[tpmoney]

Let's say you're paying your devs $100k / year. All in costs on those devs are probably $150k or so. That means your $1m / year will fund 6 full time developers with a little left over. This podcast from the CEO[1] says their engineering team was 4 people and the remaining staff is the 3 owners, the 1 remaining engineer, and one part time customer support person. So assuming every full time person was costing $150k in salary and other costs, you're already over $1m / year before you pay for any other expenses.

$1M / year is a lot of runway when it's just you. It's a lot less runway once you're paying other people's livelihoods too.

[1]: https://adams-morning-walk.transistor.fm/episodes/we-had-six...

[AlecSchueler]

The question is still why you need multiple devs worth 150-250kpa to maintain a CSS library.

[andruby]

The question isn't "what is the lowest cost that a CSS library could be maintained for"

The question is rather, how can the most popular UI system (especially for AI models) have a healthy business model?

Think of the immense value that Tailwind is bringing to all the companies and developers using it. Surely there should be a way for the creators to capture a small slice of that in our economic system.

[Maro]

> the most popular UI system (especially for AI models)

Like others earlier in the thread I'm symphatetic to this company/project, but your code/project being referenced often in AI output in itself doesn't imply that the thing needs to be a business.

bash, curl, awk, Python code with numpy imports, C++, all sorts of code is constantly being generated by AI, doesn't mean curl or numpy should be its own company, or that the AI Labs need to fund them.

As other fave written, making $1M+ already feels like a lot, maybe this shouldn't be a company, just 1-2 people who have a great time supporting this thing. I wonder if curl or awk have that kind of funding even..

[FelipeCortez]

> doesn't mean curl or numpy should be its own company, or that the AI Labs need to fund them.

you'd be surprised

https://numpy.org/about/#sponsors https://curl.se/sponsors.html

[Maro]

Great point, thanks for making it. Following onward, NumPy has a non-profit called Numfocus who is behind it:

https://numfocus.org/

https://projects.propublica.org/nonprofits/organizations/454...

Apparently they have an annual budget of ~$10M. From the contributors, it's easy to recognize the Chan Zuckerberg Initiative (so Meta), Google, MSFT. This is great.

Having said that, I'd still say that $1-2M for a CSS library seems more than enough. Not everything needs to be "scaled"..

[patates]

When very important tooling does not have very impressive funding, you get the xkcd 2347 situation very quickly.

[rurban]

Not very important. Just sugar for webdevs.

Change the pricing model and you'll better off

[no-name-here]

That’s the All Modern Digital Infrastructure relying on a dependency a Nebraskan has been thanklessly maintaining since 2003 one: https://xkcd.com/2347/

[apublicfrog]

> The question is rather, how can the most popular UI system (especially for AI models) have a healthy business model?

My question is why does it need one? Most web libraries I've used for the last few decades have not had any corporate structure and certainly haven't made a profit. They're done because someone wanted to showcase their skills and others got involved to help, or for fun or because a company who does something else built them internally and decided to open source.

We don't need to apply capitalism to everything. Not everything needs a profit and scale.

[rapatel0]

Profit is the life blood of a business. It’s what pays for, mistakes, new ideas, responding to changes in the market. It tells you your are doing good things and that you are doing them well

It’s the engineering tolerance that allows a company to operate and remain reliable.

It’s amazing to me that engineers don’t understand this concept.

(Clarification, not talking about excess profits)

[apublicfrog]

I think you've missed my point. Most of the libraries I'm talking about are not part of a business. And they didn't need anything to pay for mistakes, new ideas, etc.

I understand companies needing to profit, my question is why does an open source library need a company?

[tpmoney]

> I understand companies needing to profit, my question is why does an open source library need a company?

Because people like to eat and have homes and not everyone wants to work full time on someone else's code and then come home and work full time on their own. Because paying people for the work that they do is a good thing.

[AlecSchueler]

So the millions of dollars are going towards marketing and suchlike you mean?

[andruby]

I don't understand. Does Tailwind spend money on marketing? Afaik they mostly spend on salaries and probably a bit on hosting/infra.

[sonofhans]

If you can find a way to do it better or cheaper you’re welcome to try. No one else has. Don’t think it’s a small problem. The number of user agents and platforms supported by Tailwind would melt plenty of larger organizations.

[AlecSchueler]

This doesn't really answer my question and is quite a flippant response. I didn't claim I could do better, I'm asking why they need so many resources to do what they do.

[exceptione]

Maybe we accidentally found a more meaningful chance for having a discussion about LLMs.

As CSS is limited in scope, ultra-well defined, testable and declarative, this should be a home run for LLMs.

[reassess_blind]

It is. That’s why Tailwind had to lay off 75% of their staff.

[AlecSchueler]

But they're still struggling for money.

[reassess_blind]

Yes, they’re struggling because a large part of their business was selling the pro product of pre-built themes, pages and components and whatever else.

Now, LLMs have all but killed that side of their business. The latest models are incredibly good at writing Tailwind, to the point where no one is buying the pre-builts.

[mexicocitinluez]

Nah, Tailwind is way more important for LLMs than vanilla CSS.

Models work in contexts. If my context is "my entire app's styling", then it's going to be really difficult to write styles in line unless it's already pretty perfect.

Tailwind doesn't have that problem. It's local. I can define a single theme and KNOW FOR A FACT how something will look before it even touches my code. That's the beauty of utility-like libraries.

I stopped working in marketing and advertising (which DID need custom styles), and went to strictly app dev where my needs completely changed.

[IsTom]

> limited in scope, ultra-well defined, testable

Are we talking about the same CSS?

[mexicocitinluez]

lol People don't realize that Tailwind democratized styling for a lot of people who didn't want to or didn't know how to write CSS. We're not going back to writing hand-crafted CSS with or without LLMs. LLMs, by their nature, work better with Tailwind since it needs a much smaller context to make the right decision.

[doodlesdev]

   > We're not going back to writing hand-crafted CSS with or without LLMs.

A lot of us have never stopped writing hand-crafted CSS. Also, in my experience, Gemini 3 Pro is an absolute monster at writing layouts and styling in pure CSS with very basic descriptions of what I want (tested it while I was experimenting with vibe coding in some sleepless night LOL).

There are still a lot of developers who loathe using Tailwind and avoid touching it like the plague. Handwritten CSS still offers more opportunities for optimization and keeps your markup much cleaner than spamming utility classes everywhere (I understand the appeal of rapidly iterating with it, though).

[jpalomaki]

Agents are not yet very good at figuring out how things look on the screen.

Or at least in my experience this is where they need most human guidance. They can take screenshots and study those, but I’m not sure how well they can spot when things are a bit off.

[tpmoney]

Well they clearly don't "need" that many devs just to maintain it, since they just laid off most of their devs. But "need" and "want / have the revenue/work to hire and sustain" are different questions. I've never worked a single development position where there wasn't always more work to do and not enough people or time to do it. It appears they previously did have the revenue, and presumably had the work. Now they don't have the revenue, and so they had to let people go, and some of that work will go undone or take longer.

[toddmorey]

It was more than a library of prewritten css, though, they did quite a bit of engineering work on tooling (speeding up the code scans and dynamically creating custom classes, etc). I respect the team's productivity.

This is more a question about the business model of open source, which has always had some challenges. I don't think you can support OSS with premium templates, training, and support once the knowledge is baked into LLMs.

[troupo]

They don't only make TailwindCSS. They also make a large collection of components and templates at https://tailwindcss.com/plus

[robertjpayne]

Yes but Tailwind Plus has a flawed business model, AI was not really the reason nobody bought it, it's that it's a lifetime purchase and that shadcn + LLMs has eaten their cake left right and central.

If LLMs didn't exist but shadcn still did, do you think people would pay and use Tailwind+ or shadcn?

[omnimus]

Tailwind UI is tool companies buy to save dev time mostly on internal/back office tools. It's usually bought per project. The math is pretty easy - if it saves you few hours of devtime you buy TailwindUI. Shadcn and bazillion other similar things are certainly competition but TailwindUI is very broad and of high quality so why not pick the nicest version.

The problem is that Tailwind is extremely portable (thats why it's so popular) and since LLMs have been fed all TailwindUI code... people using LLMs don't even have to know that TailwindUI exists they just get some Tailwind styled components. They would probably look pretty confused if you told them you used to buy these templates.

[d1sxeyes]

What’s the problem with the lifetime purchase?

[hennell]

It's the difference between one-off revenue and recurring revenue. If you're making new components, making new changes for the new version, adding new css and browser support it's hard to keep going with only income from new customers.

[corobo]

It takes the recurring out of recurring revenue, 100% churn

[plagiarist]

I am wondering why are there three owners for a commercial CSS library?

[whatevaa]

You have one developer. He gets hit by a bus. Now you are fucked.

Having at least several people in critical role helps protect against busses.

[MobiusHorizons]

Sponsorships are a supplemental income stream, though, right? They have paid services in addition as I understand it. So covering several full time developers seems pretty good sponsorship wise, when the maintenance should be fairly simple at this point given the maturity of the offering and the tech stack. It’s not like they have to keep up with security vulnerabilities or a mobile version update churn.

[tweetle_beetle]

They just sell lifetime licenses to extra content at a fixed (relatively small) fee.

> Because every project is different and the way independently authored pieces of code interact can be complex and time-consuming to understand, we do not offer technical support or consulting.

https://tailwindcss.com/plus

[agloe_dreams]

The answer really is that they were spending an amount of that money on devs who were working on tailwindUI / Plus - their paid product.

[leetrout]

They were posting a job for $250k last year.

[solarkraft]

That’s an incredible amount of labor. What were they spending it on?