I was just thinking the other day, "If Wikipedia became a paid product, would I pay for it?"
Really, what would happen if it became a for profit product? Would I pay for access to the greatest single repository of information the world has ever seen? Or would I shrug my shoulder and let Google tell me that I actually meant to query something else, and trust that the top placed link is really the best?
Its a different line of discussion, but one which is being wrestled with all over the internet. How do you compensate content creators who are not your direct employees for producing content on your platform.
I used to buy encyclopedias too, paper first and later on CD. But let's be honest here, there is no coming back from Wikipedia. The amount of informations on every single subject, the details, the fact that's its extremely dynamic (yes this new Usain Bolt world record is updated faster than than the actually record time), this is incomparable.
It's still interesting to note that the Wikimedia Foundation is getting more cash in donations than it requires to operate. If it wasn't a non-profit, it would actually turn out a profit.
I don't know what their current cost structure is, but assuming they don't currently pay contributors in a meaningful amount, going for profit would mean they would probably have to, and COGS would go way way up. The current income probably wouldn't cover it.
Why would they have to? Contributors do so voluntarily. This is no different than Quora or Stackoverflow which are both for-profit and don't pay for content.
It does turn a profit. You can be a non-profit and still turn a profit. You just don’t get taxed for generating income the same way, since we assume the profits help the greater good; since, we as a people, (not greed)(ideally) deemed it to be so.
The question “what kind of car does Jimmy Wales drive?” Popped into my head after reading this. Is it ok if he drives a Tesla from profiting off Wikipedia? What about a super car? Would anyone care?
It is probably even worse, people assume that they get what they pay for, so some people won't even look at the free options.
It is happening with me right now. I am searching for a Jekyll theme, and there are a bunch of free ones, but because time is money I will only try the paid ones.
Maybe Apple and Google need to give in and let us have free time-limited trials? Nobody wants to go back to the 90s software "You have 22 of 30 days remaining" popups every time you start an app, but is what we have now any better?
On iOS you can already do it. Apple has added a way to sell in-app purchases that cost 0. Your application can use such a purchase to unlock features, and then re-lock them when your trial period expires. The purchase manifest guarantees the timestamps and that it was made only once. The user would then have to buy a "real" in-app purchase to unlock the features.
Problem is the long tail user retention story is incredibly bleak on mobile. It's considered an achievement to get them to last a week, never mind a month.
That sounds like the "free trial" system working. Someone looking for an app for X purpose installs 10 of them and deletes all but one - giving very low average metrics.
It's more like the "one" they keep is also free, and marginally less useful or slightly more irritating to use.
Mobile app stores are a cesspool of terrible ideas and naive people who think if only they give away their time developing free software some company will notice and bless them with a job. The few gems to be found are a stark contrast.
By giving it away for free only for personal rather than commercial use, as lots of software used to be in the last millenium? But of course what's going to happen is a certain F/OSS advocacy org will shout "this isn't F/OSS as it doesn't obey our no-purpose-limitation criterion" and you and your software will be treated as a pariah and avoided like the plague and even more than pure closed-source software.
Fine, so you write a bunch of code, and release it under an NC license and a paid commercial license, no problem. Then someone comes along and submits a PR. What now? Will you accept their code and monetize it, despite the fact that you still deny that person the right to monetize your code, or will you refuse all PRs?
The problem with these proposals is that it creates an unequal relationship; some developers have more rights just because they started the project, even if others have contributed significantly.
I have a few things that I plan to release at some point. Small stuff, commercial use probably isn't a big issue, and I have no real plans for monetisation as this is hobby/interest stuff not the day job.
My intent is to go for full-on AGPLv3 and not accept PRs initially. If someone sees a bug they can tell me "if these options are picked there is an infinite loop" or "you've missed input sanitisation here" and I'll put together a fix for it. That way all the code is mine, I have full control, and I don't have to worry about treading on someone else's toes if I relicense later. When/if I later decide upon a more permissive license, or to try monetise through commercial licensing, or if I chose another route entirely, then changing will be frictionless.
In the meantime anyone wanting to work on it themselves because I'm too slow, they don't want to work my way, or "just because", are free to fork (assuming they are happy with AGPL), or contact me to negotiate other terms, or write something themselves from scratch to deal with the itch (again: these are not Big Things, I certainly don't expect to be releasing the NextBigThing™).
More likely it discourages others from contributing significantly. The PRs requiring copyright assignment end up being minor fixes, and anyone willing to develop a bigger functionality will just start their own project.
Idk what is so difficult about defining "commercial use". It is a concept even a child can understand. I acknowledge the goals others have for their licensing choices, and hold RMS in high esteem for his principled stance and achievements; I even release software under LGPL and other licenses. But I don't think OSI have an answer for the new software crisis where F/OSS developers are doing all the work, just to line the pockets of the three big cloud providers which obviously isn't sustainable. It also isn't in line with the ethics of eg the FSF to lock customers in to a walled garden built by F/OSS. In this sense, it's not me who brings up ideological arguments, but OSI who insist and hold-up their Open Source definition whenever someones dares to make a living with his/her work.
You can always release your software under the AGPL (which is an OSI-approved license)--which would effectively prohibit its use by cloud providers. Probably very few others will use it either but that's the tradeoff you make.
Careful with AGPL. It has yet to stand in court, and is actually the license mongodb has left behind in favour of their own SSPL because it turned out not strong enough, and not preventing eg. AWS to sell mongodb as a sevice without releasing any of their provisioning and monitoring sw (though I'm not sure AWS actually uses mongo or, like Azure with CosmosDB, a proprietary API-compatible product instead). The amount of confusion regarding AGPL shows in eg. [1], and I've also heard the opinion it might not be enforcable because it wants to restrict third-party software. IANAL, but I wouldn't rely on AGPL, and not many people would use AGPL software anyway.
I don't really disagree. No one knows how "viral" (to use a loaded term) the AGPL is. But non-commercial is the same problem. I've chosen to be comfortable with using NC CC-licensed media for anything I'm not being directly compensated for. I won't use it if I'm being paid directly to give a talk or produce a marketing document but I will for a conference talk I'm giving for "free" even if it's a conference I'm attending as part of my job.
But I know other people who think that using NC content even on a personal blog that uses Adwords is a strict no-go. Or you try to define NC in terms of US tax code which is fraught for a lot of reasons.
ADDED: Personally I think the latest round of the licensing wars is overblown. If you want to use a more restrictive license, go for it. Just don't expect the development model benefits of open source software. Of course, most open source doesn't reap those benefits either because vibrant communities are relatively rare.
I believe dual-licensed or freemium models aren't after contributions so much as they want to use OSS as a marketing instrument, or for being even considered. This is also the main criticism - that they use OSS networking effects to pull a bait-and-switch strategy. OTOH, the alternatives come with their own problems: that their incentive isn't aligned with their customer's. For example, Red Hat, easily the biggest and most successful OSS company (or at least it used to be before the IBM buyout), thrived on releasing a Linux which is (intentionally?) such a complex beast that you'll need a support contract anyway.
I don't get the concern; if the AGPL is invalid, then the software user is even more screwed: they have no license to use it at all. Wining such a dispute would be a pyrrhic victory.
When you're doing it for money. It's not difficult at all IMHO. Note that the new-fangled, earlier discussed licenses mongodb and others are using are targetting the case where your business is hosting sw for your customers (like cloud providers do); commercial use is ok when you self-host the software.
It's worse than that, people don't value it very highly if it's free.