Console

Open source vs commercial (Appsmith & Retool)

with David Mytton & Jean Yang

S01 E06

2021-08-12

Appsmith (open source internal tool UI builder) & Retool (internal tool UI builder), a devtools discussion with David Mytton and Jean Yang.

Episode notes

Episode 7 of the Console Devtools Podcast, a devtools discussion with David Mytton (Co-founder, Console) and Jean Yang (CEO, Akita Software).

Tools discussed:

  1. Appsmith - Open source internal tool UI builder
  2. Retool - Internal tool UI builder

Other things mentioned:

David: Welcome to the Console Devtools podcast, a show all about interesting developer tools. I'm David Mytton, co-founder of Console.

Jean: And I'm Jean Yang, CEO of Akita Software an API observability startup. In each episode, we'll discuss two interesting developer tools. We're keeping this to 15 minutes. So let's get started.

David: This week we are discussing two quite similar tools, Appsmith and Retool. Both of these are internal UI builders. Developers have spent years and years building internal CRUD apps that just update and modify database records, and that really isn't making the most of what developers can actually do.

Both of these tools provide interface builders that allow you to connect into existing databases. Particularly interesting is the fact that they both proxy queries to the data source, so the results aren't stored, which is really great for security, whether you're running in the cloud or open source, deploying it yourself.

The primary difference between Appsmith and Retool, is that Appsmith is released on an open source code base and you can deploy it yourself or you can use their cloud version if you like, whereas Retool is based on a proprietary code base. Although they have a self-hosted version, which is deployed in a container, you can also sign up to their cloud version and they host everything for you.

There are some implementation details that are different between them. For example, with Appsmith, you can't just drop all of the configuration into source control. It's stored inside a MongoDB database. You can export it to JSON, but essentially everything's within the Appsmith data store, whereas Retool does let you do that. You can put everything into a Git repository and manage it there, but that's only available on their enterprise version, which I found a little bit disappointing.

Essentially they make it really easy to build interfaces to query data sources, whether that's S3 or GraphQL or connecting to Google Sheets or some other data source. You might want to build an internal interface for support teams or customer success, something that perhaps developers are not going to be building for you.

Jean: I'll just chime in here to give a few examples of why building with  these internal tools is cool. You can make a Stripe refund interface, or you can make an interface to manage your MongoDB really easily. These are things that companies customize internally all the time, but it takes their internal tools team quite a bit of time.

David: One of the nice things about Retool is it allows you to query all the sources with SQL, even if they aren't databases. Appsmith and Retool both expose the UI so that everything is a JavaScript object. If you need more customizability, then you can just get into the code and you can even create your own react components with Retool. The main difference then is the philosophy behind how these are developed, so I thought it'd be interesting to have a discussion about that.

Jean: Initially, when we wrote our show notes for this David, you had said, "I think the main difference is Appsmith is open source and Retool isn't." And first I thought, "Oh, well maybe it's really just about self-hosted or not”; because if you're doing an internal tool, you don't want some random company to be seeing all of your data. You don't want that kind of friction for an internal tool, because normally you don't have that kind of friction. Does it really matter that Appsmith is open source? I wanted to get your views a little bit on this, David.

David: It's really a philosophical difference between how the companies are approaching their development. I think now with Appsmith, they are talking about how they are developer first. For them, that means being able to modify the code and make customizations.

Retool are really trying to build everything for you so that you don't have to make any code changes. I think we've all been there where we had to build these terrible interfaces that you build once and they do the job. Then maybe you're getting requests all the time to improve them.

It's not much fun as a developer building these interfaces. Having that UI builder just drag and drop the components really does make sense. But coming down to owning your data, being able to run all the infrastructure internally, that's something that I suppose you could do with both of them, because Retool, they offer a Docker version that you can just deploy on your own servers. It comes down to being able to actually modify the code.

Jean: I will say that this is what I discovered as well when I dug into it, because when I looked at the Appsmith GitHub repository, I saw 119 contributors. I cross referenced this with a number of employees on LinkedIn, which is much, much less. I actually went quite deep into the contributors list just to see what they're contributing. It does seem like people are submitting small customizations or small fixes that they're finding useful to them. This is something that I hadn't thought was the high order bit to making a tool like this developer first though maybe it's obvious, but this was very interesting to me.

David: I suppose it comes down to the idea of being able to make changes when something doesn't quite do what you want it to do. I know when I've been playing around with different window managers on Linux and I've come across a bug, I've been able to get into the code and make a few changes. I've never really made anything massive. It's always just been a few tweaks here and there just to improve the experience.

In contrast, when I’ve found bugs in proprietary software and reported them to the vendor, if it's a large vendor, it tends to go into /dev/null and you never hear from them ever again, and the bug doesn't get fixed. That is opposed as the advantage of the open source side of things. Have you ever done that with anything you've been running?

Jean: I actually haven't. Maybe this is because I'm not a good open source community contributor. There have been small instances where I've suggested changes to docs or something like that. I had to install something a particular way or something like that, but I don't think I've really been part of a social open source community.,so I haven't gone and contributed to any of these communities myself.

David: Docs are a classic example, because they're often out of date or there's some very small thing that just doesn't quite work. It feels nice to be able to submit a request and have it accepted, and then you get your little avatar showing you as a contributor to the repository which is quite nice.

Jean: I was really impressed with Appsmith when it wasn't just docs. It seems like substantial changes went pretty deep into the system.

David: When I was putting together the notes for this, I labeled Retool as commercial and Appsmith as open source just to distinguish them. Your comment was that any tool with an enterprise version is commercial and not open source. So what did you think about that?

Jean: I run my own startup. And when we think about a startup going to market, open source is just one of the categories. I think if open source counts as a go to market category and a tool has an enterprise version, I would say this is just its commercial strategy. We have this philosophical question, if an open source community came about because of a tool's desire to become a commercial tool, is it open source? I think you, along with my digging into the contributors list, have convinced me that it is still very much in the open source philosophy and people are benefiting from it. It's more developer friendly this way.

Jean: Yeah, I think this distinction between open source and commercial, if we're making it for the purpose of signalling to developers that you can change it, it's available to you, the source code is available to you at all times. That makes a lot of sense. But I think if it's this, these people are trying to make money off of it and these people aren't, they're both making money off of it and they both deserve to make money off of it. It gets us into this other conversation, who pays for open source. It was interesting to reflect on that labelling.

David: The idea of open source has a lot of connotations around it, which perhaps don't match with the reality anymore. There are certainly loads of different approaches to releasing code, whether you just throw it out there with an open source license and that's it, all the way through to the classic example of Linux being the foundation to pretty much every computing platform that exists today.

Jean: There's this interesting cultural shift going on where people are starting to realize, "Hey, someone is paying for open source. If we're getting the software for free, how are these software developers actually living?" These days, we know that we should be paying our artists. We should be paying our creators. We should, most of all, be paying our open source developers. I think the ‘who pays for software’ story has had a few iterations on in the last few decades. We’ve reached an interesting point now.

David: When you deploy open source, it might be free as in the license is free, but it's not free as in your time. That's where the value of SaaS comes in because you're paying someone else to run the infrastructure, deal with backups, monitoring, reliability, and keep it up to date. Everyone's always on the same version. It just makes it a lot easier.

Jean: Yeah. I had this blog post:  “Why aren't there more programming languages startups?” And I meant programming technologies more generally. A lot of people said, "Well, the reason is because people won't pay for a compiler." But if we think about programming technologies more generally, most things are much less easy to operate than a compiler, which is fairly small and complex. It's not that there are all these different operational contingencies you have to make for it. But for most SaaS software, when the hard work is actually in running it, if you start using an open source tool and you scale it up, the inevitable outcome is you pay them to run it. What does that mean for open source then has been what I've been wondering.

David: We've seen these companies actually make significantly more money by offering those cloud versions where everything is run for you. You still have the option to deploy it on your own infrastructure if you really want to, but you're paying so that something that isn't your core business is handled by someone else.

Jean: Right. My question about Appsmith was what percentage of their serious users, by serious I mean they have some scale and have quite a bit of traffic on their internal tools, are using a self-hosted version versus the enterprise cloud-hosted version?

David: Because if your internal tools go down or there's a problem, then you really want to have control of the ability to fix it. If your entire customer support team or your billing team are unable to process things for your customers, that's a problem if you're relying on a third party provider. Right?

Jean: And I also wanted to bring up something else I noticed that might be a result of many different possible things.

Retool seems quite a bit further along than Appsmith. I thought their primitive constructs for building these pages was quite a bit cleaner. Their docs were a bit more extensive. On the one hand it could just be Retool is further along as a company. They have more people etc.

I also wonder if the fact that Retool isn't focused on developer first open source community spread allows them to focus on some of this other stuff. Especially when you're early as a start-up, you have to focus on your priorities. What I want to point out is open source versus closed source or commercial proprietary, how ever you want to call it, is not just a philosophical difference, because I think it really does have material impact on how the product evolves. I wonder how much of Retool's increased maturity in other domains as a result of their decision not to be open source.

David: That's a really interesting point. You mean Retool able to put more resources into building app functionality and polishing the UI. Appsmith, in contrast, have had to think about how people are going to deploy this? How do we make an update mechanism? How do we make sure it works?

Jean: Right, exactly.

David: Certainly a challenge when you're deploying code just to your own environment, you have complete control and you can always make sure everyone's on the latest version. You can manage all your software updates. Sending what is essentially packaged software out to all your users is a real pain because people don't want to update.

Jean: There's also the growth strategy question, because if Retool is not necessarily developer first, it's selling into these bigger organizations. They're getting bigger checks, which allows them to put that money back into improving their product, improving the slickness, smoothing out those rough edges. Whereas if you're going developer first, a bunch of your users aren't actually putting that money back into your system. That could make your growth slower. This is something I wonder about the differences in parts of their maturity.

David: All right. That's it for this week. Please tell your friends about the podcast and send us some feedback. Message us on twitter @consoledotdev.

Jean: Hopefully we said controversial enough things in this episode that you'll be really mad and want to talk to us. We want to hear from you. Our views on this are evolving. Everyone's views on this are evolving. You can find us on Twitter. You can find me on Twitter @jeanqasaur, David at Twitter @davidmytton and my company Akita on Twitter @AkitaSoftware. See you next week.

David: Please disagree with us. See you next week.

David Mytton
About the author

David Mytton is Co-founder & CEO of Console. In 2009, he founded and was CEO of Server Density, a SaaS cloud monitoring startup acquired in 2018 by edge compute and cyber security company, StackPath. He is also researching sustainable computing in the Department of Engineering Science at the University of Oxford, and has been a developer for 15+ years.

Jean Yang
About the author

Jean Yang is CEO of Akita Software. Jean earned her PhD in software correctness and programming language design from MIT and then became a professor in computer science at Carnegie Mellon University before she started Akita to build the future of API observability.

About Console

Console is the place developers go to find the best tools. Each week, our weekly newsletter picks out the most interesting tools and new releases. We keep track of everything - dev tools, devops, cloud, and APIs - so you don't have to.