I’m Alex from Parabola (https://parabola.io). Parabola is a visual programming tool for creating functional data flows that everyone can use. It’s entirely drag-and-drop, handles data sizes much larger than a traditional spreadsheet, calculates everything live, and can run your flows on a schedule of your choosing.
I used to work in strategy consulting, doing data analytics for SMBs and Fortune 500 companies. The amount of time wasted on menial tasks was astounding. Things like cleaning data, generating custom reports, creating human workflows to solve shortcomings in third party tools, etc.
Non-technical people have to rely on doing things manually or using fragile spreadsheets that solve only part of their use cases. We don't think everyone should have to learn to code in order to work productively (https://medium.com/parabola-labs/not-everyone-needs-to-learn...). That’s why we’re building the tools to help people stop wasting so much time on manual processes, and instead focus on their actual core competencies.
We’ve made some interesting decisions in the design of Parabola’s visual programming “language”. We’d love to hear what you think and are happy to discuss our thought process in the comments.
But, I am very bummed out every time I see a startup where SSO is that feature. SAML/Okta, fine, whatever. But, it'd be really nice for any little startup to be able to let their users log in with G Suite and get two factor and all of the other security niceties that come with it. Letting folks adopt security best practices shouldn't cost them money (and isn't really core to your value prop). Just my two cents.
Ps. With GDPR coming into effect, and all the surrounding brouhaha, this is particularly important as many of your use cases are going to be piping around and manipulating user data.
Programming is total freedom to do what you want and these tools try to make it look like you cab drag and drop data flows together and it will just work in a optimal way. And it will at first, until your first special or odd requirement needs to somehow be implemented in that tool.
First class functions.
Visual lexical closures.
User defined blocks including control structures.
Macros and special forms.
Call with current continuation!
Adding Machine Learning Blocks to Snap!
What would you say differentiates this service from say, Zapier?
>"drag-and-drop data transformation pipelines that comes with app integrations at both ends"
Parabola is very focused on data transformation/manipulation. Our best customers end up with hundreds of steps in their flows, essentially creating full-fledged programs. As part of this focus on data transformation, we primarily operate in a batch-processing fashion. We also perform calculations live while you work on your flow, similar to a spreadsheet or a python notebook.
Zapier on the other hand is great at connecting with TONS of tools, and makes integrations between those tools really smooth. It operates best in an event-based paradigm when you don't need to do much transformation/calculation in between connection points.
There's probably a small amount of overlap at the edges, but in general it's pretty clear when a use case will be a good fit for a tool like Zapier or when it's a good fit for Parabola.
Disclosure: recent ex-employee of SnapLogic.
So, I looked at your solution. It seems like a prettified version of Node-red with backup, js, and lack of ecosystem.
So, what's your buy-in? What're you doing better than a spare VPS running node-red? Not everyone is going to install and maintain a server.... but you're here on HN. I'd say that 95% or higher could get Nodered up and running here.
What’s your buy-in?
What are you doing with node red that can’t be done with a text editor and npm?
They're both in the realm of "graphical programming language in the browser". One is a pay-per-$something, and the other is libre/gratis.
And sure, you could use backend js to write scripts to do $thing, but requires a lot of setup to work with different APIs. So absolutely no, js+text editor is nowhere near the same.
If I've saved 1.5 hours, it's paid for itself, and I've been freed up to be productive on other things, like watching YouTube.
As others have alluded to, the challenge will be when users need to do something non-standard with the data. From your comments, it sounds like everything is built in JS. Is your multitenancy isolated enough that you would be able to build a JS scripting engine? That is, to enable power users to drop in some custom logic when they need to, by writing JS against a simple API.
How long ago did you launch? How did you find your first customers, and are they happy with the product?
Also would be keen to hear the story of your journey to product/market fit. Did the product always look like this? Did you talk to users as you built it? It’s such an ambitious project, I’d be scared I was building the wrong thing for months while getting to launch. But it seems like you’ve already hit product/market fit, so kudos for that.
Congrats on the launch, this looks awesome! Would love to learn more about how it was built.
Given there are definite limitations with browser based JS and user’s systems, we’re now a fully cloud based compute environment. As users make changes in our React UI, we send the work to a queue and then distribute it by parsing the flow tree across various node worker servers so steps can run in parallel (if your flow is built that way). Each step is self contained and we then use sockets (powered by PubNub) to notify the UI of changes.
AWS Lambda is fairly reliable for running untrusted functions (as well as trusted functions with carefully managed inputs) at scale for customers, but any function provider should work well for this purpose (as long as their container lifecycle meets your needs).
The problem is that someone without a technical background won't initially see the value in a tool like this and I doubt non technical companies/individuals will need as much automation to justify it's price.
Edit: I think a much more sensible pricing would have been to offer a desktop app and then charge users for the automations they want online on a consumption basis.
I'll send a message on the site chat!
Where I work we're currently replacing everything that's SSIS with .NET services. I get the idea that taking the code out of it seems easier, but I've found it usually just leads to frustrating limitations. For our latest data migration we're moving things out of some SQL-based ticketing system, online accounting services, and FileMaker into Dynamics 365. Sure, it's not pretty to look at, but we can do some deep manipulation and mapping to existing entities in Dynamics that is frankly a chore in any sort of other migration tool.
To handle stuff like scheduling, endpoints, etc we use Azure functions.
python-based version for tabular data processing & management, ecom oriented mostly (so basically op's product), myself.
Target market is something that UIPath currently addresses is it not?
How does Parabola handle authorization issues? Is it mostly through the right API?
I am teaching Python to office personnel who balked at the high price of UIPath but might find Parabola pricing easier to accept.
Take a look at Microsoft's logic apps, it is slightly more generalist, and not so focused on data processing but not nearly as polished as Parabola.
RPA tools seem to be all the rage, but questions around accountability, auditability and other security aspects still stand out.
Also, the UI for flow is absolutely atrocious once you start adding large numbers of conditionals.
The question I have is - can it outperform the command line?
(see https://news.ycombinator.com/item?id=17135841 for more)
I almost always use awk/mawk for such things.