Meta can call Llama 2 open source as much as it likes, but that doesn't mean it is Meta's newly released large language model Llama 2 is not open source. Yes, I know Meta AI people proclaim: "Llama 2 [is] the next generation of our open source LLM available for research and commercial use." So what? It's not. I can say I'm a master carpenter too, but that still won't change the fact that if I try to strike …
For Zuck, open source is just another marketing phrase. For developers, it's the rules of the road .... Steven J. Vaughan-Nichols
Steven, SMARTR Large Language Model Machines do neither recognise nor abide by any rules of the road .... which does most definitely make for a vast series of really interesting new postmodern developments ..... which truthfully is much more than just extremely disruptive whenever defaulted surprisingly creative.
> "Open source means that developers and users are able to decide for themselves how and where to use the technology without the need to engage with another party; they have sovereignty over the technology they use.
NO! It does NOT! This is deliberate obfuscation and exactly why RMS makes the point about the differentiation between Open Source and Free Software.
Open Source merely means that you can see the source code. Free Software is specifically about freedom.
Some in the industry have been pretty disingenuous by claiming that they are the same thing. They most certainly are not.
This post has been deleted by its author
The article quotes Stefano Maffulli, the OSI's executive director - if it were [OSI's definition of Open Source], it wouldn't have any restrictions on commercial use (points 5 and 6 of the [OSI's] Open Source Definition). - I've added the appropriate qualifier where he left them out.
Notably, OSI's list of open source licenses includes the GNU public license(s). So actually OSI's definition of open source is wider (more free?) than what the GNU public license(s) permit.
(Yes, you can find examples where not using a GNU license has hurt open source development - I'm not denying that - "Open" includes freedom to risk self-harm.)
Meta license, however, includes Meta-benefiting constraints that set it apart from all OSI licenses, including the GNU public license(s) .
Any attempt to use it for business includes great risks if successful.
My first thought was "Who the fuck are OSI and what right do they have to decide what is open source or not?" I really don't care whether they "approve" a licence or not, that means nothing.
Did OSI even exist before "open source" became a thing? Of course not. By definition, open source means I can declare anything I produce as open source and can choose to create whatever licence I want so long as it's actually open, with or without their approval.
As for Meta, yeah, it all seems a bit of a case of their marketing people hijacking a term for their own use. On the other hand, who can sue and on what basis?
They're just a group who published a definition and certify some licenses as being compliant with, or not compliant with, that definition. You are free to ignore them. However, there are two risks in doing so. The first is that some of us agree that their definition is a good one and prefer compliant licenses. This doesn't mean that we need to see it on their list, because it's usually pretty easy to read a license and see for ourselves whether it meets their relatively small list of requirements. However, when I see a license that doesn't meet those few requirements, I'm usually less pleased with the decision, and that tends to make me less likely to contribute to the project. If you don't care about that, no problem.
The second is that some courts have acknowledged that "open source" is not a term you can apply to anything you want. It has been held to have a specific meaning, and a company that claimed to be open source without following those requirements was successfully sued for false advertising. That was one court decision, and it may not stand if other cases come up, but it does indicate that the definition from the OSI has some acceptance which grants it some legal validity.
The judgements referred to free and open source. They were not allowed to advertise their software as free and open source I am assuming that means FOSS rather than just open source.
"Defendants and their owners, principals, agents, managers, officers, directors, members, servants, employees, successors, assigns, and all other persons acting in concert and participation with them (collectively, the “Enjoined Parties”) ARE HEREBY ENJOINED from:
2. Advertising, promoting, representing or referring to ONgDB as a free and open source drop-in replacement of Neo4j Enterprise Edition distributions"
@doublelayer BTW thanks for the link.
You are correct. The OSI are a self-appointed group who have no deity-given right to police what is and what is not “open source” or “free” or whatever.
Releasing code that is not subject to an “OSI approved licence” does not mean it’s not “open source”
I’m not saying whether Faecesbook’s licence is or is not open source, and quite frankly i couldn’t give a shit one way or the other, but i really object to OSI’s opinion being quoted as gospel
No. Freedom to control what others do with the software isn't freedom. The "freedom" aspect is what the third party can do with it.
So actually, GPL is the least free of the "free licenses"
How the hell can you say it guarantees more freedoms when it specifically limits them? As I said, more "control" guaranteed to the license holder isn't freedom - it's EXACTLY the opposite.
> So actually OSI's definition of open source is wider (more free?) than what the GNU public license(s) permit.
Although, if you look at the SPDX list they allow you to compare the OSI's opinion with that of the FSF and there are cases where the FSF tick a licence but the OSI don't (e.g. Apache 1.0).
> Meta license, however, includes Meta-benefiting constraints that set it apart from all OSI licenses, including the GNU public license(s) .
> Any attempt to use it for business includes great risks if successful
Well, leaving aside that there has long been antagonism between Open Source and business (many devs taking the view that OSS isn't meant to be propping up business: if you intend to use it to make money, damn well pay!) could just clarify those risks - you appear have knowledge I don't, care to share?
To show you the gross depths of my ignorance, the only risks I know of, so far, are:
(a) you can't build an LLM trained on this - that isn't a risk, it is a straight up requirement.
(b) if you make loadsamoney from using this, you have to switch to a paid licence (which isn't a novel licence term in and of itself - game engines, for example, have long had this term). *Great* risk? Not just something you put into the figures for the business plan on day one? You aren't sure what the cost will be? That is no different than future costs for any of your costs: you could even buy a future on it!
(a) you can't build an LLM trained on this - that isn't a risk, it is a straight up requirement.
All the LLM co's have similar use condition on their LLM. Just a thought, but I expect that will eventually be challenged in court because
- the original training material is not copyrighted by the LLM trainer, so they cannot deny others to the right to use it freely (compare to Genius vs Google - and note, Genius was paying copyright royalties).
- is is legal and free to use the output of a software black box for development purposes when developing competing software - of course it is not legal to copy the implementation of the black box, but that is different. (compare to Oracle vs Google).
> (b) if you make loadsamoney from using this, you have to switch to a paid licence (which isn't a novel licence term in and of itself - game engines, for example, have long had this term). *Great* risk? Not just something you put into the figures for the business plan on day one? You aren't sure what the cost will be? That is no different than future costs for any of your costs: you could even buy a future on it!
I expect any startup,even if using Llama2 to get up and running, would have a plan B ready so that they did not depend solely on Meta should they be fortunate enough to to get half a billion users.
Otherwise, they better hope to get bought out by Meta on Meta's terms - or maybe Meta would just use the startups success as a template and release their own competing product.
I don't think Meta is doing something "immoral" - what they are offering is a free breadboard for development or research, and that's better than nothing.
It depends what definition you are using, but I'm sure you already know about the OSI's definition, which is much more expansive than you claim, including requirements to provide the right to distribute, to make derived works, to not discriminate or place requirements on the user that would limit those rights significantly, etc. Has that stopped everybody from claiming to be open source? No, it has not. However, neither have people been stopped from declaring something "free software" when they do not meet the FSF's definitions either. I've linked a case above where a court agreed that the term generally requires more than the source being readable. People who are interested in clearly defining how much rights a certain license provides have a lot of terms to describe different levels, and their use of "open source" is not the same as "source available", a term whose common use is much closer to the situation you describe.
Open Source merely means that you can see the source code.
You can get the source code of proprietary software (usually by paying and/or signing a license), but it is not "Open".
"Open source is source code that is made freely available for possible modification and redistribution."
https://en.wikipedia.org/wiki/Open_source
Free Software is specifically about freedom.
"Free Software" is a term chosen and used to describe a rather viral form of Open Source, where the licensor *requires* anyone who redistributes their software, is obligated to offer up the source code to any changes they made, under the same license.
You may say that is to the public benefit, but it is by definition less "freedom" than open source, using any sense of the term.
'"Free Software" is a term chosen and used to describe a rather viral form of Open Source, '
"vaccine" has a lot more positivity around it, but virus does describe it better - the infectious nature it. I just can't think of a good clean positive metaphor for open source of the GPL variety!
If you actually read the OSI definition here (https://opensource.org/osd/), and don't worry because it's refreshingly concise, it's remarkably similar to the FSF definition.
Although I completely agree that "open" is too vague, and is therefore vulnerable to misinterpretation and abuse.
I also completely understand that Raymond was trying to distance himself and the community from what he perceived as a sort of quasi-religious extremism, but the word he settled on was a very poor choice IMO.
"Open" is merely a state that exposes the contents, and therefore makes those contents visible. There's nothing in that which confers any rights, and that's a crucial omission. If I leave my front door open, you can see inside my house, but that does not somehow give you the right to enter my property.
Freedom, OTOH, explicitly confers rights. As the Llama 2 community has discovered, this discrepancy makes a big difference.
The other problem is that neither OSI nor FSF have any legal power to police the misuse of the words "open" or "free", so you can expect the professional liars at megacorps to peddle whatever bullshit they want with complete impunity.
The best we can do is complain about this blatant propaganda, and hope that somebody with the power to do something about it actually cares.
I think we mostly agree on this, but not so much with this part:
"Although I completely agree that "open" is too vague, and is therefore vulnerable to misinterpretation and abuse."
Sure, "open" isn't very clear, but neither is anything else. "Free software" is more often used to mean something else, simply because it is also the right phrase to describe software which I'm not going to charge for, even if I won't give anyone the source to it, hence all the speeches about speech and beer. The developers of software for which there is no charge don't always refrain from using the term, and there isn't another one that is easily understood by the general public.
This has led some people to start using the term "libre" instead. This doesn't have a second definition, so points there, but it's also not an English word. It works for people who speak French, Spanish, or one of a few other primarily Romance languages in which that's a word, but those who do not and want to interpret it as an English word will have a bit more trouble with it. It also isn't at all clear about what kind of license terms qualify as libre, I have not seen a "Libre Software definition". For example, some people I know thought that libre software meant that software would have a stricter license that placed additional restrictions on what a distributor would do, and used it as a contradictory word to free software. In their misconception, they had proprietary, for which you'd probably need to buy or negotiate a license, free which you could use with relatively little effort in software you intended to be free, and libre which you could use only if the licenses were compatible because they would likely place so many restrictions on a work that they couldn't practically work together. I took issue with this definition, and the group eventually decided that libre didn't mean that, but it doesn't make it any clearer what qualifies and what does not.
Meanwhile, both "free software" and "open source" have definitions which are easily understood, not too difficult to use to check whether a license qualifies or not, and at least known by most people who work in this field. I can't see a better way to classify things except if the FSF and OSI trademark those terms and attach them to their definitions. If they did so, I imagine at least some people would be unhappy with them.
The OSI are a useful (but declining in usefulness, btw - see SPDX, below) collection of known licences, but they do NOT define "Open Source" - they define what they have based their list upon, that is the sum total of it.
Once you have invoked the OSI list, any reputable report should make it clear that the OSD is *not* some separate measure of openness but is merely the document the OSI themselves drafted (cribbing from Debian, but it is now a distinct document) and use to qualify who will join their list.
Even this Register article may (will!) mislead (by the uncritical quoting of Erica Brescia) the less knowledgeable into believing that this LLaMa licence has failed on two independent counts, not just one.
Whether you wish to take their definition on as your personal definition of OSS, in all of its particulars, without fault or dispute - that is your privilege, but it does not force anyone else to do the same.
There was a LOT of shouting when the OSI was created, because it was seen to be making claims on behalf of software authors it had never spoken to (as in, the vast, vast majority of them - have *you* had your letter from the OSI, introducing itself and wondering if you'd like to have them represent you?). That was never settled (there was no vote): the OSI does NOT have any power over "open source" as a whole or as a concept, no matter what it claims. Generally, the shouting has died down - except on the OSI's own front page - and we get on with things without really giving them much thought, day to day.
The OSI was useful as a repository of well-known licences, but that is all it has ever done: it has not, and to this day, does not offer anything useful in terms of analysis of the applicability of material under one licence or the next (go look up its copy of GPL2 and of MIT licences: note the (total absence of) links to their careful discussion of the pitfalls of one against the other).
Nowadays, I would point people at the SPDX list in preference to the OSI list: SPDX ids are what you want to quote these days and they provide a tick against OSI, if that is useful to you, as well as the FSF "free/libre" acceptance, which OSI can not be bothered to do. Any system to help you judge between, juggle responsibilities of, the various licences (such as creating you Software BOM) is going to use SPDX notation, so best to get straight to it.
BTW the OSI claim "The “open source” label was created at a strategy session held on February 3rd, 1998 in Palo Alto, California, shortly after the announcement of the release of the Netscape source code." Bollocks. Open source as a phrase was in use before then. Even Wikipedia agrees. The OSI has a very grandiose view of itself, very suitable for articles of this sort.
"BTW the OSI claim "The “open source” label was created at a strategy session held on February 3rd, 1998 in Palo Alto, California, shortly after the announcement of the release of the Netscape source code." Bollocks. Open source as a phrase was in use before then. Even Wikipedia agrees. The OSI has a very grandiose view of itself, very suitable for articles of this sort."
And, both from experience and confirmed in the Wikipedia article, the concept of open source dates back to the early mainframes. Likewise, the 8-bit home computer "revolution" introduced large numbers of people to "public domain" software which often, by definition, allowed you to read the code since, at least in those days, a fair amount was published in BASIC to run on the ROM BASIC interpreter most home computers of the era had. Even then, individual authors often created their own limited licencing, often as simple as, "if you re-publish this or re-use the code, credit me in the new release".
Also, as evidenced by the proliferation of "open source" licences, and the frequent "licence wars", different people have different definitions of what "open source" means. Even those licenses listed by OSI will have proponents that disagree that some of the other licenses should even be on the list because they are not "proper open source" by their favoured licence conditions
Once again, more open source FUD being sown by a megacorp. Forget the debates they want us to waste our time with about nitpicking distinctions between "free" and "open". "Open source" these days is, and should be, commonly understood as a synonym for FOSS.
People should push back against corporate stunts like this, and IBM's Red Hat shenanigans, shamelessly designed to assert control in the pursuit of locking in profit streams. Claims of software being "open" should only be accepted if source is available at no cost for use in any damn way you want.
If it's not it should be called "restricted source".
Good rallying cry, but like so many rallying cries, too blunt: great for carrying people along on an emotional ride but after that...
We have F(L)OSS - if you only want to support that, good. Do so with all the gusto you can manage. Although I'm not sure you haven't contradicted yourself (see below).
But trying to force:
> Claims of software being "open" should only be accepted if source is available at no cost for use in any damn way you want.
Well, you have just kicked all of the GPL variants out: you certainly can not use those *any* damn way you want, that is entirely the point!
This certainly isn't closed source. It isn't restricted source (as you get with many products: buy a binary licence, you get one copy of the source to read for documentation/debugging).
One of the basics of Open Source that is often ignored (or even derided, curiously by those that loathe the GPL) is: nobody is forcing you to use it. It came with with conditions you don't like - don't use it. Or talk to the author and arrange a different (usually just a paid-for) licence (which is being offered here)
@Howard Sway "Claims of software being "open" should only be accepted if source is available at no cost for use in any damn way you want."
Like MS are using opensource code any damn way they want, when using opensource code to train CoPilot.
Of course not, because opensource does not mean "use in any damn way you want" and neither does FOSS.
This post has been deleted by its author
Is LLaMa 2 really that important?
It has a shit licence - but if it isn't really worth using what does that matter?
Maybe this is a good thing to have happened, if it makes people (in particular companies) think twice instead of blindly jumping onto the bandwagon to get all the free gifts (which usually turn out to be rubbish anyway).
Open source doesn't mean "uses OSI licence" and only somebody very stupid would think it does. It's like Apple saying "it can only be a computer if it has an Apple logo from the factory". It's a very arbitrary and more importantly very *incorrect* line in the sand that makes all sorts of false assertions.
Also if nothing else 'Open Source' versus 'open source' - note which one they use.
FWIW though people are going to use what they can get and judge for themselves and play these models off against each other and (rightly) completely ignore the Responsible Use Guide.
Oh and not for nothing but the OSI model is failing because people be getting greedy - both megacorps and individuals (and maybe rightly so - but it is all falling apart).
The article says that if your startup grows big, you need a special license from meta. That is not true. The timeline is important. The only thing hat matters is the number of active users on the release date of Llama 2. So if your startup is small today, you should be fine.
> If, on the Llama 2 version release date, the monthly active users [...] is greater than 700 million monthly active users in the preceding calendar month, you must request a license from Meta...
Whether it’s “free software,” or “open source,” or even “free open source software,” I wasn’t gonna pay any which way.
Personally I’d just pirate it and move on.
That’s the way the computer industry works anyway!
(Bill Gates is famously known to have made sure that every copy of MS-DOS was easily reproduced by its own Xcopy command. To make sure the user base grew as fast as possible…)
I think your confused / changing the definition of open source.
Open source has always meant just that, no more and now less. Even proprietary software could be open source (there was a time when buying a software license included the source code).
You seem to be talking about FLOSS and FOSS software, that's software that is free and open source software, that is “free” as in freedom.
See gnu.org/philosophy/floss-and-foss.en.html
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2025
