Hacker News new | comments | show | ask | jobs | submit login
On paying for software (seths.blog)
186 points by gbugniot 15 days ago | hide | past | web | favorite | 274 comments



One of the reasons that I switched from Linx to OSX was so that I could pay for more of my software. Why? Because then I more of the software I used could be maintained by someone who had the time to dig into bugs and UI problems and to fix them. But in Linux, couldn't I just edit the source myself? Realistically, no. It takes a tremendous amount of effort to source-dive in a totally new project in a language I never use, especially without someone willing to give me a walkthrough of the architecture and fundamental models of the program. It is waaaay more efficient for these to be fixed by an engineer working not in their spare time, but as their full-time job.

A developer targeting OSX knows they have an audience of people willing to pay him money so he can spend his whole day doing usability tests and his evening watching a little league game.

I do miss strace though.


This is actually completely backwards in an ironic way. I have found in the past that when I made a program closed source, for sale, all further incentive to do anything to the program became connected to its revenue. Like, "it hasn't generated any meaningful revenue, so I cannot justify spending another N hours on it unless those hours lead to sales."

Most software that you can buy has little or no revenue, which means you will be one of a scarce set of suckers paying for it. That meagre revenue doesn't pay for the developer resources to improve the program.

Paying for software is like an insurance premium against bugs and issues in that program.

Paying for some unpopular software with few users is like buying insurance from an insurance company that insures only a handful of other policy holders.


Alfie Kohn’s Punished by Rewards provided hundreds of sources, tested and re-tested proof, and description of experiments proving over and over again that for humans, extrinsic rewards kill intrinsic motivation.

If you want your child who loves drawing to stop loving it, reward them. If you want them to stop learning, tie a reward to their learning.

It’s very simple, really.

Being able to receive sustainable income for doing what you love doing, is a different formula though. Subtly, but different.


I haven't read the material in a while but I think this was sensitive to framing. That is, I don't think it's exactly true that attaching a reward to something harms performance — I think one has to frame the task as being done for the purpose of the reward. If I am remembering this correctly, then emphasizing someone's intrinsic motivation while also providing an extrinsic reward should be ok.

If you happen to remember the material well enough to clarify on this point I would appreciate it :)


If the reward is conditional on performing the task, it triggers the loss of intrinsic motivation.

It is basically a way to hijack the intrinsic reward mechanism and externalize it.

I imagine if you provide support for he activity, regardless of outcome, and regardless of it occurring or not, then the intrinsic motivation may remain in place.

I am not speaking from memory though; I myself would have to look at the book.


> If you want your child who loves drawing to stop loving it, reward them. If you want them to stop learning, tie a reward to their learning.

Is there a way to achieve the opposite?

> Being able to receive sustainable income for doing what you love doing, is a different formula though.

What is the difference?


To “achieve the opposite”: give up on your agenda for their relationship with drawing, or whatever interest of theirs you want to promote.

Support it when requested, but give support only when requested.

Think of your child as being driven by their enthusiasm and interest. They don’t control these forces and you do t either. Trying to force them kills them. Trying to direct them suffocates them.

So, be there, celebrate what they celebrate, lend an year to their doubts, feel free to offer a bit of support but never insist on it. Watch it grow and maintain the soil; that’s all you can really do.

Compliments are also rewards by the way ;)

I highly recommend taking a look at the book, it’s in most libraries.


So, don't buy hours or drawings, but do buy pencils, paper, brushes and canvases, and show an interest, but don't say OMG what a gorgeous drawing every single time.

Exactly.

It’s their relationship, their art, their journey. Don’t try to become a main character in it. It’s OK to be a benefactor, but don’t try to amplify it or change its course.


I use Linux and most free software projects allow donations. I even donated to projects I don't actually use, like FreeBSD. I donated money to many free software projects, Ardour, Mozilla, LibreOffice, OpenBSD, etc. This is clearly a false dichotomy.


Relying on the good will of others does not work, period. The fact that it works sometimes is not something that one can use as the reason to quit their job and start an "endeavor" (you can't really call it a business).

Can we please dispense with this notion and start realizing one simple fact: people will typically only pay for something if you make them pay.


> Can we please dispense with this notion and start realizing one simple fact: people will typically only pay for something if you make them pay.

Things to appear to be changing for the better though. As an examply, Gina Häußge is working full time on OctoPrint (3D printing software) supported entirely by donations:

https://www.patreon.com/foosel

There are probably others too, I'm just not aware of them and have only really started looking at Patreon recently (for a project I'm involved with). :)


Again, donations, etc. may work for some creators, but they will not work for all. There are always going to be products that are "Wow, I'm donating to her !", but the vast majority are more of the "plumbing" variety that are much, much harder to monetize in ways other than "pay me".


You seem to forgot that most independent developers don't make money, period. The fact that your app is open source or not is only marginally related to the fact that it earns you a living. Thousands upon thousands developed applications for desktop or mobile or whatever and simply cant' make end meet. Believing that because your app has a sticker price on it will make you money is an illusion.

I hire some dev interns every year. For the past 5 or 6 years, they all were using Sublime Text as their main editor. And none of them paid it, they simply clicked the nagging dialog away. To all of them I had to explain that it would be fairer either to use a free alternative like Atom or VSCode, or pay the fee (or learn to use a real editor like Vim or Emacs :). People go through incredible pains not to pay, even if your program isn't free. Even if they're programmers, and spend all day using your program.


No, I didn't forget or ignore the fact that a lot of software products fail to make money. I went through several failed software product attempts before building one that stuck, and it's part of the process of getting to know what one's forte is, as well as getting good at running a software company and making money from it.

But, I would seriously argue the proposition that open sourcing one's software does not harm your ability to make money from it. Exhibit A is how often you see a mass of "They want money ? We're switching/forking !!!" type of posts and comments whenever the inevitable "we need to start making money from this project" point hits with an open source project.

As for your example, you're kind of making my point for me. ;-) It's a serious issue, and one that I'm trying to make as much as possible: as software developers, we're seriously harming our own economic well-being as a group by being overly-altruistic and conditioning (what would normally be paying) customers into thinking that our work is without value. It doesn't matter what the reality is, but free=no value in the eyes of consumers.


Any software I have used and found to be a tool I've relied on, I have purchased a licence for at some point in time. Granted this is when I've been comfortable financially. If it comes down to eating or paying for a software license is obvious. But to keep being a tight ass after you're more than able is just crappy. Especially since as a dev you should be able to relate to those dedicated to their craft & realize they too could be struggling financially. I do find smaller orgs more enticing than say a Photoshop license as an example, of the pricing is high for using on occasions. That said being a solely design or front end type that's something you'd use and rely on to make money on. It's really that case (for me at least) if it's used for profit it should be profitable for the creators. However I've also been a musician too who believes strongly on the opposite. Music should be free to individuals and paid for up corporations. The same model is great for open source like redhat (centos) or Magento that offer enterprise license that mean dedicated support. My point being as an individual it's hard to justify at times, but for organizations it should be obvious to licensed.

Having paid for sublime and PHP storm, both have vim modes. But I prefer nano (yes I'm one of those lol ;-) )


Sure. But the original statement was that it does not work. eg 0% success

Whereas clearly it can, as there are the occasional success stories. So, it's more than 0%. Maybe not a high percentage yet though. ;)


Yes, but that statement was followed by more details as to exactly what I meant. A model that only works for a small percentage of people does not work.

My point is that you can't start a business based upon some vague hope that someday, somehow, someone will possibly choose to donate money to you. You need a reliable stream of income, and one that will keep coming long after the shine has worn off the original product and now you're slogging through complicated maintenance and upgrades.

You want to know why the JS ecosystem keeps suffering from the "shiny penny" problem ? This is one of the reasons.


>>My point is that you can't start a business based upon some vague hope that someday, somehow, someone will possibly choose to donate money to you. You need a reliable stream of income, and one that will keep coming long after the shine has worn off the original product and now you're slogging through complicated maintenance and upgrades

What part of that doesn't apply equally proprietary software too?


I see the biggest difference being that with commercial software you know when to stop and move on to something else.


Ahhh. Yeah, now I understand what you mean.

For an already successful OSS project though, it might be a way to turn that into an ongoing income stream to then continue improving the software.


Gina Häußge is the only one making reasonable money on Patron for a software project, last I checked. I exchanged several emails with her about this very topic. Despite her success in fund raising, even she was concerned about the long-term viability of this sort of funding.

I also am the author of an active Open-Source project. I accept donations but they have never amounted to much.


There do seem to be others. The author of Vue.js seems to be doing ok as well:

https://www.patreon.com/evanyou/overview

And yep, I definitely agree. The long term sustainability of projects via Patreon clearly hasn't been established yet.

But, at least some _are+ actually now making decent money, so we can find out how sustainable things are over time.

Hopefully it mostly goes well, but yeah... it's unproven thus far. ;)


Mastodon's main dev too: https://www.patreon.com/mastodon


Sindre Sorhus has a successful Patreon as well:

https://www.patreon.com/sindresorhus

but he's kind of a unicorn (pun intended). I am sure this would not work for everyone.


And $3500 per month is not what a lot (most?) people here would consider a successful financial model even given some consulting etc. on top of that. It depends on where you live of course but bringing in $50K per year (with no benefits) or so in donations is not something a lot of professionals--especially those qualified to work for major software companies--would consider a sustainable income stream.


He looks to be located in Bangkok:

https://github.com/sindresorhus/ama/issues/414#issuecomment-...

The creator of Vue.js (in sibling comment) has a Patreon account with a monthly income of $US15k+ (at time of writing this).

That seems to be more in line with "standard" professional rates. :)


The recurring online payments/donations economy is just starting out. Give it some time to pick up steam and get a better UI, I believe it will become more popular.


Just noticed you're the primary author of CAMotics.

I'd heard of it before, but haven't yet tried it out. I'll try to get time to do so, as I've recently started helping out a different CNC project (github.com/synthetos/g2).

They kind of look like they might complement each other. Not 100% sure yet though. ;)


Yes, I'm friends with the Syntheos guys. It can work with CAMotics. You might also be interested in another project I work on, the Buildbitics Open-Source CNC controller.

Thanks. :)

Looking through the Buildbotics stuff now and some of the video's. Looks nifty!

Is the motion control done by TinyG2 or g2core? If so, would it be ok to add some appropriate links to our "Who uses g2core?" page?

https://github.com/synthetos/g2/wiki/g2core-in-use


Ahhh, looks like something custom instead:

https://github.com/buildbotics/bbctrl-firmware

No worries at all. :)


We did start out using the TinyG firmware but have since replaced it with our own code.

Thibault Duplessis also works full time on lichess.org only with donations.


I have been really surprised with how often she posts Patreon updates! It makes sense now as I didn’t realize she was full time.


In this house I have 2 tablets 2 phones 1 router 1 laptop 2 desktops all running linux it appears to be working for me. This isn't to say that more money for creators wouldn't benefit everyone.


The problem with open source is, for some reason, the developers feel no sense of obligation to the end users or the people who donate to them. MacOS devs do and will go out of their way to make sure their software is not just usable, but a sheer pleasure to use in every way possible. Yes, there are exceptions (I'm looking at you, Apple, with your Xcode and your Finder) but for the most part that's just the way things are in the MacOS app ecosystem.

Open source, on the other hand, is like trying to shave with a razor blade with no handle. It's a hard sell to get people to donate to projects that they know are going to cut them at some point.


> the developers feel no sense of obligation to the end users or the people who donate to them.

An open source project creates no obligation on the part of the developers. Open source licenses explicitly indemnify the authors. Receiving something for free and expecting something more is the essence of the entitled behavior that frustrate and burn out many open source developers.

A donation is a voluntary gift. Like any other gift, a donation to an open source project is understood to have no strings tied to it. Giving a dollar a month to a project doesn't entitle you to round-the-clock support. If you expect something more, have a conversation with the developers to work out a plan.

P.S.: we (https://sheetjs.com/) build and maintain open source projects (our largest is https://github.com/sheetjs/js-xlsx) and we turn down donations precisely because many people expect something in return. Instead, we offer actual paid support plans and additional features.


> An open source project creates no obligation on the part of the developers.

I meant a subconscious obligation, not a legal one. When somebody gives you a gift, it puts pressure on you (subconscious and social) to reciprocate. You see it all the time with youtube streamers, who frequently cater their content to their subscribers wishes to keep them happy. And to clarify, the streamers don't do this because their users are unhappy or protesting. They do it because they feel a sense of obligation to give their subscribers what they want because the subscribers were generous enough to give the streamer money.

For whatever reason, this phenomenon is completely lacking in the open source community. Instead, the opposite seems to occur. OSS developers are actively hostile toward their users, both vocally and how they react to criticism that their software isn't user friendly or respectful of their user's time. Anybody who has spent hours trying to set up a piece of OSS and then made the mistake of asking the developer for help can attest to this.


You're asking the developers to do unpaid tech support for you, while insulting their work. You seem hostile to them, it's no surprise they aren't nice to you.

Most moderate sized projects have at least decent documentations, some channel for support, and there are a range of third party options to consult. The developer generally should not be who you contact.


> For whatever reason, this phenomenon is completely lacking in the open source community. Instead, the opposite seems to occur. OSS developers are actively hostile toward their users, both vocally and how they react to criticism that their software isn't user friendly or respectful of their user's time. Anybody who has spent hours trying to set up a piece of OSS and then made the mistake of asking the developer for help can attest to this.

Not my experience at all. Well sometimes bugs gets closed for no apparent reason etc but I've rarely been shouted at and I think more often than not they try to help.


Your response is the customer relations equivalent to "it works on my machine." Ignoring the problem isn't going to make it go away.


Well; if the difference between my machine and your machine persists in the long run it certainly might be interesting to find the differences that makes it work on my machine.

And - if this approach that you are showing here is representative for how you communicate it might explain a bit ;-)

Edit: and yes, my approach has worked for years so either it is something I do that you don't or I'm just consistently lucky or something :-)


Typical paid proprietary licenses also explicitly indemnify the authors.


A 'donate' button isn't a business model. Some open source projects have a business model built around paid support.

By all means please donate to projects you want to support. But 'donation' implies you aren't getting something in return. Otherwise it would be a 'purchase'.


I have received numerous donations in the past. They were never accompanied by any specific request to fix something or make a feature. So even if I wanted to do something in the software out of a sense of obligation to those users, I would not have known what that is.

Realistically, a $100 donation doesn't get much of my time anyway.

It's more of a pat on the back; people like what you're doing, and the way you're doing it; so keep going. (Don't do anything differently, just more of it.)


I actually find open source developers to be far more passionate about their projects than developers of commercial software. This makes sense to me, as they work on these projects because they want to, not because they have to.


Passion isn't a barometer of anything but whether the developer likes his job. It certainly doesn't translate into better or more user friendly software.


In OSS I tend to see the opposite. The more “passionate” developers tend to be passionate about some implementation detail (like respect for a particular standard, the Unix philosophy, Stallman-compliant licensing, etc) without giving a crap about the actual user experience in day to day usage.

I see more passion in paid software in macOS when I see a beautiful UI because it shows that the developer actually wants me to use and enjoy his creation day to day.


It is more that well run project has testers and analyst and manager that will open tickets for what you are not passionate about and missed. So, you have to fix it whether passionate or not, it is more about result then whether you feel like doing it.

Then, there are also badly run commetial projects, which are crap despite being paid for.


Passionate about making the installation process seamless on every distro? And all the other mundane, tedious tasks that are expected by users used to commercial software?

My OSS project, which I haven’t worked on in ages to be fair, I never made any effort to make it work anywhere but on Debian for example. I know some people used it in Suse and RHEL so it can’t have been too painful but I never had that as a goal.


This is the other issue with 'linux on the desktop'. The linux desktop is massively fragmented. Notably this goes for desktop environments and package managers.

This makes it hard to target 'linux' in general with polished software. The issue of package managers seems on its way to being solved by containers. However, the issue of desktop environments is much more thorny. As you can't really abstract away the metaphors with which a user communicates with his PC.


Depends on the individual I guess. :)

Years ago, I went through a phase getting the installer working well (cross platform) for a PostgreSQL replication project. FreeBSD, NetBSD, Solaris, RHEL, ... there might have been others too, it was a long time ago. ;)


The problem with passion that it may not be focused to what the end users might most need.

Think about OSS software you've written and contributed to. 99% of time it is to scratch your own programmer centric itch.

Usability usually takes a backseat in OSS model as not many usability experts are working for free.

Same goes for documentation. It is autodoc or nothing.


> The problem with open source is, for some reason, the developers feel no sense of obligation to the end users or the people who donate to them.

I always go out of my way to help users who showed gratitude to me, either with a donation (those are very rare) or by being extra nice to me last time I helped them.

But obviously, users who insult my software and myself every time they have an issue come last on my list of priorities.


> The problem with open source is, for some reason, the developers feel no sense of obligation to the end users

This is as much true as it is for developers of proprietary software.

There are many examples on both sides that make your statement false, also as many to make it true.


"Open source, on the other hand, is like trying to shave with a razor blade with no handle. It's a hard sell to get people to donate to projects that they know are going to cut them at some point."

This sounds whiny and more importantly it isn't true its hard to get people to pay for things that they can have for free full stop.

Don't make things more complicated than they are.


I would say the blade analogy is fairly correct.

OSS isn't polished many times. The installer might be missing. Documentation is out of date or incomplete. The plug-in manager requires using the file system explorer directly.

All of that cuts. OSS can be death by a thousand cuts at times.


> This sounds whiny

Are you an OSS developer by chance?

> its hard to get people to pay for things that they can have for free full stop.

The Apple third party app ecosystem proves you wrong. Users are willing to pay for good software, even if there are free alternatives available.


You failed to prove me wrong or in fact to understand the point I was making.

People are disinclined to fund open source projects that they can get for free.

If you rely on voluntary donations you will find yourself poor.

They may be willing to pay for value added only if getting something for nothing isn't an option.


> You failed to prove me wrong or in fact to understand the point I was making.

Then communicate your point better. Losing the hostility would be a great start.


Speaking of exceptions to this...they really need to fix their damn multi monitor support. It’s one of the few things Windows is far better at. I get that a small number of Mac users are even affected by this, but man would I appreciate that a lot more than Dark Mode. And I love Dark Mode.


What issues have you experienced with their multimonitor support?


It takes a long time to wake from sleep, sometimes over 30 seconds with monitors flickering on and off. Which monitor displays which desktop will change after unplugging and plugging back in, or even after sleeping and waking back up.


I've had issues on this with windows when using DisplayPort Daisy chaining. Now I've switched to using separate cables.

The point being that anecdotally, windows doesn't have it down either.


I believe part of my issue might be that I'm using 1 HDMI port and another mDP to HDMI converter. From what I've read it seems like people tend to have these problems when using converters. So I guess if I were running directly off of mDP I'd probably be having a better time.


turn spaces on


The things that people choose to pay money for, in software, drive an incentive system that results in polished user experiences.

The things that people choose to donate for, in software, drive an incentive system that results in ever-increasing code quality and security and a bunch of other things that developers themselves care about, and feel that users should care about, but which don't actually affect application polish very much.

It's not even down to the companies choosing to do this or that with your money. It's a response to incentives. They know that, under a paid model, they'll keep getting money only if they do X; and under a donation model, they'll keep getting money if only they do Y. And so they prioritize X, or they prioritize Y, regardless of which one they think should be prioritized.

If you want X from your software, you should advocate for an ecosystem of paid software businesses. If you want Y from your software, you should advocate for an ecosystem of donation-supported software organizations.

If you want both... I'm not sure. Maybe support both ecosystems, and encourage the donation-driven organizations to put their work into "cores" or "engines" that can be wrapped into polished experiences by the paid companies?


Try to pay for a mortgage with donations.


99% of paid apps developers can't make end meet anyway. Another false dichotomy. Creating an app and sticking a price on it isn't a business model any more than creating an app and asking for donations. Both are almost certain failure. But people constantly hear about this guy or that company that made some huge success and became rich; nobody talks about the millions who failed and whose apps were simply forgotten.

Oh, and BTW, the Red Cross, Doctors without Borders, Greenpeace and many other NGOs live, AFAIK, mostly on donations. So that definitely can pay the bills, but the same reasoning applies: it's only true for a small minority, and you can't count on it.


> Oh, and BTW, the Red Cross, Doctors without Borders, Greenpeace and many other NGOs live, AFAIK, mostly on donations.

They do, but they also get sponsoring from many governments across the globe, do continuous street actions trying to get people to donate for them´, and after a certain age most people end up leaving to some positions with more stability.


These are already supported by corporations. They pay millions vs your handful of dollars.


> It is waaaay more efficient for these to be fixed by an engineer working not in their spare time, but as their full-time job.

You miss the possibility of simply offering to pay a developer with the relevant expertise to fix one of these bugs for you.

I can tell you from experience that there are FLOSS developers out there with sufficient expertise and professionalism to do this.


If I'm running an enterprise, perhaps then I can budget enough time and money to communicate about and pay for that and deal with the uncertainty inherent in software schedules.

In my personal life...that sounds like a stressor that I'd be willing to pay money to avoid.


Sure but this sounds like an opportunity for something like “Uber for FLOSS support.” Or perhaps it would be more like OkCupid.

But you could pay money and your FLOSS problems get researched and solved for you by a developer who is experienced in this area.

In practice I can see this as a forum where people make minimal effort posts and skilled people give them quality support. Just as I was forming the previous sentence I realize Stack Overflow and Reddit kind of do this. Maybe it would just be nice if people found a good way to add payment to the process.

Anyway, it’s not as nice as not having the bugs in the first place, but FLOSS has other advantages. There will continue to be people who choose commercially supported software, but the more we can close the gap on quality with FLOSS the better.


This has existed for a while: https://www.bountysource.com/

It hasn't been a tremendous success story so far - but these things take time.


That's different. I'm talking about participating as a user in a FLOSS mailing list, getting to know the people involved, then asking one of the devs if they can fix a bug or add a feature for you in exchange for money.


The issue is that even if he managed to find a good developer to pay to improve some FLOSS software, it would come out very expensive in the end, first because the UI frameworks & libraries you get in the Linux world are nowhere near what you get on macOS, so the developer will require more time, and also because you won’t be selling that software later on to make any kind of money back since it’s FLOSS, and even when it’s proprietary and not FLOSS, the potential market for Linux software (let alone paid Linux software) is very small.


Hmmm, that seems to forget that there's a large amount of Open Source software available on macOS too. :)

With the UI frameworks and libraries, what's your take on things like Qt then, which is cross platform and pretty decent for developers? At least the C++ ones anyway. :)


Yes but keep in mind people who work on FLOSS are interested in it for more than money, so the person being paid may be willing to work for a little less (or invest more time) knowing that the work that they do is helping FLOSS. So both people can share the cost burden here. We are talking about two people passionate about FLOSS that one is using it despite the problems (and willing to pay) and one has taken it upon themselves to learn how to fix it.


I could be wrong here, but as I recall, there's a psychological effect where an offer of money can reduce the value a developer assigns to their own at-will effort because it shifts the paradigm from noble to economic ends.


> In my personal life...that sounds like a stressor that I'd be willing to pay money to avoid.

Exactly, and a single citizen typically doesn't have the means to commission a software project, or even non-trivial fixes to an existing software project.


Expect 500 euros per day for a cheap developer in Eastern Europe. Maybe similar for Indian outsourcing or Latin American.

Expect over a 1000 dollars per day for a qualified developer in a first world location.

Point being. No individual can pay that much and the develop should rather get a day job unless the software has high earning potential.


Which developer in a first world country has 30k a month in salary? $1000 a day is beyond ridiculous!


For a qualified contractor? $100 per billable hour is extremely reasonable. They're not on the clock all the time. When I did consulting work (not writing software but tech-related) I charged way more than that although it was usually based on jobs rather than hours.

ADDED: And, if you're contracting through a larger firm, they're adding various overheads associated with managing the project, etc.


They are standard rates for contracting and outsourcing firms.

500 euro a day is about 100k a year, accounting for 200 work days in a year.

Remove the margin of the firm, pension, healthcare, sick leave, company taxes, office costs, benefits... and what's left is the gross salary.


> No individual can pay that much and the develop should rather get a day job

You could create a cooperative of users that hires one or two developers to work for all of them.


this is probably not too dissimilar from starting a business that grows a collection of users willing to pay fees to receive custom patches / support for the open source software


What in the... You could donate to the FSF, OSI, or directly to the developers of most projects. Did you even bother looking?


If you donate, there's no guarantee that everyone else is donating. So you end up supporting the free riders (https://en.wikipedia.org/wiki/Free-rider_problem). Which is a disincentive to donate.


I conducted dozens of software evaluation and procurement rounds for a ${BIGCORP}

One of the factors that made FOSS much more promotable was a sticker-price for acquisition and / or support. The actual amount didn't matter, psychologically though it needed to show that it had 'worth'. Ifnitnwas free, the business was worried thatvit was just someone's weekend hack. If it cost $5 per user then it was great value.

Not once did anyone raise the free-rider problem when we paid for FOSS. Frankly businesses are so focused on getting stuff done internally that the fact that someone else might indirectly benefit is irrelevant.


That problem seems to me like it isn't telling a whole story. There have always been free riders of many kinds in society. But society isn't built around a single variable. People who get a free ride in some way often are supporting other free riders in other enterprises.

Basically if you are strict about trying to eliminate all free riders, you're going to break society in a vain effort to maximize fairness in lieu of effectiveness. Society works because everyone assumes that even if some things are unfair, most will manage and the ups and downs will average out over time.

In the example given on the wiki page, people who swoop in and take the last minutes left on a parking meter that someone else paid for but didn't stay long enough to use are supposedly able to break the system. But I think that trying to lock the system down tight and completely prevent that behavior could feasibly be more damaging to the health of the system than simply tolerating it as an occasional but mostly harmless exploit.

There's a few studies which show that some piracy and other free usage might actually help an industry's profits. Key word is "some"--there is obviously a threshold past which would become very damaging. But nonetheless, if one can set aside their outrage over unfairness (which is very understandable for somebody who has poured their life and soul into working on an IP or paid a lot to license it legitimately) then one can probably stand to benefit even from the thieves and freeloaders if they manage it correctly. https://www.fool.com/investing/general/2014/06/13/does-onlin...


No, you support the person you donate to. That the "free-riders" are supported too is only a secondary thing. The main aim of FOSS is to share intellectual effort with the world anyways.


I am well aware I can do that. But I don't have enough money to ensure that there is enough donation to ensure that there is enough support for those working on the projects to focus on it full time. The systems-level change of allowing developers to use copyright to require people to pay for their work comes much closer to ensuring that.

> "did you even bother looking?"

Also, when someone is building a product for money, it incentivizes them to proactively look for feedback on the quality of their product and to value the feedback of frustrated users. When someone is building a product without a business goal focused on pleasing someone else, they are far more likely to be either passive in waiting for people to report things on mailing lists or to be actively annoyed with the reports and to occasionally express that frustration in a way that belittles users.


> When someone is building a product without a business goal focused on pleasing someone else, they are far more likely to be either passive in waiting for people to report things on mailing lists or to be actively annoyed with the reports and to occasionally express that frustration in a way that belittles users.

This seems (to me) like a cultural thing. In this instance, by "cultural" I'm meaning it depends on the culture of the Open Source project.

Many projects do seem to have developers that aren't really hmm... "user friendly" might be correct description. But not ever project is like that.

The PostgreSQL, SQLite, and DB Browser for SQLite Communities are all pretty user friendly.

As in, it's pretty rare to see a display of bad attitude towards someone reporting a bug, asking a simple question, or doing pretty much anything else in good faith.


With the exception of bounties, donations do not translate to fixing an obscure big or adding a feature. I’ve even had some features which I’ve implemented fail to make it into the upstream because it’s not the way the maintainer(s) would have implemented it, despite it being highly requested and following their development patterns as much as possible.

The attitude in the opensource community about pull requests and donations is sometimes frustrating when you realize that a contribution you’ve made may never benefit the community.


A donation doesn’t reasonably entitle you to expect anything. A purchase does


Does it? In either case the code is already written. Your payment doesn't mean there will be new features.

For many businesses, and at universities (where I work), there isn't an easy way to donate, particularly to random people.

However, it would be easy for me to spend, say £50-£100, for a support licence for software I frequently use.

I would love it if Microsoft added this to GitHub -- opt in of course.


This logic is flawed. FOSS software doesn't need to equal software that no one pays for. You want to support FOSS development? There are tons of projects which take financial support, feel free to contribute. You can also support companies which directly work on developing FOSS.


This only makes sense for software that you use frequently and don't mind repurchasing everytime Apple breaks them with OSX upgrades.


As an indie dev of desktop software, I have grown tired of complaints that my products aren't free as in speech and beer. I now reply to such criticism by linking to this post: https://fman.io/blog/dear-comrade/


Great blog post - thanks for writing it. I've had the same experience with people asking for an MIT license on dependabot-core (the public repo that holds the core logic for my SaaS business, but which has no license and a comment saying those who wish to use it for commercial gain should contact me instead). I already give the product away for free to open source projects and anyone using it in a personal capacity, but for a small number of people that's not enough.

Free software is a wonderful thing - it's fantastic that our community shares its work so widely and generously. It is not, however, the only, or the best, model for creating innovation at all times. I'm able to work full time on Dependabot because by providing it as SaaS I can pay myself a wage and by owning the codebase I can protect myself from a larger player copying my work. There's no doubt in my mind that enabling me to work on the product full time is a better outcome for the world than a license that would let others distribute and profit from it.


As a general point, please put a license on your contributions if you put them on github. Don't much care what it is, and I don't mind paying, but don't make me do a song and dance to see what your price is.

Especially if you are already giving it away to open source projects -- my bet is most people who see "call me" as a clause will move on, even people you are actually trilled to help.

AGPLv3 would be a decent suggestion, but even your own license wouldn't be bad either.


Totally agree on licenses and transparent pricing in general, but I think you've missed the context here.

If a commercial entity wants to add the core functionality of my product to their offering then there's definitely going to have to be a negotiation first. AGPLv3 wouldn't be appropriate, as the commercial entity's code being open source isn't a material consideration for me.

(The license text I was referring to is here: https://github.com/dependabot/dependabot-core#license. I could hire a lawyer and put a proper, proprietary license on the repo but I'm bootstrapped and don't think it's necessary.)


> by owning the codebase I can protect myself from a larger player copying my work

I find this statement confusing. In what sense do you not "own" your codebase if you release it under the GPL? And if the software is released under the GPL, large players have no advantage, because they have to release all of the changes they make to your code if they're using it to compete against you.

What you're saying is that restricting and hiding your codebase allows you to make more money. And there's nothing wrong with that, but it's not right to spread falsehoods about free software licenses.


That's not what I'm saying, and I don't think I'm spreading any falsehoods about free software licenses here.

Building a business is about distribution as well as product. My product is pretty good, but my distribution, compared to a large player with an existing user base, is woeful. If I licensed under GPL I would be giving explicit legal permission to any big player (who was open source) to take my product for free, and compete with me solely on distribution.

Licensing my product in the above way wouldn't be good for me financially, and as a result would decrease my likelihood of working full time on Dependabot. Worse still, it would disproportionally decrease my returns from working on the product, as opposed to distribution.

Thanks to copyright law we don't need to hide our codebases in order to charge for our work. That's a great thing, and there should be acceptance of code that is developed in public but without a permissive license. It's a perfectly rational decision, and creators shouldn't be put under pressure to give their development work away for free instead.


   Thanks to copyright law we don't need to hide our codebases in order to charge for our work. 
I don't understand that part ?


Before patents, copyright law, and all that, secrecy was the only way to stop people copying your stuff. That was kinda crappy, and people went to great lengths to obfuscate their IP, expending lots of effort on something that had no benefit for their customers / the world.

Intellectual property law changed that - it made it possible for people to openly disclose their inventions whilst maintaining control of them (normally with some conditions). It was a really, really good thing for society, as it dramatically increased the returns to "thinking work".

I can probably dig out a link if you're interested.


How is copyright / patents / IP making companies' code bases open to public or it's secrecy is not as guarded as before (as implied by your quote) ?


Especially since copyright predates software or you know computers.


Citation needed.


   In what sense do you not "own" your codebase if you release it under the GPL? 
There is no concept of property for information. By hiding your source codebase and the way it is being built (the same logic applies to your DAW music project, your video software editing file project etc...) you are making life harder for people who try to replicate features of your software, understand techniques behind it or pretend to be you to sell the software instead of you (others examples are welcome).

So I can understand that the GP /feels/ like he owns it more, even if it is just a feeling.


There is very little in the world of software that's hard to replicate. And there's almost nothing that does not have any F/OSS alternatives.


But still "harder", it puts a heavier barrier of entry for others to enter the same area as you. The fact that there are FLOSS alternatives has nothing to do with what we're talking about


I very much agree, especially on your point that the world is better for you being able to work on your product.


> We have been unable to find a license that accurately fits Dependabot's needs (suggestions are welcome)

How about dual-licensing as CC-BY-ND (using the bot as-is, even for profit) and CC-BY-NC (making modifications, but not for profit)?


I don’t understand when people complain about software that isn’t free. You want it free? Go make it yourself and give it away...oh, you can’t make it yourself? Maybe that’s why it’s not free.


> oh, you can’t make it yourself? Maybe that’s why it’s not free.

That's not a complete argument. Anyone can mop the streets, but we still pay for their services.


Until someone actually do it, and better. Good luck monetize your proprietary alternative then!


You're perhaps forgetting that people use price as a value indicator. If you take FOSS and package it for sale then people may well buy it who wouldn't "use that shitty free software". Humans are complex beasts.


Or you know, the FOSS alternative might still be 100x weaker than the propietary incumbent. I'd love to ditch Adobe products but Gimp and Inkscape don't quite cut it.

"Goodbye to your propietary business" applies when you have a commoditized product that's easy to replicate feature-by-feature. In which case this is not a free vs propietary problem - another propietary competitor could leave you in the dust just as bad - it's a general vision and execution problem.


iTunes


Love this!


Amazing how you have to justify getting money for work. Reminds me of the Nomad List article.

"Why isn't Nomad List free?"

https://nomadlist.com/help


I can understand how you feel and I also think your blog post is correct in many ways. However I think software should be reasonably priced which is most often not the case.

And also people can't pay everybody 5-10 dollar/euro for everything, we would quickly need to pay thousands of dollars/euros yearly to access/use everything. That's why I think sometimes having another way of "paying", like through ads or maybe some kind of crowdsourced project is a good alternative for some things.


And then it comes around full circle when to get advertisers you need to market them properly to specific people who would be most likely to buy their products. Then you start collecting data on users to figure out who would match with what, then the customer becomes the product.


Maybe you don't need everything.


It would be competitively priced if everyone started charging for their software and the market filled back up with participants. The concept of free software has decimated the software industry by entrenching existing players and allowing them to exist unchallenged, so of course it has resulted in higher prices from the remaining companies.


Maybe you just need fewer $5-10 dollar/month things than you think? At that price you could have 100-200 of them before “quickly” needing to pay thousands of dollars yearly.


100-200 services would add up to $1000/mo not per year.


duh, I am an idiot. Too early.


I have 1278 packages installed on the computer I'm using now it takes 11 items 5-10 bucks a month each to spend 1000 dollars annually.

Thanks open source.


You release fman on Linux, so you use at least some open source libraries. I wonder which ones are those and what is your policy of transferring a share of your earnings from your project to them.

Also I suggest anybody using this software with a paid subscription to evaluate this bit from the EULA:

> fman is provided as-is. You are using it at your sole risk. We give no warranty of any kind. You agree that we shall not be liable for any damages related to fman.

Source: https://fman.io/eula

Also, stop this FUD already. Anybody can release software however they want, free, libre or not, but this attitude that I quote from your link is plain wrong.

> We all like free stuff. But in a world where everything is free, we are the product.

We're talking about FOSS software here, not about Facebook or Google. If you don't want people "bashing a little guy who's trying to innovate", you can quit spreading misinformation about people whose gratutiuous libraries and kernels you use (yes every platform you release to uses FOSS somewhere).


Love the link.

It is easier to find people willing to pay for software when focusing on Apple and Microsoft platforms.

How has been your experience regarding consumers per platform?


Thanks! :-) Apple users are more willing to pay in my experience.


I love that post. Also, just bought a license (and subscription for updates). Product looks awesome. Installing right now and am going to use it today!


Wow. Thanks!!


I've just tried fman and I love it.

I've never heard about this tool before. I'm pretty sure that you would be able to sell more if you market it a bit more. Thanks for the software!


That's great to hear. Thanks!


As for the security (against illegal copying), I'm thinking having a basic stuff - so it won't be very easy to crack, but also it shouldn't be very hard, so people in 3rd world countries are able to use it - these aren't the loss sales anyway, and also help in popularization. What might be a reasonable price in the west, might be worth of monthly salary for some living in poor countries - I wouldn't want to charge anything to such people.


Maybe you can create a page on your software website "$softwarename key generator" or something like that. And on that page you can give away a free key for those that are going to pirate it anyway. Bonus points for having detailed statistics about the wannabe pirates.


Not sure, that might be too easy - maybe some potential buyer will use it as well?


That's indeed a possibility. But if the potential buyer is searching for cracks, I'd say the chances of him buying are very low. Maybe offer a heavy discount, or a non-commercial license.

(sorry - I little off topic) Just curious, how did you come to your current pricing? (I saw your blog post about sublime)


I'm not sure if you mean the prices themselves ($18/12) or the scheme "perpetual license + optionally updates for $x/year". Feel free to hit me up on twitter @m_herrmann.


I was thinking mostly about the $18 license. More generally though, your process on pricing in general would be cool to learn about. For instance - sources of inspiration from "the literature" (ex. Strategy and tactics of pricing), past experience, user price sensitivity testing etc.

thanks - appreciate your response and its cool to see you building and selling your software!


I see. I arrived at the $18 through trial and error. It's very new though. It used to be $14, which was way too low. Following common startup advice, I raised to $28. Sales halved so there was no advantage. So now I'm trying $18.

Right on -- thanks!

Interesting that you got downvotes for a statement like this on a forum supposedly rooted in the business of software.


Because most want to get paid, not to pay for the tools they use.


Exactly, there is some cognitive dissonance going on.


It's because obviously the shit I need to get my shit done should be open source or free by now, sheesh it's 2018 not in the dark ages however of course the serious business need I am fulfilling is novel and very deserving of a good chunk of cash.


To put it more concisely: Everybody wants to commoditize their complements.


The content of the link is FUD. He portrays FOSS as if it was like the exploitation schemes the ad supported free stuff, saying: "But in a world where everything is free, we are the product."

I did downvote him, although I'd not disagree with him (or anyone else) preferring to produce closed-source software. He could've justified himself w/o telling lies about the other side.


Nice idea! It reminds me of Sublime.


Paying for software may be moral, but it certainly isn't natural. As the game's developer, you're the scare resource. Making copies of the finished game itself costs approximately $0. Copyright makes your game artificially scarce so that you can charge for it. It's a phony concept that exists to support an unnatural business model. The real business model should be players paying you directly to develop and update the game they want to play for as long as they want to play it.


> but it certainly isn't natural.

So what? Nothing about software development is 'natural'. I don't even know what definition of 'natural' you're using here.

> It's a phony concept that exists to support an unnatural business model.

It's a social construct that we've adopted in society because it allows for things we like.

> The real business model should be players paying you directly to develop and update the game they want to play for as long as they want to play it.

This is a valid business model for some games, but there are many games that would not exist without such a model. I'd prefer that those games exist instead of insisting that reality should be the way that fits nicely with my arbitrary mental models.


>I don't even know what definition of 'natural' you're using here.

Copying data is a fundamental operation that all computers are able to perform and it costs nothing. Therefore, data is naturally easy to copy.

Software doesn't need to be manufactured like physical goods do: the software construction step is fully automated by compilers and build systems. The cost of all this is approximately $0. Networks aren't free because of bandwidth costs, but they still make it really easy and cheap to copy data to a lot of computers.

Since games are software and software is data, it is rather easy to distribute them to an unbounded number of machines. Even if you come up with some complicated copy protection scheme, people only need to crack it once and produce a DRM-free version of the game that will work everywhere and won't do stupid things such as require an internet connection. Even cryptographically secure platforms such as the PS4 eventually got cracked and the first thing people did is enable the execution of unsigned software.

>insisting that reality should be the way that fits nicely with my arbitrary mental models

That's exactly what copyright does.


> Software doesn't need to be manufactured like physical goods do: the software construction step is fully automated by compilers and build systems.

Source code surely is manufactured. I write it with my hands, using years of knowledge and experience, spending a lot of time on it.


>Source code surely is manufactured. I write it with my hands, using years of knowledge and experience, spending a lot of time on it.

Right. We not only write it with our hands, but with our brains too, as well as with the effort we put in over years to gain the skills needed to write the software in the first place. Otherwise any newbie programmer could write any of the software in the world, including complex ones.

This reminds me of that story of a senior retired engineer who was called in to diagnose a problem with some expensive, complex machine that he had worked with or operated a lot earlier (and/or maybe had invented). He walks into the room, bends down near the machine, taps it a bit, then makes an X mark with chalk near the bottom left corner. "There's your problem", he says. They open the machine, check that area, and sure enough, the fault is found and fixed. They ask him what his fee is. "$50,000", he replies. "What?", they ask. "$50,000 for making a chalk mark?". "No", he says. "$1 for making the mark. $49,999 for knowing where to make it."

https://quoteinvestigator.com/2017/03/06/tap/


Exactly. You're providing a service. The digital goods, once produced, can be replicated infinitely at extremely low costs. Nobody can create copies of you or your expertise.

In order to pay for your software development services and turn a profit, the digital goods you produce are made artificially scarce so as to enable their sale. However, that does not change the nature of those goods. Copying them is still trivial.


> Copying data is a fundamental operation that all computers are able to perform and it costs nothing.

That is not true. Running, maintaining, etc the IT infrastructure is costing money, electricity etc. It doesn't come for free.


Marginal and transaction costs are very low.

But you are right on the principle : storing, copying and transferring data always costs energy (in physics, temperature, energy and change of entropy are linked together)


OK, maybe you still have a point. What then, would be the threshold for saying: now, at this point, the costs of X are not marginal anymore but substantial enough to be taken into account?


According to Wikipedia, the best selling video game is Tetris: it has sold approximately 170 000 000¹ copies. If copying Tetris takes 1 second, the computer draws 400 W of power and electricity is rated at 30 cents/kWh, it would cost approximately $5667 to create 170 million copies of Tetris.

¹ https://en.wikipedia.org/wiki/List_of_best-selling_video_gam...


Depends on what you are thinking about ? Such cost always need to be taken into account, but IMHO in general we must always differentiate material things (that are formed from matter and its organization) and information (i.e. concept of entropy, ideas, thoughts, songs, language, jokes etc..) which have temporary physical supports but exists apart from them


Good distinction, I agree. This leads me to this: If, for example a song, is not written down anywhere, but some brains contain it, this means there some people who could reproduce this song by singing it, this song would still exist. Is a brain a temporal physical support unit in our sense?


Yes of course. As long as we have electricity in our brains we can store and process information. The very precise implementation of this would require to study biology (it's already interesting to see that we run on electricity like computers!).


Ok, cool, now, how is the number eight, for instance. Does it exist only because human brains have it in it? Or do numbers exist in the universe outside our brains? Can we answer this? This would be an important step in IP rights and all that…


It's a whole philosophical question...I would say yes maths and humans abstractions exist independently !


“Marginal costs” are the additional cost for each additional unit delivered. They can be high or low, but with modern, non-custom software they tend to be extremely low.


Suppose that a copy takes 1 second to complete and the computer's power supply unit draws 400 W of power during that 1 second. How much does it cost to complete that copy? Hawaii has the most expensive electricity¹ in the United States, averaging at 29.23 cents / kW * h. Let's use 30 cents / kW * h.

If we have 400 W sustained for one second, that equals 0.000111111111 kW * h. (400 Watt * second) * (30 cents / kWh) = 3.33333333 × 10^-5 U.S. dollars. In Hawaii, that one copy would cost $0.0000333333. If one were to make 100 000 copies of that same data, it would cost $3.33333333.

¹ https://www.electricchoice.com/electricity-prices-by-state/


This is a very large strawman you've constructed.


Please explain what the problem with what I wrote is.


> So what? Nothing about software development is 'natural'. I don't even know what definition of 'natural' you're using here.

I'm quite sure they're referring to artificial scarcity. Making copies is very easy, making games is very hard. Therefore it makes much more sense to charge for the development of the game rather than the copies. Making abundant resources artificially scarce is either frowned upon or flat-out illegal in pretty much every other industry.

> there are many games that would not exist without such a model

How do you know?


Okay, so I'm going to charge for my development time - how exactly am I going to divide that up among the customers, many of which don't exist yet, without over-charging anyone and how are you going to make sure I don't do so in a way that doesn't "illegally institute artificial scarcity" ?

The issue here is that there seems to be a fundamental misunderstanding of how markets work. Toro could choose to cut production of their lawnmowers by half and double the price, and this would cause "artificial scarcity". However, it would also be perfectly legal and downright stupid of Toro because the competitive lawnmower market would punish them dearly for it.


> Okay, so I'm going to charge for my development time - how exactly am I going to divide that up among the customers, many of which don't exist yet, without over-charging anyone and how are you going to make sure I don't do so in a way that doesn't "illegally institute artificial scarcity" ?

It's not my job to invent business models for other people, but all Kickstarter-type funded software is a good example of one possible way. Another is maintaining and updating a gaming server and charging for access to it, like World of Warcraft does.

> The issue here is that there seems to be a fundamental misunderstanding of how markets work. Toro could choose to cut production of their lawnmowers by half and double the price, and this would cause "artificial scarcity". However, it would also be perfectly legal and downright stupid of Toro because the competitive lawnmower market would punish them dearly for it.

If you had a device to copy lawnmowers, but the lawnmower industry fought back to make copying lawnmowers illegal, then you'd have a working comparison.


The software industry had/has a lucrative, self-sustaining business model that was very much the same as every other business on the planet. I'm a part of that industry, so I do have a business model. IOW, I'm not the one trying to tell someone that they shouldn't charge for their efforts/products, so it seems to me that the onus on coming up with a workable model is on those that have been trying to institute the "everything is free" model for some time, not me.

As for your other comment: I believe there's a saying that covers this:

https://en.wiktionary.org/wiki/if_my_aunt_had_balls,_she%27d...


> Nothing about software development is 'natural'

Why do you say so?

> I don't even know what definition of 'natural' you're using here

I think that he's talking about the ease of copying software: it's quite hard to make people pay for something so easy to duplicate..


"You cannot go against nature Because when you do Go against Nature It's part of nature too." -- Love and Rockets, "No New Tale to Tell"

In the scientific sense, everything is natural. Truly "unnatural" things are like supernatural things -- stuff that exists only in fairy tales or horror stories or some metaphysical religions.

Saying that something is unnatural is typically a value judgement meaning that it is unhelpful, damaging, or against a goal considered inherent in some area of life, and so on.

Otherwise, a more valid if unfortunate use of the word "unnatural" is as the common short-hand for "human-made" or "artificial", similar to the way that "animal" is used as a short-hand for "animal that isn't human". Of course propulsion labs, parliaments, and property laws are absolutely as natural as mushrooms, mollusks, and manatees, but in addition to being natural the former are also human-made. In this truest and widest sense, everything that is artificial is also natural.

Other uses of the words "natural", "unnatural", and "nature" should be informed by this broadest and truest one, especially when making judgements about laws or norms or behavior. This might sound pedantic, didactic or even obvious at first, but with some consideration it may reveal itself as a surprisingly useful and clarifying point to keep in mind.

Rather than proclaiming copyright laws unnatural -- a value judgment masquerading as an objective fact (perhaps unintentionally) -- it would be more fruitful to ask questions such as, "How does copyright differ from other property laws?", "How are property rights justified?", "What are the differences and similarities between physical things and ideas?", and so on.

Yes, it's nearly free to copy digital works, which therefore lack the same level of scarcity that we experience in the rest of our lives.

Yet the creators and consumers of intellectual and digital works have little choice but to create and consume those works in the broader context of the world of physical scarcity in which society is embedded. It's sad, but you can't stop burning calories and taking up space just because you are consuming digital media.


> > Nothing about software development is 'natural'

> Why do you say so?

Because building even the simplest computers requires large-scale industrial processes. All of this is based on a social/economic system that required the intelligent design of many many people and organizations.


> The real business model should be players paying you directly to develop and update the game they want to play for as long as they want to play it.

They have to find your game and know they want it first, and it might not be too obvious to other people why they might want your game until it is already far enough along that it might as well be released. And by then it might be too late, unless you've made a game that can be endlessly added to.

First impressions count for quite a bit, so if you make your game available before it's really ready, when there's only a small amount of content, etc, then people (who might have loved your completed game) might give it a try, decide 'This game is broken and there's no content' and then never bother giving it a second chance.

Not to mention that indie game developers (especially those that get funding ahead of time, like on Kickstarter) tend to overpromise and and underestimate how much time and money those promises will take. See: Mighty No. 9, No Man's Sky, Star Citizen. Even the mega-hit Shovel Knight burned through all the money it crowdfunded and the developers had to work for zero salary for four months before the game was released.

Finally, the moment you make your game available for others to play, you're instantly on the hook for technical support, which could take a large amount of time, especially if you're one of only a handful of developers, precious time that could be spent developing the game instead.

So long story short, if developers want to get paid as they are developing a game, they will have to spend an inconsequential amount of time marketing and supporting an unfinished, broken buggy game, which could easily take 60% or more of their time (a percentage I often see people claim is needed for the non-software side of their software business).

It's a business model that works in some instances, but it doesn't work for everyone.


No other business endeavor on the planet operates on "production time" when pricing products. Products are priced according to their value. Given a reasonably competitive market, that value will be accurate and fair for customers, and the customers will be able to find a similar product that does most or all of what they want at the price point that they can afford.


Products are scarce. Data isn't.

What if everyone had a Star Trek replicator? Would they still pay for products according to their value? That's what a computer acts like.


You would probably pay for: the replicator, the recipe,the ingredients and the energy used by the replicator. However, Star Trek has a different economy from ours.


Star Trek has no economy except for its fictional one. Their economy of fabricators is itself a fabrication. At best, it represents an ideal.


Which is why the original argument is pointless.


Yes, and software products are the sum total of all of the effort put into developing, advertising, maintaining, and supporting them. The fact that they are distributed as data is an inconsequential fact that obscures what they actually represent, and that is very much a human achievement no different from painting a picture or building a couch. There are tooling costs and all sorts of ongoing fixed costs, just like any other business.

As for Star Trek: when that day comes, we'll deal with it then.


>>The real business model should be players paying you directly to develop and update the game they want to play for as long as they want to play it.

This may be an acceptable model for games-as-a-service, but how is anyone supposed to make money on single player puzzle or story-driven games? I enjoyed westworld, but I have no desire to replay it. By this same logic, it's also 'unnatural' (whatever you think you mean by that) to pay an author for a book. Some software products don't necessarily fit with your proposed update cycle, there are a significant number of games that get developed, released, and that is it aside from bug fixes.


> but it certainly isn't natural

Neither are money or laws, in that sense.


Remembering a chat with my landlord just a few days ago, I can only say ...

Indeed. Neither are money or laws. Or promises, for that matter.


Are you saying that the "phony concept" of copyrights shouldn't exist? Authors and artists (and software developers) would have little incentive to produce in such an environment.


Yes.

Lack of incentives didn't stop the artists of the past. Many were able to secure patronage.


You need to consider survivor bias. We don't have data for the number of "artists of the past" who were stopped because they couldn't find patrons. How many Da Vincis and Mozarts had to work in fields or row galleys rather than work as artists?

It is really, really difficult to build a career around patronage. A few people can do it. Most cannot.

The fundamental flaw of the "software should be free because duplication costs are zero" argument is that it ignores the total economic picture of software development. It's like saying "gasoline should be free because it costs next to nothing to run the pump motor at the gas station". Without someone's significant investment of time and money, that software (or that gasoline) is not going to exist. For software, the means of production is not a simple file copy, no matter how easy it is to invoke that COPY command.

We've been paying craftsmen for their work for many thousands of years. Software isn't any different if you take an honest view of the whole process.


You make some good points. Concerning whether works should or should not be free, I'm not sure. I think people should get paid somehow. How to solve that problem is not clear to me.

What I've been observing, though, is people acting as if the software is free because copying is practically free. This happens all the time and nobody even thinks about it. Friends share programs, music, movies, pictures, documents and all kinds of data among themselves and they may not even realize the fact they're infringing copyright. I see my colleagues sending each other copyrighted scientific articles via messaging applications every week. Nobody really thinks about some journal's copyright, it's just "those people need this file and I have it, so I'll send it to them". Companies have been trying in vain to police this behavior for years. I don't think they're ever going to succeed. I think that's a sign this system is incompatible with current technology. What good is a law if you can't enforce it?

>It's like saying "gasoline should be free because it costs next to nothing to run the pump motor at the gas station".

I don't think this is the same thing. A more apt comparison to software would be "gasoline should be free because the pump is magical and can somehow duplicate existing gasoline, creating new matter in the process, and the cost is so infinitesimal we don't have coins worthless enough to pay for it unless we purchase a whole tanker's worth of fuel". I'm willing to bet that such a magical pump would drastically change how people perceive the value of gasoline. Even though creating that first volume of gasoline required a substantial amount of money, the magical pump can simply duplicate it endlessly.


You're never going to stamp out casual copying. We know that doesn't work. Even in a healthy market there will always be some. Rampant copyright infringement is a sign of an inefficient market; supply is limited (e.g., geo-locked) or the price is too high. But there's always going to be some person who rationalizes liberating a $2 game because "you can't steal bits". We've been over the arguments. I've written copy-protection code, I've gone out for beers with self-professed software pirates. None of this is going away.

In your patched-up analogy, someone has to pay for that first billion dollar gallon of gasoline. Someone did pay for the exploration, drilling, extraction, transportation, refining and marketing of the fuel; they paid for equipment, for people's salaries, for mineral rights, for the overhead of keeping a company running. It's a lot of money; it's an investment, and the investors expect a return. Take away the return and you'll find that none of the exploration, drilling, refining, etc. are going to happen. Have fun starving in the dark.

Software is no different, except that the "transportation" costs are close to zero. None of the other costs are zero, but the cheapness of that COPY command is all people really think about. They don't want to think about the engineers, the animators, voice actors, IT staff, management and rooms full of equipment necessary to create the product. They don't want to know about buying servers, spending money on marketing programs, creating sales pipelines, paying taxes and permits and making sure companies can hire and retain the folks necessary to make more product. I guess it's inconvenient to the rationalization.


Most of the “software” mentioned in the post, are actually services (slack, zoom, tidal, etc). For many people, it is easier to accept to pay for a service rather than software.

While I do use paid services, I usually “pay” (donate) for free and open source software. Why? It helps me get my work done, support is usually good, I set the price and I know that people who aren't as fortunate as I (being able to afford to buy software) can use it for free and get the same opportunities as me. :)


The marginal cost for software is zero, the marginal cost for services is non-zero. This puts a floor under the price. People don't accept that they have to pay, rather they are given no choice.


Often.. but for a lot of software businesses, the difference between being software or software as service is a choice, and cost is often a minor implication of the choice.


Right, mentioning Keynote and services like Dropbox (as done in the article) in the same way isn't quite right. Yes the marginal cost for another user on Keynote is zero but certainly not Dropbox or any other software as a service. If 2 people use Dropbox is may cost almost nothing but if thousands use it you either need a business model because costs could be significant or (like some of the big players) have another business group that subsidizes the running cost.


I have experience selling to companies and I’ve found that even when they realise they need and want to pay for a piece of software they have absolutely no clue how to figure out what it’s worth.

Consequently they engage in endless hardball negotiation and invite other suppliers to provide quotes (even when your offering uniquely meets all requirements).

In the end the challenge is to get customers comfortable that they’re paying a fair price for what they’re getting. That however is much easier to say than to do.


If you haven’t already, stop talking about what it took to create it and start talking about how much value they’re going to get out of it.

Figure out how much time / money it will save them. How much that is all worth. Tel them your software is less money than that.

I have the luxury of knowing I have the best product, but even if you don’t, you can’t give in to the hardball shit. Play hard to get instead. Dare them to go buy the cheap thing and let them feel like you know they’re making a mistake and you’ll see them in a few months when that doesn’t work out. It pays to be cocky in those situations. It starts to make them feel like you must know something they don’t.

It’s the only way. (I sell one of the most expensive software products there is, with loads of cheaper competitors)


This is a great point.

No one wants to be ripped off. Software basically can't be objectively priced. You might be selling £50k per seat software that competes with £50, or free. It sucks to be making that decision.


>Software basically can't be objectively priced. //

Can anything be objectively priced, how?


I can buy apples at a somewhat objective price by comparing the price provided by other sellers in nearby stores. Also the price will be mostly affected by supply vs demand.

OTOH, softwares prices can be affected by too many parameters.

Then again, IT industry is a child when compared to say, mechanical or Civil industries.


Your first method just overprices/wrong-prices the whole market. I'd argue that supply-demand doesn't give fair pricing (which presumably is what's meant [in part] by objective), if you can make something very cheaply does that mean you objectively deserve to be very rich simply because you've got a corner on that market?


I would not have a problem paying for software if what I was paying for didn't suck. Case in point: I have never been very happy with Win 10 since I got it, but in recent weeks it has become completely unusable, mostly because of their constant upgrades and behind-the-scenes processes that slow my machine down to a crawl. Out of frustration I wiped my drive and installed Ubuntu. Granted, Linux has it's share of problems but still works much better for my uses. So, one OS is paid and one isn't. Guess which? :/ My personal gripe is that it seems to me, when it comes to basic customer service, we the customers are on the bottom of the totem pole. Software companies spend more time worrying about how to reassure their shareholders than they do about making sure we have a usable, hassle-free product, and I've been seeing this trend getting worse in recent years. So until that thought-process changes these people will not get my money. As far as I'm concerned, at this point anyway, FOSS is the lesser of two evils.


> So, one OS is paid and one isn't.

Quoting from https://en.wikipedia.org/wiki/Canonical_(company):

Canonical achieved a small operating profit of $281,000 in 2009, however until 2017 it struggled to maintain financial solvency and took a major financial hit from the development of Unity and Ubuntu Touch, leading to an operating loss of $21.6 Million for the fiscal year 2013.[19] The company reported an operating profit of $2 Million in 2017 after shutting down the Unity development team and laying off nearly 200 employees. The company now plans to focus on its server and professional support solutions, which have proved to be most profitable. By shifting resources away from Ubuntu Desktop and cutting less profitable products and services, Canonical plans to maintain solvency and achieve long-term profitability.

"one isn't" (your words) to the extent that they had to switch focus away from the desktop OS to stay afloat. Much as people would like to pretend otherwise, the almighty dollar/euro/yen/etc. is still what matters most.


I'm not convinced that unity every added much value to ubuntu it seems like they could have spent a small fraction of the money customizing gnome/cinnamon or actually shipping plasma with up to date deps and come out way ahead both for users and monetarily.


Ubuntu gives you a very clear option to pay for it when you download it.

If it works for you and you want it to keep working, making a donation/payment may not be the worst idea.


Exactly, I always paid in some form to the distributions I had on my computers.


Even Stallman doesn't have anything against paying for software. Seth's point 1 is reasonable. Point 2 is about paying for support, not paying for the software itself. Point 3 is no point at all, it's just rephrasing point 1 (ie marketing is about the business, not the software).


I see the use of OSS in government services as almost a moral imperative-but the first point seth makes is crippling this "you need money to write the software in the first place"

Take voter management software - the preparation of polls, counting of results etc. all this software behind it is afaik propriety but used exclusively by the public sector - which seems mad, and even risky when we look at electronic balloting, but can one persuade any government to kickstart the process of building new? i cannot

Edit: thinking of this, what chnage would a kickstarter project stand to build a base of oss government software?

(I know things like GDS and 18F exist but they are not covering the full range of services - see http://oss4gov.org/manifesto


The software might be free, but nobody acquires mission-critical software without a support agreement -- which is certainly not free -- unless they don't value staying employed. We might not like Diebold (or whatever name they go by now) but _someone_ has to be an accountable party for supporting electronic voting systems.


free as in beer is irrelevant - free as in speech that i can read, check verify and is what matters - so i would envisage (and encourage) a network of "certified in using votermachine-oss" contractors that can be called upon or paid for, in a retainer fashion, and most importantly some means of priming the pump


Why is it so hard for developers to make a full-time living selling their open-source software? Are the available options stacked against them?

Thousands of developers make money building products or services for clients using open source software. But what if you don't want to be a consultant and simply want to charge a business for using your open source product?

The Free Software Foundation (FSF) says that you can sell your free software by charging for distribution (for example via CD-ROMs). Physical media made sense in an era of dial-up modems and low-bandwidth connections, but no-one can realistically charge for distribution today - not least when so many projects are hosted for free on GitHub. See: https://www.gnu.org/philosophy/selling.en.html

So what are the alternatives?

- There's the dual-licencing approach e.g. Red Hat. (Naturally doesn't work with permissive licences. Requires a strong copyleft licence like the GPL).

- The open core approach e.g. GitLab.

- There's also the option of charging for support only - I'm guessing this is an unattractive and unappealing option for many developers who'd much rather be building their product.

- Or there's the Kickstarter or Patreon approach which simply isn't a stable, sustainable long-term solution.

There have been some alternative licence approaches such as the Fair Source licence (fair.io). This is neither open source or closed source but allows developers to charge for their software.

The large market for Wordpress plugins is an interesting case study. Wordpress is published under the GPL v2, but many Wordpress plugins are not open source. No one polices this, but those plugin developers are selling their software to businesses who are willing to pay for the features of those plugins.

So who here in 2018 has made a success selling their open source software? Did you follow one of the models above? And who has struggled to make it work?


As someone that was part of an effort to do this in the early .com days, it is a lost hope trying to sell open source software to consumers.

While selling to enterprise can be made via consulting and training, most consumers just want a one time shot paying for their software.

Well, if you do that with commercial software, usually the only way to avoid paying is piracy.

In open source world, they can just go to another guy that is re-packaging your work and making it available for free, thus killing your source of income.

Which is the major reason most desktop software shops are going the store model, hoping that higher control will prevent that kind of workaround (which doesn't work with rooted devices anyway), or forcing consumers to migrate to a SaaS like in the good old timesharing days.

I might add that FOSS developers are the first to blame, as contrary to most professions, many go to great lenghts to avoid paying anything, while feeling entitled their costumers pay them.

I am overgeneralizing here, but you see it every time someone comes up with licensing changes as means to be able to keep working on the product that they love, right away the forum gets flooded with free beer alternatives and vocal complaints how everyone is going to migrate to something else.


I don't think the primary problem is cost, but that when you build your product tightly coupled with a propretary product that is hard to migrate off of then the vendor of that product can shake you down, or they can go bankrupt or what have you. It makes your business vulnerable, and no one wants that.

Also your production environment crashing because a license expired. Fun times.


I agree with this. That's why I avoid most web and SAAS platforms for things nowadays. I make an exception for things where I can get my data off their platform in a format I want pretty easily (i.e. I will use Google Docs because I can easily download the files as DOCX or PDF files) or the data I'm storing with them is mainly settings or preferences (I.e. Youtube knowing I'm subscribed to X publishers).

I've been burned a few times and lost data just because a company goes out of business, or I decided I wanted to use a different platform and all my data from before was locked in to this other proprietary system.

So yeah, if you're coming up with this hot new productivity/task software and I can't either download the data in an open format or at least use the software on the desktop, 100% offline? No thanks.

And in general I hate subscriptions unless I'm getting access to libraries of content, i.e. Spotify, Netflix, or it's just for updates. By that I mean I get the software and a year's worth of updates when I paid for the license, and it's mine, you can't take access away from me, I have it for as long as I want, and I only pay again when I decide I use the software enough and the updates are compelling enough that I really want to update.

Which is another reason why SAAS and Web apps don't really work for me, because they almost invariably charge a monthly fee and there's nothing to download so I just lose access the moment I stop paying. There are exceptions, but those are rare (mainly web hosting and cloud storage).

It's for this reason why I refuse to pay for Adobe Suite, because you only have access for as long as you pay the monthly subscription. Screw that, my copy of Adobe Illustrator pre-subscription still gets the job done fine thanks. I might have upgraded by now if they hadn't switched to a subscription model, but now I'm going to keep using that as long as humanly possible, then search for an alternative.


In all fairness, the software that you buy today is almost certainly going to stop working at some point in the future as you upgrade your OS etc. So, unless you can download data into some reasonably open format, you will eventually lose access anyway.

That said, I don't disagree with your basic point. In the case of Adobe, I use one of their programs fairly heavily. But most everything else is once in a blue moon. As a result, their Suite subscription doesn't work for me as I'm perfectly happy running something from a few versions back. Subscriptions only make sense for things you use pretty regularly.


Yes, but this is a symptom, not a cause, of a non-competitive market. You get stuck with only one option in markets where most of the other players have left the market because they can't figure out a way to make money in that market anymore. With competitive markets, switching costs are typically optimized away by the players that are interested in poaching one another's customers.


You really aren't replying to the person who actually tried to make money in open source and related some actual experience of why it didn't work.

Businesses actually don't give two sh*ts about being "tightly coupled" with a proprietary product. That's a sysadmin thing. It makes sense to the "us" of the world, but businesses don't care.

Look at any area or field where billions or trillions of dollars worth of transactions still go through systems running on IBM iSeries, or mainframes. It's just the cost of doing business.


What do you think of the viability of providing source code with the software that you sell? It would be open source in the sense that you could alter the software and contribute patches, but you had to pay for the software first.

It would have to eliminate forking by necessity. I don't see that as being a downside, though. I think the open source community commonly forks code to save time when the correct technical decision should really be doing ground-up rewrites.

I think the primary issue would be giving away the 'secret sauce' so to speak. But a motivated competitor could still decompile and reverse engineer an application. Would giving away the source code make it easy enough to do that it makes a big impact? like how Android apps are so trivially easy to clone that Chinese competitors can clone an app in seconds.


This is how the Delphi and Turbo Pascal 3rd party markets have worked for a long time: you pay for 3rd party components/libraries, but they almost all come with source code or an option to purchase the source code for an additional charge. It works really, really well, especially for larger organizations that buy from smaller organizations that want to ensure the continued availability of the software and/or audit it for security purposes.


Yeah, there's a contingent that won't ever be happy with a commercial arrangement.

Just thinking about the way that version control was done for Linux before Git. The author of the system was happy to have people using it as long as no one reverse engineered the product that he was actively selling to customers. The reaction from the community was, "thanks for contributing your hard work...now I'm going to make a clone."

There's a sense of entitlement that really turned me off of open source *nix for a long time. That's just me though. I'm over it, it won...OSS won...free beats everything in the end.


Yes, I agree with what you say.

I actually think the dual-licencing model seems the most realistic for many developers who are selling to a business. Ironically, it's because a strong copyleft licence like the GPL is unattractive to many businesses that they naturally turn to the commercial licence which allows them to make changes they don't need to release publicly.

I'm sympathetic to what Fair Source is trying to do, but even that garnered a lot of negative reaction among developers (at least if you read the discussion on Hacker News).


Why is it so hard for developers to make a full-time living selling their open-source software?

Do you really expect people to pay for something when they can get it for free?

Making software open source means you are either selling something complementary (hardware, other software, services) or else begging for crumbs.


I don't know much about these things and I'm making this up as we go but I wonder if we can use AGPL or a stronger version of AGPL. Basically, you pay for a subscription or everything on your servers that touches this software must be free and open source. Should work for researchers and students and there is a continued revenue stream. For example, require any software that talks to a rest API also be foss. With more people paying, the subscription can slowly become cheaper and more accessible. Thoughts?


Thoughts?

It depends. In my own business I decided open source 'customers' are more trouble than they are worth so I do very little.

Just the fact that people are willing to pay for my software is a strong validation that I'm doing something useful and on the right course.


Many companies already do this. See Mobicents, which we use at my current job.


Let's look at your examples about Red Hat and GitLab. Managers pay for service contracts from those companies so they have a "throat to choke" if things go sideways so it's not simply their neck on the line. From a CIO's point of view, a service contract is an insurance policy against standing before the CEO or board and saying "I have nobody but myself to blame, I have nobody but my team to fix the issue -- and they don't know how." Typically it's a start-up or a company where the CEO/founders are driving the technical decisions where you find IT using CentOS, PostgreSQL, etc without a support contract... and when there's enough revenue or investors buying that "throat to choke" is part of due diligence in a growing business.


>- There's the dual-licencing approach e.g. Red Hat. (Naturally doesn't work with permissive licences. Requires a strong copyleft licence like the GPL).

Red Hat doesn't dual-license and sells subscriptions (support, updates, backports, KnowledgeBase, etc.) for open source offerings that use a variety of different licenses including permissive ones (e.g. OpenStack).


With regards to WordPress, it’s the lack of policing on the OSS-ness of a plug-in — as you mention — helps create a viable market, but many of the most successful plugin devs sell a premium version tied to a service in some way. Whether the service is necessary is often beside the point, the SaaS aspect can be used as a shield to obfuscate the open source code.

I also think with WordPress in particular, the target customer doesn’t know how to compile the plugin source if they find it. It’s easier to pay than to figure out how to get it for free. Now, whole rings of people sharing premium themes and plugins exist, but the malware often added to those items can turn people off of that method. It’s easier to just buy it.

Honestly, support seems to be the most long lasting option — see RHEL and Canonical and Oracle with their Linux distros. You can use centos instead of RHEL (and could before red hat took it over, for that matter), but you’d need to deal with the the problems yourself or have a person on staff to deal with them. Especially as stuff is more cloud-based,it’s easier to just have a support contract.

The problem with this model for the indie developer, of course, is that the act of offering support has a cost too. Time, extra staff, logistics — and that can be harder to do than the closed-source perpetual license model, where you offer support but it’s understood the transaction is primarily for the software. And just speaking for myself, if I buy a piece of software, I’m not necessarily going to bug the dev every time something breaks b/c I was really buying the ability that use the software that would otherwise not be available to me. But if I am explicitly paying for support, I’m going to be more entitled, even if it’s for an OSS product. Because if I don’t get support and it’s OSS, why didn’t I just compile it myself and save money?


Interesting, that Fair Source License seems like the perfect example of how your software can simultaneously be open source software but not free software. It doesn't matter if it's open source if I can't modify it and deploy it to however many computers I want to.


This is kind of timely, as setting up a Patreon account is something we (DB Browser for SQLite / sqlitebrowser.org) did a few days ago:

https://www.patreon.com/db4s

For us it's an initial experiment, as we had no idea at all if anyone would decide to become a Patron.

However, it's um... 3 or 4 days since the account creation and already some people have become Patrons. Kind of nifty. :)

Other people have been at it longer, and are definitely having a measure of success with it. For example, seeing Gina Häußge being able to work full time on OctoPrint gives us hope for the future. :)

https://www.patreon.com/foosel


> There's the dual-licencing approach e.g. Red Hat.

This bit is confusing to me. AFAIK Red Hat doesn't use a dual-licensing approach.

Is it something they've recently introduced?


Redhat owns Centos if I recall correctly. Prior to that centos was just a rebuild of all the open source parts of redhats software.


Yep, Red Hat (note - not "RedHat") does employ several of the core CentOS team these day, although it started out externally.

> ... just a rebuild of all the open source parts of redhats software.

That seems like super weird description of things. :)

All of the Red Hat OS software is Open Source. The only significant pieces CentOS don't include are the Red Hat branding and artwork:

https://www.centos.org/about/

As far a dual licensing goes... that doesn't seem related to this? :)


To name a few, I’m glad to use and pay for: Overcast, Feedblitz, Discourse, Zapier, Dropbox, Roon, WavePad, Bench, Nisus, Zoom, Slack, SuperDuper, Mailchimp, Hover, TypeExpander, Tidal.

Those are services, not standalone software. (Tidal may be standalone, but it's from 1979.)

It's not like people are still buying Walter Bright's C++ compiler.


At least 3 of those are desktop software.


Great post.

On the same note, paying for software has psychological benefits for the person as well, which I wrote about recently https://rolandas.blog/posts/id-rather-pay-for-software-than-...


I believe corporate red tape one of reasons to not-pay for a software licence. If you download the trial or non commercial edition, works for you, you would never go through chain of approvals to buy a cheap software.Meare player is an example that you can think of. Araxis merge, Visual studio, Visio are some to name a few.


Depends on where you work. I've been in shops where developers had a certain amount of money -- $500 or $1000 -- to buy whatever work-related tools they deemed to be helpful in getting their jobs done. Those tools could be software (Beyond Compare and ReSharper being popular ones) or hardware (monitors!) and up to that amount there were essentially no questions asked (the only caveat being if you spent on stuff that was not needed you risked ruining it for everyone else so you ran the risk of a getting a Code Red from the dev team if you wanted to spend wastefully). As much as companies spend on finding and keeping good developers, dropping $1k/each per year for things that make them happy is chump change.


>Ubuntu gives you a very clear option to pay for it when you download it.

If it works for you and you want it to keep working, making a donation/payment may not be the worst idea.

Yes, agreed. However, one of the points I think I'm trying to make is that that paying for a piece of software doesn't automatically make it better quality. I think that a lot of people in the free software world do try and put as much effort into making a good product as the ones in the paid world. Even under the pressure of not knowing how they're going to pay their mortgage. :/


Bruce Eckel had some ideas about a better model to fund open source projects. https://bruceeckel.github.io/2016/06/20/a-model-to-fund-open...


I was thinking about this recently, maybe FOSS isn't quite ideal for the developers.

As long as nobody is compensated, it is very fair system. But if somebody gets compensated for their work, then if their work depends on other people's work for free, the other people not being compensated is not fair.

For example, consider somebody producing Youtube videos using FOSS. They are compensated, yet the authors of the tools they use are not (necessarily).

FOSS basically started as a radical revolt against copyright system, which is somewhat dysfunctional, but its purpose was to make compensation fair.

But maybe a better copyright system would actually help. Say limit copyright to 5-10 years. The short term would remove the huge gap between doing copyrighted work and doing work for free, and make them more compatible with each other.


>FOSS basically started as a radical revolt against copyright system

FOSS depends on the copyright system in order to operate. If you don't have rights to created works you have no rights to impose reciprocal obligations on downstream users.


It kinda does, but it's an open question whether it's a requirement.

In recent years, many programs became BSD-like licensed and it didn't impede the availability of their source code. Similarly, in the world of Minecraft, many mods started closed-source but free, and eventually became open-sourced, because it turned out to be beneficial on its own.

I actually like the free software philosophy, but it might be that the copyleft clause is not required if there is enough rational self-interest. However, in the current copyright system (where the copyright lasts effectively forever), this self-interest commands to hoard the source code (for the duration of the copyright).


> I actually like the free software philosophy, but it might be that the copyleft clause is not required if there is enough rational self-interest.

At least RMS’s form of the Free Software philosophy sees preventing non-free software, not encouraging free software, as the highest goal. In this version of the philosophy, tools that prevent even a small amount of non-free software being made are valuable even if they don't do anything to encourage free software (and even if they impede it somewhat.)


Proposal to shorten copyright to 5-10 years wouldn't prevent philosophy of free software to function.

It would be additional incentive to release the source code after it enters public domain.


> Proposal to shorten copyright to 5-10 years wouldn't prevent philosophy of free software to function.

It would certainly not be helpful to the implementation of the philosophy of the part of the Free Software movement (which includes RMS) that wants to prevent non-free software from being created to the extent possible.


I am not sure what you mean. RMS himself expressed many times that the copyright term should be shortened.

Remember, the copyright was created in order to promote creativity. If the term was short enough, then lots of useful software could be created based on the source code that is in public domain which was previously monetized by being closed source. This additional software would be an economic win for everybody.


There's a good argument that, given the success of the open source development model, permissive licenses are better because they put fewer roadblocks in the way of participation. And I tend to be in that camp. No argument about the problems of copyright terms though I'm not sure how relevant that is to software relative to software copyrightability generally.


Yes, in a way, that's true. You can make free software more compatible with copyrights by turning it into open source.

But you can also remove roadblocks in copyright to make commercial model closer to open source. Which is what I am talking about, and it would make it also beneficial for the open source developers.

After all, if the open source model benefits everybody, why should only developers make the concession?

More

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact

Search: