Hacker News new | past | comments | ask | show | jobs | submit login
Show HN: Deta – A cloud platform for building and deploying apps (deta.sh)
81 points by abdelhai 15 days ago | hide | past | favorite | 60 comments

Hello HN, co-creator here!

With Deta we are exploring ways to make the cloud less of a monster.

As a dev who worked in sales and marketing, I always had ideas for small web tools to improve my own and my team's workflow — but I always faced hurdles, like setting up a server, configuring an http proxy and implementing auth.

With Deta, we are trying to take away these repetitive tasks and allow you to simply build your web app, not the environment it runs on.

We would love any feedback / questions / comments you may have!

AppEngine and Firebase tackle the same frustrations. They both worked out pretty well. Have you used firebase? You should at least try it out and explain where it falls short.

Firebase is great, but there’s plenty of space for yet another simple cloud solution.

Providing a decent SQL database would be a good feature compared to Firebase. And tons of little things that avoid reinventing the wheel everytime I have a new idea

Are you running your own hardware or on top of a bigger cloud?

If it's your own hardware, where is the datacenter and how many do you have?

We use AWS for now and are available in 5 regions. It makes more sense for a small team like us :)

That makes a ton of sense. Thanks! Does that mean you'll make it possible to access other AWS services?

Yes, kind of. We put a lot of emphasis on offering a great developer experience/docs/SDKs (even for new comers and front-end devs), so might abstract some things.

Our upcoming product is a file storage/CDN service that's backed by S3.

The code will look bit like:

    from deta import Drive
    files = Drive()
    files.put("myresume.pdf", "./resume.pdf")
From inside a Micro, this will just work and have the right permissions set.

For enterprise customers, we could offer direct access to AWS resources and Deta could run in their own AWS account.

This sounds cool but one bit of feedback in terms of prioritization

> DetAI (pretty soon) – No-code neural nets to auto-fix bugs before they happen.

I (and I suspect others as well) don't really need no-code auto-fixes for code. What I want is simple workflows for training WYSIWYG AI models. I'm not interested in writing python to express the model's computation graph. All I want are some blocks that I can glue together into a tensor graph and then deploy it for training and inference.

> All I want are some blocks that I can glue together into a tensor graph and then deploy it for training and inference.

Something like that? https://peltarion.com/platform

(no relation)

I looked at the website and I couldn't find a pricing page anywhere. I tried looking for a site map to see if that would lead to a pricing page, but there's no site map page either.

This is great. Thanks!

Sorry, this a poorly placed joke :/ No AI planned!

I'd definitely remove it, it does not come off as a joke.

Re: "All I want are some blocks that I can glue together..." This is a bit over my head, any chance we could discuss this on Slack/email/call? I believe we might be able to help.

I just meant I want an editor that can help me express dataflow graphs but instead of the usual programming nodes I want to work with AI building blocks like convolutions and batch norm. This page on TensorFlow has some example graphs: https://www.tensorflow.org/tensorboard/graphs.

But I assumed the AI part was serious so if it's not then I think this currently would be out of the product scope so I wouldn't worry about it. Hope that helps.

I understand. Apologies again for the misleading joke!

No worries. Most good ideas start as jokes.

I think you might be underestimating the fact that your marketing copy is a good proxy for your product commitment in the long run. I am having difficulties adding you to the short list of "platforms to be considered when starting out a new project" mainly for this reason alone..

Best of luck though, i just wanted to give honest feedback

Thanks :)

This sounds like one of those platforms that's neat in concept, but will fall apart fast for complex requirements.

Ya, after reading the page I would think the developers agree. They even wrote "Deta is building a cloud for the developers with less build bells and whistles". Sounds like a simpler version of Heroku? But why not use Heroku?

They also make it sound like the problem they're solving is that YAML is hard... but it's really not

We do more than Heroku in some cases. Our plans is to offer more convenient abstractions for _developers_, where they can just ship code and we do the rest (imagine netlify but for the backend).

YAML/K8S/AWS make sense for complex backends, not for our target use cases (small tools and utils).

Calling something both production grade and beta at the same time seems like a contradiction to me

The pricing page is hilarious:

Explorer: $0

Adventurer: $5

Enterprise: MONEY

It was funny when I read it but not sure if it is completely appropriate for attracting Enterprise customers.


Please don't be a jerk in Show HN threads—or any HN threads, but Show HNs have special rules: https://news.ycombinator.com/showhn.html.

Sharing your work puts you in a vulnerable position, especially in a community like this one where projects are often at an early stage. It's important that people be able to do that without being mocked or put down.

It’s rude

Hi, thanks for posting this. I have two simple lightweight use cases and I would like to know if those fit to this platform or not.

1. A simple Python script with two-three dependencies that needs to be run everyday once. Basically I am looking for a crontab in the cloud.

2. A very lightweight script that needs to be run 24/7 and would not consume more than few MBs of RAM ever.

None of these two needs an HTTP endpoint.

> 1. A simple Python script with two-three dependencies that needs to be run everyday once. Basically I am looking for a crontab in the cloud.

This is possible and a target use-case. We offer cron and you could get rid of the API endpoint.

> 2. A very lightweight script that needs to be run 24/7 and would not consume more than few MBs of RAM ever.

We do not offer long running jobs atm.

Ping me mustafa@deta.sh and happy to onboard you.

I can't find how to setup cron at your site. Can you please point that out to me?

Looks like you can get this accomplished with a Heroku instance or a digitalOcean box. I current have a script that runs 24/7 on DigitalOcean that costs me $5/month (might not work for you if you're looking for a free solution).

Linode also has a 5$ tier. No issues with them thus far.

You are currently sending the shadiest verification email I ever encountered.

It's the default Cognito template. We are too focused on the actual product that we missed this one – will be fixed soon :)

Now I am curious how shady it is.

The sender is no-reply@verificationemail.com (managed by AWS). We are working with them to have an @deta.sh.

Are you planning to add a SQL database? I don't want nosql

Got the same feedback. NoSQL is great for some ideas. But not all of them!

SQL is not planned yet but we would like to offer it at some point.

I take it SQLite ain't already bundled with Python, then? Seems like that'd be an easy win (and SQLite's almost certainly "good enough" for Deta's target audience).

can you tell a bit more how deta compares to similar (?) services like firebase, netlify, vercel?

(don’t worry if there isn’t a great “usp” answer, just interested in learning more)

I’m personally quite excited about getting the journey from idea to something to tinker (and shareable) back down to how I remember it from a few years ago where I just live-edited files on a server. I tend to over-analyze all the various tech stack and deployment aspects too early :(

For hackathons I’m currently also using repl.it and glitch to get immediate results.

edit: kudos to getting this shipped/launching, can relate to how difficult it is to get an initial version out!

Hi — another co-creator here, thanks for the feedback.

With respect to your first question, I believe abdelhai answered it in another thread above.

We totally empathize with the experience of getting something to tinker and share rapidly up and running.

We actually started out our journey with our own cloud editor. It was essentially a worse repl.it or glitch but with a built in database and files. What we learned from our own talks with devs is that many of them really love their local development workflow and their own text editors. We decided or original approach was the wrong one for us but still want to provide the instant experience and rapid feedback that Glitch and repl.it do so well.

A couple USPs that might not be apparent in our marketing materials:

If you type 'deta watch' from the CLI, your locally saved changes will deploy in real time to your Micro(server). All you need to do is hit 'refresh' in the browser or make a new HTTP request.

We provide a real time 'Visor' which caches and logs incoming requests and responses. These can be replayed from within Deta against your latest code. Users have found this very useful for developing things like webhooks.

Having a database is a useful advantage over Vercel.

$0, $5 for the first two tiers? That sounds way too cheap.

For $5 you can get an entire VPS from DO. You can also get similar services from Firebase for less. How is $5 for a service that offers a very very basic NoSQL API and simple Auth API cheap?

You have to provision and maintain your VPS. As far as I can tell, Deta isn't in the same line of business, they provide a managed infrastructure.

For sure you can already find cheap ways to host your systems. I still think that if you're a small team that want to compete in this domain you shouldn't try to compete on price, they are doom to fail if they go in that direction against already established businesses.

As someone who seem to match their target audience, I can say that $5/month is too cheap for me, that doesn't really inspire confidence. Of course that's just my opinion.

So you're a little bit of firebase a little bit of Herou? I guess consolidating managed services can have value.

Yes + dev-friendly SDKs. That's what we believe.

We changed the URL from https://www.deta.sh/billion-ideas to the project home page, which is more appropriate for Show HN.

We also changed the title from "Show HN: A Cloud for the Next Billion Ideas", which is too baity for HN. I always advise startups (especially YC-funded startups when they prepare their launches) to rigorously take out any sales and marketing language from their text and titles. HN readers are averse to it.

The basic rule is: Don't try to sell them, try to interest them. If you sell them, they will close the tab. If you interest them, they will sell themselves.


Honestly, we felt that our description (cloud platform) doesn’t match our USP, and we thought that our phrasing conveys our message/ambition more succinctly.

No worries. I'd be happy to change the title to something accurate and neutral if you want to suggest better wording.

Isn't this pretty much what Heroku does? Why would I choose Deta over Heroku?

We offer a higher abstracted cloud. With Deta, you do not VMs/container, you just push code and it just works.

No need to load balance, etc. We also have a built-in fast NoSQL DB that works without configuration.

And many other dev-oriented conveniences.

Very cool platform, but I can't actually sign up!

This was down for a bit but should be fixed now, sorry!

Why did you link to the ideas page?! There are 839 words on it and I have no idea what your product is or does.

After looking at the docs[1] and front-page, you're a serverless lambda/functions/zeit with all the normal limitations, but a pretty decent tech-specs for the free and paid tier.

What makes you different? I really like the lack of vendor lock-in, but you need some more killer features to stand out.

1. https://docs.deta.sh/docs/micros/about

Did you not make it to the bottom, which includes a very specific and useful definition of the services they offer?

No, because it started explaining why computers are good and I didn't need a recap.

Well, it got your attention so it kinda works :)

We see Lambda as a building block. For creating web apps you need more than a runtime, in most cases you need a DB, login and permissions, UI, etc.

For small ideas/tools (our target use cases) we offer you most of what you need with as little config as possible (ideally none) – things just work. For example in your Deta Micro, you can just import our Database (Deta base) and store data and it just works, no config required. Soon we will launch Deta Drive, then email, then cron... you get the idea.

We also have permissions built-in, so you can give your colleagues a run-only access.

We also focus more on the developer experience more than others, devs should feel as a first class citizen in the cloud. We think this will free them to be more creative.

We have lots of other features that we will announce soon: real-time debugging, online editing, etc.

I hope that was helpful.

We've changed the URL now - see https://news.ycombinator.com/item?id=23631761

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact