Slashdot Mirror

That said, proprietary code can be open too.

No, it can't, by definition. If it's not available to everyone, then it's not "open" in this context. To quote the OSI, "Open source software is software that can be freely used, changed, and shared (in modified or unmodified form) by anyone". The essential missing part here is that sharing must be allowed, and the sort of commercial arrangements that get you source to proprietary code don't allow that.

Proprietary software that makes source available to customers has some of the properties of free software. But since it can't satisfy all of the GNU project's four freedoms, it's not appropriate to refer to those products as free software either.

http://opensource.org/osd, criteria 6
Ofcourse, there's no legally official definition of the words "open source", but that's pretty much the definition the whole world uses.

People often equate open source with GPL so one might also be quick to assume he's talking about developing proprietary software and profiteering off of open source but that is not always the case; GPL often presents itself as a pain for open source developers as well who may want to license their software under an even more liberal license such as BSD, MIT or even the WTFPL and wish not to taint their software with the ideological restrictions and burdens GPL places upon its users.

People often equate open source with GPL so one might also be quick to assume he's talking about developing proprietary software and profiteering off of open source but that is not always the case; GPL often presents itself as a pain for open source developers as well who may want to license their software under an even more liberal license such as BSD, MIT or even the WTFPL and wish not to taint their software with the ideological restrictions and burdens GPL places upon its users.

> Other nations can distrust anything they wish, but they have not other useful alternatives than to deal with us, they are our bitches.

That is true in case of some technologies like chips which are expensive to independently develop for less rich/advanced nations. But a good deal of software stuff is quite replaceable, with minimal pain. There are open source solutions or foreign services that are only slightly behind proprietary or US hosted solutions/services. The current surveillance situation simply incentivizes the alternatives and bridges that gap.

Peru did an open source requirement for government work some time ago and other governments were looking at similar stuff. Microsoft wrote that famous letter, 12 years ago, defending proprietary companies; something which is quite indefensible now.

http://opensource.org/docs/msF...

They simply did not have enough incentives until now. This isn't rocket science; its mainly a policy decision. China is developing its own Linux-based OS and has already replaced western social media services and search engine with its own etc. etc.

There is already that project that this will cost us $180 Billion in the near future.
http://blogs.wsj.com/cio/2013/...
Let's see if it will bear out.

The GPL does not say you cannot bundle binaries with trademarked stuff (logos). You can, you will be required to give away the source. But this still does not interfere with any trademark rights.

Not 100% sure for v3, because there are at least clauses about patents (one of the big differences between v2 and v3: You need to allow people to use the patents needed by the software), but afaik this is only about patents and not about trademarks.

The other point is fair use. You can use the binary, get the source, build it, use it again ... you may only not distribute it with logo/name which associates it with the vendor.

Now look at some very free license:
http://opensource.org/licenses...

> Neither the name of the copyright holder nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission.

See?

I wonder what RMS would think of a more "business-friendly" license where a commercial entity selling software could take software that's publicly-available, modify it, and then distribute that to paying customers, but not back to the community, but where the license required them only to distribute the modified source code to those same customers, however the customers were not allowed to distribute it themselves.

Any kind of selective-treatment, for people or organisations is, in Free Software terms, a Bad Thing. (Well, with the exception of the actual owner - licensing decisions will always be theirs.)

You'd fail Debian's Chinese dissident test. You'd also fail to provide Stallman's Software Freedoms 2 and 3. The FSF probably wouldn't even consider it to be 'semifree'.

As for whether it would count as Open Source: not by the OSI's definition, no. You wouldn't even meet the first criterion.

Funny, because redistribution is listed as point one in the Open Source Definition.

I personally prefer the ISC license: http://opensource.org/licenses/ISC

Specifically because it does not have the advertising clause - i.e. I do NOT want 3rd parties to be using my company name in their advertising.

*--jeffk++

Not open source? The source is available for download here.

You can't compile it yourself. You have no idea what is in the source.

You certainly can compile it yourself; I built it on my old Linux iBook G4 (PowerPC), since there were no binaries available for that platform. As has been discussed above, it does have a weird license, but it is absolutely open source.

Grandparent probably refers to Open Source Software, which is a formally defined term. It's not enough that you can merely read the source, you have to be able to redistribute it and any changes, too.

And the 1988 MIT license is among the OSI approved licences: http://opensource.org/licenses/alphabetical . The oldest OSI approved GPL license is the 1991 GPL 2.0, the oldest OSI approved BSD license is the 1998 3-clause BSD license. The older BSD licenses are not among the OSI approved ones, because they are clusterfucks, not because there was no OSI at the time.

Bullshit.

It's a variation of the BSD/MIT license, which has fewer restrictions than the viral GPL. It's functionally equivalent to the most permissive Creative Commons license, only requiring attribution. It's explicitly listed on the OSI website as an approved license.

"Permission to use, copy, modify, and distribute this software and its documentation for any purpose, without fee, and without a written agreement is hereby granted, provided that the above copyright notice and this paragraph and the following two paragraphs appear in all copies."

The only people who object to this type of license are GPL bigots who want to impose their version of "freedom" on everyone else.

That is why you stick with 2-clause BSD and avoid the fancy ones which talk about advertising or endorsements. It is difficult to find a standard license with wider compatibility than that.

Consider if the FSF merges with a different organization with a different charter; like the often more corporate-friendly Open Source Initiative.

I don't see how a merger between FSF and OSI would pose a problem. The Open Source Definition published by Open Source Initiative is worded nearly identically to the Debian Free Software Guidelines on which it was based, and each of the OSD's conditions maps to one of the FSF's four freedoms.

Would be nice if such a guarantee could be written into the license itself.

I agree. But given how some countries appear not to recognize a dedication of a work to the public domain as irrevocable, charters are the best we have.

Actually, even RMS refers to the BSD and Apache licenses as "GPL-compatible free software". So the GPL and other two popular licenses, BSD and Apache, are all free software by the Free Software Definition. The difference is that GPL is a copyleft license and the Apache and BSD licenses aren't.

Why are the Apache and BSD licenses becoming more popular than the GPL? Because free software has grown up. Where I work, we would not dream of implementing the whole software stack from scratch. We use lots of open-source libraries. My company's legal department is allergic to the full GPL because they want to keep open the option to do exactly what the GPL is designed to forbid -- make a proprietary product using open-source code. Usually our code is custom developed for a specific client but we might want to re-use that and/or make a general purpose product some day.

So, for us, using Apache/BSD licenses is easy. It's almost frictionless. Legal is comfortable with them, and pretty much all we have to do is include the license file and do a quick audit to make sure we've complied with it. GPL is much harder for us to work with because we have to justify to legal why we're signing away the rights before the product is even developed.

The whole point of the Open Source Initiative, as I understand it, is to promote adoption and use of free software. It turns out that copyleft is {sometimes, often} a barrier to that in the business world. So I would say that "open source" (aka non-copyleft) has simply beaten "copyleft" in the marketplace.

Copyleft was a brilliant idea but non-copyleft free libraries are what I use in day-to-day development work. And I say that as a dyed-in-the-wool, sandals-wearing, free-as-in-freedom, latte-sipping, corporation-hating hippie wannabe.

It is your problem if everyone else is using the authoritative definition and you are using your own made-up one.

Open source is a standard term in the industry, with a well-defined meaning. If you opened a restaurant and sold "egg creams" that consisted of...cream with an egg, you'd piss a lot of people off, even if you could proclaim "technically, I'm correct." Technically, you're not, because the term has a well-defined meaning that you're deliberately ignoring. Technically, what you'd be is an "asshat" (a not-as-precisely defined term, but one that certainly covers the situation). Same if you misuse the term "open source", especially on a technical site like Slashdot.

CC-BY-NC-SA is not open source, not by the traditional definition of open source. The NC part of the license is the problem. Open Source licenses should permit commercial redistribution, and this is in fact part of the first criterion given in the definition of an open source license:

The license shall not restrict any party from selling or giving away the software as a component of an aggregate software distribution containing programs from several different sources. The license shall not require a royalty or other fee for such sale.

(emphasis added) The NC portion restricts selling the manual. It isn't a free cultural work either for the same reason.

The answer is easy, and I think it applies to all distributions:

To be in the distribution, the licence of your project must fit the , currently version 1.9 or later. This means that the licence most likely also has OSI approval and can be found on the SPDX list. Beyond that, you also need to make sure that your package is compatible licence wide with the licences of all your dependencies.

To be available for a distribution, you only need to take care of the latter bit, and you can choose any licence, including non-FLOSS commercial ones. I, however, will not look at, review, debug or build that package without being paid for it outside of the scope of my work on the distribution.

I am a packager for a major GNU/Linux distribution.

The answer is easy, and I think it applies to all distributions:

To be in the distribution, the licence of your project must fit the , currently version 1.9 or later. This means that the licence most likely also has OSI approval and can be found on the SPDX list. Beyond that, you also need to make sure that your package is compatible licence wide with the licences of all your dependencies.

To be available for a distribution, you only need to take care of the latter bit, and you can choose any licence, including non-FLOSS commercial ones. I, however, will not look at, review, debug or build that package without being paid for it outside of the scope of my work on the distribution.

I am a packager for a major GNU/Linux distribution.

Actually, there is a whole bonanza/plethora (depending on how one looks @ it) of Open Source licenses out there, and so there is quite a variety. The above are just some of the more popular ones. As for GPL, there is a whole mess of issues about combining it w/ any licenses - not just proprietary licenses. The FSF really muddies the waters by having all the categories os copyleft/non-copyleft, Free/non-Free, GPL-compatible/GPL-incompatible and so on

As leader of an open source project

Apache POI, funded by MS .

So? Ad hominem much? Ex Concessis? Reductio ad Hitlerum? Not that I agree with the article, but c'mon, this is as stupid as the article in question.

Here, buy this book (Damer's "Attacing Faulty Reasoning: A Practical Guide to Fallacy-Free Arguments"), read it and then try again.

As leader of an open source project

Apache POI, funded by MS.

Yes, opensource. http://opensource.org/licenses/ms-pl/ See?

Slashdot uses the OSI logo as seen on this very story, so I wonder what the rules are on that.

The OSI web site FAQ says:

Can I use your corporate logo on my web page to link to you?
        Yes. You can always use a trademark in a truthful manner to refer accurately to an entity.

What about logo usage not linking to OSI?

Well, I read about Nominative Use and ... don't understand.

The nominative use test essentially states that one party may use or refer to the trademark of another if:

        The product or service cannot be readily identified without using the trademark (e.g. trademark is descriptive of a person, place, or product attribute).
        The user only uses as much of the mark as is necessary for the identification (e.g. the words but not the font or symbol).
        The user does nothing to suggest sponsorship or endorsement by the trademark holder. This applies even if the nominative use is commercial, and the same test applies for metatags.

Seems like dilution to me, but IANAL, etc.

Also, it says the symbol can be used for linking to the OSI website.

Finally, it seems that the logo is to be accompanied by the text, "We recommend using the Futura Md BT Medium fonts as complementary fonts to the OSI Logo."

Having rambled on through all that, I have to assume Slashdot is in compliance and I'm too tired to make sense of it all.

Slashdot uses the OSI logo as seen on this very story, so I wonder what the rules are on that.

The OSI web site FAQ says:

Can I use your corporate logo on my web page to link to you?
        Yes. You can always use a trademark in a truthful manner to refer accurately to an entity.

What about logo usage not linking to OSI?

Well, I read about Nominative Use and ... don't understand.

The nominative use test essentially states that one party may use or refer to the trademark of another if:

        The product or service cannot be readily identified without using the trademark (e.g. trademark is descriptive of a person, place, or product attribute).
        The user only uses as much of the mark as is necessary for the identification (e.g. the words but not the font or symbol).
        The user does nothing to suggest sponsorship or endorsement by the trademark holder. This applies even if the nominative use is commercial, and the same test applies for metatags.

Seems like dilution to me, but IANAL, etc.

Also, it says the symbol can be used for linking to the OSI website.

Finally, it seems that the logo is to be accompanied by the text, "We recommend using the Futura Md BT Medium fonts as complementary fonts to the OSI Logo."

Having rambled on through all that, I have to assume Slashdot is in compliance and I'm too tired to make sense of it all.

There's open, and there's free. The two are not necessarily the same thing, but many people confuse the two.

I don't see how you're helping.

A lot of engines have been made "open", with sources that can be made available.

Microsoft, to their credit, calls this "source available", and not "open source". Open source was deliberately coined and popularized with a definitive meaning in 1998.

There's open, and there's free. The two are not necessarily the same thing, but many people confuse the two.

I don't see how you're helping.

A lot of engines have been made "open", with sources that can be made available.

Microsoft, to their credit, calls this "source available", and not "open source". Open source was deliberately coined and popularized with a definitive meaning in 1998.

There is a whole bunch of OSI approved licenses that allow the redistribution of software to be restricted

No, there isn't. "Free redistribution" is the first of the necessary criteria for qualifying as Open Source by OSI's definition, and the second criteria also specifies that source code must be distributable too.

Why also, do people totally miss the point of FOSS and focus on price rather than freedom of choice?

Well, I don't know, but if it's not a big point why is it #1 on the list?
http://www.opensource.org/docs/osd

Depending on which flavor you drink, it's also third out of four here.
http://www.gnu.org/philosophy/free-sw.html

You ask why do people ignore all the other points? Honestly? Why would a non-technical person give a damn, and why should they? Expanding the scope more, why would most technical people care, they are not all software developers, don't all have the skills to utilize a project's code, or consciously choose not to have the responsibility of inspecting and modifying a free project's code because that has dick to do with operations.

The OSI's definition of Open Source, at least the _bottom_ half of their wishlist _tries_ to appear sincerely altruistic - "No Discrimination Against Persons or Groups/Fields of Endeavor" etc. The Free Software Def. fairly clearly says "do what I want and give it away".

Neither mention anything about quality, or improving the state of the art, things people DO care about and will PAY FOR.

However, I personally am more interested in the ability of organise my desktop in such a way that maximises my ease of use

How software functions has _NOTHING_ to do with this. Write your own software that does what you want. I can say that _fully_ within the spirit of free open software, so you can't complain. Or are you really telling me you wrote a custom patch for complex desktop software, and maintain it or fight with the original authors to accept it, just to organize your desktop... for your "ease of use"? Why would other people care?

Uh, if you look at the Open Source Definition, the first point they make is about Free Redistribution. Very surprisingly, the Source Code comes second. It's a pity, b'cos had the OSI definition not insisted on Free redistribution, it would have served the needs of open source better w/o raising converns about the financing of such projects.

There is very much a difference. A lot more licenses are considered 'open' than 'free', even while containing provisions that make it more palatable for the software creators. Look at the 2 lists that I linked to. The ones mentioned as 'open' are simply listed and categorized by their utility, w/o any judgement calls about their ethics, while GNU is more interested in listing them according to their purity i.e. similarity to the GPL itself. So it's small wonder that projects would pick an 'open' but not 'free' license - there are simply more of the latter than the former, and w/o onerous restrictions on software creators.

Contrary to myth, the Open Source movement, rather than the FSF, is the reason we have such major open source software such as Open Office, Firefox, Chromium, Apache, Android, and so on - if you notice, most of them are not GPL, and even Linus has decided not to make his kernel GPL3. If anything, companies like Sun went for things like the CDDL because it is not GPL. Oh, and before anyone says 'Android is Linux', Android is released under an Apache license and not GPL 2 nor 3.

I'd credit the likes of the OSI in helping popularize the Open Source model and bringing it to where it is. Unlike the FSF, it is not hostile to corporate interests and prefers to promote the advantages of this development model, rather than moralizing about the 'ethics' of 'Free Software'. Speaking of which, what is this 'community' that RMS, and you are talking of? People typically buy/download for free/copy software that they want to use, and use it. Most people don't, and won't, tinker w/ source code, nor pay someone else to tinker w/ them - if a software doesn't work the way they need, they either look for alternatives or workarounds.

ESR mentioned some of that in the 'Cathedral and the Bazaar', where he noted that worse than the confusion over the word 'free' was the perception that the FSF was down and out hostile to business. I'd say that that perception is accurate - name me one company (not non-profit organization like FSF) that Stallman endorses. As I've pointed out several times in the past, Freedom 2 of GNU is the poison pill in the GNU charter that makes it the most business hostile model. If a company, otoh, is fine w/ distributing its source code to its customers, but restricts re-distribution further downstream (for the obvious reason that they want to sell to those downstream potential customers themselves, and not have the value of their work diluted by other people who put no effort into it simply distributing it for free or their own profits), then they are more likely to find a sympathetic solution from OSI than FSF, who probably wouldn't give them the time of day.

There is only one case that I can think of where 'free' is a better idea than 'open'. It is the case of when a company is releasing support software for a competitor, like the recent story on /. about a TI employee writing FOSS drivers for QCOM in his free time the same way that he was writing FOSS drivers for TI in his work time. In such a case, it would be a good idea to use something like GPL3, just so that QCOM cannot make use of a non-employee's unpaid work and then include enhancements after making that proprietary. While it would have been perfectly ethical for them to do it w/ their own paid employees, it is somewhat unethical for them to do it w/ work done by employees of their competitors off the clock.

Although a lawyer is important, there are several licenses available. I am not aware of any open source license that allows for liability against the author. Most proprietary software also disclaims liability and claims that the software is not suited to any purpose. I am not sure why there is a worry about someone making money from it, especially if you do not intend to provide support. Providing it to the community and allowing someone else to pick it up and support it for a fee might be considered a win.

If nothing else, a lawyer could use one of the existing licenses as a boiler plate and add whatever additional terms are needed in this case.

If you are actually a contractor instead of a county employee, this whole things gets more complicated. In particular, you may have a restricted rights clause in your contract deliverable that could prevent this scenario or require additional customer agreements.

Um, I don't know what definition of "OSS" you're using, but I'm using the standard definition, and what you describe does not meet that definition. If he "forbids re-distribution" as you suggest, that violates criteria 1, Free Redistribution.

Yeah, he can distribute source without making his code open source (this used to be quite common in the Unix world), but that's not what he asked about.

BSD license = We take your code and use it for free and we do not tell anyone about it

2-clause BSDL:

Copyright (c) ,
All rights reserved.

Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met:

- Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer.
Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution.

Now I'll probably get hate for this, which will be ironic and sad since /. is supposed to be libertarian, but WTF I don't care.

The following is not hate. Just a sincere and considered response.

Slashdot is not "supposed" to be libertarian, except perhaps in the fantasies of Slashdotters who are libertarians. The philosophies of open-source software and libertarianism do have a non-empty intersection. But that doesn't mean that open-source advocates (or Slashdotters) are necessarily libertarians in the majority.

We ALL know why GPL is going down, its because TINSTAAFL and with GPL V3 RMS has gone so damned anti business he's scared away too many folks. [...] Now ironically if FOSS truly WAS a community and collective effort then right about now a large group of devs, users, and businesses would get together and hash out what the problems are and fix them, basically cooperate for the betterment of all, and if RMS didn't want to participate they'd just fork which is the standard way that FOSS routes around damage.

In fact, this has happened already, many times, and will continue to happen.

But sadly RMS doesn't want a democracy, he wants a dictatorship.

Opinions of RMS's ego aside, in the end who cares what he wants? The FOSS movement can route around RMS just like it routes around damage. (See above.)

But if there is zero ways for businesses to get a ROI then they simply will stay away and surprise! That's not good for anybody. look on the desktop scene where Mandriva is DOA and Canonical won't be far behind.

So what? Companies come and go, including open-source ones. Those that survive are the ones that figure out how to make money. And plenty of open-source companies are thriving.

The simple fact is you NEED companies to pay for all the work that needs doing, without that pay you end up with the "busted shitter" problem where nobody wants to do the lousy jobs like bug fixing, QA, regression testing, writing decent docs, so they just don't get done.

Companies can certainly make important contributions to some of the things on your list. But you ignore the powerful contributions of the user communities. Companies can do what users don't want to do themselves. But the reverse is also true.

Maybe its time for a new license, one that respects your freedom to tinker while accepting that those that pay for something have a right to get paid for their labor? Something like "You are free to look at and modify the code, but if you distribute you have to pay for it"?

You want a new license? Write your own. But first, check the list of existing open-source licenses in my link above.

I would choose a license that is very easy to understand in a few sentences, like the BSD license.

Ever considered the Sleepycat license? It's a copyleft like the GPL, but it's short like the BSD license.

I'm not to blame if you use different definitions of concepts than the ones commonly accepted on this community.

Here, "open source" means this. Whether the code is obfuscated or not is irrelevant to that.

Don't forget that Apple got their OS for free too (most of it, anyhow).. since its primarily based on BSD, which they use and then of course contribute little back to the open source community, unlike Google who makes significant contributions to many open source projects

Oh really?

It looks like Apple contributes quite a few open source projects.

OpenDOS source code derivatives bay be redistributed only for "non-commerical purposes" [sic] and thus it is Open Source but not Free Software.

Umm... this is just wrong. Open Source is a trademark of the Open Source Initiative. Their criteria for open source software explicitly prohibits restrictions on fields of use (such as the one no commercial activity clause in OpenDOS's proprietary license).

Think of Open Source as a form of brand identity for Free Software. They are both useful terms, but the latter lacks (1) a logo, (2) a single, specific formalization, (3) a license certification process, (4) important legal protections, and (5) marketability [because of the whole "free as in beer"/"free as in freedom" thing].

I would consider any of these
http://www.opensource.org/licenses/
an open source license, since others do.

Yes, BSD can sometimes be substituted with the GPL, since it is more permissive, and by that I mean that you can do more with BSD licensed source code, for example merge it with business source code in one binary.

However, the same is not true for program binaries distributed under the BSD license, because these may contain components that you will not get a license for.
So, if you want to improve a binary that includes BSD as a customer, you are bound to violate licensing terms.

The BSD license is not the main target of my arguments in the previous post, but other licenses are, which are completely incompatible with the GPL, both regarding source and binary.

The GPL concept comes from a time when a few companies like IBM, which were used to selling "big" mainframes, also owned the software on these machines. If these companies had successfully cornered the personal software market, even more people still would be stuck in using proprietary software.

The GPL was written by RMS and the Free Software Foundation about ten years before the term "Open Source" was coined. The Free Software Definition and Open Source Definition are nearly identical and generally, if something is Free Software, it is Open Source and vice-versa. However, the FSF and OSI have deep philosophical differences. The primary difference is that the FSF considers software freedom to be most important, while the OSI considers the pragmatic value of source code availability to be the most important. I'm convinced that both FSF and OSI have important roles to play, as do permissive and Copyleft licenses. However, I can't agree with the proponents of the term "Open Source" who claimed that it was less confusing than "Free Software," as posts like yours make it clear that two competing terms make things more confusing rather than less.

If the GPL was intended to be temporary, it was only until copyright was no longer relevant to software. If you need an example of the relevance of Copyleft in general and the GPL specifically, you need only look at Android. Google has made it clear that they prefer permissively-licensed software so that they can choose whether or not to release source whenever they want. They exercised that power with Android 3 (Gingerbread) by releasing no source at all with the exception of Linux, which is under the GPL v2. The fact that many companies "take open source seriously" is no reason to abandon Copyleft but rather the exact opposite.

File and track licenses

Done and done.

Of course, if they include the source code, and possibly rename the app, perfectly legal to do...

See points 1, 2 and 4 of the Open Source Definition http://www.opensource.org/osd.html

According to RMS himself, founder of the FSF, the Apache and BSD licenses are not free software licenses because they specifically violate the second value; the freedom to modify, which requires source code.

That's only in the case where the program is not open source:
The program must include source code, and must allow distribution in source code as well as compiled form.
http://www.opensource.org/osd.html

So if the program is open source, then by definition you have the source code and if that code is licensed under a BSD license then it most certainly complies with RMS's Free Software freedoms.

Tell me, since when does open source (which doesn't mean what you falsely imply it does) explicitly say they can't do what is exactly within the apache license, dumbass?

It doesn't become open source until you actually release that source code, if MS said the Windows source was under BSD license but never released the code would you consider it to be open source?
Up until now Honeycomb was closed source for the obvious reason that the source code was not released, hence not open.

Except from the Open Source definition:
The program must include source code, and must allow distribution in source code as well as compiled form.
http://www.opensource.org/osd.html

Just to be clear the only one of the GP's assertions I agree with is that Honeycomb wasn't open source (until now) and I wouldn't even say Google is 'evil' for not releasing it earlier, in fact I entirely understand and support them doing so, their reasoning made perfect sense. But it's obvious that Honeycomb was closed-source up until now.

If you ship free software as defined by the FSF or by the Debian project, or you ship open source software as defined by Open Source Initiative, then do you "ship" only the copies you make, or do you also "ship" the copies people make from those copies?

Open Source code and model are different things.

Android, the Software, is (now) Open Source, as defined by the OSI, since it fits the definition.

Android, the Project, may not be Open Source, provided a single definition of the model is accepted.

If Google has claimed Android to follow an Open Source model, I agree with you, but calling Android simply "Open Source" is still correct.

The Apache license (as well as BSD licenses) make no requirement that the source be provided, so they're not really "free software" licenses in the "freedom" sense

The Free Software Foundation appears to disagree with you.

They're certainly not opensource licenses since they make no requirement that the source be open

The Open Source Initiative appears to disagree with you.

They're basically just free-as-in-beer distribution licenses; you can share the binaries all you want, but you can't really change them, because nobody is obligated to give you the source to go with those binaries

And how did they create those binaries? By compiling the source code, which was Apache or BSD licensed. The derived work may be proprietary, but the ASL or BSDL code it open.

Isn't there some form of restriction here in the license, are they allowed to make a closed source derivative work, seeing as they're the original authors? What open source license (if any) was Growl formerly using?

I know some licenses require all derivative works to be open source, but I'm definitely not expert on open source licensing.

Strangely, though the previous Growl Source page had a link to a tarball, the current download page only has non-link text "Growl source code." under "Developer Downloads." The Growl Developer Documentation page says:

Growl is distributed under the conditions of the BSD license. The Extras are BSD licensed as well. Example applications are in the public domain.

The Growl license does seem to be a permissive BSD-style license. This means that nobody using the source has any obligation to provide source to anyone and can use it for any reason as long as they include that license with binaries. This is why OSX contains significant amounts of BSD code and Windows a smaller amount but neither Apple nor Microsoft is obligated to release the source for any of their changes.

Licenses that require source to be made available to those who get binaries are called Copyleft licenses. This is why Google must release the source for changes they make to Linux as part of Android, but they are not required to release changes to other parts of Android which are under permissive licenses.

Both permissive and Copyleft licenses can be used for software that is both Free and Open Source. Since Growl seems to have been under a permissive license all along, there's nothing stopping those who control the web site from making it proprietary and never releasing source again. OTOH, there's nothing preventing anyone else from forking the last released source just as Perry Metzger has done. A fork would probably be required to use a different name. Since those who control the Grow website seem to have removed all download links to source (even older versions) and banned Metzger from the mailing list, it may indicate they plan to keep future versions proprietary, though that's not entirely clear at this point.