Open source, closed wallets, big profits – nobody wins the OSS rock, paper, scissors game

There's much talk of the Open Source Sustainability Problem. From individual developers to Google's White House lobbying, the issue seems simple but intractable. Is the willingness of volunteer coders a solid enough basis for the long-term health of essential infrastructure? This is, of course, balderdash. It's not an open …

[Pascal Monett]

Not a bad idea, but

Here's what is likely to happen :

1) OSS coders who want to get paid will get fed up not being paid anything, and will simply stop updating their code

2) Companies using said code will start fretting about outdated code, or will wait until everything breaks to wail about how they didn't see that coming and OSS is not reliable

3) Somebody will step in and say "for a price, I can update this", and they will require a licensing deal, which companies will gladly fork over

4) Situation normalizes to "free" OSS code, not paid for ever, and "commercial" OSS code, which requires a license and gets updates, but can be forked and any approved coder can contribute (how that happens needs to be determined)

I don't believe in unicorns and I doubt very much that companies are going to wake up to anything until it slaps them in the face - especially when it means spending money that does not go towards CEO/board bonuses.

[Anonymous Coward]

Re: Not a bad idea, but Not the real problem

Ultimately, some way will be found to get money to the random person in Nebraska, be it through tithe or tax or license.

But the real problem is not keeping code actively maintained, it's reviewing old code to identify problems that need to be corrected. Developers, in general, want to look forward, not backward, and the result is that problems are found after they are being exploited as attack vectors or, worse, break the internet.

[Anonymous Coward]

Re: Not a bad idea, but Not the real problem

> But the real problem is not keeping code actively maintained, it's reviewing old code to identify problems that need to be corrected. Developers, in general, want to look forward, not backward, and the result is that problems are found after they are being exploited as attack vectors or, worse, break the internet.

Open Source is a system based on trust: as with many other things, it was created at a time when black-hat hacking wasn't really a thing; it was certainly long before the exponential growth in processing and networking capabilities made it ever easier to perform bulk attacks.

And for all that the "open" aspect of open source has a number of benefits, these very much rely on people having the time and motivation to actually review the code.

Sadly, there's far less money floating around for white-hat reviewers who want to fix things, as compared to black-hat hackers looking for exploits they can monetise.

So what we've ended up with is people building stuff atop vast mountains of libraries, with dozens of layers of dependencies and abstrations.

For instance, we use node for our JS libraries, and the node_modules directory contains over 170mb of libraries and associated code/data.

Practically, there isn't a way for any individual company to review all that code - or any changes made to it over time. And while you can do things like version locking, that carries it's own issues.

And there's further edge cases, beyond that.

For instance, we had someone working for us, who built a module to parse some stuff, and decided to publish it as a open-source module.

Which is well and good, but since they've left the company.

I'm not expecting this person to try and haxxor the gibson, or anything similar. But as a general principle, the idea of automatically pulling in third party code produced by someone with insider knowledge definitely makes me nervous...

[unimaginative]

Re: Not a bad idea, but

"4) Situation normalizes to "free" OSS code, not paid for ever, and "commercial" OSS code, which requires a license and gets updates, but can be forked and any approved coder can contribute (how that happens needs to be determined)"

How can you reconcile "requires a license" with "can be forked"

I think the solution is dual licensing. GPL or AGPL together with an alternative paid for license. Some people already do this.

[sabroni]

re: This is, of course, balderdash.

Yes. Yes, it is.

Hoping everyone will start to play together nicely is not a plan.

[jake]

Re: re: This is, of course, balderdash.

See: Cats, herding.

[Arthur the cat]

Re: re: This is, of course, balderdash.

Hoping everyone will start to play together nicely is not a plan.

It is. It's a cunning plan. In the Baldrickian sense.

[TDog]

Tragedy of the commons

And how does this prevent the above? Bad users will make a relative competetive advantage by not contributing with no loss to themselves. There is a reason why taxation is compulsory and also why most of accountants spend significant time understanding and using the most "efficient prayers" to the system. Some small subset of creators find their products make more and more demands on their time and are used excessively, beyond their expectations. Most don't. The religious model given is in many ways apt; few gods are bothered by the prayers of many.

And I have yet to notice as stringent expectations placed on gods as maintainers of FOS software. In fact it is close to the opposite - when things go wrong we call it "an act of God". When software goes wrong it can be and has been called any damn thing whatever but there is an expectation of responsibility and FIX IT NOW.

[Greybearded old scrote]

Re: Tragedy of the commons

Thing about the tragedy of the commons, it's a fairy tale. The real commons lasted for centuries. They weren't salami-sliced by people pushing their luck, they were seized outright by the local lords.

Consequently only those who went to business school believe that this is a sensible way to behave. (Well, OK a few exceptions.)

Now, how do we dispose of those damned astrologers in the economics department?

[Arthur the cat]

Re: Tragedy of the commons

business school ≠ economics department

The term "the tragedy of the commons" was invented by a biologist, Garret Hardin. The idea was debunked by an economist, Elinor Ostrom, who won the 2009 Economics Nobel for doing so.

Economists aren't a problem(*), it's the cargo cultists in business schools and the managers with MBAs they produce that should be disposed of. All the MBAs I've worked with have been worse than useless, with their common sense surgically removed and replaced with whatever damn fool idea was fashionable at the time.

(*) Politicians' {mis,ab}use of their work is.

[Anonymous Coward]

Re: Tragedy of the commons

It wasn't my my intent to debunk your comment but it made me curious enough to lookup https://en.wikipedia.org/wiki/Tragedy_of_the_commons where it says

The concept originated in an essay written in 1833 by the British economist William Forster Lloyd, who used a hypothetical example of the effects of unregulated grazing on common land (also known as a "common") in Great Britain and Ireland. The concept became widely known as the "tragedy of the commons" over a century later after an article written by Garrett Hardin in 1968.

before going on to say

Although open-access resource systems may collapse due to overuse (such as in over-fishing), many examples have existed and still do exist where members of a community with regulated access to a common resource co-operate to exploit those resources prudently without collapse, or even creating "perfect order". Elinor Ostrom was awarded the 2009 Nobel Prize in Economic Science for demonstrating exactly this concept in her book Governing the Commons, which included examples of how local communities were able to do this without top-down regulations or privatization.

if !w is to be believed it was invented by an economist, popularized by an ecologist and later refined by another economist. The full article is rather informative.

[veti]

Re: Tragedy of the commons

The description on Wikipedia even specifies that the TotC is something that happens when there is no effective community governance. That would be the refinement added by Ostrom.

I think the lack of community governance can be assumed in any domain related even tangentially to software.

[Greybearded old scrote]

Re: Tragedy of the commons

Prolly too late for anyone to see my reply, but here goes.

My point was based on the idea that Economics and Business have a similar relationship to Science and Engineering. Except for the latter pair making considerably more sense.

[JoeCool]

First, you need to demonstrate "competitive advantage" ....

from not paying for FOSS.

Large companies are internally, some of the most inefficient organizations known to man. Externally, they are ruthlessly capitalistic organisms. Quantitavely, paying for sowftware won't affect the bottom line; it's simply a matter of priorities. That's part of the refence to CEO salaries.

[jmch]

Re: First, you need to demonstrate "competitive advantage" ....

"Large companies are internally, some of the most inefficient organizations known to man"

Absolutely this. It's incredible, when you are inside the walls and can see beyond the shiny website and glitzy marketing, how much of it is all held together by string and gaffer tape. How many projects are delayed leaving a team twiddling their thumbs for months because another department they depend on is trying to save a few thousand quid on a contract. How key people spend months arguing about architectural ideas while nothing gets built.

etc etc

[Lusty]

Re: First, you need to demonstrate "competitive advantage" ....

100% agree. I often think that the time taken to decide a server name is the square of employee count in man hours. A one man IT team will just call the system Bob, a multinational team needs a whole schema and justification with a “discovery phase” to make sure nothing is missed.

[Anonymous Coward]

"A one man IT team will just call the system Bob"

A one man team will have a few servers to take care of, a large company may have hundreds if not thousands in different locations...

[Anonymous Coward]

Re: First, you need to demonstrate "competitive advantage" ....

And as one who has done quite a lot of work for them, the BBC is one of the most spectacularly inefficient organisations in existence. The level of waste there is depressingly amazing. When they say "Any cuts in funding will impede our mission" they mean "We can't send 300 people to Glasto this year".

[Anonymous Coward]

Re: First, you need to demonstrate "competitive advantage" ....

As someone who's also done quite a lot of work for Sky, ITV and several independent studios...

The BBC is very efficient in comparison, especially as Sky et al don't have malicious and/or incompetent politicians meddling anywhere near as often.

[Dan 55]

It seems the Internet still hasn't got the hang of payment for content (whether it's an essay or code, it's all the same thing), which suits the big corps very well thank you so why would they want anything to change or come up with a payment system?

Unless you count the Brave wallet and tip jar, which is crypto-currency and therefore planet burning, so that's not a solution either.

[jonathan keith]

Micro-payments

Axate seems to be making progress on the whole micro-payments for content thing, which I believe is the only thing that will ultimately make the web both private and self-sustaining.

[SImon Hobson]

Re: Micro-payments

I'd be curious about the reasons for the downvotes. But I have to admit, I'd not heard of them.

[jpo234]

The decision to use a certain Open Source component is usually up to individual developers who usually don't have budget authority. They just do it. If there was some kind of contract with payments involved, things would slow down to a crawl: That would require an aqucisition process that might not even be suitable for most projects (a Patreon account doesn't fit).

[Yet Another Anonymous coward]

No it would be quite simple

The open source contributor just has to pay to take the company software buyer on a golf trip....

[steelpillow]

Governments

Government departments are becoming increasing users of OSS, and the more enlightened senior civil servants are even becoming vaguely aware of the fact.

But often, they have arcane rulebooks which forbid them from helping to develop or fix a given product, or even OSS in general, for some absurdly obsolete reason (Insistence that everything their staff write during working hours must be Crown copyright, for example).

Yet all around the world, governments spend billions on developing and supporting vital infrastructure, while their taxpayer-funded research institutions frequently help out private enterprise in their side of the development.

A little realism and engagement, pushed down from our political and Departmental leaders, could go a very long way in making OSS deliver on Departmental needs - which include such goodly things as long-term stability and support.

We have seen it on rare occasions, such as when the American NSA released SE Linux. But it needs to become the norm, not the exception.

[jpo234]

Re: Governments

It's not just Governments. If money and contracts get involved, things get complicated everywhere. This adds so much friction, that it suddenly might not be worth it.

In public organizations that would usually trigger requirements to issue an open tender. These requirements are there for a good reason, but they don't fit the open source model.

[xcdb]

Re: Governments

And herein lies the ultimate problem with financing OSS.

Hell, even if govt 'solved' the problem by requiring fully supported software stacks, you'd end up with more or less the same arguments of unfairness or horribly-structured dependencies.

Look what has happened with broadcast music - most songs ended up around 3 minutes long 'cos that's what the station was prepared to play. Streaming services (spotify etc) pay for music on a different basis and so - magically - artists have changed their offering to maximise revenue and so tend towards having many more shorter tracks. Leftpad would only be the start...

[alain williams]

Re: Governments

most songs ended up around 3 minutes long 'cos that's what the station was prepared to play

I though that 3 minutes was/is the play time of the vinyl 45 RPM 7" single.

[Anonymous Coward]

Re: Governments

3 Minutes is probably the average but playing time on my small collection of singles ranges up to 5 minutes ("Have You Seen Her", by the Chi-Lites).

[jake]

Re: Governments

The history of vinyl is interesting enough that any self-respecting geek/nerd should probably look into it ... It is a form of WORM drive, after all :-)

Suffice to say here that the "3 minute single" was an artifact of the recording technology of the late 1800s cylinder media. This time meshed with the later 45s 7" format and its attendant recording technology.

Interestingly, Tom Petty once claimed that without this "limit" he would never have made it as a recording artist because his attention span was so short that he couldn't write or perform tunes much longer than that.

Better grove cutting techniques lead to roughly doubling the recording space available (for you pedants that's quadrupling, when you take into account mono vs stereo). The longest "single" in my collection is Dylan's "Like a Rolling Stone", at 6 minutes even.

_{That's Bob, not Thomas ... given El Reg's readership I see potential for confusion.}

[martinusher]

Re: Governments

Don't forget the 50 years or so of 78 rpm records. These were typically in two sizes, 10" and 12". The run time of a typical 10" disc was about 3 minutes which is where the 7" RCA discs come from -- like the 78s they were focused on autochangers that would feed stacks of discs to be played, thus making up albums (and explaining the big hole in the middle of a 45).

(The RCA 45 was sold as higher fidelity than the CBS 33. Its not just the speed difference but the more accurate tone arm tracking due to the small disc.)

[Ian Johnston]

Re: Governments

Plenty of autochange record players worked with small spindles. RCA really just wanted to sell players on which you couldn't play Columbia records.

[jake]

Re: Governments

Yes. Like many such things in this world, the intentional incompatible differences came from the diseased minds of Marketing.

[Denarius]

Re: Governments

So records were initially made by cutting grooves on nice round slices of tree from forest groves, Jake ?

Thre was I thinking 3 minutes was all a teenager could last doing one thing.

[jake]

Re: Governments

Don't be silly, cellulose foam scratches, warps and cracks far too easily. They started with a more sensible metal foil over cardboard, but soon shifted to a damp-proof wax.

[tiggity]

Re: Governments

I have a single at over 17 minutes (Grendel by Marillion) - though it is a 12" single

[jake]

Re: Governments

I think you'll find that is what we called an EP, not a single.

[Irony Deficient]

I thought that 3 minutes was/is the play time of the vinyl 45 RPM 7″ single.

George Harrison’s Isn’t It a Pity (the B side of My Sweet Lord) is 7:10 — I think that that was published as a 45 RPM 7″ single in late 1970.

[jake]

Re: I thought that 3 minutes was/is the play time of the vinyl 45 RPM 7″ single.

Yep. I had never noticed that ... I learned something new today, thank you. Have a beer :-)

Note, however, that in the original US release[0] from 1970 it's not the B-side, rather it is the flip side. That particular single was released as a true double-A side, with neither claiming A or B ... or 1 or 2 for that matter. The later British release called MSL "Side 1".

[0] Only played twice ... once when I bought it[1], and then promptly filed it away as "garbage, but collectable", and once again many years later when I digitized my entire collection. My British copy is unplayed to this day. I don't think it's even been out of its sleeve until just now.

[1] More like maybe two quarters ... I seriously doubt I listened to the entire cut of either side back in '70. Did I mention it's garbage? Both sides. Complete trash. Stoner music, at best.

[Anonymous Coward]

Re: Governments

@xcdb

Quote "Look what has happened with broadcast music - most songs ended up around 3 minutes long 'cos that's what the station was prepared to play. Streaming services (spotify etc) pay for music on a different basis and so - magically - artists have changed their offering to maximise revenue and so tend towards having many more shorter tracks".

Seriously? Where are you getting your "facts" from? Tiktok?

Here is my Spotify. spotify:user:ishtiaq (may need an upper case "s" am not sure)

Go on, look, there are plenty of tracks lasting longer than 3 minutes.

I would tell you how the three minute thing came about, but I can't be arsed. But it was nothing to do with radio stations.

[Yet Another Anonymous coward]

Re: Governments

A certain ex-member of PinkFloyd was notorious for splitting his contribution into 2 parts so he got double the royaly share

[Kane]

Re: Governments

"A certain ex-member of PinkFloyd was notorious for splitting his contribution into 2 parts so he got double the royaly share"

Careful with that axe...

[Kane]

Re: Governments

...Eugene...

[Anonymous Coward]

If you're a OSS dev, do you try to make it possible to get paid?

(AC to protect identity)

The large company (multi £billion turnover) I work for has a whole department which approves / signs-off on architecture and, particularly, the s/w choices made therein. No project can go ahead without this architecture being approved and rubber-stamped. On the face of it, this is to try to ensure consistency and promote re-use (etc etc). All of the software we write is for consumption by external clients (other than the obvious internal tooling etc)

In reality, this entire department is staffed by lawyers and procurement people. There are, as the article says, carte-blanche approvals for certain types of licence (Apache, MIT) but anything with a copyleft can be a bit of a problem to get approved ([L]GPL especially).

This department is 100% concerned with "If our client sues us, do we have somebody to counter-sue or deflect to?" and "if we use <X> on this project, can we be sued for violating IP developed on previous projects"?

These guys (and I interact with them all the time) would much rather pay for software than use open-source - because paying for it comes with certain contractual obligations which can be enforced with the help of m'learned friends. They frequently lament that even if they wanted to pay for a vital piece of software, how do they cut a cheque to developer_and_his_dog@github when there is no commercial entity to deal with? There's no way of being able to expense any donations that I might want to make, and no way for the company to do it.

Apache have, for many years, something vaguely resembling the skeleton of a model that could work. Incubating projects vs top-level projects. Perhaps if there was clever lawyers at ASF who could craft a licence which required payment when >£threshold is reached for a top-level project and the ASF can disperse the payments - because the ASF are easy to deal with commercially. I use 'ASF' as an example, it doesn't have to specifically be them. The YouTube model (I know, I know) might also have similar inspiration: you're only allowed to "monetize" videos when <thresholds> are reached

Companies won't volunteer, but that doesn't automatically mean that all of them will try to avoid payment - legal and risk and procurement people in these companies actually want commercial paid agreements in place but there's no way of doing it

Until OSS devs and maintainers have, or are part of, some commercial structure which the byzantine procurement of megacorps can deal with - i.e. is able to handle how these megacorps *buy* - then this situation is not going to change any time soon.

TL;DR if you want to get paid for the OSS you develop and maintain, then you have to be able to 'engage commercially with the people who pay you, backed by a licence where there's sufficient compulsion for them to pay you.

[Anonymous Coward]

If you're a corporation using OSS, why have you made it so hard to pay the devs?

Not only do you want them to write your software for you, you also want them to run themselves as a corporate entity because you can't find a way to pay them. Why should the maintainers bend to meet the needs of the corporation? Why can't the corporation change its ways?

TLDR; if you want to derisk your software pipeline, then you have to 'engage non-commercially with the people providing it for you', backed by regular income so there's sufficient compulsion for them to continue.

[SundogUK]

Re: If you're a corporation using OSS, why have you made it so hard to pay the devs?

"Why should the maintainers bend to meet the needs of the corporation?" Because otherwise they're not going to get paid.

[Michael Hoffmann]

Re: If you're a OSS dev, do you try to make it possible to get paid?

You save me from posting effectively exactly this:

The last 5 organisation I worked with or for in the last 3 or 4 years all pursued what you described: OSS was allowed by *license type*: as long as they could use the hell out of it, make money with it without paying *and without having to give back*, it went on the pre-approval list. Anything else went through legal and frequently denied.

Gently pointing out Linux and its license caused blank stares: in the heads of beancounters it has fully become commercial software "because we pay license fees, don't we?" (they meant subs to Redhat...)

[Charlie Clark]

Sounds very much like the music industry

The proposal sounds similar to how the music collects and distributes royalties. This works better in some countries than others.

However, the biggest problem is reconciling the open source approach with product liability when the two are, in a sense diametrically opposed.

[Cederic]

Re: Sounds very much like the music industry

It'd be even more broken than mandatory music royalties. After Linux sucks up 90% of the revenues (on the grounds everyone and everything uses it), the Apache foundation, Red Hat and 2-3 others take 90% of what's left. After that the people distributing will say "It's not cost effective to try and recompense everybody else" (and be entirely correct, because transaction costs will be higher than the payments) and nobody will get anything anyway.

Then Governments will step in, make it compulsory, and you'll end up having to pay some profit making royalty distribution service money to use your own software and get nothing back.

[jake]

Re: Sounds very much like the music industry

"The proposal sounds similar to how the music collects and distributes royalties."

The music industry distributes royalties? News to me. They collect them, sure ... but distribute? Not so much. Look up Janis Ian's comments on the subject ... Try this link:

https://archives.janisian.com/reading/internet.php

Or was that your point?

[Charlie Clark]

Re: Sounds very much like the music industry

That was my point: sounds great as an idea but experience with it varies a lot. But the liability issue is also something that needs addressing; first for software, in general (I've got a € 30000 bill I'd like to send to Microsoft for their fiasco that is Exchange) and then for open source. If accepting payment means accepting liability then lots and lots of people will wisely say no.

[Anonymous Coward]

Re: Sounds very much like the music industry

I was going to make the same comparison. If you're an organization which uses music - from the largest concert venue down to the smallest hair salon which leaves the radio turned on for its customers - you have to pay a public performance licence fee. You can play whatever music you like, but the money collected by the PPL is in principle distributed to artists based on usage stats (radio plays, streams etc). In practice most of it goes to record labels rather than artists. This means there's also a huge secondary industry around rights management, analysing and tracking usage, all taking their own cut.

I suspect that any OSS usage-based fee collection would end up with similar organizational bloat. Take IANA, which performs the job previously done by a single person in his spare time (Jon Postel): this is now a multi-million-dollar organization which exists primarily for its own self-aggrandisement and for the benefit of its many committee members.

I also observe that every large software project eventually ends up having to have a "foundation" or other organization to manage it. Therefore I expect the vast bulk of money collected would be distributed to these foundations, and not to individual software authors or contributors. How much will trickle down to those authors? Probably not much.

[aaaa]

Re: Sounds very much like the music industry

The French music industry is an interesting example of how this could be done:

https://www.npr.org/2021/01/11/954994402/how-france-is-helping-its-artists-during-the-pandemic

Things like CD's are taxed and the money distributed to professional musicians who are not currently working, because the French society values having musicians.

Similarly if a country valued having software developers/FOSS, they could tax all commercial software development (a new VAT rate maybe, or like national insurance for staff employed as software developers) and use it to pay 'unemployed' software developers (FOSS developers).

So Red Hat staff would still need to be paid by Red Hat. Apache foundation staff would need to be paid by Apache Foundation. It's just the 'volunteers' who don't have enough free hours left in a day for actual employment who would benefit, ie: the people who do the actual coding.

The devil is in the detail - but it could be done.

[doublelayer]

Re: Sounds very much like the music industry

That's not a good idea in my opinion. In order to do that, you run into lots of definitional issues. For example, which of the following people gets funding, and how much in each case:

1. A developer who writes a project important to many people in their free time, but has a job at a tech company writing other code as well.

2. A developer who maintains code, fixing bugs and the like, but didn't create the project and doesn't add new features.

3. Someone who writes documentation only.

4. A translator who localizes open source software.

5. A person who spends a lot of time on a project they came up with, with few other users.

6. A person who commits occasionally, but not very much.

7. A person who frequently switches what projects they're working on.

I don't know how France's music system works, but that already sounds like a bad idea (I write music as a hobby, but I don't think you should pay me to do that). Open source contributions are perhaps even harder to assign a monetary value to.

[jake]

I have been contributing to the FOSS world since before BSD was BSD. Quite frankly, I have never thought about getting paid for it for one simple reason: It doesn't matter.

Read that again, it's important: It doesn't matter.

I wrote code, created patches, chased down bugs, wrote documentation, and all the other bits & bobs that go into FOSS because I am extremely selfish. I wanted it to work for ME, my way, in my time. Once it worked the way I wanted it to work, it solved a problem that I had, which more than paid for the time and effort that I put into it.

Then I released it to the wild, without caring if anyone else needed it. It's MINE, it scratched my itch ... now, if you have the same itch feel free to make use of my scratching post. No point in you re-inventing the wheel to do the same job ... and better, it frees you up to work on something to fix another itch.

Thankfully, other people have many other itches. In aggregate, we have created something useful. Without money bags getting under foot.

[Charlie Clark]

And BSD shows how things can and should work… then the VCs discovered the internet, invented the dot.com hype and basically broke everything: since, many business models have been based on some form of excessive arbitrage.

[Bitsminer]

A lot of FreeBSD changes are freely given by commercial companies using it. Eg Netflix but also many others.

Those companies have figured it out. Let's encourage the others to adopt the same method.

[Michael Habel]

Thats great, and all... but, some of us like to eat, and keep a roof over our heads. This pholosophy only works (or, so it seems to me), if your tackling your seven year "itch", as a hobbiest. Otherwise it completely passes me by, as to how you could make a living at it.

[doublelayer]

Depending on how you want to make a living, you have to decide the way you want to do this and structure your work accordingly. If you use a license that says the user can use, modify, distribute, and sell your code without receiving permission from you, then they're not required to pay you. If you use a license that says that they must pay you to do those things, then you probably won't get assistance from others. If you started with others' code, you may be required to adhere to their license. If voluntary donations aren't sufficiently secure for your comfort, you may want to do something that gets you a more stable income stream, but the consequences of your choice are your responsibility.

I have a library I wrote to solve a problem for me. I considered releasing it as open source, but I didn't. One reason is that there's already a better version out there (mine is only better if you have to run on an embedded device with very little memory). But another reason is that I thought maybe I could turn this into a commercial project. In order to leave that option open, I didn't give my work away. If I did, I shouldn't expect companies to come to me and pay me for the thing I just gave them for free.

[Michael Habel]

Which again brings us back to the front. As most of these weekend bedroom keybord warriors types have their day jobs, and or other responsabilities, ost of which will either sooner or later take them out of the game, and such software that existed, either gets flushed, or forked off.

[Charlie Clark]

People making a living from working on open source is likely to remain forever an exception. It's not where open source came from and it's not what it's about. Research, particularly in academia, is a reasonable comparison. A lot of it goes on without specific budgets but some things won't happen without them. And the history of the lack of research in vaccines and antibiotics over the last 50 years demonstrates similar weaknesses in our approach, with the bugs and viruses essentially discovering unpatched bugs. And, no, despite the current hype around COVID, we haven't got a solution for this either.

[Anonymous Coward]

offer support

form an LLC and offer support this also provides a means in which big corporation can pay for or sponsor your work

[Neil Barnes]

Thank you, El Reg

You reminded me of a couple of contributions I wanted to make.

[Neil Brown]

We've switched from occasional donations, to a monthly donation

We're only a relatively small business, but my plan is that, by switching from donating when I think about it to making it a monthly thing, I am more likely to do it, and I factor it into our budget.

I've also started writing up to which projects we've donated / paid for something each month, mainly in case it is of interest to others.

Deciding which project gets what is tricky, but my thinking is that "doing something" is, in this case, better than "doing nothing", or just carrying on as we were before.

[oiseau]

What?

It's not that o Organisations don't like paying for things even when the benefits are nebulous.

There you go.

Made it sound more like reality.

O.

[ecofeco]

Re: What?

Organisations don't like paying for things.

There. NOW it's fixed.

[Ian Johnston]

How would the FOSS contribution be assessed? To take the xkcd canonical example, is the person in Nevada keeping 200 lines or 200,000 lines of code going?

[johnbrownsschooldays]

Something like https://github.com/ossf/criticality_score might be a starting point?

[Charlie Clark]

Not really. There are two separate things to consider: recognition of the work done so far; maintenance going forward. Any kind of fee-based maintenance is likely to introduce an element of liability, which while sorely lacking in the software industry, will need addressing at some point.

[Anonymous Coward]

Eventually, FOSS is recognized as a religion...

And religions always are very bad at managing money for what their ideal should ask them to do - what really goes to charities is tiny fraction of the money they collect. Most of it still goes to their "CEOs and executives" - call it a Pope and Cardinals, or whatever you like - and for self-aggrandizing building excessively expensive HQs and branches. Recently, the local church spent 100k euros to replace the altar and the lecterns, because the old marble and wood ones were not in the priest's liking... although still perfectly fine.

And the fact that FOSS is better for long-term security is plain belief - especially since we see how much old code is around, even if updating it would be "free" - just upgrading and testing software and devices still costs money, even when the libraries are "free", so it does not happen. Moreover the fact that a library is commercial doesn't mean you can't have the source code too - I always used libraries from companies that delivered the source too. Sure, we paid for that too.

The fact the a lot of FOSS code feels much freer to break backward compatibility because you "can always recompile/upgrade everything/ecc. ecc." in some ways make it even worse, as updating the libraries may require extensive code changes. And older version become desupported soon, much faster then Microsoft OSes. Sure, maybe you can find a random guy in Nebraska to maintain them - but how many do? The fact that something is possible doesn't mean it makes sense, nor that it happens. Just like the many eyes perusing FOSS code - Linus belief was exactly just a belief too.

And still nothing ensures the that latest and the greatest will run on older devices - the same that may still run Windows XP for lack of alternatives.

It's time to stop to believe "FOSS will make the world better" just because. We have to look at the cold reality - people need an income, and there are not magic ways to pay them - schemas developed for the poor are not the right one, unless the target is still to be able to exploit a large number of developers paid a pittance.

[heyrick]

Re: Eventually, FOSS is recognized as a religion...

Funny. Linux is quietly running on a myriad of embedded devices (which most people probably don't even know about), a silent revolution that the corporate behemoths didn't stand a hope in hell of matching.

Your broadband router, your smart connected camera, perhaps your printer and PVR. And Linux begat Android which is on most people's mobile phones these days.

FOSS has made the world better, and quite demonstrably so. The problem is that it's popularity has attracted the sort of leacher scum that the corporates themselves would complain about if it were their products. Take, don't offer anything in return, and worst of all of there's a problem then be all too willing to point the finger of blame at some random guy who they didn't give a toss about at any point in the past.

The problem here isn't with FOSS. The problem is how corporates use and abuse it.

[Anonymous Coward]

"Linux is quietly running on a myriad of embedded devices"

Sure, outdated version of Linux are running on myriad of embedded devices. And being FOSS doesn't automagically keep them up to date. And it's there only because it's free - there were embeeded OSes before, just how can you compete with "free"?

"FOSS has made the world better, and quite demonstrably so."

Really? In which ways? Making companies pay developers less? Letting Facebook, Google and Amazon amass an enormous power? The world looks worse to me, actually.

"The problem is how corporates use and abuse it."

And do you really believe that giving something away for free could end in a different way? Again - it's just some kind of religious belief sustaining FOSS - and like all religions, it's designed to exploit the most for the benefit of a few...

[heyrick]

Re: "Linux is quietly running on a myriad of embedded devices"

"And being FOSS doesn't automagically keep them up to date."

Being commercial doesn't automagically do that either. Apple has given up on my iPad Mini. Microsoft has given up on XP. Many Android vendors gave up on your device before you even purchased it.

"Really? In which ways?"

Perhaps more reliable devices due to a reasonably stable base to begin with? Or maybe less expensive devices because nobody had to factor "write an entire operating system" into anything. Not to mention, with some devices you can throw away the manufacturer stuff and put your own in its place. Such as OpenWRT, for example.

"Letting Facebook, Google and Amazon amass an enormous power?"

As much as I'm all for bashing those three, I think you'll find their use of open source is more a "when it is useful to them" rather than a significant factor in what made them the entities that they are now.

Google, for example, started out life as a faster ad-free alternative to AltaVista and the like. This was important in a dial-up world. As line speeds increased, they started adding in some "sponsored" things, and, well, it grew into an empire.

Facebook? Built on the backs of people willing to blather every thought that passed through their vacuous minds. The lure of analysing data points for targeted advertising wasn't far away.

And Amazon, originally the world's biggest bookstore, now sells pretty much anything (and far too much of that is counterfeit crap). That empire was built upon somebody smart enough to figure out how to streamline online selling to the point of making it as addictive as gambling, and organising a logistics network to ensure it all actually worked.

"do you really believe that giving something away for free could end in a different way?"

I'm far too cynical. The question isn't so much whether or not it could have ended in a different way, but more what needs to be done to alter the current trajectory.

Once upon a time the concept of open source in any sort of serious context was practically unknown. Enough people had the same general vision, even if there were disagreements (BSD vs GPL, for example), that open source became a reality. And rather than being a useful program that had its source in the archive in case you wanted to fiddle with, it instead became entire operating systems. Ones that are arguably more solid and hardened than the typical commercial offerings.

"it's just some kind of religious belief sustaining FOSS"

You might almost have a point if you replaced FOSS with GPL, but it's not necessarily a valid point with FOSS, as that spans the gamut from the rather restrictive GPL to the "I don't give a shit" three line BSD licences, and many such as the person above, who make their contributions in order to scratch their own personal itch. To quash their favourite pet bug. And, look, you can actually do that with an open source project. Spot a bug, look to work out how and why, develop a fix, submit a patch. Let's see my long list of iOS/Android/Windows bugs get fixed in a similar manner. Oh, wait, nobody gets to see that so no third party fixing is available. You just gotta rely upon the in-house developers fixing things, but that's a pretty big hope as in my experience they're too busy fucking with the UI for no good reason to spend time looking for annoying bugs to fix.

There is no easy answer, sadly. But calling FOSS a religion? That's not an answer at all.

[Anonymous Coward]

"Being commercial doesn't automagically do that either. "

It pays more developers, at least. And when devices gets more expensive, people ask for better support cycles as well.

"Apple has given up on my iPad Mini. Microsoft has given up on XP."

Just like Linux has given up on older kernels? There's a lot of open source software desupported and even abandoned as well. Things get obsolete and are replaced. Still, there are better chances that XP software runs on Windows 11 that old Linux software runs on a recent version, and even recompiling it could be not so easy even if you've the code.

"Perhaps more reliable devices"

LOL! I'm seeing worse and worse code and applications everywhere. Linux is creating a lot of developers who have no clue about desktop software, for example, and even web sites are becoming a mess of ill-designed UIs and badly written back-ends that chain together with glue and tape a plethora of little pieces code. But hey, they run on FOSS!!!!!

Even OpenWRT supports a limited set of devices, and may lack hardware acceleration on many of them. Who wants cheaper devices when they do less, and what they do, do badly? Just because the OEM can spend less and save on developers reusing some code found around, and who care about security? All the IoT bad stuff we've seen till now runs on FOSS, doesn't it?

"you'll find their use of open source is more a "when it is useful to them" rather than a significant factor "

No - it's a very significant factor - they can exist at that scale because they don't have to pay for most of the software they use. Also they are able to greatly reduce the investment on a lot of software they use because they can exploit free development - or at least greatly share that investment.

And don't look at Amazon the seller, look at Amazon the cloud provider....

"Once upon a time the concept of open source in any sort of serious context was practically unknown"

Was it? Most of Unix systems, and mainframes in general, came with full source code, even fully printed. The money was in the hardware, not the software.

It's not a surprise that the very concept of FOSS came from the university environment used to Unix and to its code - which wanted more when the wind changed -, and not from the PC commercial sector were competition was much stronger, and money were made selling software, not hardware nor selling services. Now that selling services became the easiest way to make money, once again they don't want to pay for the software....

[Anonymous Coward]

Re: "Being commercial doesn't automagically do that either. "

LOL! I'm seeing worse and worse code and applications everywhere. Linux is creating a lot of developers who have no clue about desktop software, for example, and even web sites are becoming a mess of ill-designed UIs and badly written back-ends that chain together with glue and tape a plethora of little pieces code. But hey, they run on FOSS!!!!!

At risk of what did the Romans do for us, is there any FOSS application software which is actually better than the commercial alternative? MS Office is miles better than LibreOffice. Sibelius is miles better than MuseScore. Matlab kicks Octave's sorry butt. Audacity is a toy. VLC is a bug-ridden, feature crept horror. There is -effectively - no FOSS CAD/CAM software. Avid laughs at OpenShot.

I'm a Linux user and the only commercial software I use is Matlab, on a work licence. All the other applications I have are second-rate.

[jake]

Re: "Being commercial doesn't automagically do that either. "

It's not the application software that FOSS is good at. Rather, FOSS is good at making the tools that allows commercial software to be developed.

With that said, how's your favorite LAMP stack doing? I'll bet you a plugged nickle that not a single piece of it is commercial software, and yet without it your entire pointy-clicky Internet Experiance would disappear in a puff of newly emptied drive space.

[Richard 12]

Re: "Being commercial doesn't automagically do that either. "

How about web browser engines - all three of them are open source. Webkit, Chromium and Gecko killed the closed source ones so completely that even Microsoft couldn't justify keeping IE alive.

Web servers too. Apache is far better than IIS, for example.

Aside from that, FOSS is the glue that holds the world together. Commercial software tends to sit at the endpoints (user and hardware), with FOSS providing the interoperability and interconnectivity.

In a lot of cases, the commercial software is "merely" a skin around the FOSS software that actually does the tasks. Which is fine of course.

[Howard Sway]

Yet nevertheless, there are plenty of OSS creators who'd like to be paid

Well, if you want to get paid for something, it's up to you to find someone willing to pay for it, you can't just expect someone else to dream up and implement a way for this to happen for you.

On the other side of the coin, if you want people to write software so that your requirements are met, you will probably have to pay for it.

OSS may or may not be a way of doing this, but it's up to OSS authors how they work, and what conditions they might agree to, and there are as many reasons why people do it as there are people doing it. Many people, including myself, don't think about money at all because it's just making available software that I've already written that might be useful to someone.

I have a feeling that we're going to have to watch as many years being wasted in debating this as we did with all the endless arguments over the GPL. When the answer really is that everybody's going to have to work out for themselves what they want to work on and how they're willing to work / get paid and find the way that's best for them. The one-size-fits-all models are rapidly becoming a thing of the past.

[Ian Johnston]

Re: Yet nevertheless, there are plenty of OSS creators who'd like to be paid

The attitude of OSS developers reminds me of the London tube buskers who glare at you if you walk past without paying. They chose to put their work out there for free, and they are in no position to grumble about people who take it for free. If you're that good and want paid, get yourself a contract.

[jake]

Re: Yet nevertheless, there are plenty of OSS creators who'd like to be paid

I think you'll find, if you actually take the time to look, that it's not the vast majority of FOSS developers that are bitching about not getting paid ... rather, it is the non-dev paperwork-pushers at the various Foundations that are bitching. Many of these people think they are owed a living simply because they are somehow attached to a project, despite the fact that they can't code their way out of a while loop.

[heyrick]

They have no problem spaffing millions in

You neglect to mention, the people decide upon where money will be spaffed are invariably also the beneficiaries of said cash cow.

[Filippo]

I strongly doubt that any solution that is based on corporate goodwill can work, because there is no such thing. I think the problem can only be solved by compulsory payments, either by making licenses non-free, or, far less likely, through governmental action of some sort.

I'm not even banging the "corporations are evil" drum; rather, I believe corporations are simply not moral or ethical entities at all, and should be considered more like forces of nature: you can often get them to do something useful, if you're careful; sometimes you have to contain them, mitigate them, or extinguish them; expecting them to follow morals or ethics is utterly pointless.

[wub]

"I strongly doubt that any solution that is based on corporate goodwill can work, because there is no such thing."

It might be useful to divide corporations into publicly traded and privately held, then address the question of good will. There is no possibility of goodwill in a publicly traded corporation, since management has a fiduciary obligation to maximize shareholder benefit. Passive shareholders rarely have a voice, let alone a persuasive one.

Private corporations are unlikely to do things differently, but occasionally their shareholders may agree to accept reduced tangible benefit in exchange for some form of goodwill.

[Richard 12]

Not the only thing

There are many examples of publicly traded shares crashing because the company did something "morally wrong" (but not illegal), even though such moral turpitude arguably (or even clearly) did create larger profits.

Some of these have resulted in lawsuits, which makes them easier to find if you search. *cough meta*

[F0ulRaven]

Open Source started as a free way of sharing useful code in a way that didn't end up being locked away in a proprietary way.

There never was any expectation of having to support the code, so why are we having this discussion?

If you want it fixed, make it into a tender, and post it, maybe on Reddit.

It will then get fixed and shared, and someone will get paid, and the process easily fits into the corporate process of paying for things.

Why are we trying to make things difficult?

[heyrick]

Because the mentality is pretty much: you made the code, it's broken, you have a moral obligation to fix it.

Accordingly, people are like "wait, what?" and "you want that, cough up coin".

[cdegroot]

OSS with license clauses like "only usable by small companies" will never qualify as OSS, and for good reasons, I think.

What I've been doing is slap either the extremely strict AGPLv3 on things (so you're at least always forced to contribute back) or - a pretty good way to scare off company lawyers - just drop stuff in the public domain. The latter is too nebulous for legal types to recommend and smart companies will avoid it.

Theoretically, that forces organizations that do not want to play ball to contact the author, and then things like dual licensing can happen.

There is a reason, I think, that the GPL forces sharing and has some idea of how authors can make money and that's why the big companies all love the "permissive" licenses like Apache, MIT and BSD. And there's a reason that the FSF, for all its flaws, sticks with "Free Software" and does not like "Open Source". OSS makes it just too easy to be a bad player.

Also: Google lobbying the White House is a surefire way to get to a bad outcome in all of this.

[gerryg]

Let's look at one of the largest eco-systems, KDE

I found the discussion a bit abstract.

People still try to stigmatise KDE's decision to use Qt despite the evidence. It's supported by about eight commercial organisations and it has a regular tip jar request.

It seems to cover all the bases discussed in the article and gives every appearance of doing OK.

Muse Score, a bit more niche, has recently got a lot of flak in its attempts to get some money for reasons I haven't quite managed to fathom.

I don't know how much Nebraskan code is out there OTOH Libre Office seems a bit upset.

The signal to noise ratio in this discussion is very difficult to measure

[Anonymous Coward]

Re: Let's look at one of the largest eco-systems, KDE

Muse Score, a bit more niche, has recently got a lot of flak in its attempts to get some money for reasons I haven't quite managed to fathom.

Because it's owned by a Russian organisation and because it wants to charge people for access to scores for which it does not pay.

[Blackjack]

As said before, move to a FREE FOR NON COMERCIAL USE ONLY model. If you make money using it? Better pay.

As some other poster said, change FOSS for "FFNCUOSS*"

* Free For Non-Commercial Use

[doublelayer]

You can do that if you want, but there's a reason that doesn't fall under either the FSF's or OSI's definitions. I agree with them. If you do that, I won't contribute to the code or donate, because it gives the author or owner too much power to restrict the rights of the user. The whole "free speech not free beer" bit has gotten old, but it's still accurate in this case; letting the author decide that someone can't use it because of who they are or how they intend to use it is a very different philosophy and one I will treat differently.

[Blackjack]

Free speech and paying for a service are two different things.

[doublelayer]

That's overly literal. It's a phrase example from a well-known speech, clarifying that the word "free" means freedoms to use and modify. I put quotation marks around the phrase to clarify that I was referring to that meaning.

If payment to the original author is required for certain kinds of uses, then the freedoms to use and modify are not absolute. Those restrictions are already often seen in proprietary licenses of all kinds, from the simplest noncommercial license to the most defended product of Oracle's lawyers and the most arcane formation of weird legal terms. They are perfectly possible and acceptable, but they mean something clearly different and users will treat them very differently. That is the point I was making.

[Boris the Cockroach]

So long

as its free corps will use it

Largely on the basis of "M$ want 150 000 dollars and 0.05% of sales revenue for use of their audio codec, but this FOSS codec will do the same and we dont have to pay anything, which one do we use?"

Then complain when the guy doing the FOSS codec decides to dump doing code and takes up painting instead.

Plus I suspect a lot of corps take the source code to that audio codec and tweak it to exactly fit their needs and dont bother releasing it...

[scubaal]

Free riders

Of course the fundamental problem for any human system based on trust is that of free-riders.

Why would I pay if I can get the same benefit without paying?

In corporate terms why would I pay to support this software if I dont have to - when in doing so I make it available/supported for my competitors.

Is it a charitable donation - will I run my business on software supported by charity?

Who do I sue when it fails and/or cant be fixed?

Caveat - I dont *personally* think like this :)

But this is the underlying premise of market based capitalism.

[Richard 12]

Sueballs are a red herring

Has anyone successfully sued Microsoft or Apple for software that's not fit for purpose?

The key advantage of relying on something FOSS is that you can fork it and pay someone of your choice to fix it if it breaks or isn't the right shade of teal.

With commercial software, if it breaks and the supplier won't fix it, you're usually screwed - even if you have the source, the licence agreement often prohibits you doing much.

[BlokeInTejas]

TANSTAAFL

There ain't no such thing as a free lunch.

If you as an individual or a corporation choose to build your business on software that someone else wrote and offered with no warranty, no way to get fixes within a guaranteed time, and which you personally didn't test to within an inch of its life to the same standards as internally-developed stuff - then you're an idiot.

If you want stuff you didn't pay for to be reviewed for errors, but you're not prepared to do it or pay for it to be done - then you're an idiot.

What may fix the whole and complete nonsense of "yeah, it's software, but it comes with no warranty of correctness whatsoever" is somebody who embodies software in their product getting sued to within an inch of their lives - or beyond - for supplying stuff that didn't do the job it was clearly supposed to do.

I'd guess that this will eventually come about in the automotive space, because cars can kill people.

If there were a real threat of serious financial damage being done to you when you ship product embodying flawed software and said software fails, then perhaps things would change.

Until then, it really and truly doesn't matter. Stuff that matters has financial implications.

[jake]

Re: TANSTAAFL

"If there were a real threat of serious financial damage being done to you when you ship product embodying flawed software and said software fails, then perhaps things would change."

Microsoft has been doing that very thing for decades It hasn't affected their bottom line any. I'm certainly not going to worry about it as a SOHO who utilizes FOSS code with every client I have. But then I make sure the code I provide fits the needs of the client ... I don't simply stuff an entire shovel-ware distribution on a beige box and call it a "router". Or worse. a "IoT fridge" ... or even worse, an "IoT baby monitor" or "home alarm system". Etc.

[bigtreeman]

Heavenly bank account

He's got twenty million dollars

In his Heavenly Bank Account

All from those chumps who was

Born again

Oh yeah, oh yeah ....

FZ

[Robert Grant]

Started good, but then

seeing their work be exploited

Sigh.

[vincent himpe]

here is some free software

after a couple of years : can we get some money for it ? we like to eat too.

THAT are the problems ( giving away stuff for free and needing food ) They are, highly likely, mutually exclusive. Once we solve that the market will thrive.

[captain veg]

the problem isn't FOSS

The problem is that closed-source is often closed forever. Even when the company that owns it goes bust.

Escrow ought to be a legal requirement.

Or, if that's too burdensome, just open-source it. It doesn't necessarily make it "free".

-A.

[ecofeco]

Why is anyone surprised?

Whocouldaknowed big business would steal from people? /s

It's a very noble thing to make shared code and software and intellectual property. It is utter, inexcusable naivete to think it would not be exploited with no recompense.

[fpx]

Thus confirming the vision of Richard Stallmann. If companies can take your work for free without giving back, they will.

[Alan Brown]

Even with GPL, companies frequently steal code and sell it as their own. The only difference is that you can use copyright law against them if they do

[Alan Brown]

MBA seagulls

The problem is simple

MBAs ask "how much do we need to pay for this?" and when told "nothing, but if we don't there won't be further development", they shut down after the first word and refuse to pay anything

That's the primary reason why OSS devs don't get paid. Even if there's a payment system in place it's rare to get any offers until AFTER a critical piece of software development is shut down and a company is desperate

It's not just businesses either. Even in University positions when I asked for budget to throw at XYZ project - critical to our infrastyructure - to ensure it kept being worked on I'd usually be shut down

Inevitably the developers would close the source and move to $expensive support models, leading to manglement griping about costs, but not listening when informed they had an opportunity to keep costs down by doing the right thing in the first place. This has just happened with one of the better asset management platforms (GLPI)

About the only way OSS gets done reasonably well is when it's funded for a purpose by an institution and the results given away but once they lose interest in it, it becomes abandonware just like anything else (only with source available)