I understand your feeling, but I used it in a startup I work for. As I was responsible for this feature (customers uploading their csv files), with PP I did this task in half a day when it could have took me a few days. I told this to my boss and I asked him to donate to PP from the company, but I never heard about it again, so I will ask him again on Monday.

Thanks again for this library.

See, I love hearing stuff like this (but also am sad for the realities of donations from companies). Thanks for your comment!

Donations from companies are notoriously difficult to push through for any but the smallest/newest companies. Any more established ones tend to have rules and bureaucracy built up around the process of making donations (if they do at all), and individuals within usually have little direct influence.

Paying license fees for tools or software, on the other hand, is commonplace enough that small amounts are usually at a manager's discretion with little oversight. Offering to send an invoice for a 'license' that doesn't offer any additional benefits may be a way to simplify things for companies like this that want to pay to support the open source they use.

I second this: Open-source projects, please send invoices so companies can bury them in their accounting. Even postgresql.org let me down with a "the only invoice is the CC receipt from paypal", which triggered a specific discussion with my accountant about the legal limits of donations (France).

It can also be straightforward to get individuals in to give a talk, or training, on what a bit of software can do. I've done that in the past where paying directly for the software or service itself isn't an option for whatever reason.

I third this.

An invoice is enormously easier than making a donation. Most large scale organisations give even low level managers an authority limit (up to X amount) where they don't need to engage the full Contracts and Procurement process. If you're asking for less than $1k per annum, you'll find a lot of large scale organisations will happily pay. And the people using these tools are often very supportive of open source, and want to support these projects. It's in their interests; you maintain the library, they get the benefits of having bugs fixed.

You also make Caddy, right? I've been impressed by how you encourage donations to that project. The copy on the caddy site is totally different from nearly all other OSS that asks for donations, very well done.

Does it work? Do Caddy donations and support contracts cover your bills?

Thanks. We recently moved away from a donation model, and we now seek sponsorships and subscribers: https://caddyserver.com/pricing

Donations never did cover more than a few meals per month. That was nice, but sponsorships and the Engineering Package allow me (and my business partner) to dedicate more time on development and invest back into the community for the project's growth. Still not at the break-even point yet, but I'm hoping that won't take too long...

> We recently moved away from a donation model, and we now seek sponsorships and subscribers

Ahyeah, so the sponsorship part is what I meant by "better copy for getting donations" :-) Hope you'll go break-even soon!

I wonder if a site that sells minimal support for open source projects as a broker / support tracking site passing the majority of the money to the developer/ project would work for companies.

I use papaparse, it's been pretty awesome, so that thank you for making it.

Have you considered offering support contracts, or a dual licensing approach? I don't mean to butt in and offer a bunch of unsolicited advice, just more curious because I might be trying to do similar stuff with some of my projects, so always on the lookout to see what works and what doesn't.

Thanks again for making papaparse it's really fast and has really helped me get some projects off the ground a lot faster than I would have otherwise!

Hey, thanks! I'm glad you like it.

I did consider a dual license, however I don't want to fragment the project, given that it's pretty much "done" and someone would just use a fork from before the license changes anyway. As for offering paid support, I could definitely still fall back on that if I need some extra income, but it takes time. So right now I'm leaning on my graduate research.

One method I have found somewhat successful (but I am still flushing out) through another project, Caddy, is selling sponsorships. They give me some flexibility to put more time into the project to benefit everyone, rather than a single customer.

Monero the crypto currency does this. I've seen them fund every thing from Ph.D. Research and paper writing, to application development and bug fixing. It seems to work well for those who really believe in and use a project. I'm surprised more people don't try this model. Release the core code and let users vote and fund which bugs to fix and which features to add. I find it an interesting and refreshing change from the free for all of whiny depressing masses who demand all sorts of crazy, hard work from OS developers without paying dime one for it. Maybe it will be the model of the future, who knows!

I truly yearn for the day when I can pay rent with my GitHub street cred.

Makes me think, there was that old Y Combinator experiment of letting HN posters vote for funding in a small funding round. Combine that notion with a karma-weighted transparent proposal voting fund ("DAO" style) and something interesting might happen?

Is it incredibly popular because it is uniquely useful or primarily because it was open source under the MIT license? To put it differently, would people have started using it if you started with a closed source solution, or would someone else have emerged with a better open source alternative?

It's perhaps unfortunate, but such a library does indeed become very popular for the very reason that it is open source and free. I haven't used this, but if I ever needed in-browser CSV parsing, I'd use it. Were it not free, I'd find an alternative or roll my own library with the bare minimum functionality needed for my project. Paying for anything related to CSV just doesn't "compute", for lack of a better phrasing.

This applies to things like date pickers and html5 charts as well. While these could be separated into a separate category based on the fact that their markets are oversaturated, the result is the same: why pay when there are perfectly usable options that are free, or can be done in-house on the cheap if necessary?

I'd like to think it's both. The license definitely helped drive up adoption very quickly, but it still does some things that no other library does. Still, I'm sure it's just a matter of time before <insert any proprietary software here> has an open-source alternative. (Not that that's bad or a defeat -- the proprietary stuff can still be successful -- but it is a reality.)

You can still monetize it right? Look at what happened so far as a market validation and build a "better" paid variation of it that includes support and maintenance. Wouldn't that be possible?

Could learn from the Pharma industry where they use techniques like changing the formulation or introducing a new drug administration route to "renew" the patent.

Instead of debating it, if it were me, I would just give a people a reason to pay and see if it works.

You could have the same product presented as: Lil Papa "recommend for individual use" - FREE/DONATE Fat Papa "recommended for business use" - 1 MILLION/month (the same thing but we send you one emoji a month and give you a proper invoice)

Looks like you already have people trying to pay you and they could easily do it if there was invoiced paid option.

If folks can make money off of "Laundry Balls" there is no reason you shouldn't off of work like papaparse.

I'm a huge fan of the two-clause BSD license, but I can also understand that regret. I've always wondered how you would monetize something like this, though. It always seems difficult to me how you ensure that you're being paid without an army of lawyers (the Microsoft model) or limiting it to some sort of app store (the Apple / Google model). What would be the best way to do it, if you had it to do over again?

Given that it's client-side JS, I'd probably just use an honor system. I'm not really a licensing law guy, but: Give it a license that businesses would not like to use (GPL?), then sell proprietary licenses to waive some of the requirements of GPL.

But again he or she above you doesn't really want to enforce that. Enforcement of the GPL is a big problem. Who has the time and energy and money to sit around looking for violators? And then if you do find them you have to lawyer up and that costs a lot of money. You dont have any time or money to write code. The reality is you either accept people are going to use the code you OS and just let that code go (BSDL) or you keep it closed source. I like the idea I just posted above of a model of releasing the core code but letting actual users vote for and fund what bugs and new features get added to the free core code.

If the options are 1) permissive open source license, zero income, or 2) dual license with minimal additional effort and get some fraction of users paying you, then do you really need to care about actively enforcing the license?

Worrying about enforcement is like not opening a store because you are concerned about chasing down shoplifters.

Closed source also need enforcement. What is she/he going do if companies download it from a warez site? Who has time, energy and money to spend on lawyers to hunt down those people, but do not have the same time, energy and money if the license conditions happens to be GPL?

> What would be the best way to do it

Look at highcharts, you can use it commercially without anyone noticing it because its all client side JS but they are doing fine in terms of revenue because enough people pay for the software/support

Great home page, I've never seen that conversational readme style executed so well.