SugarCRM 6 Released, But Is It Open Source?

darthcamaro writes "SugarCRM markets itself as a professional open source company and this week released version 6 of its Sugar platform. But the main new feature is a new user interface that isn't available to users of the community version — it's only available to paying users. No they don't claim to be open core either, they claim it's all open source, even if you have to pay for it. '"Open source doesn't mean free and was never really meant to mean free," Martin Schneider, senior director of communications at SugarCRM, said. "Open source runs through everything we do, it enables us to be transparent and gives customers more power. We are an open source company and it's why we're better than proprietary companies."'"

Comments

He's right

[iplayfast]

There's nothing about open source that means no cost.

Re:He's right

[mwvdlee]

Not directly.
But (AFAIK) if you pay for an open source (as OSI defines it) product, you are allowed to copy and give it away at no cost.

Re:He's right

[n3v]

OSI?

Haven't you ever seen the Venture Brothers?

Re:He's right

[snowgirl]

Not directly.
But (AFAIK) if you pay for an open source (as OSI defines it) product, you are allowed to copy and give it away at no cost.

That depends upon the license used. With the GPL and BSD, then you're absolutely right.

However, there could be an open source license that doesn't allow this. Find a counter-example is left as a problem for the reader.

Re:He's right

[shaitand]

If there were it wouldn't be open source. Anything that would prevent this would prevent you from modifying and distributing your modified version. That is the core of open source and without that ability a license isn't open source.

Re:He's right

[NNKK]

Have you tried actually reading the first criteria of the open source definition?

http://opensource.org/docs/osd

Re:He's right

[shaitand]

Someone else already pointed to http://opensource.org/docs/osd but he pointed to requirement 1. It's actually 3.

Re:He's right

[snowgirl]

You deny Section 1, but claim Section 3.

Section 3 only requires the right to distribution to be granted when there is a modification or derivative work.

Section 3 cannot be used to justify simply copying the source code and sending it out to anyone.

Re:He's right

[snowgirl]

Section 1 only requires that redistribution rights be granted to people when made a part of an aggregate work from multiple sources.

This means you are not guaranteed the right to just copy the code out to anyone and everyone... without created an aggregate work.

Re:He's right

[BigDXLT]

I think the trouble is in overloading the term "open". It's certainly not closed source. Someone witty from marketing should come up with a name for these semi-open source licenses. Paid source, perhaps?

Re:He's right

[shaitand]

When did I deny section 1? I just said it isn't the relevant section.

Change the title and you'd fall under section 3, package it alongside the source to any of the required libraries and you'd fall under section 1.

Any way you slice it, there is no license that could in any practical manner prevent free redistribution and still be open source let alone free software.

Re:He's right

[shaitand]

I think Microsoft came up with one, shared source.

Re:He's right

[Eivind]

There cannot be a counterexample, not if you use the conventional definition of "Open Source". The two common definitions for that, is the one from the Open Source Initiative, which states as one of the 4 freedoms that must be guaranteed for a license to qualify as open source that: (and I quote)

"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."

You can look at FSFs definition of "Free Software" if it suits you better, but it says the same thing, though in more words.

The freedom to redistribute, is one of the central freedoms in Open Source. If that freedom is limited, for example by insistence that you -must- charge for such redistribution, the software in question, does not qualify as Open Source. (nor as Free Software)

Re:He's right

[h4rm0ny]

Has anyone here actually read the article (I know, stupid assumption). SugarCRM has a dual licence. There's a "Community Edition" and a "Professional Edition" (also an Enterprise Edition, but that's not different from Professional - it's just the support offers sort of thing as far as I recall).

Now the Professional Version is obviously not "closed source" because it's a great sprawling PHP application so they have to give you the source. But that doesn't make it "Free Software". It requires a licence on a per user basis. In contrast, the Community Version is what we call "Badgeware". You can download it free, you can deploy it free with whatever users you like and you're free to make and distribute plugins and such for it. But you can't remove the SugarCRM logo and weblink for example. (In fact, there are some amusing little attempts to prevent people from doing that in the code, e.g. the legal notice that comes up if you alter the SugarCRM image doesn't appear as text in the files, but encoded as base64).

Anyway, there's an open sourcish community around the Community Edition that write tools for it. But, IMO, the whole thing doesn't feel very open sourcey. What it comes down to is not an issue of programming, so much as it comes down to business needs. SugarCRM has a system of "modules" - pluggable business entities such as Contacts, Product Lists, Accounts, etc. The great big difference between the Community and Professional versions is that the Professional version comes with additional modules. And for most businesses (I would say), they're modules that you really need. There are various other bits and pieces like the Professional Edition supports workflows whereas the Community Edition does not.

What it comes down to, is that SugarCRM has a community edition which serves as a good bit of PR, a hook to get in new users and a source of occasional free bug-fixes. But most serious businesses - the ones who actually are potential customers - will end up needing the features of the non-Free Professional Edition. There are attempts to replicate some of what the Professional Edition does in the Community one, but from what I've seen they don't really compare and of course the company itself isn't helping much because primarily they want people to buy the Professional Edition to get those features. Their forums are also littered with unanswered technical questions. If you're a paying customer and you file a support request with them, you get fixes (in my limited experience with them), but if you're a Community type asking questions on the forums, you take your chances. It would also be pretty difficult to make any substantial changes to the code base because you're always tailing the Professional Edition which SugarCRM control. So if you write a wonderful new thing for it (the do-it-yourself Open Source way), expect there to be a good chance that it will be incompatible shortly.

I actually quite like the model of a free version of software and then a paid-for pro version with extra coolness. It's a model that works well. But when you combine that with Open Source, it becomes a little more dubious (maybe) because there's the possibility that you use the name of Open Source but create a system where in practice, people can't meaningfully participate and it's primarily a hook into the paid version. This is where I feel SugarCRM are. I have no doubt that there are people using the Community Edition for business purposes, but I think what I describe is the bird's eye view of the situation.

Re:He's right

[Richard_at_work]

That's *an* open source definition, not *the* open source definition.

Re:He's right

[RoFLKOPTr]

If there were it wouldn't be open source. Anything that would prevent this would prevent you from modifying and distributing your modified version. That is the core of open source and without that ability a license isn't open source.

The term "open" means the source is open for you to view and perhaps compile for yourself. That's it. The problem with today's open source movement is that people now automatically assume that because something is open source, it must be free. There's more benefit to being open source than just being free. You can view the source and see exactly what's powering the program, run your own audits, find your own bugs, and make sure that what's running on your machine is exactly and only what you want running on your machine. Being free happens to be an extra perk for most open source software, but it is definitely not required.

Re:He's right

[Eivind]

http://opensource.org/docs/osd

Requirement #1 is "Free Redistribution", i.e. that you have the freedom to, without limitation, redistribute the software.

Would you like ketchup with that ?

Re:He's right

[h4rm0ny]

No, Snowgirl is right. Though I see you got your +1 Sounds Confident mods. What you are describing is Free (capital 'F') software, also known as "libre" software. Open Source does not mean anything other than you have the source available. For example, so you can inspect it for security reasons, so you can make in-house only changes. Free Software however, is what you get under the GPL. Free Software must of course be Open Source, but the inverse is not the case.

Hear that booming tread? That's Richard Stallman walking up your drive with a four hour lecture of the philosophy of information sharing. I'd hide under your desk if I were you. ;)

Re:He's right

[snowgirl]

I imagine this is why people hate lawyers...

everything is couched in such terms that they know are unassailable, and even when a reasonable argument can be mounted against it, there are still the technicalities introduced intentionally at the start to ensure that they can't be wrong.

I cannot claim that the vast majority of people reading what I wrote will not make invalid assumptions about what I meant, and thus make me sound like I was wrong, but seriously...

The OSI requirements for open source license definitions would allow me to make a license that prevents people from distributing unmodified copies of my program in anything but aggregated products.

Essentially, only avoiding the specific case of people pawning off my direct unmodified works individually. Which is the only thing I objected to in the first place.

Re:He's right

[snowgirl]

Note my above comments that have all been unfortunately marked "troll" or more accurately should have been marked "rules lawyer".

I can produce a license that the OSI would approve that does not allow for the immediate action that I objected to right here: buying the original, taking the source code, and redistributing it singularly... without any modification or derivative work.

Re:He's right

[drsmithy]

If you can cite a requirement that an open source license must allow indiscriminate distribution, then I will willingly eat my words.

It would be far more interesting if you could cite even a hypothetical license that meets the commonly accepted criteria of open source but manages to disallow "indiscriminate distribution".

Re:He's right

[harlows_monkeys]

That's nice. However, there are other sections besides section 1. They aren't there just to help with the feng shui of the site.

Re:He's right

[snowgirl]

http://opensource.org/docs/osd

Requirement #1 is "Free Redistribution", i.e. that you have the freedom to, without limitation, redistribute the software.

Would you like ketchup with that ?

Only in an aggregate work. Not as an individual work... which I believe the original objection was about.

Re:He's right

[snowgirl]

It would be far more interesting if you could cite even a hypothetical license that meets the commonly accepted criteria of open source but manages to disallow "indiscriminate distribution".

The license does not allow redistribution of the original code in any form except when included in an aggregate work. (Satisfies section 1)

The license requires that distribution of the original source code cannot be done, but that modifications and derivative works may be distributed as patches. (Potentially satisfies section 3)

Alternatively, the license requires that the original code may not be distributed unless modification is made or a derivative work made. (Definitively satisfies section 3)

Thus, this prevents the specific claimed action, of simply buying and then distributing unedited.

Re:He's right

[Chrisq]

That's nice. However, there are other sections besides section 1. They aren't there just to help with the feng shui of the site.

Naturally, and I'd be happy to cover any of the other sections.

Section 3 provides that one can freely distribute modifications and derivative works. We weren't talking about a modified or derivative work, we were talking about redistributing the original.

So I change a few labels then redistribute .......

Re:He's right

[harlows_monkeys]

Try section 2.

Re:He's right

[Chrisq]

http://opensource.org/docs/osd

Requirement #1 is "Free Redistribution", i.e. that you have the freedom to, without limitation, redistribute the software.

Would you like ketchup with that ?

Only in an aggregate work. Not as an individual work... which I believe the original objection was about.

OK so I package it with a picture of your mum sucking my dick

Re:He's right

[ksandom]

Section 1 only requires that redistribution rights be granted to people when made a part of an aggregate work from multiple sources. This means you are not guaranteed the right to just copy the code out to anyone and everyone... without created an aggregate work.

How did this get marked as troll?

Re:He's right

[tapanitarvainen]

No, Snowgirl is right. [...] Open Source does not mean anything other than you have the source available. For example, so you can inspect it for security reasons, so you can make in-house only changes

No. It used to be fairly common to sell software with source code, with explicit restriction that it may not be redistributed: source was only provided for in-house use. That is certainly not open source.

Open Source does imply the right to redistribute, and that's explicitly allowed in every OSI license, snowgirl's legalistic quibbling notwithstanding: the definition referred is not a license or any other legally binding document, and if someone really tried to make a license that explicitly forbids redistribution of the program in unmodified and non-aggregated form, I'm sure ISO would reject it - possibly clarifying their definition, if they thought it was necessary - but the intent of the definition is clear enough, whether or not it appears legally watertight.

Re:He's right

[tapanitarvainen]

I can produce a license that the OSI would approve that does not allow for the immediate action that I objected to right here: buying the original, taking the source code, and redistributing it singularly... without any modification or derivative work.

I don't think you can. Feel free to try though. OSI is not bound to your reading of their definition - or any other outsider's reading of it for that matter, and they can ignore obviously unintentional loopholes, or amend the definition to close them if they choose.

Re:He's right

[Inconexo]

Can you provide an OSI-certified example?

Re:He's right

[h4rm0ny]

No. It used to be fairly common to sell software with source code, with explicit restriction that it may not be redistributed: source was only provided for in-house use. That is certainly not open source.

I remember those days, believe it or not. :) No, that really was (and still is) open source. What we have going on here, is a group (the OSI) attempting to re-define Open Source to be synonymous with Free Software. Perhaps one day they will succeed and we'll have to start referring to any open source software that isn't also licensed under something like the GPL as "shared source". One day, perhaps, but (best Aragorn voice) it is not this day!.

Seriously, if you want to refer to software that is both open source and includes the right to distribute and modify, call it "Free Software" like the FSF, or "Libre" software. It's nice, unambiguous, is an existing term and doesn't confuse half the software world which is still filled with people like me who recall Open Source meaning only that the source code is available.

Re:He's right

[Anonymous Coward]

Looks like you're technically right from reading the OSI criteria (and I blew a couple of mod points to counter the 'troll' mods) but you're standing on a very small patch of ground.

Your theoretical licence would be super-easy to work around with trivial aggregation or trivial modification, as others have pointed out. It appears to be counter to the aims of OSI, but I'd guess that since it's such a feeble loophole it's not worth plugging.

Re:He's right

[mr_gorkajuice]

Please read your own quote in its entirety.

"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."

It's quite obvious that you only care about the bold part. Snowgirl is correctly pointing out that there's more in there:

"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."

So yeah, add your own Hello World application and you're most likely good to go. However, you are failing to prove that indescriminate copying is per definition allowed in open source context.

Re:He's right

[AlXtreme]

ie. they use the "open core" model like what the submitter hinted at: have an open source portion of your system but for any practical use you need to buy the 'professional edition'.

El reg had a small discussion about this a while back due to in part this blog post.

IMHO this is a business/marketing decision that will alienate open source fans. I have had to use similarly-licensed software in the past, don't think I'll actively pitch in with new features or submit bugreports for either the open or licensed version. You can make a business out of open source alone, you don't have to handicap your open source version to make your professional edition look better.

Thankfully this tactic should quickly lead to forks. For niche markets the community will need to rally in order to support the fork, but for that to happen you need a strong community, which tends not to develop around an open core product.

Re:He's right

[tapanitarvainen]

No. It used to be fairly common to sell software with source code, with explicit restriction that it may not be redistributed: source was only provided for in-house use. That is certainly not open source.

I remember those days, believe it or not. :)

So do I (yeah, I'm old). Did you really use the term "open source" then? I'm sure I didn't, the vendors certainly didn't, actually I'm pretty sure I never heard the term back then.

Of course I may have missed someone using it, but it certainly wasn't a common term.

As far as I can tell, the very term was invented as a generic term for freely redistributable source - as a substitute for "free software", which had too heavy political and philosophical connotations.

Seriously, if you want to refer to software that is both open source and includes the right to distribute and modify, call it "Free Software" like the FSF, or "Libre" software. It's nice, unambiguous, is an existing term and doesn't confuse half the software world which is still filled with people like me who recall Open Source meaning only that the source code is available.

Can you point out any references to "Open Source" that predate the current common meaning (that includes free redistributability)?

Re:He's right

[Richard_at_work]

Their definition is *a* valid one, regardless of the fact that they 'invented' the term (if they even did), they do not get a monopoly on it.

Re:He's right

[mr_gorkajuice]

While it's easy to work around and create versions that can legally be redistributed, it will still be slightly altered versions. They might be practically identical by all means and standards, or they might include malware. Anyone looking to get the product for free (gratis), legally, will get "something slightly different". That fact alone might be enough to make some people prefer the real, authorized, payed-for deal, thus giving the potential license restrictions SOME degree of actual purpose

Re:He's right

[phatcabbage]

Citations for that claim?

Re:He's right

[icebraining]

Wrong. "Open Source" is defined by the OSI and besides letting you view the source, it must let you distribute it for free.

http://www.opensource.org/docs/osd

Re:He's right

[dotancohen]

I actually quite like the model of a free version of software and then a paid-for pro version with extra coolness. It's a model that works well. But when you combine that with Open Source, it becomes a little more dubious (maybe) because there's the possibility that you use the name of Open Source but create a system where in practice, people can't meaningfully participate and it's primarily a hook into the paid version. This is where I feel SugarCRM are. I have no doubt that there are people using the Community Edition for business purposes, but I think what I describe is the bird's eye view of the situation.

Although I agree with you about the model of a free version of software and then a paid-for pro version, it never works out in practice. Just look at Star Office and the mess it caused when people tried to commit patches to Open Office. Sun would require the patchers to give them copyright on the code, and very few would do that, hence go-oo was born. Oracle is making an even bigger mess of it, even though Star Office has been dropped and is now part of Open Office.

About the only good implementation is Red Hat, and that is because they play by the FOSS rules. Even Novell tried a variation on the free/paid route, not playing by FOSS rules, and look at where Suse is today.

Re:He's right

[samoanbiscuit]

The term for such as that, I believe Microsoft has termed Shared Source.

Re:He's right

[icebraining]

This whole discussion is moot, because the community edition is badgeware - you can download the source for free on their site, and the Commercial version isn't Open Source*, because it require a per user license.

* as defined by the OSI. I don't respect any other definition

Re:He's right

[nstlgc]

That's just the definition of Open Source according to the Open Source Initiative. Nothing more, nothing less.

Re:He's right

[icebraining]

You can't do neither, because you need a key for each copy of the paid version - you pay by the user. It's not OSI approved in any way, it's just a marketing abuse, imho.

Re:He's right

[LingNoi]

This isn't wikipedia, do you own research. You can start by not being lazy and looking at the main OSI site.

Re:He's right

[samoanbiscuit]

I thought go-oo also required copyright assigned to them when submitting patches (I agree they're not as retarded as Sun when it comes to accepting patches).. Why is Oracle making an even bigger mess of it? By attaching it's name to the project? And what exactly did Novell do different than Red Hat that makes Suse such a 'dog'?

Re:He's right

[Anonymous Coward]

Has anyone here actually read the article (I know, stupid assumption). SugarCRM has a dual licence. There's a "Community Edition" and a "Professional Edition" (also an Enterprise Edition, but that's not different from Professional - it's just the support offers sort of thing as far as I recall).

Now the Professional Version is obviously not "closed source" because it's a great sprawling PHP application so they have to give you the source. But that doesn't make it "Free Software". It requires a licence on a per user basis.

It doesn't make it "open source" either. It makes it proprietary software with source code available.

Search for "open source etymology" or similar; it's an artificial term with an artificial meaning.

Re:He's right

[samoanbiscuit]

Strawman much? All that could be done is a work with a single bit added, and a requirement that each time it was copied/passed around, this bit were 'flipped' so to speak, such a trivially achievable loophole around the license would allow trustworthy copies of the work in the community... Comparing two works, showing only one bit of difference, would be enough security for most people...

Re:He's right

[gmhowell]

People hate lawyers for many reasons. This is just one of them.

Re:He's right

[gmhowell]

Software licenses, GPL and BSD especially included, are all about legalistic quibbling.

Re:He's right

[misiu_mp]

There are no 'a' definitions of open source. There is only one. The one that grands the user full use, modification and redistribution rights. Regardless who has the trademarks or inventions. Thats the core. You cant go around changing meanings of words according to your personal view. If you want other meanings there are other words, such as 'shared source'. OSI is just an organization which the community trust to be able to recognize open source licenses. It doesn't matter how they define it. If they start using a different definition, they might lose the trust and thus their purpose.

Re:He's right

[misiu_mp]

No. It used to be fairly common to sell software with source code, with explicit restriction that it may not be redistributed: source was only provided for in-house use. That is certainly not open source.

I remember those days, believe it or not. :) No, that really was (and still is) open source. What we have going on here, is a group (the OSI) attempting to re-define Open Source to be synonymous with Free Software. Perhaps one day they will succeed and we'll have to start referring to any open source software that isn't also licensed under something like the GPL as "shared source". One day, perhaps, but (best Aragorn voice) it is not this day!. Seriously, if you want to refer to software that is both open source and includes the right to distribute and modify, call it "Free Software"

Those old programs weren't open source just by the virtue of being accompanied by their source code that was open for view. Open Source in the meaning that become public in the 90's always included freedom to modify and/or distribute. If there were other meanings, they were local and are now superseded. The main difference between Free Software and Open Source is that Free Software requires you to use a free license on the derived work, and Open Source only gives you the right to do so. This difference in point of views is the reason for the creation of the term Open Source. That's also why Open Source is not necessarily Free Software.

Re:He's right

[misiu_mp]

No need to do that. #2 clearly states: "The program must include source code, and must allow distribution in source code as well as compiled form".
Note: "The program must (...) allow distribution in (...) compiled form". No mention it has to be derived.

Re:He's right

[Richard_at_work]

Uhm, I have absolutely no need to do any research at all and as such I am not being lazy - if you are referencing my "if they even did" comment, then I stand by that *regardless* of what their own website says.

Re:He's right

[Alex Belits]

Open source was not a term that was in use before the the open source definition. It was described as "comes with source code" or "source code available" or similar.

No.

Open Source, as applied to software, was never in use before Eric Raymond/Open Source Initiative started using it. The term "Open Source" was invented to be distinct from more strict definition of "Free Software". Please note that similar definition used by the other OSI founder, Bruce Perens, is called "Debian Free Software Guidelines" and does not mention anything "open" -- when it was created, the term Open Source Software did not exist.

The term "Open" (without "source") was routinely applied to software that is most definitely closed source (for example, everything by Open Systems Foundation), and has absolutely nothing to do with Open Source.

Re:He's right

[Richard_at_work]

Again with the "citation needed" rubbish - its not a claim, its an opinion and I do not need any "citation" for an opinion.

Re:He's right

[vegiVamp]

> it's a great sprawling PHP application so they have to give you the source

That's a misconception. Zend offers the possibility to only provide encrypted PHP to your customers.

Also, the whole open-or-not merely comes down to what license they use. Nothing in GPL, for instance, prevents you from saying "we won't give you this bit of functionality unless you pay".

Here's the GPL, very briefly:

* IF your distribute a program, THEN you MUST also provide the source for free
* You MUST allow people to modify and/or redistribute the program, and they must then too comply with the GPL

Nothing says that they have to give the code to people who haven't paid them for the commercial version, given that everyone has to pay for the commercial version.

Thank you, kdawson, for another utterly useless post.

Re:He's right

[Richard_at_work]

Hate to break it to you, but no they do not - the OSI do not hold a trademark on the term "open source".

Re:He's right

[pinkushun]

I'm not too sure that paying entitles you to redistribute the source, even if you are entitled to the source as a paying customer. Can you provide any real world instances of this, for our reading pleasure?

Re:He's right

[Hognoxious]

if someone really tried to make a license that explicitly forbids redistribution of the program in unmodified and non-aggregated form, I'm sure ISO would reject it

On what grounds is it their call to make?

Re:He's right

[fuzzyfuzzyfungus]

Because "customer" does not necessarily imply "end user of product" just "buyer of product".

Say, for instance, we have a Product X that is Open Source, in the sense of being redistributable; but only distributes source to customers who buy the binary version(as is compatible with OSI or Free Software licences).

Say, for instance, you run a Linux distro, and your customers are whining about "Why isn't Product X in the repositories?". Assuming you think that it will make you more in additional support contracts/boxed copies/OEM integrated sales/whatever you just feel like doing it, it is completely economically rational for you to buy a binary copy of Product X, obtain your copy of the redistributable source, and then package it for your distro.

It would be similarly logical, though requiring more organization, for a coalition of small customers to club together and buy a single binary copy, then share source copies between all members of the club.

Re:He's right

[tapanitarvainen]

if someone really tried to make a license that explicitly forbids redistribution of the program in unmodified and non-aggregated form, I'm sure ISO would reject it

On what grounds is it their call to make?

Sorry, typo: I meant OSI of course.

Re:He's right

[Chriscypher]

And this dual Community and Professional edition has backfired with their implementation IMHO.

I have tried to use the Community Edition for internal use, but what I have found is that all pages are notoriously slow and the modules they provide are missing essential functions. Imagine an application with 4 second page loads and a "project module" where tasks can have no dependancies, but 'trust us' the Pay Version makes it all better.

They seem to spend a lot of time trying to figure out how to wall their garden rather than providing a decent calendar, etc etc. They certainly could use more community developers to refine and extend the software, but they seem to be against the community on many levels I haven't the energy to digress into.

Re:He's right

[gtirloni]

SugarCRM has adopted the Open Core model. Open source version - nice for a try, useless for deployment "Professional" version - what you actually was expected. Sounds like shareware uh? I doubt they have a active community of contributors around it either.

Re:He's right

[gtirloni]

With "open core" there is a conflict of interests. The sponsoring company has to carefully select which features will be open source and which will go into their "professional" edition. I've still to see a company that gets it right because it's impossible. You'll get a lot of pissed customers when they find the open source version cannot be deployed. It's either open source or proprietary.

Re:He's right

[gtirloni]

For a software to be "open source" you MUST make the source available for free. Period.

The problem with open core companies is that in most of them, the "open source" (or kid's) version is not usable providing only a minimal set of features. They will then pitch their "professional" edition as it came with "add features". In fact, it comes with essentials features that wouldn't possibly deploy in production without.

Open core companies are trying to set feet in both open source and proprietary worlds and will fail in both.

Re:He's right

[orasio]

That's *an* open source definition, not *the* open source definition.

That sentence is right. The problem is that you can use that reasoning to question anything. It's right, but useless.
That's *an* Internet Protocol, not *the* Internet Protocol.
That's *a* World Wide Web, not *the* World Wide Web.

People use conventions in order to communicate. When someone uses "open source", and knows what he is talking about, he means OSI open source. There is the possibility that that someone does not yet know what open source means, or that he is trying to twist its meaning, but there is a general convention about what we mean when we say "open source".

There was the same problem with "free software", and that might lead to confusion ("software libre" in spanish doesn't leave any doubts). I don't think anyone who knows enough can honestly use "open source" meaning something other than OSI open source.

Re:He's right

[hey!]

I actually installed the community edition of SugarCRM a few years ago. My take on it is that it's got the right problem (a big, big, big part of creating a great piece of software), but that the system design and implementation is painfully amateurish. The database schema was an incoherent joke, the code meandering, verbose and inarticulate. There's no reason for code (even PHP) to be that bad.

Obviously a huge amount of work went into the thing. The kind of work you do when you've got a poorly thought out system and real customers to satisfy. The thing about that kind of work is that if you hack away at a system long enough *in response to customer needs*, eventually it will fill those needs fairly well. Being badly engineered doesn't preclude providing value to users. We certainly found it useful, but whenever I had to fix a bug or tweak something, I was constantly amazed that the system worked at all.

Now we all know there are two schools of thought about software development: the incrementalist (make the software work even if it is ugly) and the purist (make it elegant even if you have to rewrite it). The reason these two schools persist is that they are both right in different situations. There are times you have to live with less than elegant, and times when you have to bite the bullet and do major rewrites. I think most successful programmers balance these impulses, tidying up and refactoring as they fix bugs and meet customer's needs. The sign of a skillful programmer is that the more he works on a body of code, the simpler and more elegant it becomes. But when you have a gawdawful mess like SugarCRM, it makes no sense to invest anything more than occasional trivial effort unless you're willing to commit to a complete fork. You'd have to do major refactoring unless you were willing to spend all your time hacking your way through cruft, and the SugarCRM folks probably wouldn't because they actually understand all that unnecessary complexity.

Overall I'd say that SugarCRM is a useful, but mediocre piece of software. If you can live with its limitations, it is an asset, particularly in a small business where you have to introduce management to the novel concept of CRM before getting them to part with money. SugarCRM is not much of an asset to F/OSS, because it's not likely to attract many talented contributors to the core system, yet discourages them from developing competing solutions because it is "good enough" for not-too-demanding users.

Re:He's right

[shentino]

Speech versus Beer

Re:He's right

[silentcoder]

>However, there could be an open source license that doesn't allow this. Find a counter-example is left as a problem for the reader.

Any license like that would violate the FSF's 2nd freedom. To be free software it MUST grant that right. The same requirement exists as part of the open-source definition.
Since open-source is trademark (at least, it was going to be - I'm not up to speed on whether the trademark application was approved) it would be a trademark violation to call something open-source unless it's license is on their approved list.

All the licenses that are approved by either the FSF or the OSI permit free distribution and approving a license which didn't would violate the founding rules of either organisation so it's not going to happen.

Re:He's right

[silentcoder]

http://www.gnu.org/philosophy/free-sw.html
See freedoms 2 and 3 = e.g. it cannot be free software.

http://www.opensource.org/docs/osd
See article 1 of the open source definition. Therefore it cannot be open-source either.

Conclusion: do you think your words would be tastier if you add some ketchup before you eat them ?

Re:He's right

[silentcoder]

>I never said anything about practicality. I said that it could be possible to construct a license that fit the open source definition, yet still did not allow indiscriminate redistribution. (That one can restrict it to only aggregate redistributions means not indiscriminate.)

If this actually gets done - I promise you the very next DAY the OSI would change the definition to remove it, they certainly wouldn't grant it "approved" status either way - on the grounds that it CLEARLY violates the spirit of the definition.

And while you could still in theory match the open-source definition - you'd be flat out on your arse at meeting he FSF's requirements so you definitely won't be free(libre) software. Freedom 2 specifically refers to unmodified free distribution (and freedom 3 to distribution of modified versions).

If the rift between the two movements actually split on this point. I think a LOT of people would follow Bruce Perens' lead. Yes, the man who WROTE the open-source definition in declaring "it's time we stopped talking about open-source and start talking about free software and ethics again."
He stated as much in a mail to the OSI mailing list a few months ago. His reasoning was that open-source was a term to focus on the technical merrits to get free software a foot in the door of business. It's there now and the goal achieved, but the ethical requirements that led to the CREATION of free software in the first place still exist, and it's time to start talking about that again because businesses ARE sold now and it's time to think of the users.

Personally, I never went to the open-source side. I've stuck with free software for the sake of freedom since well before 1998 when the term open-source was coined. I will keep sticking to it, but it's nice to see good folk like Bruce starting to think about the Stallman-ethics of this situation once more. I wonder how much things like the Novell-Microsoft deal had to do with that ? The same deal that made Jeremy Allison quit his job ?

Re:He's right

[shaitand]

"Ok, I'll clarify for the intellect impaired"

If your intended meaning matched the stance I reiterated we are on the same page so why continue to beat a dead horse? Ohhhh... I get it, you just wanted to vent your frustration about being wrong with a petty verbal slight.

If you are going to be that blatant you could at least try for something with more teeth. You just called me "intellect impaired" and then clarified with a sentence that means something entirely different than what you actually wrote.

To use your own words. I leave determining whether the individual who read the sentence as written or the individual who can't construct a sentence is "intellect impaired."

"I never said anything about practicality. I said that it could be possible to construct a license that fit the open source definition, yet still did not allow indiscriminate redistribution. (That one can restrict it to only aggregate redistributions means not indiscriminate.)"

Perhaps you need to look up the definition of discriminate. There is nothing about tossing a library or changing a source comment that forces one to be discriminate in their distribution.

In any case, it doesn't really matter what YOU intended because you were replying to me and therefore your comment inherits its context from mine.

"would prevent you from modifying and distributing your modified version. That is the core of open source and without that ability a license isn't open source."

Which is exactly the requirement set forth in sec 3 of the OSI requirements.

Re:He's right

[shaitand]

"Essentially, only avoiding the specific case of people pawning off my direct unmodified works individually. Which is the only thing I objected to in the first place."

This wasn't a thread about what you objected to. This was a thread about forking the enterprise product once a paid customer got the source. You claimed a license could have been used that prevented that and still be open source. That is false. No license that is OSI FSF approved could prevent forking.

Simply removing the trademarked branding from the project would constitute a derivative.

Re:He's right

[silentcoder]

>The license requires that distribution of the original source code cannot be done, but that modifications and derivative works may be distributed as patches. (Potentially satisfies section 3)

That we have precedent on. The original Minix license said the same and it was neither recognized as open-source OR free software. Sheez now I feel old.

Re:He's right

[MBGMorden]

Now the Professional Version is obviously not "closed source" because it's a great sprawling PHP application so they have to give you the source

Don't be so quick on that one. I don't know all the details on it, but our webmaster recently purchased a mailing list app (yes, I know the futility in this when there are free ones out there, but it wasn't my decision) written in PHP. The source came encrypted and it came with some special software setup to decrypt the code to the web server as needed to keep you from opening the code and looking at it.

Now I'm sure with a little work you could still get at the source, but still, there are PHP apps out there that can and do purposefully obscure the source from the user.

Re:He's right

[drinkypoo]

But (AFAIK) if you pay for an open source (as OSI defines it) product, you are allowed to copy and give it away at no cost.

However, there could be an open source license that doesn't allow this. Find a counter-example is left as a problem for the reader.

You (and the GP) have completely misunderstood the meaning of Open Source. It's not your fault; Bruce Perens and the OSI have been lying to you about what it means literally for years. The GP's statement ("as OSI defines it") is at the heart of the problem; Bruce Perens does not have the right to redefine the term "Open Source", and by extension neither does the OSI. That's because Bruce Perens did not invent the term. As per the link, the earliest use of the term "Open Source" that I can find pertains to Caldera (I know, right?) OpenDOS, in 1996. Perens claims to have invented the term in 1997 but my research indicates that his first recorded claimed use was in 1998. So not only does he not have a good claim to the term, but his claim is not even as good as he says it is.

"Open Source" means that those who receive the binary can receive the source, that is all. The right to redistribute is part of the rights of Free Software, not Open Source.

Can we finally put this dog to bed, please?

Re:He's right

[MBGMorden]

"Open Source" existed as a concept LONG before OSI came around. They can't claim the authority to redefine the term as they see fit. If I start the "Free Speech Initiative" and define free speech as any speech that doesn't offend another person then that means diddly squat to the rest of the world, even if I do get a few zealots to subscribe to my newsletter.

Re:He's right

[drinkypoo]

Open Source, as applied to software, was never in use before Eric Raymond/Open Source Initiative started using it.

That is a lie, or at best, you are mistaken. But stating it as fact when you are wrong makes you a liar of a special sort, the kind who runs his mouth when he does not know what he is talking about.

The term "Open Source" was invented to be distinct from more strict definition of "Free Software".

No, it is very much the other way around. The term "Open" has been used in computing since long before ESR's problem with a printer driver. Bruce Perens attempted to steal the term for himself (as did several other people who claim to have invented the term at the same meeting, see my above article) and used it to found a minor cult of personality. He continues to champion this appropriation of the term in an attempt to increase his cachet to the levels of RMS or ESR.

The term "Open" (without "source") was routinely applied to software that is most definitely closed source (for example, everything by Open Systems Foundation), and has absolutely nothing to do with Open Source.

Open means interoperable. Full stop. Caldera used the term "Open Source" to mean "you can get the source" over a year before Bruce even claims to have invented the term, and almost two years before anyone else claims that he invented it. Your view of history is revisionistic at best and if you continue to promote it I will conclude that you have a vested interest in Perens being named as the coiner of the phrase. Please further note that ESR does not claim the term, only to have been present at the meeting at VA Software where it was supposedly invented, a full year after I can find the term being used in a press release from Caldera.

Re:He's right

[sribe]

As far as I can tell, the very term was invented as a generic term for freely redistributable source - as a substitute for "free software", which had too heavy political and philosophical connotations.

From one old guy to another: you are correct. Further, the people arguing with you are being either really stupid, or like Schneider, disingenuous. When source is provided with a license under a paid contract with all sorts of restrictions, that is "source code", as opposed to "no source", and there's nothing "open" about it.

Re:He's right

[drinkypoo]

Any way you slice it, there is no license that could in any practical manner prevent free redistribution and still be open source let alone free software.

One last time to link my article (Since amazon is not functioning on my site right now, I do not in any way get paid for any kind of clicks or views) and then I will stop: The supposed Open Source Definition does not define Open Source. The term predates the OSI. I have proven this conclusively; the use of the term in the sense in which redistribution is not guaranteed (though since you own your patchset you can always send that around, as in the case of qmail) is clearly Open Source:

Individuals can use OpenDOS source for personal use at no cost. Individuals and organizations desiring to commercially redistribute Caldera OpenDOS must acquire a license with an associated small fee. Source code for proprietary third-party components of Novell DOS 7 will not be published.

Further, Caldera does not even claim to have invented the term: "Caldera's OEM and Channel Partners can utilize the open-source code models for DOS and Linux to create low-cost, custom applications and enhanced utilities for vertical and niche markets." So OpenDOS was free but not Free software and yet was still Open Source, in that you could get the code and look at it. And because you retain the copyright on your patchset (has this been tested? is an overlay for someone else's artwork which is only meaningful when laid over their artwork a derivative work?) you can distribute that, so everyone who has PAID for the software can receive most of the benefits of Free software. For Caldera, Open Source was indeed the selling point of Linux (see above snippet) and not Free Software.

Unless you can somehow show that Caldera was abusing the term by bringing out a yet older citation, something which I for one would very much like to see, it will have to be accepted that the OSI was created well after the fact and is not entitled to redefine Open Source, which is the real reason they have never properly applied for a trademark. There will simply be prior conflicting use demonstrated, and the OSI's cachet (and that of Bruce Perens, who has most personal stake in this falsehood) will be deprecated.

Re:He's right

[drinkypoo]

Can you point out any references to "Open Source" that predate the current common meaning (that includes free redistributability)?

Yes. This predates Perens' erroneous claim of invention of the term by a year, and "everyone else's" claim of when it was invented at a meeting in which Perens was included (but everyone there thinks they invented the term, which is I suppose reasonable considering that no one there invented the term) by two years.

Bruce Perens has done an excellent job of convincing the masses that Open Source means what he wants it to mean, but the reality is that it meant source code access but not rights of redistribution before the OSI existed, before Bruce claims to have invented the term, and so on. Unfortunately, Even Wikipedia buys into this revisionist view of history which can be debunked by any asshole who can use google. And yes, I am that asshole, and I did use google, and I wrote the last-linked article which expands on this theory.

Can someone figure out who wrote that Wikipedia entry? It might be interesting.

Re:He's right

[mqduck]

The main difference between Free Software and Open Source is that Free Software requires you to use a free license on the derived work, and Open Source only gives you the right to do so.

You're wrong. That's the difference between copyleft and non-copyleft free/open source software. The Free Software Foundation maintains a list of licenses it considers free software, many of which do not require you to release derivative work as free software, including the BSD licenses.

Re:He's right

[medlefsen]

"Open Source" as a term was coined explicitly to serve as a better brand name for Free Software. It was intended to be a more business friendly, pragmatic version of Free Software.

The Open Source Definition found on the OSI site was how the term was introduced to the world. Even if you believe that the term was being used before that and with some different definition (which I haven't seen any evidence of, but it's possible), this is the source of the term as it's currently used.

What's really going on here is that proprietary companies hoping to cash in on the Open Source brand are attempting to broaden the popular definition to include things that aren't really open source.

• http://en.wikipedia.org/wiki/Open-source_software#History
• http://opensource.org/history
• http://www.catb.org/~esr/open-source.html
• http://news.slashdot.org/article.pl?sid=98/02/09/213900

Re:He's right

[Richard_at_work]

When someone uses "open source", and knows what he is talking about, he means OSI open source. There is the possibility that that someone does not yet know what open source means, or that he is trying to twist its meaning, but there is a general convention about what we mean when we say "open source".

I'm sorry, but that's just arrogant. So if someone doesn't use your exact adopted meaning, then they either are ignorant or malicious? Your opinion is showing.

Re:He's right

[afabbro]

Now the Professional Version is obviously not "closed source" because it's a great sprawling PHP application so they have to give you the source. But that doesn't make it "Free Software".

FYI, once can distribute closed-source php apps. They have an encrypting scheme. Some template makers for Wordpress, etc. do this.

Re:He's right

[Zerth]

I vaguely remember some guy that distributed the source code to his software, allowed you to patch it and allowed you to distribute the patches.

But you couldn't distribute the patched software, the user had to apply the patches.

Re:He's right

[icebraining]

I don't care who invented it first - it's the current definition, accepted by most people in the area, and anyone using it with another meaning and not explaining that clearly is simply a scammer.

If I start the "Free Speech Initiative" and define free speech as any speech that doesn't offend another person then that means diddly squat to the rest of the world, even if I do get a few zealots to subscribe to my newsletter.

Except that's not the case, as the OSI's definition of Open Source is currently the widely accepted one.

"A few zealots"? Please, you're embarrassing yourself.

Re:He's right

[drsmithy]

No. It used to be fairly common to sell software with source code, with explicit restriction that it may not be redistributed: source was only provided for in-house use. That is certainly not open source.

The ability to modify (and by implication, observe) source code is absolutely the most important aspect of "open source" philosophy.

Being able to subsequently give it away to all and sundry is a distant second, and the insistence on it being a key part of "open source" tarnishes the whole philosophy, and pretty much brings destroys any ethical and moral high ground it might have, boiling it down to: "we want stuff for free" (as opposed to "we want to be able to fix problems and/or make improvements").

Re:He's right

[Attila Dimedici]

Section 1 only requires that redistribution rights be granted to people when made a part of an aggregate work from multiple sources.

This means you are not guaranteed the right to just copy the code out to anyone and everyone... without created an aggregate work.

Right, it just means that if I want to distribute the software as part of a collection of software (say Sugar and Linux) I may do so, or the software is not "open source" according to OSI.

"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."

Re:He's right

[cg3001]

Your right, thats why some of the most seasoned developers of sugar and the modules went on to code Vtiger and left the sugar forums..and coding for them due to them wanting to close up the source, they have wanted to do that from the start...but just didn't have enough code changes to make it happen..

Re:He's right

[c0d3g33k]

It's not arrogant - it's a prerequisite for clear and un-muddled communication while keeping the latter reasonably streamlined and efficient. People agree by convention that in a given context a phrase or label means something specific so that they don't have to repeatedly specify exactly what they are talking about. When discussing software, open source generally stands for OSI open source. If you are not following convention, then you should really state explicitly that you mean something different. If you don't, you either don't know that there is a convention and what it means (ignorance) or you *do* know and are deliberately using a different meaning without saying so (malicious). Orasio was exactly right.

Re:He's right

[Richard_at_work]

Yes it is arrogance - you are once again forcing your chosen definition on to other people, whether they like it or not. Just because the OSI have defined it in a particular way, and you (and others) have chosen to use that definition does not make it the only one, nor the overriding one regardless of your opinion on perceived convention, because no such convention exists - the very fact that we are having this conversation proves that.

Re:He's right

[h4rm0ny]

That's a misconception. Zend offers the possibility to only provide encrypted PHP to your customers.

Yeeeesss. I considered whether I should add qualifications to what I said when I wrote, then didn't bother. Hey - it's Slashdot! No-one's going to fuss over the details! :D

I'm curious as to whether you've ever used Zend's encrypted PHP. The code obscuring would certainly be annoying, but far from insurmountable. The encoded PHP I don't have experience with. I would guess, given the nature of PHP deployment, it would necessarily be easy to build a decoder. For those reasons, I kind of skipped over this possibility in my reply. Curious for comments from anyone who is more familiar with this tool

Re:He's right

[h4rm0ny]

I actually installed the community edition of SugarCRM a few years ago. My take on it is that it's got the right problem (a big, big, big part of creating a great piece of software), but that the system design and implementation is painfully amateurish. The database schema was an incoherent joke, the code meandering, verbose and inarticulate. There's no reason for code (even PHP) to be that bad.

I have more recent experience. My opinions are similar, but not quite as extreme as yours, so perhaps the code base has improved somewhat. The code is self-modifying which I don't approve of for security reasons (and don't forget to back up your webroot if the code has changed itself!). The database schema is... not inelegant, but certainly atypical. SugarCRM will create new tables to represent new modules (either downloaded, or ones the user creates) and preserves relationships between tables in another table. Keys are all based on timestamps. So the database schema is dynamic and relationships reflected in the database data rather than in the code (i.e. if a Contact module can be related to an Account module, there's no line of code the looks up Contacts for Accounts exactly, but a row in the relationships table that says Contacts are Many To One to Accounts, or along those lines, obviously I'm simplifying a little just to show the principle). Before any DBAs get upset at that, I'll point out that SugarCRM is exclusively MySQL, so half the things that this makes it difficult to do you wouldn't be able to do with MyISAM anyway. ;)

The code base itself is a little erratic from what I saw. I don't rate it the best structured code I've ever seen, but I have seen worse. There are some howlers in the API though, I know that.

I'd be interested to know how well it scales. I suspect it scales adequately to a certain point so long as you throw sufficient hardware at it, but that due to the design, there will be a tipping point where things start going downhill quite quickly.

SugarCRM seems to serve its target market quite well - small to medium business that need some quick to deploy and easy to learn CRM software and don't want to get into coding something themself of course. I don't know that it would meet the needs of larger players. There's a possibility that it might become the MySQL of the CRM world. MySQL started off pretty flaky pre-5.0, but is now decent enough that you can make good use out of it and it's undeniably successful. But there could be a PostgreSQL out there ready to pounce too (PostgreSQL had excellent design from the start, but took a while to get up to speed. Now it matches MySQL punch for punch in performance and retains its more robust and feature rich capabilities). I don't know who PostgreSQL might represent in this analogy. Vtiger has been mentioned elsewhere, but I've not examined that one in much detail. Looks like it has potential.

So basically I agree with you, but I'd say it's not quite as bad as you describe any more. Time will tell. By the comments in this thread though, they might be better off either ditching the Free Software pretense or biting the bullet and going fully Libre. They'd need to move to a fully support-based income stream, but they might find big gains in terms of uptake and developer contributions.

Re:He's right

[shaitand]

No free software is something else altogether. Open source software does not just mean the source is "open", open source software is licensed under an OSI approved open source license and there are requirements.

http://opensource.org/docs/osd

If it doesn't meet those requirements it isn't open source and the source while it may be available ISN'T actually open.

Re:He's right

[h4rm0ny]

I agree with everything in your post, but just for reference it was Richard Stallman who had the problem with the printer driver, not (Everybody Loves...) Eric Raymond.

Re:He's right

[shaitand]

Open source is OSI. FSF free software is a different standard with different requirements.

You sound like a replay to the days of Microsoft trying to pawn their license as "open source". A battle they lost and ultimately caved to pressure and renamed to "shared source."

"It's nice, unambiguous, is an existing term and doesn't confuse half the software world which is still filled with people like me who recall Open Source meaning only that the source code is available."

Misremembering. Unfortunately, the world is indeed still filled with confused people who authoritatively claim inaccurate information. Did you know humans only use 10% of their brain?

Re:He's right

[jd]

Free as in Freedom NOT as in beer. And, frankly, the OSI can go take a running jump. Although I do not regard Richard Stallman as god (demigod, perhaps), I do regard the premise on which he bases his definition of what is free as being fundamentally sound. The OSI is ethically suspect and morally dubious at best. At worst, who does it actually represent? The EFF handles all the legal matters and the various consortia that have sprung up (such as the FSF) represent the special interests of the groups involved. The OSI adds... what?

Re:He's right

[shaitand]

I wouldn't disagree. Caldera used the term correctly, to refer to Linux and Free Software. As far as I know the term was originally synonymous with free software and redefined and formalized with OSI standards. This was in response to businesses that simply provided source access in some form abusing the term.

But free software is the original open source so redistribution is still implicit. There was nothing in what you quoted to indicate otherwise.

Re:He's right

[rsborg]

Overall I'd say that SugarCRM is a useful, but mediocre piece of software. If you can live with its limitations, it is an asset, particularly in a small business

Quite an insightful and valid assessment, as I have worked on it. The major work I ended up doing aside from configuring the basic objects involved lots and LOTS of jQuery to make the UI more responsive and add in non-Sugar JS widgets (which were fed json with a seperate data model layer I wrote myself). Overall, this suite really pales in comparison to Salesforce, but is useful if you need something good OSS wise, I'd go with a smaller lightweight CRM like Fat-Free CRM (RoR), or small/medsize CRM like SalesLogix (Sage).

Re:He's right

[warrior]

You are free to do that. Some, such as corporate customers or those just concerned about security, may want the original binaries as compiled by the company that is selling them. As snowgirl pointed out, you are not free to take those binaries and redistribute them at will.

I like this. This is the type of open source business model that I think a lot of people have been looking for. We want our software to be open. We also want to feed ourselves and our families.

Re:He's right

[malzfreund]

But if open source implies free, isn't the term FOSS (free and open source software) a pleonasm?

Re:He's right

[mwvdlee]

Yes, just because 99% of Open Source users communicate using the OSI definition of Open Source does not make that 1% (rounded up) wrong if they choose to use a different definition that only they understand.

It DOES make them wrong if they think they can adequately communicate with the other 99% though. Oh, and please note that it's not you that decides whether you communicate well, it's the 99% that doesn't understand you, that gets to decided on that.

Re:He's right

[harmonise]

The OSI hold a trade mark on the term open source

No, they don't. They failed to apply for a trademark on the term. They hold trademarks for "OSI", "Open Source Initiative", and their logo, but they do not hold a trademark for the term "open source".

Re:He's right

[harmonise]

When someone uses "open source", and knows what he is talking about, he means OSI open source.

"And knows what he is talking about?" What could mean anything. Talk about being non-specific. There is no agreed upon definition of what "open source" means. You prefer your OSI-related definition. Others may disagree.

Re:He's right

[ShieldW0lf]

When someone uses "open source", and knows what he is talking about, he means OSI open source. There is the possibility that that someone does not yet know what open source means, or that he is trying to twist its meaning, but there is a general convention about what we mean when we say "open source".

I'm sorry, but that's just arrogant. So if someone doesn't use your exact adopted meaning, then they either are ignorant or malicious? Your opinion is showing.

"'When I use a word,' Humpty Dumpty says, in a rather scornful tone, 'it means just what I choose it to mean -- neither more nor less.'

'The question is,' says Alice, 'whether you can make words mean so many different things.'

'The question is,' says Humpty Dumpty, 'which is to be master -- that's all.'"

Re:He's right

[Alex Belits]

That is a lie, or at best, you are mistaken. But stating it as fact when you are wrong makes you a liar of a special sort, the kind who runs his mouth when he does not know what he is talking about.

There is a very fundamental difference between a word being used once in some conversation, and a term that is published and accepted in general use by a worldwide community.

The only reference in the linked text that can in any way be verified, is to Caldera press release that reads:

Caldera enables its partners to integrate stable, low-cost Internet devices with the most powerful Internet server platforms and commercial add-on products. Caldera's support for the open-source code model provides partners with maximum flexibility in providing more complete product and service solutions.

Individuals can use OpenDOS source for personal use at no cost. Individuals and organizations desiring to commercially redistribute Caldera OpenDOS must acquire a license with an associated small fee.

This is very far from anything that "Open Source" meant at any point, and in this form it was not used by anyone before or after that release.

The term "Open" has been used in computing since long before ESR's problem with a printer driver.

"Open" was, and is used to describe things that have nothing to do with open source.

Open means interoperable. Full stop.

"Open" is not the same as "Open Source".

Caldera used the term "Open Source" to mean "you can get the source" over a year before Bruce even claims to have invented the term, and almost two years before anyone else claims that he invented it.

Caldera did not establish any kind of definition -- it used the same words once, to describe something completely different. It did not produce any kind of established meaning, either in its own documents or as a term accepted by anyone else.

Your view of history is revisionistic at best and if you continue to promote it I will conclude that you have a vested interest in Perens being named as the coiner of the phrase. Please further note that ESR does not claim the term, only to have been present at the meeting at VA Software where it was supposedly invented, a full year after I can find the term being used in a press release from Caldera.

I care about preventing subversion of established terms that describe ideas and principles important in the modern culture. Any particular person's role in this is insignificant compared to the effect of establishing the term.

Re:He's right

[Shadow Wrought]

I imagine this is why people hate lawyers...

Having worked with lawyers for a decade now I can say, yes. Yes it is. Except that the hatred of lawyers tends to increase exponentially the further their opinion is from yours :-)

Re:He's right

[snowgirl]

"Essentially, only avoiding the specific case of people pawning off my direct unmodified works individually. Which is the only thing I objected to in the first place."

This wasn't a thread about what you objected to. This was a thread about forking the enterprise product once a paid customer got the source. You claimed a license could have been used that prevented that and still be open source. That is false. No license that is OSI FSF approved could prevent forking.

Simply removing the trademarked branding from the project would constitute a derivative.

Oh, I'm sorry, I thought the original objection was to the statement: "But (AFAIK) if you pay for an open source (as OSI defines it) product, you are allowed to copy and give it away at no cost."

But then, my copy-paste might be wrong.

There's no talk of forking here. Just a simple direct copy with no modifications, nor derivative work.

Re:He's right

[snowgirl]

I will definitely grant you that there is no way that such a hypothetical license as suggested to be possible would pass the muster for "free(libre) software".

But if a company is using "open source" as a claim, then you need to consider it under its claims, not what its claims are not.

Not like the SugarCRM PL is approved by the OSI anyways... it could potentially be as restrictive as Microsoft's license...

Re:He's right

[Kitsune Inari]

There was the same problem with "free software", and that might lead to confusion ("software libre" in spanish doesn't leave any doubts).

Perhaps "freedomware" could be used? There's also the term FLOSS.

Re:He's right

[snowgirl]

I started the objection thread to begin with... thus all your comments inherit their context from me.

Your statement says: "he pointed to requirement 1. It's actually 3." In this statement, you explicitly deny requirement 1 as applicable in this situation.

Then I said: "You deny Section 1, but claim Section 3." Which carries the same intended context as your shorthand phrase.

"discriminate" means (in this context) "To make distinctions." That one can make the distinction that only modified or derivative works may be distributed is a discriminating factor. That one can make the distinction that the original code may only be redistributed in an aggregate work, is a discriminating factor.

Thus, an OSI approved license need not be required to allow indiscriminate reproduction.

Get off your horse... my narrowly worded objection is still not invalid... if you read it generally, and read a bunch of context into it that was not intended, then woe for you.

Re:He's right

[snowgirl]

http://www.gnu.org/philosophy/free-sw.html
See freedoms 2 and 3 = e.g. it cannot be free software.

http://www.opensource.org/docs/osd
See article 1 of the open source definition. Therefore it cannot be open-source either.

Conclusion: do you think your words would be tastier if you add some ketchup before you eat them ?

I never made any claim about "free software", (evidenced by my use of the term "open source license") and would have been quite open in agreement that a "free license" would guarantee a right to indiscriminate redistribution.

However, section 1 requires indiscriminate redistribution only for aggregate works. Thus, it is not indiscriminate redistribution itself (as it has a limiting factor).

I've been eating nice tasty chunks of words that I didn't ever say or imply, however I most certainly wouldn't enjoy them more with ketchup... I've prepared them too well to desecrate them with the accursed ketchup.

Re:He's right

[snowgirl]

Try section 2.

Section 2 only covers the requirement that source code must be transmitted along with the product itself. There is no guarantee of indiscriminate redistribution contained in this section. (However, there are no limiting factors to the scope of redistribution.)

Re:He's right

[snowgirl]

That's nice. However, there are other sections besides section 1. They aren't there just to help with the feng shui of the site.

Naturally, and I'd be happy to cover any of the other sections.

Section 3 provides that one can freely distribute modifications and derivative works. We weren't talking about a modified or derivative work, we were talking about redistributing the original.

So I change a few labels then redistribute .......

With you having done that, I could now potentially make trademark claims against you.

Re:He's right

[snowgirl]

Can you provide an OSI-certified example?

Likely not. It's an esoteric corner of the OSI definitions, and unlikely to be covered by the few examples available.

My statement though wasn't that it did exist, but rather that it "could" (emphasis in original.)

Re:He's right

[snowgirl]

You may argue that these satisfy section 3, but how are you getting past section 2? "The program must include source code, and must allow distribution in source code as well as compiled form" (emphasis mine).

Section 2 only applies when the product is transfered.

If you limit who gets the compiled product, then you only have to supply the source code to those individuals.

Section 2 contains no limiting factors, nor guarantees of indiscriminate redistribution. It only sets the terms upon distribution.

Re:He's right

[BLKMGK]

No officer I wasn't "speeding" because you see MY definition doesn't match yours and don't you go forcing your chosen definition on me either! Yes, that sounds workable....

His point about accepted terms for clear and un-muddled communication was a good one. If these folks have chosen to create and use their own definition that no one else recognizes that's their business but it's not good business...

Re:He's right

[snowgirl]

>The license requires that distribution of the original source code cannot be done, but that modifications and derivative works may be distributed as patches. (Potentially satisfies section 3)

That we have precedent on. The original Minix license said the same and it was neither recognized as open-source OR free software. Sheez now I feel old.

I've looked through a bunch of stuff, and the "new Minix license" most certainly is a derivative of the BSD license and most definitely free and open source... was the original Minix license ever actually proffered to OSI for approval?

If it wasn't, then there really isn't precedence. However, if it were, then I'd like to see what they said.

Re:He's right

[Kitsune Inari]

The main difference between Free Software and Open Source is that Free Software requires you to use a free license on the derived work, and Open Source only gives you the right to do so.

Uh, not really. That's the difference between "strong copyleft" and "weak copyleft". For example, both the GPL (strong copyleft) and the BSD (weak copyleft) licenses are recognised both as Free (by the FSF) and Opensource (by the OSI).

Actually, are there any licenses that are Free but not Open or viceversa? I've heard that it's possible for a license to be in that situation, but I don't really know of any.

Re:He's right

[Hognoxious]

I didn't notice the typo. So my question still applies.

Re:He's right

[silentcoder]

Minix's license was changed for version 3. I doubt very much it would have been offered under the original license as minix2 was under prominence before the OSI existed, before Linux really.

The license however was one of the main reasons linux took off and stole most of Minix's users- it was deemed unacceptable by the community. This is why Tannenbaum changed it later.

Re:He's right

[pr0f3550r]

The simple truth is that OSI is the de facto open source standard. If you don't believe that then perhaps this applies to you.

It is true that OSI does NOT hold the trademark which would make them the de jure open source standard. But the simple fact that they applied for the trademark shows greater investment and clout that other who simply seek to redefine the term to suit their own immediate needs. This gives OSI an even greater claim to the de facto status.

Additional basis for this claim is that vast majority of projects and software that claim to be 'open source' subscribe to these policies or licenses recognized by OSI as 'open source'. Other non-software users of the term may reference the term but there is some fluidity to it which is why OSI does not *own* it de jure.

Re:He's right

[jbolden]

I'd go with the DSFG, section 1

The Debian Free Software Guidelines (DFSG)

Free Redistribution

The license of a Debian component may 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 may not require a royalty or other fee for such sale.

Re:He's right

[NNKK]

You are extremely confused about how the law works. Trademark law is used to prevent someone from using your names and marks to pretend they're selling your product (e.g. I can't make a stereo and slap a "Sony" label on it).

There is absolutely nothing in trademark law to prevent me from taking software under an open-source license, changing the name and logos, and redistributing it freely. This is why CentOS can take Red Hat's product, strip out/replace the Red Hat trademarks, and distribute it for free.

Re:He's right

[styrotech]

As someone who was only first coming into contact with *nix and open source in the late 90s, I had sometimes wondered about how widespread the term "open source" was before Bruce/Eric etc officially defined it and whether it ever meant something else.

I don't know about anyone else, but to me your blog probably had the opposite result than what you intended. It clarified to me that before Bruce and Eric popularised it, practically nobody used the term open source much and certainly nobody has gone on record trying to properly define it. You really seem to have a large chip on shoulder about Bruce Perens for some reason.

I personally don't really have any problem with the OSI definition being regarded as the actual offical definition now.

Re:He's right

[harlows_monkeys]

Section 2 only covers the requirement that source code must be transmitted along with the product itself. There is no guarantee of indiscriminate redistribution contained in this section.

Yes, there is stop. Stop playing stupid games. First, you ignore section 2. Then when it is called to your attention, you only read part of it. Read the whole of section 2.

Re:He's right

[LingNoi]

See here why that doesn't matter.

Re:He's right

[shaitand]

And? That wouldn't prevent you from downloading the GPL'd source from the linksys support page and distributing it to others who hadn't bought linksys routers.

Re:He's right

[shaitand]

"open source" complies with the OSI requirements, simply making source available to someone does not meet the requirements. "free" means it complies with the FSF free software philosophy, again, simply allowing redistribution isn't enough.

"You've mixed up "open" and "free". Different terms meaning different things."

Not at all. You are just mistaken on what those terms mean in this context.

Re:He's right

[shaitand]

Open doesn't imply free. Simply being able to redistribute doesn't mean the software is free. "Free" in the free software sense means FSF approved license. "Open source" means OSI approved license.

Simply providing source code with no additional rights is neither since it wouldn't meet the requirements.

Re:He's right

[Eivind]

The question has a larger context. The parent of snowgirls claim was "if you pay for an open source (as OSI defines it) product, you are allowed to copy and give it away at no cost." which she disagreed with, claiming there could be Open Source licenses which do prevent you from doing that.

Which is factually incorrect. If a program is Open Source (any license that qualifies), you CAN infact "copy and give it away at no cost". Yes, it's technically possible you might have to include some random other program when you give it away, to make it an aggregate. But notice that there's no requirement that you have -written- this other program. I.e. I can include a copy of GNU hello, and be good to go.

So she's nitpicking.

For all practical purposes, a Open Source program that you have received -- CAN be copied and given away at no cost. Depending on how you read the rules, you MIGHT have to do some token action in addition, but that's a point that's not really important for the overall argument.

Can I give it away for free, or do I need to do 30 seconds of work and THEN I can give it away for free. That's a distinction without any substance.

Re:He's right

[Eivind]

Technically (perhaps) true, but irrelevant to the practical argument, which was that the price of a Open Source product in the free will tend to gravitate towards zero rapidly, because even if you sell the first copy, the recipient is free to copy and give it on for free.

Even -if- a Open Source license that required aggregation or modification was in force, that would not change the essentials of the argument. Okay, so I distribute the thing as a .tgz containing the original program plus a copy of Gnu Hello. Or I modify a single -comment- in the program, and distribute the 'modified' program.

But fine, technically such demands are restrictions. So you're right, technically: One cannot nessecarily copy it without ANY restrictions, but one CAN copy it without any MEANINGFUL restrictions.

If we're talking of the practical results of the rule though, the two statements are identical. If I have a copy of a open source program, I *can* give you, and everyone else, a copy at zero cost. Whether or not I need to run a single wget and a single tar command before doing that, doesn't really change the fundamentals of the situation.

Re:He's right

[ShieldW0lf]

Being able to subsequently give it away to all and sundry is a distant second, and the insistence on it being a key part of "open source" tarnishes the whole philosophy, and pretty much brings destroys any ethical and moral high ground it might have, boiling it down to: "we want stuff for free" (as opposed to "we want to be able to fix problems and/or make improvements").

Are you sure it doesn't boil down to "We have no desire to do the violence necessary to prevent people from redistributing your shit, we have no desire to fund the violence necessary to prevent people from redistributing your shit, but if you decide to try and carry out the violence necessary yourself (like that's ever going to happen), we're quite happy to do the violence necessary to lock you in a box for the rest of your life."?

That all sounds quite ethical to me.

Perhaps there's a bit of "Even though we are among those individuals wise enough to avoid your traps, we find it offensive the way you rope people into dependent relationships by getting them reliant on knowledge which you have intentionally rendered arcane, we find it threatening the way you wield those enslaved people, and for the sake of their suffering and for the sake of our safety, we want to break your control and leave you no more relevant than a man who sells wagon wheels for horse drawn carriages." in there... who knows.

That sounds quite like the moral high ground to me...

Re:He's right

[Inconexo]

Well, the OSI invents a term and says that "Open Source" is software that complies with A, B and C.

Who the heck can come and say that "Open Source" is something different?

You can provide other terms, specify other type of software, but changing the definition? How?

Re:He's right

[snowgirl]

The program must include source code, and must allow distribution in source code as well as compiled form. Where some form of a product is not distributed with source code, there must be a well-publicized means of obtaining the source code for no more than a reasonable reproduction cost preferably, downloading via the Internet without charge. The source code must be the preferred form in which a programmer would modify the program. Deliberately obfuscated source code is not allowed. Intermediate forms such as the output of a preprocessor or translator are not allowed.

I'm not going to play argue the semantics with you, because you're clearly insisting that section 2 states something that it does not. There is no limiting clause, or clause guaranteeing redistribution.

I've already been forced into a corner confining my position to one that says that a license could at best confine redistribution of an unmodified stand-alone copy.

Your claim that section 2 guarantees indiscriminate redistribution must fail... it only states terms that must be allowed for distribution.

Re:He's right

[snowgirl]

You are extremely confused about how the law works. Trademark law is used to prevent someone from using your names and marks to pretend they're selling your product (e.g. I can't make a stereo and slap a "Sony" label on it).

There is absolutely nothing in trademark law to prevent me from taking software under an open-source license, changing the name and logos, and redistributing it freely. This is why CentOS can take Red Hat's product, strip out/replace the Red Hat trademarks, and distribute it for free.

I mistook "change a few labels" for meaning in the source code. (Sorry, "label" has a specific meaning in computer source.) As some modification of the source code would be required as per my hypothetical license.

Switching the marks, and logos, would of course remove any trademark claim I had.

Re:He's right

[snowgirl]

Most certainly, unfortunately, without the original minix license being put to the OSI, they never made any claims about it, so it fails as a precedent.

I don't disagree that it would likely be argued by many as insufficient, but Microsoft's license was granted approval by OSI, so "open source" is a really unacceptable term to the "free software" community... even when approved.

Re:He's right

[silentcoder]

That particular microsoft license has ALSO been approved by the FSF. The requirements are identical. It's the GOAL of setting them that's different. The difference between free software and open-source is in their goals not in their practise. They do exactly the same things in exactly the same ways - they just do it for different reasons.
This is why they two communities generally collaborate on practical matters even as they argue on philosophical ones.

Re:He's right

[vegiVamp]

Not necessarily. PHP must at some point interpret the PHP code as lower-level instructions, so it should be possible to provide those lower-level instructions as the "encrypted" file.

Compare it to, say, the intermediate code that .net uses. And yes, that is still edible, but so is assembly. You won't have the benefit of clear function and variable names, so it'd be a bother.

(I'm not saying that *is* how it works, mind you - I've seen it used, but never did so myself)

Re:He's right

[vegiVamp]

> Hey - it's Slashdot! No-one's going to fuss over the details! :D

If I hadn't seen your UID, I'd given you the ymbnh treatment. This is Slashdot, the details are gonna be used to rape you to oblivion :-)

Re:He's right

[dotancohen]

> I thought go-oo also required copyright assigned to them when
> submitting patches (I agree they're not as retarded as Sun when
> it comes to accepting patches).

I did not know that. I've never submitted a patch. It sounds ironic, and I see no mention of it on the site:
http://go-oo.org/developers/

Can you tell me where you found that info?

> Why is Oracle making an even bigger mess of it? By attaching it's
> name to the project?

I actually like the Oracle name. But they've reassigned OOo engineers from bug fixing (I am in contact with Oracle for licensing so long as Issue #5556 gets fixed) and they've taken the MS Office ODF compatibility plugin and started charging for it:
https://shop.oracle.com/pls/ostore/f?p=ostore:product:8843910539649667::::P3_PPI,P3_LPI,P3_METRIC,P3_TERM:3710062267511641485310,3760869190131631757316,Application%20User,_Perpetual

> And what exactly did Novell do different than Red Hat that
> makes Suse such a 'dog'?

I never called Suse a dog.

Re:He's right

[harlows_monkeys]

And now you overlook section 4...

Re:He's right

[drsmithy]

The license does not allow redistribution of the original code in any form except when included in an aggregate work. (Satisfies section 1)

Trivially circumvented simply by including the code with anything else (let's say, an introductory README), thus creating an "aggregate work".

Thus, this prevents the specific claimed action, of simply buying and then distributing unedited.

It does not. In no small part because you're completely ignoring section 2: "The program must include source code, and must allow distribution in source code as well as compiled form."

Now, I'm not going to disagree that the focus on nearly all OSS licenses is, in fact, "getting stuff for free" (rather than having "the freedom to change things"), but that ship has long since sailed (well over a decade ago). While the ability to redistribute is only a relatively minor aspect of the purported philosophy of the OSS "movement", it's considered an essential "feature" of all OSS licenses, to the point that not allowing it means you're not "Open Source" and the premier "Free Software" license (the GPL) was engineered (and then re-engineered) to basically _compel_ it.

Re:He's right

[orasio]

I am not trying to hide my opinion (WTF?).
Yes, that is what I mean, that you could honestly misunderstand "free software", but using "open source" meaning "visible source" is either ignorance or malice.
In general, it's the first, in your case it's most probably due to professional trolling.

Re:He's right

[snowgirl]

Section 4 has three parts.
1.) Distribution of modified source code may be restricted as long as patches may be distributed with the original source for application in build process.
2.) Distribution of software from modified source must be allowed.
3.) There can be the requirement that modified source must use a different name or version number.

Now, to cover each one.

Part one only qualifies for modified source code, not for unmodified source code, which was the subject of the original objection.

Part two only qualifies for modified source code, not for unmodified source code, which was the subject of the original objection.

Part three only qualifies for modified source code, not for unmodified source code, which was the subject of the original objection.

Re:He's right

[snowgirl]

There are two licences that OSI approved but FSF lists as non-free: NASA Open Source Agreement and the Reciprocal Public License. Both of these are “send-back-ware” where changes have to be sent back or made available to the upstream developer. Many debian developers agree such payments-in-kind are non-free (OSI’s OSDefinition is based on the Debian Free Software Guidelines), but I think an early OSI advisor thought they were a good idea, so those approvals look like an OSI bug to me.

Source

So, differences do exist in practice.

Re:He's right

[snowgirl]

I've covered this before. Section 2 is ambiguous as it does contains neither limited factors, nor expanding factors.

It is rather a procedural requirement.

Section 2 may be fully complied with, even when distribution of the original unmodified source code is restricted exclusively to aggregate works.

Section 1 requires an "aggregate software distribution", so simply adding a README is not sufficient to be in compliance.

Re:He's right

[snowgirl]

Looks like you're technically right from reading the OSI criteria (and I blew a couple of mod points to counter the 'troll' mods) but you're standing on a very small patch of ground.

Your theoretical licence would be super-easy to work around with trivial aggregation or trivial modification, as others have pointed out. It appears to be counter to the aims of OSI, but I'd guess that since it's such a feeble loophole it's not worth plugging.

I will agree with anything and everything you have said in this comment.

I will gladly admit that I'm only "technically correct" and that it's a shaky piece of ground. However, it is a tenable position...

As stated, I should have been marked -1, Rules Lawyering rather than Trolling.

Re:He's right

[shaitand]

My other comments aside. It doesn't matter what the term originally meant. TODAY the term open source universally and without exception refers to using an OSI approved license.

Using it in any other way TODAY is an attempt at fraud regardless of what anyone might think the term meant historically.

Once upon a time a faggot was a bundle of sticks. Today it is a term to refer to a homosexual (with derogatory overtones). The only time you'd ever hear it refer to sticks/wood is as part of a comic relief act.

Re:He's right

[silentcoder]

>So, differences do exist in practice.

This is, however, an incredibly rare situation. In fact my choice of words was a direct quote from Richard Stallman's article on the subject: http://www.gnu.org/philosophy/open-source-misses-the-point.html

I would further say that the rarity of these cases are why they even persist. Those licenses are used only in a few rare cases and I don't think they represent a major issue. In most cases you WOULD send your changes to the original project anyway unless you had a very important reason to create a fork (perhaps the original project shut down, or were changing license - I once forked a program of my own after the company I worked for insisted on making the next version non-free. I quit my job and used the last free version to base a new free project on, as it happened my project still exists while nobody has even heard of the old one in several years - this is usually what happens).
But these licenses DO interfere with one of the core freedoms the FSF demand. The right to distribute changes yourself- and there ARE times when you need to. More-over those licenses would prevent you from using parts of this code to write something else from.

Frankly - I am surprised they were approved as I'm not sure they actually meet the OSD requirements, but that's not my call to make.

Ultimately though these cases are quite sufficiently rare that I can understand why Stallman says "The difference between the movements are philosophical and not practical". Nearly all the time that's right, the rare exceptions are quite insufficient to make it a good idea to refuse to cooperate on projects, or for somebody who believes in free software to refuse to use a program because it's author prefers open-source.

Re:He's right

[snowgirl]

Frankly - I am surprised they were approved as I'm not sure they actually meet the OSD requirements, but that's not my call to make.

As the author of the source material states, he sees them as "bugs" of the OSD.

But then, this whole argument all rests upon "technicality" anyways... :)

Well..

[ak_hepcat]

Sure, if it's open source, then one paying customer can take the source and fork it back out to everybody else for gratis.

That's what open source means.

Trying to disguise commercially licensed software as open source is setting yourself up for failure.

Re:Well..

[OverlordQ]

Yes, the community edition is AGPL, but they dont say anything about what license the Pro/Enterprise versions are.

Re:Well..

[glob]

> Sure, if it's open source, then one paying customer can take the source and fork it back out to everybody else for gratis

correct; mysql have the same business model, and there are places where you can legally grab the enterprise editions.

Re:Well..

[ducomputergeek]

Depends, things like PA-DSS and HIPPA suddenly can throw a monkey wrench into things. We forked an opensource project with the goal of getting it PA-DSS certified so we could use it to continue processing credit cards after July 1st of this year. We've done the audit and now just waiting for the paperwork to go through to get it listed as "certified" software. However, only versions signed and distributed by our organization is certified. Like SugarCRM, we only give out the code to customers who are paying for support contracts.

They are free to download the code and they could even compile and use the code in house and be okay under PCI-DSS. However the monkey wrench comes if they decided to compile the application and then distribute their version to other parties. Technically those 3rd parties would not be able to use the software to process credit cards since the version would not be "PA-DSS certified". And while the software would still be functional, if you used the uncertified version to process credit cards, then one could lose their merchant account. And processing credit cards is a MAJOR feature of the product and too risky for a lot of businesses to consider using it without that certification.

So while one could have all the source code, the source code without the "PA-DSS Certification" certificate doesn't do folks much good in practice.

Magento is doing something similar. Only their "Enterprise" version is PA-DSS certified. The Community Edition is not. And I suspect we're going to start to see more of this as time goes forward. I'm not saying it will be impossible to do it, but it is extremely hard. PA-DSS certification requires a lot of documentation and about $25k up front to pay for auditing, the PCI-SSC, and the best part is the validation is only good for 3 years. That's either a lot of community donations or someone bank rolling the operation.

Re:Well..

[ducomputergeek]

Sort of. MySQL has a really interesting clause in the community edition, you are not allowed to bundle the database with 3rd party applications. If we wanted to use MySQL and package it with our Point of Sale software, the cost is $500 per install last time I checked. (That was with Sun, god only knows with Oracle). Hence this was one of many reasons we elected to go with PostgreSQL as part of our installation.

Re:Well..

[LordLucless]

No, that's if it's GPL. If it's Open Source, it means that each customer can dick around with the source themselves, and don't have to rely on their vendor to do it. That's what Open Source is - you get the product, you get the code too. IIRC, that's what started Stallman's crusade - he got a binary blob of a printer driver that sucked, and he couldn't fix it 'cause he didn't have the code.

Now, the most popular Open Source licenses (GPL, BSD, etc) don't work that way. BSD puts no restrictions on distribution, GPL says that you're allowed to distribute only if you also distribute the code with it. But that doesn't mean commercial and Open Source are polar opposites - proprietary and Open Source are, but not commercial.

Re:Well..

[statusbar]

So does that mean that mysql is not open source as per the osi definition clause 1!

Re:Well..

[molecular]

that's why monty eventually quit sun and made MariaDB

Re:Well..

[vrt3]

As I understand it, the problem is not with the database but with the client library. That library is dual-licensed GPL and something proprietary. There's no problem if you use that library in a GPL application, but otherwise you have to use the proprietary-licensed version. Which costs money.

If I'm right, you could work around the problem by writing your own client library (I have no idea how difficult that would be) and using that instead of theirs.

Re:Well..

[gtirloni]

Under which license was the original code ? Care to mention company and product ?

Re:Well..

[gbjbaanb]

then one paying customer can take the source and fork it back out to everybody else for gratis.

And that's vTiger, see a comparative review between the two.

Re:Well..

[LWATCDR]

You mean like this?
http://www.vtiger.com/

Re:Well..

[MBGMorden]

one of many reasons we elected to go with PostgreSQL as part of our installation.

Indeed. When I was much less experienced with databases the "easier" MySQL (I quote it because it doesn't really even seem easier now) was attractive, but now a few years later with a bit more knowledge under my belt, I can't possibly fathom choosing MySQL over PostgreSQL for a new project. For legacy stuff that would need a rewrite then sure, it might not be worth the hassle, but for something new? It'd be like starting to frame a house and actually choosing a pipe wrench to drive in the nails instead of a hammer.

Re:Well..

[BitZtream]

No, thats not what open source means.

Thats what you want it to mean, and in your head, that may even really be the diffenition, but its one that isn't shared by the majority of the people in the world.

Open Source does not have to mean everyone gets the source.

It does not mean you can distribute source given to you.

Open source means that the author will let you view the source.

Period.

Anything else you add on after that for your own warped agenda doesnt' apply to the rest of us.

The authur may charge to see the source, they may not let you modify it, or they may place other REQUIREMENTS on you just like GPL.

When are you zealots going to realize that your narrow, single minded view of OSS doesn't actually apply to the rest of us who have been dealing with OSS since before your little movement got started.

Open source

[Eric Smith]

They only give the source to paying customers. But do they prevent those paying customers from redistributing the source? If not, then it really is open source. Nothing about open source requires that owner of the code give it out to everyone, but if there are restrictions on redistribution, it's not open source.

Re:Open source

[emurphy42]

A bit of poking around indicates that the community edition is released under GPL v3 and the paid edition is released under this variation of the Mozilla PL. Someone want to dig through it and work it out?

Re:Open source

[wolverine1999]

AFAIK they don't.

This is like NASA's open source software. It's very expensive actually.

Re:Open source

[kimvette]

That's not true. If they are the owners of the project and I am not a paying customer, they are under no obligation to provide me a copy of the source.

If they claim to be open source (as defined by OSI) then they have to give me 100% of the source as a paying customer, and if it is truly open source, I can choose to distribute that same source code to others, or even create a derivative product, like the folks over at CentOS (and previously White Box Enterprise Linux) do with Red Hat (er, I mean, "a prominent North American Enterprise Linux vendor") Enterprise Linux. :)

Re:Open source

[jamesh]

"open source" is one of those terms made up of two words who's meaning appears to be redefinable to suit the needs of any given agenda. That's why terms like 'GPL' and 'BSD' are more useful as they define what the terms of the 'openness" are. On slashdot "open source" and "GPL" are mostly synonymous but not necessarily in some industries.

but if there are restrictions on redistribution, it's not open source.

Well even GPL fails at that. It places the restriction that if you distribute the binary then you must make the source available too. That's kind of the opposite kind of restriction to what you were saying but it's still a restriction in that it limits your freedom to do what you want with the code, but only in as far as you can't deny others the freedom you were granted, which is widely considered to be a good restriction.

Even Microsoft open their source to various organisations (academic mostly). I think they don't ever refer to it as "open source" though but "shared source" instead, so I guess they are off the hook.

Re:Open source

[snowgirl]

A bit of poking around indicates that the community edition is released under GPL v3 and the paid edition is released under this variation of the Mozilla PL. Someone want to dig through it and work it out?

*reads reads reads*

Section 2.1, and Section 2.2 pretty much say you can distribute the original code with or without modification indiscriminately. ... So... yeah, one should be able to simply buy it once and then "fork" it from the original and provide it free.

Re:Open source

[srh2o]

You are completely incorrect in regards to what Centos is doing and how they do it. Red Hat makes their source code to Enterprise linux versions availabel to anyone. The only things not open source are their trademarks and images. Download yourself if you like. http://ftp.redhat.com/pub/redhat/linux/enterprise/5Server/en/os/ As for the GPL well version 3 has a section that clarifies this 10. Automatic Licensing of Downstream Recipients. Even under GPL 2 the hassle involved in trying to avoid making your source generally available is more hassle then it's worth as there is simply no way to prevent your source code from becoming generally available.

Re:Open source

[ducomputergeek]

You maybe able to fork it, but you can't call it SugarCRM or use their artwork. Same thing with FireFox, OpenbravoERP, or anything else using a Mozilla style license.

Re:Open source

[snowgirl]

You maybe able to fork it, but you can't call it SugarCRM or use their artwork. Same thing with FireFox, OpenbravoERP, or anything else using a Mozilla style license.

That's typically trademark issues, etc. If there is no modification involved at all, there's no actionable offense for a trademark claim... you're simply redistributing their product. Their trademark is still being applied solely to their product, nothing more, nothing less.

Re:Open source

[h4rm0ny]

It's what Bruce Shneider terms "Badgeware" (afaik, the term originates with him). You're not entirely free to do what you want with it. You have to keep the SugarCRM logo and link in there unaltered and unadjusted. In practice this means you have to stick with their UI and you'll always have a big old footer at the bottom with their logo.

Re:Open source

[jimicus]

I don't see how that's drastically different from CentOS.

Sure, you can use the free version of a "popular mainstream distro" (as the CentOS project describes it) but enough people are prepared to pay for support that CentOS hasn't yet driven RedHat out of business.

Re:Open source

[lakeland]

Right, from what I read it's exactly like CentOS vs RHEL. Whether that's open source or open core comes down to definitions more than anything.

Re:Open source

[hweimer]

A bit of poking around indicates that the community edition is released under GPL v3 and the paid edition is released under this variation of the Mozilla PL. Someone want to dig through it and work it out?

The Pro edition is being licensed under a proprietary EULA, which even contains Bitkeeper-like non-competition clauses regarding the community edition. It that sense it is even worse than MS-EULA, which at least allows you to mess around with the few open source components Microsoft has released.

Re:Open source

[jon3k]

You're using far too narrow of a definition of open source. You're referring to the GPL specifically.

Re:Open source

[BitZtream]

No, because the GPLv3 applies to ... THE COMMUNITY EDITION, which has been well established to be open source.

It doesn't apply to the Professional/Enterprise edition, which is not under GPL.

So no, just because you paid for the pro/ent edition doesn't suddenly make its license the same as the community edition. Thats just a retarded statement.

Re:Open source

[snowgirl]

No, because the GPLv3 applies to ... THE COMMUNITY EDITION, which has been well established to be open source.

It doesn't apply to the Professional/Enterprise edition, which is not under GPL.

So no, just because you paid for the pro/ent edition doesn't suddenly make its license the same as the community edition. Thats just a retarded statement.

No, but when Section 2.1 of the SugarCRM Public License grants you unconditional redistribution rights of the original code either modified or unmodified, then I am right.

So, it's not a retarded statement, it's taken straight from the Public License that their pay code is released under.

Is this an article or an ad?

[Anonymous Coward]

Welcome article reader, we have an ad for you barely disguised as an article.

Use "gratis" not "free"

[El_Muerte_TDS]

Can we please stop using "free" when we mean "gratis". You know, when something doesn't cost anything. "free" is too ambiguous.

Re:Use "gratis" not "free"

[iamnobody2]

or how 'bout we use "open" instead of free. "free" is too ambiguous. (i'd much rather use free as gratis then free as in open)

Re:Use "gratis" not "free"

[jamesh]

"free" is too ambiguous.

And therein lies the fun :)

Re:Use "gratis" not "free"

[value_added]

The problem is that English (at least as it's commonly spoken) doesn't have different words so as to distinguish between "libre" and "gratis". So while "gratis" does have English origins (IIRC), no one uses it.

Put more simply, it would be like expecting the French to have an equivalent for "entrepreneur".

Re:Use "gratis" not "free"

[ducomputergeek]

Not funny, he has a point. There is Libre and Gratis. Libre is generally what people think in terms of GPL (sorta) or BSD licensed software (more BSD/MIT) that you are free to modify and do what you want with the source code. Gratis would be free in terms of no money needed. It's free as in Frankly GPL probably falls under the Gratis category the more I think about it because there are restrictions.

Or in simple terms: Libre = Free as in speech/do what you want, Gratis = free as in beer.

Re:Use "gratis" not "free"

[drsmithy]

Can we please stop using "free" when we mean "gratis". You know, when something doesn't cost anything. "free" is too ambiguous.

How about we stop using "free" to mean "restricted by a particular set of rules that I happen to agree with", since that's a vastly less honest equation ?

Re:Use "gratis" not "free"

[dotancohen]

Who modding this as troll? It is exactly what I intended to post. "Free" means "without cost" no matter who has hijacked the term or what contingencies they associate with it.

Re:Use "gratis" not "free"

[El_Muerte_TDS]

i'd much rather use free as gratis then free as in open

FYI, "Free Tibet" doesn't mean what you think it means.

Re:Use "gratis" not "free"

[leenks]

"You are free to go" / "You are free to duplicate this"
"This item is free"

All have similar meanings, ie there is not "cost" involved (of whatever kind - monetary, legal, ...), but the currency of "cost" is different in each case.

Nobody hijacked anything. English has been this way for a long time.

Re:Use "gratis" not "free"

[Sarten-X]

What you are referring to is actually "free of charge" or "free of cost". This is different from, for example, being "free from defect in workmanship" as you'll often find guaranteed. It's also different from "free speech" which can often cost quite a lot, such as a 30-second spot during the Super Bowl.

Re:Use "gratis" not "free"

[BitZtream]

How about not using 'free' when you mean 'open' as well.

Both usages of the word are intentional and meant to confuse, and it works well as 99.999999999999999% of the people in the world think that 'free' means no cost, and when you see them ranting about how some software package should be OSS because its free and better they mean ... no cost, not free as in open because they need to fix a bug, unless by bug you mean the lack of money or willingness to pay for software.

The use of the word 'free' incorrectly in relation to OSS is an intentional misstatement your boy Stallman and his ilk pushed to confuse their agenda with all the little kiddies who just don't want to pay for anything.

Re:Use "gratis" not "free"

[drsmithy]

Nobody hijacked anything. English has been this way for a long time.

The problem isn't the language. The problem is people trying to manipulate the language (RMS and similar "free software" advocates, in this case) to present one thing (a certain set of restrictions) as something different ("free").

Re:Use "gratis" not "free"

[dotancohen]

Exactly. So long as SugarCRM calls their product "free" and not "free as in speech" then they are fine.

Re:Use "gratis" not "free"

[PipsqueakOnAP133]

Can we please stop using "free" when we mean "libre"? You know, like normal people? "free" is too ambiguous.

What 'open source' means to Martin Schneider

[macraig]

"Open source doesn't mean free and was never really meant to mean free," Martin Schneider, senior director of communications at SugarCRM, said. "Open source runs through everything we do, it enables us to be transparent and gives customers more power. We are an open source company and it's why we're better than proprietary companies."

Translation: What "open source" apparently means to the Martin Schneiders of the world is freely given code and other contributions to THEIR product from others for which they don't have to pay a dime, i.e. leeching off the "community". Their version of open source is apparently a one-way street with all the signs taken down. It might be giving them more power than it does their customers.

This has bothered me for a while...

[Anonymous Coward]

Strong advocates of "open source" always talk about how having access to the source is a kind of freedom, and that's true. Personally, I would prefer if all software that I purchased came with the source code (and the means to rebuild it) - because this gives me the freedom to fix bugs or make enhancements myself (and also to pay someone else to do it, i.e. to avoid vendor lock-in). It's an important freedom to have, **but** it's a big jump to then say that not only should I have the freedom to see and modify the source, but I should be able to share the whole source - even the parts I didn't write myself - with anyone I want to, without permission from or kickbacks to the original author(s). That is certainly nice, but it's not a "freedom" so much as it is a privilege.

Is the source "open" just because I have access to it along with the software...? I say it is. If I can also give it away to others then it's also "free", but that would be in the as-in-beer sense, not the as-in-speech sense.

The main reason I often prefer "open source" software is because I, personally, get access to the source code - not because it's free in cost, but not either because everyone else "in the wild" can get it too.

Re:This has bothered me for a while...

[neonsignal]

Being able to share the source (both as it is or with modifications) is considered a freedom because it allows you to benefit others in the community. You still need permission from the original author (because copyright remains with the author), which is what the libre style licences provide.

Having access to the source certainly gives you more freedom than many licences provide. But freedom has more than once facet.

Now it may be that you think the author of software deserves recompense for their work, and I agree with this. But that does not mean that the correct way to manage this recompense is by placing restrictions on the users.

Re:This has bothered me for a while...

[Requiem18th]

Wrong, to put it bluntly. You miss the point of software freedom/openness. Basically it's freedom from lock-in.

Software freedom means freedom from lock-in. As long as the original vendor retains control about who runs the software, how the software is run and what the software can do then it lacks that *freedom*.

Software openness is about distributed development/ownership of the software. Everybody is considered an author of the software even just potentially so and everyone can use it as they see fit, including sharing it with others.

SugarCMR is neither free nor open, they are simply dishonestly representing themselves.

Yes sure they can come up with whatever definition of open they want, for that matter I can call say I sell holy software, or rainbow software, I just have to make up whatever definition will sell.

Re:This has bothered me for a while...

[e3m4n]

It really depends on the license. Open Source has lost any true concrete definition and carries no legal obligation and binding should someone throw the word around. At the end of the day the only thing legally binding is the License Agreement that they license the software under. If its GPL then yes, if you make a modification and distribute it you are required to include the source code of the original product in addition to your modifications. If its a BSD license then you aren't even required to distribute the source code. Calling something Open Source has evolved into something subjective the way people in Atlanta call all soda's 'Coke' even if they're drinking Mountain Dew. This sort of generalization that tends to occur is why Linus Torvalds trademarked Linux in order to prevent misuse; specifically deliberate misuse by micro$oft funded FUD websites. If the Open Source Initiative wants to protect the word Open Source and its meaning they need to trademark before it becomes the next 'Aspirin'.

Re:This has bothered me for a while...

[gtirloni]

In "open source" your right to distribute the code you received with/without modifications is a RIGHT. You don't have to ask the original developer if you can. You CAN. Now, if you want to distribute that modified version under a different license... you gotta talk to who holds the copyrights.

Re:This has bothered me for a while...

[Requiem18th]

And anyway, if they give you the source then they are not in control of "what the software can do" nor "how the software is run"

I know, I only included that for the sake of completeness about the definition of free and open. I know SugarCRM doesn't restrict you in that way. But it does result in a vendor lock-in.

Suppose for example that you want a piece of a generic software, let's say a web browser.

Later you find out that it doesn't do everything you want but, at the same time you can't move because you also depend on some features of your current browser.

One solution is to develop the required customizations in house. At your own expense, for each iteration of the browser. You can't rely on an external provider with a ready-to-go solution because that's illegal in your model.

The example I have in mind is Firefox/Flock.

You like Firefox, you developed extensions for Firefox. Then you want to move to Flock because it has the features you want AND is compatible with your extensions.

But you just made Flock illegal. You can still have Flock, if you develop it in-house. You can also have it if Flock only releases patches, but then Flock can't seek for any external help because THEY can't leek the source as per the license. And even if Firefox used some sort of MS style license were people could still see the source and hack it but only Mozilla has the right to distribute/compile/use/license it, then you are still locked-in.

If Mozilla starts charging for Firefox then Flock(the patched Firefox) will also cost money, if they raise their prices again the price of Flock will also raise.

That's vendor lock-in.

Now you are going to say, "but isn't that normal? Mozilla has the right to charge for Firefox if they want" and it's true, but you are still locked-in, you can't change browsers because you depend on the extensions you made for it.

That might not mean much for you if your see software as an end product, but in F/OSS software is always both, an end product and a platform.

People make software on Linux counting on Linux being F/OSS, people make Ffirefox extensions counting on Firefox being F/OSS.

Had Linus pulled this "trick" people would have invested themselves in a platform that locks them in.

Not that I think anyone falls for this trick, it's just a cheep publicity trick.

Also SugarCRM's licensing ensures that the equivalent of Flock will never exist for SugarCRM.

Re:This has bothered me for a while...

[cronius]

Personally, I would prefer if all software that I purchased came with the source code (and the means to rebuild it) - because this gives me the freedom to fix bugs or make enhancements myself (and also to pay someone else to do it, i.e. to avoid vendor lock-in). It's an important freedom to have, **but** it's a big jump to then say that not only should I have the freedom to see and modify the source, but I should be able to share the whole source - even the parts I didn't write myself - with anyone I want to, without permission from or kickbacks to the original author(s). That is certainly nice, but it's not a "freedom" so much as it is a privilege.

While I agree with what you're saying, I think it's also important to be pragmatic. Should software freedom only be limited to software writers? Because those are the only ones that will benefit from access to and being able to modify source code that can not be redistributed.

Example: In Norway sometime after the "DeCSS" case they applied a law that basically said it was legal to circumvent copy protection schemes, but illegal to help others do so. Creating DeCSS would be interpreted as the former, distributing it to others as the latter. So in effect, they've actually made DeCSS illegal. In my opinion, that defeats the whole purpose of the freedom to circumvent the copy protection, as very few will be able to do that from scratch without help from others.

So in other words, the only way to allow non-techies the privilege of open source is if that source can be redistributed with modifications.

Re:This has bothered me for a while...

[neonsignal]

It isn't just the freedom to redistribute the source code. It is also the freedom for to make use of part of that code in something that the user intends to distribute. Or even just the freedom to allow a third party to make bug fixes (which means giving the third party access to a copy of the code).

Incidentally, the original developer has not lost their freedom to get recompense for their work. There is nothing to stop a free software developer from charging their users; not all free software is made publicly available. Notice that this is more of a service model than a manufacturing one.

However, I agree wholeheartedly that the term 'open source' is a misleading one, and shouldn't be used to imply other freedoms.

copyright

[stiller]

It's not about the code, which looks to be covered under GPLv3. The artwork is probably just covered by copyright. Only paying customers get to use this. This is actually not that uncommon. With some other products you are required to buy a license if you want to change the branding/artwork. Doom/quake are open source, but you still need to pay for the content. Does the new GUI provide functionality the old one does not have?

Want open source?

[kimvette]

Check out vtiger

SugarCRM has been guilty of decepting customers with their "open source" claims in the past. They originally released under a modified Mozilla public license (the Sugar Public License), with requirements that derivatives remove any and all SugarCRM branding. A few enterprising folks forked it to form vtiger, which supposedly led to SugarCRM threatening to file suit for actually exercising their rights outlined under the license, and the CEO publicly lambasting the vtiger folks for actually taking SugarCRM up on their offer extended by the original SPL.

http://forums.vtiger.com/viewtopic.php?t=11
http://blog.tmcnet.com/blog/rich-tehrani/crm/sugarcrm-vs-vtiger.html
http://developers.slashdot.org/comments.pl?sid=188554&cid=15541264
http://www.zdnet.com/blog/open-source/is-sugarcrm-open-source/867

I've posted previously about sugar vs. vtiger before:
http://slashdot.org/comments.pl?sid=223770&cid=18118754 (which drew out anti-F/OSS zealots and folks who didn't bother to read the licenses fully and obviously did not compare it to the previous SPL as it was originally written and released)

Now, the SugarCRM folks may have updated their licensing to remove the restrictions about moving to the free/community edition after having used the "enterprise" edition but honestly those folks were so scummy when they threw a fit after folks actually exercised their rights to create a derivative project that I can't be bothered to check.

Does vtiger functionality stack up well against SugarCRM's enterprise version? Not exactly. However, reverse is also true; vtiger offers some bells and whistles you don't get with Sugar - but in any event, vtiger does not use a license to try to restrict using your own data in another product.

Don't get me wrong: SugarCRM is a pretty good product, but I don't like to use products made by companies which engage in deceptive practices, even when some of the product editions may be "free."

Re:Want open source?

[bigsteve@dstc]

Here's an excerpt from the current Evaluation License, copied from the SugarCRM website.

Licensee shall not bifurcate the source code for any SugarCRM open source licensed products into a separately maintained source code repository so that development done on the original code requires manual work to be transferred to the forked software or so that the forked software starts to have features not present in the original software.

That smells of "not open source" to me.

Re:Want open source?

[Anonymous Coward]

No- this doesn't smell fishy. It is fishy. It is not libre. They are restricting your rights to modify the code. Somebody should have a shitfit about it.

Re:Want open source?

[e3m4n]

Well its clearly their license agreement. If you really want to screw them trademark the phrase Open Source and then file legal notices for them to stop using the word. THAT would be the right thing to do.

Re:Want open source?

[e3m4n]

we use vtiger and while the community could use a little help with release cycles and getting smaller fixes implemented faster, its better than the free version of sugarCRM. I am not sure if you have used 5.10 but they've added the ability to place calls to asterisk and include some pop-up notifications of incoming calls showing what accounts/contacts match that callerid number. The code, by default is setup for a Thirdlane PBX Manager implementation but the modifications to use another variant are as simple as context declaration in a config file. People need to realize they are not going to get the sugarcrm group to change their stripes. The most they'll be able to achieve is to get them to stop using the phrase open source. If people want a true open source CRM they should join the vtiger community and start adding modules.

Re:Want open source?

[Anonymous Coward]

Because it's an evaluation license for the enterprise product, dumbass. The "open source" one is the community edition which is GPL3.

They do interpret GPL3 to mean you can't remove attribution in their UI though which is questionable. There is a message which will say

"This copy of the SugarCRM customer relationship management program appears to have legal notices or author attributions modified or removed in violation of the GNU General Public License version 3. Please contact SugarCRM Inc. to correct this problem."

Re:Want open source?

[kimvette]

Wow so they brought it even further from being open source.

Re:Want open source?

[kimvette]

The most they'll be able to achieve is to get them to stop using the phrase open source.

Changing it to refer to it as "shared source" would make it a whole lot more palatable. Their claiming it is open source is a flat-out lie.

Re:Want open source?

[Keruo]

It's nice that they add features like asterisk calls, but after 5 years, the bloody thing still doesn't support active directory authentication without hacking it in yourself.
Problem with vTiger probably is that they have weak project management skills. They lack the skills to set goals to certain release targets and assign meaningful tasks to work on, instead they crunch away with yet another useless feature and throw out once-a-year update which breaks 2/3rds of everything even on vanilla system.
Don't let me even get started with the code quality.. probably better just to say half of what's there is reasonable, other half is bubble gum.
The account interface lacks support for bigger companies which might have subsidiaries, and it doesn't even check if the account you're creating already exists with identical name. Same thing with contacts though I understand that there can be 10 persons called John Doe, but do they all work at the same company?
Email system? There's 4 different routines to send email from vtiger, and each sucks more than the other ad finity.
Why is the email system half-assed? For example, (perhaps it's just me, I hate micromanaging) getting email from vtiger saying that someone in sales added a phone number to certain account drives me nuts, because it wastes my time with irrelevant information.

Truly open source? Yes, but that's really required since you spend 2-3 weeks actually fixing the idiotic things to get it in semi-working state at first.

Re:Want open source?

[Just Some Guy]

or so that the forked software starts to have features not present in the original software.

Do they define "feature"? Because to me it sounds like any change whatsoever could be considered as violating the licence:

1. Adding new functionality? Clearly a new feature.
2. Fixing bugs? "Feature: fewer bugs!"
3. Adding comments? "Feature: better comments!"
4. Breaking it completely? "Feature: unexpected behavior."

Replace the word "feature" with its synonym "attribute" and see how creative you can get.

Do something else then.

[pspahn]

Honestly, it's their business model. Paying customers have a foot to stand on if they want to complain, but if you aren't paying for something, that's your fault.

Does anyone seem to remember the power of open source is that you have access to the code? In the long term, being able to adapt a business solution because you can alter the code is a huge cash saver.

Re:Do something else then.

[shutdown -p now]

Honestly, it's their business model.

That is not an issue. But them claiming that they're OSS when they really do not conform to the conventional definition is deceiving their customers.

Re:Do something else then.

[lakeland]

Is redhat an open source company? I see no difference between what Sugar is doing and RHEL - except perhaps RedHat is more open about it.

Re:Do something else then.

[leenks]

You mean because RedHat opensource pretty much everything and so do Sugar?

Public Domain

[Anonymous Coward]

Actually, anything that NASA does is in the public domain for U.S. citizens.

Re:Public Domain

[innocent_white_lamb]

Isn't that internally inconsistent? Public domain is public domain. If it's only available to a subset of "everyone", then it's something else.

Open Source Product vs Company

[heikkile]

If that source code isn't made available, then you're not an open source company.

Technically, a single company can have products licensed for both closed and open licenses - I know, I work for one. They can even offer the same product under an Open Source license, and under a different license. Owning the copyright, they can fork the product, implement some features only in one version, and release that only under a closed source license.

Of course, nothing prevents anyone from taking a version that has been released under an Open Source license, and (re)implementing the features the company only offers under a closed license. Except that it requires time, effort, and know-how.

Re:Open Source Product vs Company

[ducomputergeek]

Or one or more of the three "T's": Time, Talent, Treasure

Re:Open Source Product vs Company

[gtirloni]

Agreed. I think what's causing stress is that plain simple software companies are calling themselves "open source companies" when in fact they only decided to release a crippled kid's version of their proprietary product as open source. Another point of stress is what most people come to expect when they hear "open source": a project with code available where people are encouraged to participate actively. Most "open CORE" companies don't have a community around them and contributions are usually rejected if it doesn't meet the company's roadmap (which is fine, but they usually don't tell what that roadmap is). What most people really hate is to contribute to an open source project, assign copyright to another party only to see that party sell a proprietary version of your code.

Re:Open Source Product vs Company

[Kalak]

Calling the Community Edition a "Crippled Kids" version shows that you've not used it. We use the Sugar CE, and are quite content with it. The paid version would be nice, but it's too expensive for us. We're a small department at a university, so we're not rich, not poor either, but it's out of out price range. The CE does what it does well, and we're glad to find something this good where we can modify the source, which we need to do for our authentication system here at the university.

Re:Open Source Product vs Company

[gtirloni]

Sorry for using that adjective. BTW, I meant crippled/kid's version (crippled version OR kid's version), not "crippled kid"'s version. Little English mistake.

My point is that a company is walking a very thin line between having an open source version which includes the necessary bits vs. a kid's version.

Just thinking now, if it narrows down its target market where it can be confident that customers' needs won't deviate too much.. that would help. But most companies want to reach to everybody they can. And thus COULD have very disappointed customer on the bottom of that pyramid.

In you case they have got it right: you're happy with what you got and it suits your need. The paid version would ADD UP to that and make you even happier.

What happens in most cases is that companies in the urge to not compete against themselves (how can a salesperson sell the paid version when the open source has everything), take essential bits from the open source version. The end result is more like that of shareware than what most people would expect when they hear "open source".

Anyway, everybody is free to do as they please. I just think open core sends a conflicting message to customers and has the potential to upset them instead of get them interested.

Wikipedia page history

[Zantetsuken]

Just checked out the Wikipedia page for SugarCRM to find out what it is - the whole page is written like a marketing pamphlet where the drone that went and put the page together sat down with a thesaurus and changed literally every other word just to make the Wikipedia article sound fancier.

Re:Microsoft also release MFC and the CRT library.

[deniable]

They're careful to use terms like 'shared source.' This let's them provide some of the benefits while marketing continue to abuse open-source.

My understanding is ....

[ksandom]

that open source is simply that users have access to the source code. The license then defines what you are allowed/obligated to do with it. Making it freely available/redistributable to everyone works very well with open source since it's very hard to control who can use it and who can't. But as far as I'm aware, open source does not actually define what people can do with it. It's just saying that the source code is available to its users.

Re:My understanding is ....

[MadTinfoilHatter]

While that's what the purely semantic meaning of the phrase would imply, it's not what the OSI means by "Open Source". Their definition http://www.opensource.org/docs/osd is far more narrow. So is the FSF's definition of "Free Software" http://www.gnu.org/philosophy/free-sw.html and these are the definitions most people use...

Re:My understanding is ....

[ksandom]

Hasn't open source been around for quite some time before these?

Re:My understanding is ....

[ksandom]

I've had a bit of a sleep on it, and want to change the way I'm portraying myself here:

If you follow the link in my sig, you'll see that I have two open source projects (1 GPL, 1 BSD). They fill the gap that I needed, and if someone wants to take them further they are most welcome to do so.

I also regularly post videos about novel/obscure hacks and provide supplimentary info for free on my site. These videos are under a creativecommons.org license (mostly attribution, share a-like) because open source for these simply isn't viable or valuable. So I am actively contibuting to what I believe in and if what I do interests you, I encourage you to take a look.

So Back on topic.

While that's what the purely semantic meaning of the phrase would imply, it's not what the OSI means by "Open Source".

I feel your frustration. It annoys me when people take literal meaning over what was originally meant. While GNU has been around for quite a while, OSI has only been around since 98. Having said that, I'm sure the original inventor of the wheel didn't have in mind a lot of the stuff that the Nitro Circus guys get up to. So things do need to grow.... I'm going to end that stream of midnight babble. Next!

I read another thread somewhere on this forum that SugarCRM has been misleading before in their licensing. If that's true, that sucks. I came across a chronic example of that a few years ago, and out of spite, I reverse engineered quite a lot of obfuscated code. Fun :D , but bad that I had to.

But almost everything I've seen in this discussion has been badly one sided, and some very valid points in the other direction have been marked as troll. As such I feel the moderators should look at themselves in the mirror. Do they honestly respect the person they see?

I've had the priviledge to work at some places that see the value of open source software and want to give back. This is awesome. We've got to remember that an important portion of the software we use today has come about this way.

I've also worked at places that wanted to contribute, but chicked out because of the sort of purist comments I've read here tonight.

For our ecosystem to survive, there's got to be a balance between "for the common good" and "I've got to live". Otherwise people simply won't want to contribute.

HTH
- Kevin :)

Re:My understanding is ....

[tokul]

that open source is simply that users have access to the source code.

you missed modification and distribution rights. You can see the code, but if copyright owner does not grant you distribution rights, you can't distribute it. If you distribute code without grant from copyright owner, you would be braking copyright law.

Re:I'm not so sure about his comment

[h4rm0ny]

While it's true that open source does not mean free as in beer, it's pretty damned hard to lock out a segment of the user community (i.e. non-paying users) when the source code is released, allowing anybody to build the "missing" feature.

Heh heh! You're right in theory, wrong in practice. SugarCRM has an important system in it called "modules". These are analogues of various business concepts, so there's a Contacts module, a Product Catalogue module, et al. The Professional Edition comes with some significant modules which, I'd say 75% of companies really need. Yes, the Community Edition people can write their own versions of these modules (as far as I'm aware - I don't know if SugarCRM legally could or would come down on them for making very similar free versions), but a substantial part of the utility of SugarCRM is in these modules. So what you're talking about is a fair bit of work. Also, they'll inter-relate. So if you buy the Professional Edition and get to use the Leads module, you get the functionality that can transform Leads into Contacts, relate them to Accounts, etc. Finally, SugarCRM the company control the code base and drive it with their Professional Edition. So you might write something wonderful that they like and incorporate into the main code body, but you might also find that having spent three weeks writing a cool equivalent to one of their modules, SugarCRM shrugs, makes some changes that will break it and rolls out the latest version of the Community Edition. In theory, if you're really awesome, at that point you could create a fork and the rest of the Community Edition users would follow you. In practice, that's not going to happen.

So I appreciate where you're coming from - I expect you're an engineer of some variety and have the in-built assumption that everyone else works to maximise interoperabilty and flexibility that you do, but in fact this has not matched the reality in this instance, I'm sorry to say. :(

Regards,
H.

Re:Microsoft also release MFC and the CRT library.

[shutdown -p now]

Well, Microsoft will certainly insist so.

Actually, no, it will not. MS is careful with OSS in general, and this extends to labeling its own products. It will only be called open source if it really is open source - e.g. IronPython or ASP.NET MVC, both released under licenses which OSI considers "open source", and FSF considers "Free software".

For other cases where code is available but there are strings attached (typically this means no redistribution) - MFC & ATL, .NET class library, Rotor etc - different terms, such as "shared source" and "reference source", are used.

SugarCRM definition of `Open Source'

[Anonymous Coward]

The SugarCRM definition of `Open Source' is different than the OSI definition
-------

The Open Source Definition (Annotated)

Version 1.9

The indented, italicized sections below appear as annotations to the Open Source Definition (OSD) and are not a part of the OSD. A plain version of the OSD without annotations can be found here.

Introduction

Open source doesn't just mean access to the source code. The distribution terms of open-source software must comply with the following criteria:

1. Free Redistribution

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.

is it that hard to get?

[batistuta]

After reading so many responses I think people just don't get it. Open source just means that "someone" gets the source code. That's why why it is often needed to specify for whom this is open source. For instance, it is not uncommon for our company to buy products which are "open source to customers". So we but a product, get the source code, and sign an NDA that we cannot redistribute it. Sometimes we can modify it under the waiver for liability and such. Open source does not mean free, it does not mean that you have to redistribute changes, it just means that "someone" gets the source code. Everything else is refinement of the open source concept and includes GPL, BSD, CCL, and any of the many open source license derivatives.

so please stop claiming that open source and GPL is the same or any along those lines.

Re:is it that hard to get?

[neurocutie]

your definition of OSS is pointless and silly since nearly all software fits the criterion of "someone gets the source". Windows would be OSS under your definition since MS happily gives out the source to Windows if either you're the right kind of customer or you pay enough.

Commercial software WITH source code

[Rsriram]

Many commercial software developers provide their software along with the source code. But they do not qualify as "open source". If there is a restriction on forking the source code or maintaining it yourself, it is not open source.

"Open Source" Software is different from "Source Available" software.

Re:Commercial software WITH source code

[Fahrvergnuugen]

So at the end of the day, is there really any way to stop SugarCRM from false advertising their application as Open Source?

Open Source != Free Software

[SharpFang]

Free software is where you can take source and do mostly anything with it, including forking, releasing for free, incorporating in your commercial product and so on. (restrictions being often that it can't be made "not free", but little beyond that)

Open Source is where you have access to the source code. Little is guaranteed beyond that. It may be only so that you are allowed to audit the code and nothing else. It may be that it's expensive add-on to inexpensive binary, and you are not allowed to redistribute it. Nothing beyond "you get to see the code" is guaranteed.

Currently I am working with one open-source project by a 3rd party. We have the source code of a library they provide (and only source, and for free). We incorporate the library in our closed-source product (embedded device) and sell it, without ever providing sources to our customers. We have to pay a pretty high license fee for each item that contains that library we ship. Nobody ever claims it's Free Software.

Re:Open Source != Free Software

[ksandom]

Agreed. I love open source, and use it almost exclusively. However, when people start getting purist about it, I do feel the need to remind them of companies like redhat who had made significant contributions to the open source community. Companies like these need to make money somehow. Sometimes they do it by having a paid version which effectively pays for the free one, sometimes it's by providing a service of some form like advertising, customisations, or installation. Everything has to be created, maintained, documented etc by someone, and there's one heck of a lot of work to do and only so many students with spare time.

Re:Open Source != Free Software

[e3m4n]

you forgot the classic RMS qualifier of "Free as in speech not Free as in beer".

Re:Open Source != Free Software

[SharpFang]

Yep, but Open can be neither.

visible source

[molecular]

They should call it "visible source" in this case, because they only meet one of the two criteria generally associated with "open source":
1.) source code is accessible (visible) CHECK
2.) source code may be changed and redistributed NOCHECK
"visible source" is still a lot better than "closed source"

take Microsoft letting Russian authorities view (some of?) the Windows sources. That's not opensourcing windows, that's selectively visible-sourcing it.

visible source

[molecular]

I'd call it "visible source"

"Open source" and "free" are used to get attention

[Futurepower(R)]

It seems to me that the important issue is that many companies are using "open source" and "free" to get attention and web site hits, when they aren't really open source or free.

Is just a full work for hire?

[AHuxley]

You buy the product, you get all the code. All who have paid for the code can enjoy that access.
Spread the code outside their closed community and its MS like lawyer time for you?
Could it be called it a "source included company"? Your just seem to getting more more long term code support for your $ from day one.

Stallman was right

[Orgasmatron]

Imagine that.

vtiger

[Pav]

vtiger is a fork of SugarCRM... it purports to be the real deal.

Re:SugarCRM could do it

[misiu_mp]

What a bunch of total human crap (a worse kind of excrement than bull shit). Its _not_enough to see the code for it to be open source.

This is from MS-Public License:
2. Grant of Rights
(A) Copyright Grant- Subject to the terms of this license, including the license conditions and limitations in section 3, each contributor grants you a non-exclusive, worldwide, royalty-free copyright license to reproduce its contribution, prepare derivative works of its contribution, and distribute its contribution or any derivative works that you create.

This is from SugarCRM public licence:
2. Source Code License.
2.1. The Initial Developer Grant.
The Initial Developer hereby grants You a world-wide, royalty-free, non-exclusive license, subject to third party intellectual property claims: (a) under intellectual property rights (other than patent or trademark) Licensable by Initial Developer to use, reproduce, modify, display, perform, sublicense and distribute the Original Code (or portions thereof) with or without Modifications, and/or as part of a Larger Work; and

Finally, this is from Open Source Definition:
2. Source Code
The program must include source code, and must allow distribution in source code as well as compiled form.
3. Derived Works
The license must allow modifications and derived works, and must allow them to be distributed under the same terms as the license of the original software.

Its clear all of those give the right to modify and/or redistribute the work as source or binary.

Re:"Open source" and "free" are used to get attent

[gtirloni]

EXACTLY!

SugarCRM replacement ...

[kwandar]

two years ago when looking for a system for our 200+ person company, I looked at SugarCRM. It was "okay", but there were licensing issues. Was it really open source? There were certainly questions at the time. Then I looked at VTiger. It was better than okay (and much improved since then, although I'm sure SugarCRM has improved too) and it was clearly open source.

And this is how it begins

[rfolkker]

Now that the precedent has been set, the "alternate" meaning of the word will slowly become common place, and in time, we will learn to accept Open Source as a corporate entitlement, granted by us paying them for the rights to use their code... I was wondering how long it would take. Took longer than I expected.

SugarCRM is Faux-pen Source at best

[DoctorLard]

My thoughts: 1. SugarCRM, at least when I was using/hacking it (version 4.5 - 5), was an unmitigated pile of PHP spaghetti crapness. 2. They are a proprietary software company that periodically dumps a six-month-old, pared-down zip file of their code on a "community" of people who haven't realised they're wasting their time. 3. They use "open source" as a marketing gimmick to attract people who don't want to pay shitloads of cash for Salesforce et al. 4. They are not in the slightest bit interested in community contributed patches, code, or design ideas. 5. Periodic requests on the forums for access to their SVN repo are not even denied, it's all "oh yeah, we'll do that next week". And sorry, but all I've ever been able to get out of the vtiger developers is an inactive mailing list, a silent IRC channel, and one guy in India occasionally updating his blog with "went to some conference yesterday" every few weeks.

TigerCRM is better anyway..

[cg3001]

Sugar is awful, there is a fork of Sugar called VTigerCRM it basicly the same thing and 100% free, 100% open source and has a HUGE community. Most the sugar modules work on it, if they do not then small tweaking to get it to work.Sugar is corp this, and profit that....I been telling clients this for ages...and the few who listened are still reaping the benefits. here is the link, for those who wish to research it...http://www.vtiger.com

Mods just need to get laid

[spun]

Section 1 only requires that redistribution rights be granted to people when made a part of an aggregate work from multiple sources.

This means you are not guaranteed the right to just copy the code out to anyone and everyone... without created an aggregate work.

How did this get marked as troll?

Because someone with mod points hasn't gotten laid in a while, and the user in question has the word 'girl' in his/her nick. Vengeance is mine sayeth the sexless.

Re:KDawson Strikes Again!

[Ja'Achan]

And so we have a debate on the story, which is what the summary is intended to provoke. Well, done, I'd say.

Mod Parent up

[Kalak]

I really wish I had mod points now. The Community Edition is still AGPL *from the downloaded source code SugarCE-6.0.0.zip* with no click through agreements or any other license distributed with the source code.
The paid editions are still SPL as they were before, making 90% of this entire discussion a tempest in a teapot. (The web page says it's a CC no derivatives license and says it's for 5.5, but they are distributing 6.0 with an AGPL license, which is what counts.)

They just added the new interface as one of the features under the SPL and paid instead of the GPL. If you have a problem with that, then write your own interface under the GPL and maintain it as a patch to the Community Edition.

Open source has many definitions, and everyone here should know this by now. So Sugar calls their paid license open source. Personally I would agree, and some would not. It's not Software Libre, which is a better term. It's "paid as in beer with a recipe."

License conflict?

[Keldrin1]

The documentation for SugarCRM Community Edition is located here: Sugar Community Edition 6.0 Documentation.

License This work is licensed under the Creative Commons Attribution-Noncommercial-No Derivative Works 3.0 License (“License”). To view a copy of this license, visit http://www.creativecommons.org/licenses/by-nc-nd/3.0/ or send a letter to Creative Commons, 171 Second Street, Suite 300, San Francisco, California, 94105, USA.

This license forbids both commercial use and creation of derivative works. Now, download a copy of the community edition here. Unzip it and look at the "license.txt" file.

GNU AFFERO GENERAL PUBLIC LICENSE Version 3, 19 November 2007

So, which is the mistake?

Re:License conflict?

[jmertic]

The documentation for SugarCRM Community Edition is located here: Sugar Community Edition 6.0 Documentation.

License This work is licensed under the Creative Commons Attribution-Noncommercial-No Derivative Works 3.0 License (“License”). To view a copy of this license, visit http://www.creativecommons.org/licenses/by-nc-nd/3.0/ or send a letter to Creative Commons, 171 Second Street, Suite 300, San Francisco, California, 94105, USA.

This license forbids both commercial use and creation of derivative works. Now, download a copy of the community edition here. Unzip it and look at the "license.txt" file.

GNU AFFERO GENERAL PUBLIC LICENSE Version 3, 19 November 2007

So, which is the mistake?

This first license is for the documentation only; the Sugar CE codebase is licensed under the AGPL.

Need Recommendation

[jkabbe]

I am looking for a recommendation. SugarCRM and similar applications seem mostly directed toward sales.

What we need is an application (prefer a web application) that has built-in calendar / task / contact / email -- all linked together -- with the ability to add my own data, relationships, and screens for entering / viewing said data.

I would start with something like FileMaker Pro but 1) I'd like to go the web app direction if possible because you never know how well an app like FMP will transition to the web, and 2) it has no robust built-in cal/task/contact/email. And, of course, there are licensing fees (which I am not averse to, but I would rather not pay a yearly seat fee -- if I have to buy I'd rather just pay once for licenses).

I am open to all suggestions. I intend to build a robust and sophisticated field-specific application. But I need a good foundation to start on. I just need to identify what that foundation should be.

Open source does not guarantee better

[sherriw]

"We are an open source company and it's why we're better than proprietary companies."

This is the biggest line of marketing nonsense. Just because you are giving people the source when the buy the product does not make you 'better'. It's still proprietary if users are restricted like crazy on what they can do with that source. If they can't share or fork it. Also, providing the source doesn't mean that the app itself is better when it comes to features, usability, stability etc. By what metric are they claiming to be 'better'?

More to the point

[ThatsNotPudding]

there will rarely be a thriving Open Source community around any software that is a near pure-business play. Being open, sharing, or even treating others in a humane fashion is anathema to Modern Business. If anything, such noble traits are probably viewed as weakness and foolhardy. 'The grabbing hand, grabs all it can, all for itself, after all it's a competitive world'.

bandwagon

[bryonak]

Of course, by your right to free speech, you can call it whatever you want!

Liferay is similar

[Jagungal]

Liferay is another product like this.

It started as a nice feel open source project. There was available paid support if you wanted/needed it. Then it went to a dual release of an enterprise version and a "community version" with all sorts of promises that nothing would change. The level of marketing then also seemed to go into over drive - to get the enterprise version.

It then became obvious that the community version, is full of bugs, gets no bug fixes and is released only once a year.

It has gone from a good, usable open source product to a really cranked up commercial product. It's open source true but this is more a marketing tool than a reality of an open project where everyone can contribute.

Re:SugarCRM could do it

[Inconexo]

First of all, I am not afraid of being wrong, because all I did was asking. And I did not ask for evidencing the poster, but because I did want to know if there is such an example. Because if such example exists, the debate would change.

Aside from that, it is a common error to understand that "open source" means "open to watch", but not "open to modify" or "open to redistribute".

Open source does require these freedoms.

the re-distribution test

[cas2000]

can one of their customers take ALL of their code, including the new interface, and give (or sell) it to someone else?

if not, then it's not open source.

there are other B&W binary yes/no tests (incl. freedom to modify, freedom to use for ANY purpose), but from what I've read about SugarCRM, it would fail on the first test alone.

Yep, MySQL is not open source

[hingo]

So does that mean that mysql is not open source as per the osi definition clause 1!

The answer to your question is yes, even though it doesn't follow from your parent comment as you may think.

What the parent comment is referring to, is called "dual licensing" and means that MySQL is available with an open source license, but also under a proprietary license, and some customers choose to "voluntarily" buy the proprietary license. The reason typically is that the open source license is GPL and the customer doesn't want to obey the GPL copyleft stipulations. ("voluntary" in citations, because in my years selling MySQL I've never seen a customer who voluntarily gave away money, still, the choice not to be GPL is their choice.) This model is not seen as conflicting with the OSI, since all of the code is available as open source, the customer just doesn't want it.

On the other hand, MySQL also engages in the same practice as SugarCRM, where there is an "Enterprise" version that includes proprietary-only modules. So yes, unfortunately MySQL is not fully open source.

(Disclaimer: I work for MariaDB, a fully open source fork of MySQL.)

Re:Yep, MySQL is not open source

[statusbar]

Thank you for the clarification!

So if I write a proprietary web application for my own web site and do not distribute the web app, both MySQL and MariaDB do not force me to either pay for proprietary license and do not force me to GPL my closed source application?

Re:Yep, MySQL is not open source

[hingo]

This is another question with 2 answers :-)

The GPL places requirements when you distribute the code (in this case MySQL or MariaDB) to some other recipient (like a customer). But this isn't the case for a web application, since you just run the code on your own server. In this case you can more or less disregard the GPL completely.

But if you actually sell or give your web application to someone else, then the GPL matters. Now, in this specific case all is still not lost for those who want to stay closed source. For a client-server database like MySQL or MariaDB, your application will only include a small client library that then connects to the database server. It is the license of this library that matters.

So for instance in PHP you already have a library that is under the PHP license, so PHP apps using MySQL don't have GPL anywhere.

For other languages there exist similarly BSD and LGPL licensed client libraries, but they are not yet fully feature complete (we are working on them). Contact me at henrik dot ingo at avoinelama dot fi if you want info on these libraries.