Slashdot Mirror
How Can We Fix The Broken Economics of Open Source? (medium.com)
"The economics of Open Source software are fundamentally broken," argues Matt Klein, a senior software engineer at Lyft (who created Envoy). Here's a heavily-condensed version of his essay on Medium:
If we take consulting, services, and support off the table as an option for high-growth revenue generation (the only thing VCs care about), we are left with open core [with some subset of features behind a paywall], software as a service, or some blurring of the two... Everyone wants infrastructure software to be free and continuously developed by highly skilled professional developers (who in turn expect to make substantial salaries), but no one wants to pay for it. The economics of this situation are unsustainable and broken...
[W]e now come to what I have recently called "loose" open core and SaaS. In the future, I believe the most successful OSS projects will be primarily monetized via this method. What is it? The idea behind "loose" open core and SaaS is that a popular OSS project can be developed as a completely community driven project (this avoids the conflicts of interest inherent in "pure" open core), while value added proprietary services and software can be sold in an ecosystem that forms around the OSS...
Unfortunately, there is an inflection point at which in some sense an OSS project becomes too popular for its own good, and outgrows its ability to generate enough revenue via either "pure" open core or services and support... [B]uilding a vibrant community and then enabling an ecosystem of "loose" open core and SaaS businesses on top appears to me to be the only viable path forward for modern VC-backed OSS startups.
Klein also suggests OSS foundations start providing fellowships to key maintainers, who currently "operate under an almost feudal system of patronage, hopping from company to company, trying to earn a living, keep the community vibrant, and all the while stay impartial..."
"[A]s an industry, we are going to have to come to terms with the economic reality: nothing is free, including OSS. If we want vibrant OSS projects maintained by engineers that are well compensated and not conflicted, we are going to have to decide that this is something worth paying for. In my opinion, fellowships provided by OSS foundations and funded by companies generating revenue off of the OSS is a great way to start down this path."
[W]e now come to what I have recently called "loose" open core and SaaS. In the future, I believe the most successful OSS projects will be primarily monetized via this method. What is it? The idea behind "loose" open core and SaaS is that a popular OSS project can be developed as a completely community driven project (this avoids the conflicts of interest inherent in "pure" open core), while value added proprietary services and software can be sold in an ecosystem that forms around the OSS...
Unfortunately, there is an inflection point at which in some sense an OSS project becomes too popular for its own good, and outgrows its ability to generate enough revenue via either "pure" open core or services and support... [B]uilding a vibrant community and then enabling an ecosystem of "loose" open core and SaaS businesses on top appears to me to be the only viable path forward for modern VC-backed OSS startups.
Klein also suggests OSS foundations start providing fellowships to key maintainers, who currently "operate under an almost feudal system of patronage, hopping from company to company, trying to earn a living, keep the community vibrant, and all the while stay impartial..."
"[A]s an industry, we are going to have to come to terms with the economic reality: nothing is free, including OSS. If we want vibrant OSS projects maintained by engineers that are well compensated and not conflicted, we are going to have to decide that this is something worth paying for. In my opinion, fellowships provided by OSS foundations and funded by companies generating revenue off of the OSS is a great way to start down this path."
Just because people can build the software from source if they want to doesn't mean you can't get them to pay you for it.
That requires both a buyer and seller. FOSS is free.
http://saveie6.com/
It is working as designed. ALL software, in the long run, is a race to the bottom in terms of pricing. There is always cheaper labor, and with zero tangible resource cost (unlike computers, houses, or cars) that means it eventually falls to the lowest price for labor - often for free. Open source accelerates that, as the product is basically free to begin with, and you HOPE you can find someone to pay you to manage it. This often results in UIs that are not heavily worked over for user friendliness, because that kind of sabotages the entire "hire me to make it work" push. And because it cannot be realistically deployed without that help - it becomes of little interest to consider unless there is a lot of already-built-up demand and use in the market. A vicious cycle, that results in poor or zero income for everyone in it. By design.
Browsing at +1 - no ACs, I ignore their posts. So refreshing!
... requires developers to develop software people are actually already paying for. I've thought long and hard if I could find investors to change the AAA videogame industry from the bullshit payment models and shit service to "buy to own" and "game development as a service model".
AKA there should be enough nerds for us to basically revive 8-bit and 16-bit type AAA games as a service model (aka we build games together that we ultimately all own and the code is open) for those of us above average incomes and who are true enthusiasts, basically take advantage of enthusiast interest in technology and turn it into a "sams warehouse club" for nerds. I was thinking about this with how costco has membership. If you want to do OSS then you're going to have to do so with a product that there is a known demand for. People don't want the boring shit, they want entertainment and shit that actually is valuable enough to pay for, aka you do the shit people want and use the funds from the shit people want to do more serious stuff.
For instance game development requires tools programmers that could make dents in the CAD and Image processing industries - aka take potshots at the crappy tools made by Autodesk and Adobe. Now this is not to say that many private sector products are bad, but Open source software versions developers have no discipline because they are free, when your job or your company is on the line with the quality of your work it forces you to stand up and take notice.
Now while windows had huge problems as we all remember from an engineering standpoint, you have to acknowledge the savvy of making computers user friendly enough for people to actually want to use them. That was microsofts genius. People used DOS and windows back in the day because the market was big enough for games and other apps.
There are people that make money of the software through services (eg. Canonical and Red Hat) and there are people that simply use it for free and perhaps even make a job out of it because they want to put the effort in it. FOSS works, no need for SaaS models where everything interesting sits behind a paywall and you get a barebones demo version of an "open source" product that they still want you to contribute to. If your software is interesting enough, people will use it, some people will make money of it. If a software project is so large that it can't exist without massive financial backing just to keep it maintained, your software has gotten too big and crappy and it should die. Linux as an example exists regardless of the money that is being pumped in it. It would still exist without it. Android wouldn't exist because it's crappy, it's overly complex and most of the interesting bits (whether it's the hardware on virtually all ARM chips or the applications Google provides) are closed source.
Custom electronics and digital signage for your business: www.evcircuits.com
The point of working on something for free is the work, not some sly method of monetizing something that is labeled "free."
Might as well write an article about how soup kitchen volunteers can sinergize to maximal returns with Soup as a Service open core pricing.
Trash writing from a human being with trash ethics.
I tried, and failed to understand why OSS economics is "broken". The linked article starts off with some problem with Redis, which apparently I'm just supposed to know whatever happened with it without further explanation. Then it goes off into a long winded meandering essay about a bunch of other things. Huh?
If you can't summarize a problem in at most a paragraph, I don't see how you can convince anyone there's a problem.
Stated succinctly, What the hell is this guy talking about, and why should anyone pay attention to him? I don't see any deep rooted problem with OSS economics. It seems to work just fine, and is better than ever.
If there's a problem, it's more with the niche OSS products that get created, then abandoned. Some obscure, but vitally important library for a small market. Of course, most of the time no closed source library exists for this either... but this is more of a problem with being in a small niche than a problem with OSS.
Whether via "open core" (where all the interesting bits are closed), or SaaS (when Joe User can't meaningfully deploy it) or services (where the UI needs to be hard so Joe User needs support)?
There is nothing wrong with how open source works, it works fine. The problem is what some people want from open source.
Everyone wants infrastructure software to be free and continuously developed by highly skilled professional developers (who in turn expect to make substantial salaries), but no one wants to pay for it.
Here's the conflict, people want something for nothing and that can work out sometimes but it means you are at the mercy of people you have no control over. That said, since it is open, you can hire people that you have control over to contribute to the code. The fact that few chose to do this demonstrates a failure in leadership rather than a failure in open source.
TL;DR: you dumbass MBAs are shortsighted nitwits who deserve to bear the responsibility for every security breach that happens under your blind-leading-the-sighted leadership.
Anons need not reply. Questions end with a question mark.
Developing software is one thing. Its basically a personal devotion to solve a problem. Its just the dev in his cave writing code to solve A problem.
Business is dealing with people, LOTS of people. Who all want different things, almost none of them are a very critical concern of the dev.
Writing code and making is free is not a business. You are not selling anything nor are you offering a service. You need that middle connection. That's why Business managers are a thing and get paid to do what they do.
Do you think the top level OSS based businesses would have gotten where they are if the devs collaborated more? LOL NO
How can we fix the broken economics of free developments in mathematics ?
I generally stop reading if the opening argument starts by stacking the deck:
"but apart from better sanitation and medicine and education and irrigation and public health and roads and a freshwater system and baths and public order... what have the Romans done for us?"
"I know I will be modded down for this": where's the option '-1, Asking for it'?
There should be a license that doesnt allow any commercial software to be distributed on the same device or medium.
The same concept that debian uses to seperate main, contrib, non-free
But it skeptical it can be fixed now, there is too much money invested in capitalizing on other people work, it would be like ending slavery...
>>If we take consulting, services, and support off the table as an option for high-growth revenue generation (the only thing VCs care about)
As if high growth, and the concentration of wealth to those that drive it, were worthwhile goals for all human endeavours.
About a year ago I started a new business providing voice services to small businesses in my area (VOIP).
I specifically opted to use FusionPBX because it's completely free, so getting started is no cost other than some time figuring out how it all works.
After I got more familiar with it, I came to find the author actually provides training courses for more advanced features.
He charges a fair price for classes, and who better to learn more advanced features than the person who writes the software.
I really like this model because its free for those who just want to fiddle, and in depth education is available for those who plan to make a living from it. The author makes some money to keep the project going.
He also maintains an online resource available for those who take the in person training to keep up on the changes on the platform called "Continuing Education" for a monthly fee.
That's not gonna fix it. If anything it's going to make it worse.
People keep viewing pay software as the software authors demanding money from users. It's actually the other way around. Users paying software authors is how they signal what features they like or want. That's how users influence the direction of future software development - the software authors want to be paid more, so they make changes or implement features and fix bugs that the users want.
Without payment, open source is basically a dictatorship. The software authors dictate what features to add, what bugs get priority, what new direction the software should take. The users are powerless. This gives contributors and especially maintainers an inflated ego, which makes them even more resistant to accept user feedback and suggestions. Paying maintainers from a foundation would just exacerbate this behavior by inflating their egos even more, and further insulating them from user feedback.. (The VLC developer eventually relented after a couple years, after much ass-kissing by users, and changed VLC so you could assign the mouse wheel to something other than volume.)
If you want to fix it without having users pay for open source software, then I can think of two ways. Either you need to eliminate the egos of the programmers and maintainers, which realistically is never going to happen. Or you need to set up a system where users can pledge a bounty payment for when the project implements a feature or fixes a bug the user wants. The payment should be held in escrow (refunded after a certain time to encourage timely action), to be awarded to the open source project only if the user-requested feature is added or bug fixed. That would fix open source by giving users a say in the course of software development again (other than ass-kissing), without the stigma of requiring all users to pay to use the software.
That'll turn open source software from whatever the hell the programmers and maintainers want it to be. To something which the users actually want, and which addresses their needs and requirements. The way things are right now, open source is frequently throwing free food at users when what the users really want is water. With the project maintainers and contributors ignoring user pleas for water because making food is more fun or fulfilling.
>>> If we take consulting, services, and support off the table as an option for high-growth revenue generation
>>>
What about Redhat?
If there is any 'broken economics' it is that some disciplines appear to be more forthcoming and generous with time than others. For example, I see a lot of talented coders on projects but no evidence of Technical Writers, UX Designers or possibly Business Analysts... eg lots of volunteers for the cool bits but not so many for the boring stuff, like writing help guides! SoapUI is my favourite example of a project that has very useful free functionality and also comprehensive tutorials, guides etc - plus a commercial offering that gives a more enterprise-level product.
Isn't this already exactly how the Apache model works (or used to)?
Multiple companies each sponsored a developer to have a seat on the Apache project board giving them a controlling vote(s) on what direction to develop the project in next. All the companies contributing, and everyone else not, benefited by the resulting Apache project results. The sponsoring companies got their needs met sooner (and at all).
Many people like the "fun" of making a baby but not so many are prepared to accept the responsibility of supporting the child into adulthood. Producing software has many of the same attributes. Writing software is fun, creative. Fixing problems is a chore.
Continually updating your "progeny" to stay compatible with changing O/S and API requirements is boring. Making it compatible with the other programs in the environment is difficult and making it usable, well-designed and intuitive needs a rare skill set.
To encourage "good parenting" in FOSS, we should place more value on projects that are kept up to date. Where the authors show commitment to fixing bugs, new releases (although most FOSS could do with far fewer new features and more time spent on improving what is already there) and making it easy for the user to use - documentation, examples, explanations.
politicians are like babies' nappies: they should both be changed regularly and for the same reasons
I dislike the entire debt-based economy and has for a while. One of the problems with it is that it is based on extracting more dollars from "consumers". You can also see this in DRM for music and video for example. In my Google DoubleClick essay I mentioned Novell and Sun as examples.
If you want a great example of why the open source ecosystem is broken, look at how we value code versus documentation. We expect to get ten of thousands of lines of code - someone's labor - for free but will then pay O'Reilly $50 for the book explaining the software? So writing code has no value but writing a book about the code does? Yes, open source economics is broekn but it is really the value system of the programmers that has been warped.
The fundamental flaw is that in economics-as-we-know-it things are done purely for profit, not for getting things done. Which all too often leads to the wrong things being done, or the right things being done in a disastrously wrong way. For the economy, it's just as well as long as there is profit.
OSS economics now has the "problem" that OSS is about getting things done, not about profit. And it's about, if possible, getting the right things done the right way.
The clash between OSS and economics-as-we-know-it is indeed a good reason to think about what is really going wrong, and how it can be repaired. Hint: to bend and break OSS economics until it somehow fits into profit-based economics is probably not the answer.
Couldn't find any similar reference, and not motivated by today's Slashdot to search more carefully. However I will repeat what I believe to be the best approach:
Use a CSB (charity share brokerage) to help manage the OSS projects. Software gets funded only when enough donors are willing to "buy" the charity shares for the project. The brokerage makes sure that the project proposals are complete including the success criteria, and evaluates the finished projects to report the results to the donors and the public. The projects can be quite flexible including (1) new software, (2) new features for existing software, (3) support projects, (4) ongoing cost projects (as when a server is required to support software or features) and others. The goal is to let the donors focus on the work but the CSB earns a fraction (perhaps a tithe) from funded project in exchange for providing the project management support.
Same basic idea can be modified to support journalism and several other applications. As far as I know, it is not loose in the the wild, but I'd be quite pleased (and surprised) to learn about (1) an existing implementation or (2) your better idea. NOT interested in the usual snideness, especially from people who haven't even read what I wrote. Time's up, but ADSAuPR, atAJG.
Freedom = (Meaningful - Coerced) Choice != (Speech | Beer^2), and sad sock puppets' bad mods avail them naught.
It's the ability of VCs to monetize open source software.
Not a problem for me, or for established businesses either.
Post may contain irony: discontinue use if experiencing mood swings, nausea or elevated blood pressure.
I've worked full time getting paid quite well to work on open source.
Open source doesn't mean developers work for free. They get paid by the same people who normally pay them. It's just that cooperation makes them more efficient, so they can potentially be paid more. I'll give you one example that I have allot of experience with.
Universities (and others) are offering a lot of courses online these days. Suppose 40 institutions want to offer online courses. They each need an online system to manage those courses, an online campus. Suppose having such a system is worth $60,000 to each school. A programmer can build it in 500 hours at a cost of $30,000, so each school does it. (It'll cost them $30K and be worth $60k of value). Forty schools each spending $30k is a total of $1,200,000 spent. Later when they want other features they'll spend more.
Alternatively they can cooperate, building a modular online campus system that works well for all of them. Maybe that costs four times as much as building a system for one school - 2,000 hours, or $120,000. That's 90% less than it costs for them to each build their own. Later, when a school wants a new feature, it was already built for a different school. They just install it.
In the first scenario, the programmers each took 500 hours to provide something worth $60k. In the second scenario, 40 programmers at 40 schools spent a total of 2,000 hours. That's 50 hours per programmer to provide the same $60,000 of value to their school.
Who do you think can be paid more per hour: a programmer who takes 500 hours to get the job done, or one who gets the same job done in 50 hours?
Far from "working for free", by cooperating on an open source project we were able to provide the same value with 90% less work from each of us, so we were much more valuable and highly paid than a programmer who doesn't cooperate and use open source effectively.
You wrote #4 (we need to use the software, selling the software isn't our business), then you seemed to completely forget about that when you wrote your conclusion.
Cooperating with other organizations who have the same needs rather than having 100 different companies each build their own is better for everyone. See:
https://slashdot.org/comments....
The bits don't just come in, set themselves up, configure themselves, integrate themselves with your existing business systems, customize themselves to your needs, and maintain themselves. The software itself is one ingredient in solving a business solution with a software system.
I've worked full-time, being paid well, working with and working on open source software for many years. Some companies hired me to handle their open source software, customizing and maintaining it, being the expert on it. See this for more:
https://slashdot.org/comments....
However, at least 50% of companies and schools were nervous about having just one or two in-house experts and decided to instead contract with a third-party company to deploy, customize, and maintain the open source software. Those third-party companies have teams of developers and technicians which do nothing but work of open source software.
We already have several ways to "pay for" free and open software.
1) Dual licensing: Several high profile projects, including QT, Ghostware, and MySQL use these for benefiting both closed source commercial, and open software. There is even research on this topic: https://www.sciencedirect.com/...
2) "Patronage": Since middle ages artists depended on wealthy "patrons" to commission their work. The end result was "open" in sense they were usually presented in cathedrals, or museums, but the work was paid for powerful individuals. Now we have companies like IBM, RedHat, and even Microsoft sponsoring open source projects that benefit all.
3) Amateur work. The work itself has an intrinsic value in terms of intellectual gratification. Many people will contribute to open source projects just because they can on their own time.
The system actually is tested and works. It has produced great results, and used by billions of people today. Of course new ideas for patronage could arise, but it does not mean the current ones are broken.
MIT devs believe they are developing solutions to problems. GPL devs think they are developing products. Only one of these groups has an issue.
I spend 90% of my time getting paid to write MIT licensed software. The economics work fine.
If there is specific software you are interested in, I can post some companies that provide the business services around the software, while supporting the core development team.
As one example Moodle is an open source learning management system (online campus) with 80 third-party companies who can set it up, configure and customize it, write custom modules, host it, and maintain it, provide training, etc (whichever more of services your school or company needs). Each of these contributes funds back to Moodle HQ
https://moodle.com/partners/
OK, I'm asking in good faith... Is the AGPL v3 license in accordance with what the owner of iTextSharp says? There appears to be a huge disconnect in the licensing terms for AGPL v3 vs. what iTextSharp says.
The ambiguity is apparently so sharp that Google forbids usage of any open source software using AGPL:
WARNING: Code licensed under the GNU Affero General Public License (AGPL) MAY NOT be used at Google.
The license places restrictions on software used over a network which are extremely difficult for Google to comply with. Using AGPL software requires that anything it links to must also be licensed under the AGPL. Even if you think you aren’t linking to anything important, it still presents a huge risk to Google because of how integrated much of our code is. The risks heavily outweigh the benefits.
Do not attempt to check AGPL-licensed code into google3 or use it in a Google product in any way.
Do not install AGPL-licensed programs on your workstation, Google-issued laptop, or Google-issued phone without explicit authorization from the Open Source Programs Office.
Given this confusion, does anyone wishing to speak about the economics of open source want to actually talk about the legal amgiguities of open source licensing?
I don't know any off hand since I'm not really involved with Gimp. Here's the developers mailing list, where you can ask the people who build Gimp about who can help you, people and companies providing paid support, training, customization, or whatever you want.
https://mail.gnome.org/mailman...
I'd first get clear about what you want. Training? Do you want someone to write some custom modules for Gimp? Do you want an on-call Gimp expert?
It varies. I would say I my experience, most often a bare-bones software exists, a starting point, before multiple people start working on it professionally. Someone, either a company or person, writes something small and simple to "scratch their itch" (solve their own need), then other people find it useful and a community grows up around it. That can easily mean someone puts in 20 hours building a script to load X onto Y hardware, then eventually others spend thousands of hours building it into something far bigger, very often programmers being paid to add another feature or module.
WordPress is a reasonable example here. It was a VERY simple set of scripts. Then one guy with a web site paid anothet guy $200 to add some feature to WordPress. Another web site owner paid another programmer $150 to improve something else. The company I used to own sent out a lot invoices for $100-$400 for improvements to WordPress or WordPress modules. We and hundreds of other companies turned a simple blog script into something more complex and powerful than some operating systems.
One software project I was heavily involved with started as three guys on a forum cooperating to get about five lines of of .htaccess absolutely correct. Someonw offered me, Mike, or Chuck $75 to expand it to fit their need. It grew to a over dozen lines of .htaccess configuration, then someone added a 50-line Perl script. Years later, it was a very capable security system used on hundreds of thousands of web sites, with plug-in modules available from multiple companies.
I've been involved in multiple open source projects which followed the Netscape-Seamonkey-Firefox model. A company producing proprietary software releases their software as open source, perhaps as they either go out of business or drastically reduce their work force. Firefox doesn't have a ton of companies selling work around Firefox and Gecko but there are some and there could be more. If you want something built on Gecko, or you want someone to write and submit a Gecko patch to support something you need, there are people who will do that for you.
Several companies sell OpenWrt based routers and otherwise contribute to the OpenWrt ecosystem. This is an example of another model. Linksys created the router firmware, several companies are now contributors. The Linux kernel existed before Linksys put it in routers, Linksys did a lot of work to produce a router firmware which included the kernel.
So there are many different paths. One path includes an individual programmer writing the first part for their own reasons, unpaid. Several other paths start with programmers getting paid from the beginning.
> I'm trying to understand the business model of whoever it is that writes the free software in the beginning.
My first reply may not have been well-focused on answering this question. I did some tangents. The main thing I'd say about "whoever it is that writes the free software in the beginning" is this:
The the BEGINNING, the software may very well take less than an hour to write. Someone whips up a simple script to solve their problem and posts that script on a forum.
If several people on the forum find it useful, you can easily see how one person might add two or three lines to fit their particular model or version. Someone else notices that the 60-line program has a bug on line #22. An open source project is born - no business model needed.
Venture capitalism is completely unsustainable and broken.
> Or... An entrepreneur invests in a few programmers (not 40, which is overkill), writes the software, invests in advertising and sales, sells it
You as the customer can pay for advertising, sales, and the entrpeneurs profit, or you can just pay only the cost of the developers and skip the salesmen in their $1,000 suits. Would you rather pay for sales and sales marketing, and license audits by your vendors, or would you rather your money go to better software?