for the same with a guide and examples.

Thank you for your feedback!

Fantastic work. The on-boarding is excellent and first use is very intuitive. You've done a great job of abstracting away the complexity of defining the data models so that I was able to build and deploy an example in just a few minutes.

It's a snappy admin UI and I love the auto-generating docs. All around excellent job!

That's fantastic to hear :)

I'm so glad you liked it and thank you so much for your feedback! We are working very hard to ensure that the onboarding is smooth and the learning curve flat :)

If I am weighing up:

* Complete no code - Bubble

* Out of the box code - Next.js etc.

* Code - Node JS / Express etc.

* Canonic

Where do you fit in? Who chooses you over the other options?

Because the disadvantage of yours is lock-in / pricing / up-skilling to a non-standard way. Same problems as many others - Bubble.io etc.

There must be some killer win you get from using this.

Bubble lets people who have never coded build a rough MVP, maybe get some orders, so I can see why people fork up the $ and struggle through that UI, because the alternative is a 4-5 figure fee from a freelance programmer.

But I don't think Canonic avoids code as you still need to do the front end. (Maybe that's it? You create a front end version of Canonic next?)

So I'm interested in who will buy this and why.

Thank you for your detailed feedback,

Out of the ones mentioned, we are probably closest to Bubble. However, as you rightly said we are currently focused on the backend API building problem.

As we've been pivoting to be a more citizen-first non-tech friendly platform, frontend on Canonic is something we have been discussing and debating frequently.

The main concern with a frontend builder for us is the need for it to truly be as close to a design tool as possible. The frontend of an application frequently ends up being unique to accommodate branding and creative freedom and becomes hard to fit in a traditional drag-n-drop tool. We just want to make sure, we are truly solving that problem and not just making another frontend builder that people have to wrangle to get their designs to fit.

As for lock-in, we do want to avoid it all costs and our working on strategies and features that would allow you to migrate in and out of Canonic with ease.

Currently, Canonic is a great fit for anyone who wants to build a backends quickly and without writing code. Some level of technical know-how is, however, required. We do hope to eliminate that soon.

P.S. Canonic plans to have a free forever plan for getting started and hobbyist projects

looks pretty interesting, but as a developer why would I commit to spending time learning and using this product without even a hint of what the pricing will be? That's a non-starter.

You make a really valid point. We've been testing the product by putting it out there to get feedback and validation, and then work on our pricing strategy based on the positioning we take. Given our architecture, one of our aims was to offer a free forever plan, the other was to have an eject strategy so that you can take you project with you as a complete and functional code repository.

We feel that currently, the learning curve is pretty minimal. Given the hours it'll save one from coding, maintaining and bug fixing. However, we will definitely keep this in mind when thinking ahead.

Thank you so much for the feedback!

Without any of pricing I've already forgot about this product.

We're sorry you feel that way! We're just getting started, we hope to get there soon :)

> the fastest way to solution it

I've never heard "solution" used as a verb before. It reads rather odd to me, but the product seems interesting :)

Haha now that you mention it, it does sound a little odd! Will go through the content again. Thank you so much for your feedback :)

A trend in modern English, I'd say - though not necessarily an improvement, e.g. "he gave me a book -> he gifted me a book" and here, like you pointed out, they could easily have said "the fastest way to solve it" instead.

I agree that's true. But gift as a verb has apparently been around since the 17th century[0] and doesn't sound as odd to me as using solution as a verb.

[0] https://www.theatlantic.com/entertainment/archive/2014/12/gi...

It's true that "gifted" (which I still find a bit grating regardless of its history) provides more detail, there's some lost nuance with "gave", since you can say "she gave me a glass of water" and "she gave me a painting for my birthday" but we wouldn't say "she gifted me a glass of water".

Am I the only one who cannot load this page? It renders for a brief moment, I can just begin to see the rocket animation, then it renders as a plain blue page. Same results on Firefox 78.0.2.esr (64 bit) and Safari Version 11.1.2 (11605.3.8.1).

That's certainly weird. While we look into it, you can signup directly at https://app.canonic.dev/signup

Interesting product, pricing and deployment model would be good to understand. (Side note: What are you using for docs website ?)

Thank you for taking some time to check it out. We are actively working on that and will have that out soon :)

The documentation is based on Gatsby!

I guess the site is having HN traffic issues? Deployment takes a very long time and then always fails.

The good kind of bad problem! We're actively on it, we are hoping to restore everything back to asap. Apologies.

No worries! Keep up the good work.

Can you compare yourself to Hasura?

Yes! We're similar to Hasura at first glance given the similarities with which we approach content modeling.

The vision is to reduce the complexity (and terminology) down to a point where one would need no prior knowledge of engineering to actually apply their business logic. We’ve been slowly pivoting towards citizen first development and the goal is to get there sooner or later!

Hasura on the other hand, I feel offers a superior developer experience but that comes at the cost of alienating non-technical/citizen developers. Backend knowledge still plays a pivotal role in how well you can use Hasura.

Seems like Hasura is picking up legitimate competition

That's the best compliment we've heard all day :)

Reply on news.ycombinator.com