[seanmcdirmid]

Or the poster doesn't understand what open source really is, or their definition of open source is different from yours. Honestly, GPL screws up commercial use almost as much as a non commercial use restrictions. But I totally agree that open source are not the right words to use in this situation.

Full disclosure, I work for Microsoft.

[lucian1900]

Microsoft trying to hijack a term with an already fully established meaning is dishonest and screws up everything. This may be an honest mistake by an employee that fell for Microsoft's own marketing, but it's still not right to spread misinformation. Is the license on this list [1]? Then it's almost certainly not open source. Also, this particular license has already been rejected by OSI.

The GPL is not the only open source license. It may screw up certain commercial usage, but it never stops users from forking the software. This Microsoft license does.

1. http://opensource.org/

[throwaway54-762]

Er, technically this license doesn't stop forks -- they are explicitly allowed -- but it definitely propagates the non-commercial clause. I agree, it's definitely not Open Source.

[lucian1900]

I meant that it doesn't allow arbitrary forking. There are cases in which it's illegal to keep using it, whereas with the GPL at most you'd have to release the derivative source to your (paying) users.

[wisty]

> Honestly, GPL screws up commercial use almost as much as a non commercial use restrictions.

The GPL has a few problems, but non-commercial causes are just noxious.

Who exactly can use a non-commercial piece of software? Presumably, universities can. Can a government research lab use it? How about a military research lab? What about a contractor doing military R&D? If Exon-Mobile uses this to search for oil, is this research (thus non-commercial)?

Any clause which is likely to give your lawyer a headache trying to interpret it is probably a bad license.

The GPL has some ambiguous bits (like its interaction with interpreted code eg Python bindings to MySQL and Javascript libraries in the browser) but it's not as ugly as the term Non-Commercial.

[lucian1900]

Actually, interpreted code or browser code isn't really ambiguous. GPL is based on "derivative works" not necessarily C-style linking.

[wisty]

Is progressive enhancement of a HTML page enough to make the rest of the frontend "derivative"? What about the backend? What if they interact?

If the JS engine mixes interpreted code with the proprietary DOM, how does that work?

How about LGPL works?

But you're right - it's usually pretty clear if you think about it. There will always be edge-cases, but unlike non-commercial clauses the edge cases aren't that common.

[josephlord]

GPL has no restrictions at all on commercial use and in commercial use places no obligations on you at all.

If you distribute GPL code you then become subject to its restrictions and must provide the code.

[dagw]

However many large companies have a no GPL policy. I did a little bit of work for a large multinational engineering company most of you have heard of and was told in no uncertain terms that I was not allowed to use any GPLed code anywhere in anything I delivered. They also had a large document they made me read which listed dozens of licenses and exacting details over how I was and wasn't allowed to use software under those licenses.

[josephlord]

But could you use GPL software to produce what you delivered (rather than as part of it)? If not were Linux, Emacs and use of Android phones for any work banned?

If they were banned because of the GPL the company was doing its own stupid rules completely unrelated to the restrictions imposed by the GPL. If they weren't then it shows the difference between the GPL and the non-commercial use term of this particular MS license.

[seanmcdirmid]

Sure, I'm aware of this. My point was that GPL is almost as poisonous as a non commercial use restriction in terms of distributing software. GPL works great on the server where this isn't a concern, but doesn't work for almost any other commercial use case. Very arbitrary but at least very clear.

[josephlord]

You seem to have a VERY narrow definition of commercial use which involves selling software or being a software service provider. There are a vast majority of companies manufacturing/distributing/retailing and providing services for whom the GPL imposes no restriction at all. The MS release seems to prevent all commercial use not just commercial redistribution/binary release making it a FAR broader restriction than the GPL.

For MS and a surprisingly small (in terms of total global economy) number of other businesses selling closed source software (or software dependent on other non-Free software components) the GPL imposes very real restrictions.

I would argue that they are in no way arbitrary but have a clear purpose and objective to further increase the amount of Free software in the world. You may or may not support or want to assist this objective but it certainly doesn't feel arbitrary to me (I've taken the 'capricious; unreasonable; unsupported' definition of arbitrary from Dictionary.com as my interpretation of your meaning).

Edit/reply:

Can't reply to you for some reason. No citation but you have missed my point. I wasn't comparing Open Source Industry to closed source software industry but really the software industry to ALL Other industries (and individuals) in the world. Basically software consumers rather than producers (of which closed source companies like MS form a large part).

[seanmcdirmid]

Citation needed on how the open source software industry is so much larger (money wise) than the closed source software industry. I'm sure its much more of a mix, plus closed source software is more often sold, meaning more money is involved in say oracle DB vs. MySQL. Also, most of the OSS industry prefers apache/BSD style licenses that aren't viral.

If RMS was true to his principles, he would make GPL more viral to cover every deployment and co-deployment. But they leave this huge server hole instead. That is what I mean by arbitrary. Why a hole there and not elsewhere?

[josephlord]

I don't say anything of the sort. Please read again. I am saying that total global commerce and use of computers is greater than the software industry (not the world's greatest insight I know).

Don't want to answer for RMS. I can't think of less arbitrary way of achieving their aims.

[sedachv]

RMS didn't foresee the server hole. That's why GNU came out with the Affero GPL in 2007: http://www.gnu.org/licenses/agpl-3.0.html

[rmc]

Just about everyone who uses the term "open source" or "free software" would read "the source code can be used for non-commerical purposes" to mean "not open source".

There is basically no disagreement about what counts as an open source/free software licence. This licence is not that.

[throwaway64]

Open Source a term with a single, specific definition.

It is damaging to the ideas of FLOSS to allow stuff like "shared source" to be called open source.

EDIT: Removed reference to "open source" being trademarked, it is in fact not, apologies for the confusion.

[hermanhermitage]

No. This is a fantasy.

A trademark does not take ownership of general words or phrases from the people.

[throwaway64]

The problem with this idea when applied to this case is the motivations of people wishing to use some alternate definition. There is a real profit motive in being able to call stuff "open source" when it doesn't grant its users the same kind of freedoms the BSD or the GPL license, etc, do.

If you allow stuff like "shared source" to be called open source, you reduce the term to a meaningless buzzword, like how the term "open" is frequently abused.

citing trademarks is a heavy handed approach, but many parties with a profit motive don't care about anything except strict legality.

This is relevant in particular to this blog because Microsoft has a history of attempting to change what the term means

http://opensource.org/node/280

[skrebbel]

Correct. A trademark is just that: a mark used in trade. It does not come with the ability to define a standard definition for the mark's meaning, other than "~ is a trademark of $ORGANISATION".

Also, is "open source" really trademarked? Wtf? That's like a beef promo organisation trademarking "well done".

[nitrogen]

A trademark does not take ownership of general words or phrases from the people.

Except when that trademark is "Windows" or "Office" or "Word".

[iy56]

When you ask your officemate to open the window, does legal tell you to stop saying that Word(TM) because you're not using it according to the way Microsoft defined it?

[nitrogen]

No, but if I hypothetically created and started describing a graphical windowing system as a "Windows," and a word processor as a "Word," Microsoft would surely complain. It seems a bit hypocritical to complain about others' dilution of Microsoft's own marks, but then dilute others' marks in return.

This is a bit of a digression from the original topic, but seems tangentially relevant given that we're talking about Microsoft accidentally misusing or deliberately abusing a term that has an accepted meaning in the marketplace, the very purpose for which trademarks were created.

[Also, I should have added the following disclaimer to the previous comment: Disclaimer: Long ago, in a galaxy far away, I worked for a company that seemed to have a chance at invalidating the Windows trademark, but when Microsoft offered a large settlement, my former employer took the money and ran.]

[iy56]

Window managers are still commonly called window managers today. Word processors are still called word processors. That's the point - Microsoft hasn't taken away these generic terms. No, you can't call your competing product the exact same thing they call theirs. Why should you be able to? That'd be retarded.