Show HN: DrawDB – open-source online database diagram editor (a retro) | Hacker News Reader

Show HN: DrawDB – open-source online database diagram editor (a retro)

drawdb.app

263 points1ilit56 comments

One year ago I open-sourced my very first 'real' project and shared it here. I was a college student in my senior year and desperately looking for a job. At the time of sharing it i couldn't even afford a domain and naively let someone buy the one i had my eyes on lol. It's been a hell of a year with this blowing up, me moving to another country, and switching 2 jobs.

In a year we somehow managed to hit 26k stars, grow a 1000+ person discord community, and support 37 languages. I couldn't be more grateful for the community that helped this grow, but now i don't know what direction to take this project in.

All of this was an accident. But now I feel like I'm missing out on not using this success. I have been thinking of monetization options, but not sure if I wanna go down that route. I like the idea of it being free and available for everyone but also can't help but think of everything that could be done if committed full-time or even had a small team. I keep telling myself(and others) i'll do something if i meet a co-founder, but doubt and fear of blowing this up keeps back.

How would you proceed?

Open articleView on HN

Just for the context, I've been making (and selling) database (migration/translation) tools for 20-ish years.

What you have built can definitely become a product supporting you with a very nice income. Free version can stay free but you can sell solo licenses and team licenses as subscriptions. Paid license users will also have your support - and that matters to companies.

As others have said, commercial features can be built on top of the OSS product, like Electron desktop client, support for more databases, for views/procedures etc.

There is no downside to trying to commercialize this. Whoever wants to use what you have built so far can continue and I guarantee that many companies would be happy to give you $50/month for a more advanced version with support.

(edit for typo)

> There is no downside to trying to commercialize this

I know lots of people in the FOSS community who would stop using it as soon as you "sold out"... but those people probably don't represent the majority of potential users.

Legally, any enhanced version would need to remain FOSS anyway. It appears to be an AGPL-licensed project with 60+ contributors and no CLA, so the codebase has 60+ different copyright holders.

Monetization of a GPL or AGPL project generally involves a CLA so that the creator can offer closed-source and/or SaaS enhancements while still accepting third-party contributions. Without a CLA, the project creator would be violating the copyright license that was granted by the third-party code contributors. He/she doesn't own their code and cannot relicense it in a way that violates the AGPL.

I didn't look into the code structure - 60 contributors mean that yes, there is a problem.

People start something with good intentions, then the project grows to the point where it requires a lot of unpaid time, and there are expectations from the users. There are bills to pay and limited time in a day.

I would probably be irked if I contribute in my free time and then see my effort being used in a commercial product. Then again, in some products you may have one person doing 95% of the work and another 50 doing 5% of the work.

If OP is motivated, they can relinquish control over this project and start a completely new one from scratch with no code borrowed from the existing one but learnings can definitely be applied. The new commercial project would need to compete with the existing AGPL project - and that's a good thing. If 60 contributors are contributing significantly, FOSS project won't suffer and the OP will be able to live from something they are passionate about.

Or could they just reach out to contributors and ask them to help?

Or here’s another route: sell “licenses” regardless of the actual license. I think https://cyberduck.io/ has this: you can donate and get a key that removes the donation nag. If you do this, you can’t go after the pirates, but would you really want to spend your time on that? (Of course, I would still reach out to the contributors first, explain the situation and see if they are okay with all that.)

Good luck to the OP!

Out of curiosity, is that also true if you merely want to sell services such as support on top of it? How about customization?

Disclaimer: I'm not a lawyer.

Selling services like support or consulting is fine, since that's generally independent from FOSS licensing/copyright concerns. Anyone can do that though, not just the creator or contributors. For example, there are lots of third party database consultancies that offer support for major databases, both FOSS and commercial.

Customization is also fine, but the purchaser of the customization must still abide by the AGPL: the end-users of the customized software need access to the customized codebase. If the customized product is only being used internally (e.g. for employees of the company that purchased the customization) then there's no problem. But if they're offering it in an externally-facing SaaS, or embedding it in externally-distributed software, the customized code must be made available to the end-users of the software.

With FOSS often it's simpler to do this as "sponsored development" where someone pays for a new feature to be made directly upstream, rather than having a customized fork.

That all said, services and customization are generally difficult to scale as a business. Even with a fairly large userbase, the percentage of companies who are willing to pay for support or customization tends to be disappointingly low.

A solo developer trying to find a way to offer something for free and eat shouldn't be mutuals exclusive.

Selling out is relative to having one's living costs somehow subsidized.

Not everyone has the same financial safety nets, and likely most have to create a safety net for themselves.

Maintaining a clear line between the free size and paid should be easier here compared to other tools should the developer decide to go that way.

This seems like a really cool project.

Some observations...

First I'd get some legal advice for how to convert this idea / product into a thing that you own and have a stake in, but that's distinct from you. This means you'd be able to collect money and attribute that money to the product, you'd be able to sell it to someone else and walk away from it, etc. This legal construct is important; probably not _necessary_ but very useful.

You seem to have a reasonable community contributing to and using the project; that's your most valuable asset. Make up a brief roadmap of where you'd like to take the project in six and eighteen months, share that with the community. Think about how some projects, like Elastic.co, grew based on community feedback and how ultimately perhaps those companies felt they needed to change the company in ways that many in the original community disagreed with. There are lots of sides to that coin, and you as the founder need to think about how you'd want to develop the product.

Lastly, once you've got a clear accounting mechanism of how to track money from it and costs of it, you should just .... start asking people for money. "How much is a support contract worth for this?" "How much is it worth to you to focus on developing this feature instead of that?" Obviously -- be very cautious if people are offering money in exchange for fractional ownership of "the thing" but even that may be palatable... but make sure you've had someone representing your interests read the contract...

Are you talking about incorporating a company? Why is this necessary at this stage before earning money? Can be a waste of money.

I'm suggesting that starting a project like this should involve consulting with an attorney and possibly an accountant.

It is important, generically, to be friends or at least have a relationship of some sort, with an attorney. This is all _really_ vanilla stuff and they should be able to assess the minimum requirements and financial outlay needed in less than 20 minutes.

You want to be very clear about what the thing is, where you stop and it starts, who owns it. If you're going to be taking money for doing the thing, you want to know how to pay taxes on it, and how to value it (now, when you create it, and later, in the event you plan on selling it).

Make a plan, talk to the attorney about the plan, and understand what steps need to be taken to protect yourself and the project.

You can certainly play fast and loose; it'll probably be fine. You may have to play fast and loose if you have no resources. But there are lots of people who have gotten very badly burned by not paying attention to the rules and fine print in the contract(s).

Examples of things going badly include Redis and CCR, just off the top of my head.

where i'm currently based, every single cent that can be considered income is tracked and taxed. plus i would have to declare its origin. so there's definitely some legal set up before monetization

You may in fact decide to explicitly never monetize it, which is a perfectly reasonable decision.

You may also find that there are ways of setting the thing up so that you aren't paid (at least not right now if it's not convenient to be paid) but the thing is able to collect money and use that money to do $RANDOM_BUSINESS_THINGS like pay for hosting and maybe eventually hire a developer...

It's likely that this is all really simple stuff, to someone who knows exactly how thing work in $LOCAL_LEGAL_REGIME

It can even export as Mermaid? That's awesome. I could see that being super useful as a way to take a db schema and turn it into a chart for a github readme. Definitely bookmarking this.

Btw, the SQL import doesn't seem to work for an sqlite `.schema` dump; it breaks on the PRIMARY KEY here:

    CREATE TABLE IF NOT EXISTS "__EFMigrationsHistory" (
        "MigrationId" TEXT NOT NULL CONSTRAINT "PK___EFMigrationsHistory" PRIMARY KEY,
        "ProductVersion" TEXT NOT NULL
    );

(This is the standard migrations table that Entity Framework creates.)

W/r/t monetization, you should know there's a really powerful flowchart tool called draw.io that's already free. It's not tailored specifically towards databases the way yours is, but it can make similar graphs, so if you go too hard on monetizing you might cause people to just use that instead. Maybe there are some more B2B-oriented extensions or db integrations, or maybe a team collaboration feature you could add without compromising the core tool.

thank you!

regarding the sql import, it breaks because of the parser. i'm using a 3rd party parser to parse sql and pull out everything needed for the diagram. i'll report it to them, thanks for letting me know

Not everything has to be monetized. You will have other ideas or maybe you’ll take this further. It’s a ton of work to build a thriving business, and you may lose some of your joy and pride along the way.

You created a useful tool and have the support of the community! That’s pretty cool. Congrats!

appreciate it!

This looks a lot like ChartDB[0]. Is that correct? If so, how would you compare the two tools, or better yet, distinguish yours from theirs?

[0] https://github.com/chartdb/chartdb

hehe mine was up and running way before theirs. but we have diverged quite a bit. might be biased but i think mine is easier to jump right in. mine is more like the draw.io of database editors while theirs is the miro, yk?

LOL I don't know actually, but I will definitely give your project a go when I have to mock up my next DB schemas.

It seems like your project is more grassroots than theirs, so hopefully you can use it to make some money... Not sure how your platform works, but do you provide (paid) storage, syncing, multi-user collaboration, etc.? That seems like the predictable monetization strategy.

I experimented with this for five minutes, and so far it looks great. The one thing missing, or that I didn't see in five minutes, is that when I uploaded the schema for one of our SQLite databases, the tables came out perfectly, but the views were ignored. There are a lot of views, and we would want them in the diagram. I will definitely be coming back to it. As I say, what I saw was great.

Thank you! Yeah we don't have views yet, but will

If you wanted to commercialise this, then you could do it on a feature basis.

Create a paid version where you add more features, like support for views.

Looks really great! But regarding monetization, one word of caution that other commenters haven't mentioned so far: make sure you have a really good understanding of the licensing concerns before proceeding.

Initial disclaimer: I am not a lawyer and this is not legal advice.

Your project is using AGPL, without a CLA, and you have third-party contributors. If I understand correctly, this essentially prevents anyone from having an enhanced closed-source version (SaaS or otherwise) of your product -- even you. Although you created it, you only own the copyright for your own portion of the code. Third party contributors own the copyright for their contributions, which are equally subject to the terms of the AGPL for everyone.

So if you wanted to monetize an enhanced version, you have two options, if I understand correctly:

A) Keep the enhanced version entirely open source. You can charge for a hosted version but users must be able to self-host. Obvious downside: this leads to weird misaligned incentives where your revenue can depend on your project being difficult to install/host.

or

B) Obtain a licensing agreement from all third-party contributors since the time you switched to AGPL, or completely re-write any code where you can't get that agreement. Might not be practical.

You could try alternative monetization methods that don't rely on an enhanced version, such as support contracts or sponsored development of new features, but these are hard to scale... especially since a decent number of tech companies have a blanket ban on using AGPL projects that aren't dual-licensed.

This isn't to say AGPL is bad, just that it's a nuanced topic. There are lots of blog posts out there about the monetization implications, for example here's one from the cofounder of GitLab: https://www.opencoreventures.com/blog/agpl-license-is-a-non-...

A suggestion based on an idea that I've been trying to implement for a while: consider replacing the left panel's form editing experience with an assistive command line at the bottom.

E.g. rather than clicking table, "add field", clicking the name, typing, clicking the type drop down etc., just start writing something like:

    : users.add_column
    name: email
    type: varchar
    length: 200

etc., with some assistive auto complete

or, if you're inclined to using an LLM:

   add column email to users table, varchar(200)

or, for those who want strict sql:

   alter table add column...

A quick(ish) and obvious path to monetization would be offering a SaaS hosted version. The obvious barrier here is how to connect to databases that are private (not publicly accessible). VPN, Wireguard, Twingate etc can be utilized but still a barrier.

I own and run a DevOps consulting company and if you have interest I may be able to assist you getting the hosted platform up and running. It is built currently to be multi-tenant with projects/organizations and users supported? Gonna need single sign-on support as well.

As someone who is afraid to release personal projects I think what you have done is amazing. Id love to learn From your experience. You are being too humble but this kind of growth is very hardly an accident!

I'm not a monetization expert by any means so take all this with a pinch to salt. If money is not an urgent factor for you you can see how far you can take it especially towards "other verticals". If you can displace any existing incumbants doing so would be easier without the responsibility of supporting paying customers.

appreciate it!

Congratulations on releasing the product and its success.

You will likely have to compete with Supabase's diagram editor (there might be other competitors too), which has AI integrated. You can check it out and see if you can improve your system. Their product is open source as well, so you can review the code.

The simplest monetization path would be to keep it free and charge users for AI access. However, don’t hire a team, as it may not be viable against other big players in the long run.

This is very cool! I have some feedback:

1. Allow for adjusting grid spacing and grid point size.

2. I can't seem to find a snap-to-grid option (e.g., a magnet icon).

3. Allow docking of the sidebar to the left or right of the screen.

4. When the sidebar is hidden, the popover does not include tabs for Relationships, Subject Areas, and Notes. It would be nice to have those available.

5. Allow sidebar width customization in pixel size or percentage

I really like this app. I tried it out but when I wanted copy and paste the first table I created as a template to create the next table, the app asked me for access to read text and images of my clipboard. I don't really like doing that because it seems like your app can read anything of my clipboard at any time?

>the app asked me for access to read text and images of my clipboard yeah, that's the browser asking for permission to read and write to the clipboard

>it seems like your app can read anything of my clipboard at any time only when you paste, the app reads the latest entry in your clipboard, and only if it is a valid json object of a table, note, or subject area, it gets imported

just using the clipboard api: https://developer.mozilla.org/en-US/docs/Web/API/Clipboard_A...

Not bad, I bookmark it, its good for prototyping ideas.

If it doesn't cost you much money you should keep it free and maybe sell an advanced version aimed at companies.

There are already similar things like this in the market, you could easily create an advanced version by adding paid AI prompts that generate code.

Hey! This is awesome! Great job :-)

I make Beekeeper Studio, another open source db tool (although v different), we started charging for a pro version 2 years ago. Would be cool to connect! I can tell you what we did and what did and didn't work - Matthew (at) beekeeperstudio (dot) io

thanks! would love to connect

I tried it with a mysql project that has ~150 tables, which I have been struggling to find a good way to visualize in general, but it became extremely slow and overlapped a lot of information so it was basically unreadable/unusable.

The examples look quite nice though.

I would love this as a desktop app.

Awesome work! We've been searching for something to switch away from MySQL Workbench for years, but the one feature we need is the automatic generation of SQL migration based on a current and a new schema.

Coming... https://imgur.com/a/U8cbJ5Y. Working on versioning, based on which migrations will be generated.

that's really impressive.

Looks very very cool!

Think about enterprise features you can charge for, and see what your community thinks.

If 100 people per month can pay 50$, that's 5k which is very nice for a side project!

For myself, I absolutely would love this as a Django plugin!

>If 100 people per month can pay 50$, that's 5k which is very nice for a side project!

sounds too good to be true xd

>For myself, I absolutely would love this as a Django plugin!

after migrations, will be adding orms. or some plug in system for anyone to add whatever they want. don't know how but would be a fun feature to work on

I'm curious if you are planning for auto-layout? It's not listed in your features list but it's one of the most important and challenging to do well.

This is interesting and quite an useful tool, but what's the USP from other DB diagram editors

"A retro"?

Does this mean "a retrospective"? It's kinda ambiguous - why not type the extra characters?

yeah, we call our end-of-year 'summaries' 'retros' at work, i put it there without even thinking

Very cool. Which library do you use for the drag&drop part?

Thanks! No library. Just svg, foreign objects, and event listeners

why do I only see the warning

"Attention! The diagrams are saved in your browser. Before clearing the browser make sure to back up your data."

only at the very bottom of the home page ?

good point

Cool! Are you planning to add support for a mobile layout?

IMO the best way are open a patreon or another crowdfunding option.

i had a buy me a coffee link for a while. then idk what got into me and i starting feeling bad for the $20 someone donated

Good

Failed to import a Postgres export with all of this errors:

- "missing_fields" _text NOT NULL DEFAULT ARRAY[]::text[]

- "status" text NOT NULL DEFAULT 'RUNNING' ::text CHECK (status = ANY (ARRAY['RUNNING'::text, 'PAUSED'::text, 'STOPPED'::text, 'SUCCESS'::text]))

- DROP TYPE IF EXISTS "script_trigger_event";

- CREATE UNIQUE INDEX name ON table_name USING btree(unique_key, environment_id, deleted_at) NULLS NOT DISTINCT;

yep, parser issues. will report it, thanks for letting me know

Reply on news.ycombinator.com