I am the founder of OpenERP / Odoo and I can't believe what I just read. Just wrote this mail to RMS, hope he will fix his post:
Although we do sell a service to help companies upgrade, there is also an open source tool by the Odoo Community Association to upgrade OpenERP: OpenUpgrade. [1] There is not a single dependency in OpenERP/Odoo against proprietary software or services. We have always fought about any kind of lock-in.
Can you fix your post? it's probably not the intention of gnu.org to make FUD against others open source project.
I guess you have been misinformed by Luis since he has done a lot of misinformation when he decided to fork to promote his project.
Having said that, you should have a look at what we do. We replace thousands of proprietary management software by Odoo, converting whole governments [2], SMEs and big companies as well as NGOs. (60% of the top 10 NGOs are run by Odoo: doctors without borders, red cross, ...)
I expect rms is gonna say something about being told about "open source", heh.
> There is not a single dependency in OpenERP/Odoo against proprietary software or services.
So, do your servers just run OpenUpgrade? Are you basically just helping others with the overhead of running OpenUpgrade? Or do you have secret sauce that works stronger, faster, higher than OpenUpgrade?
Open source is a synonym for free software coined in 1998 by Christine Petersen, promoted by Eric Raymond and Bruce Perens and funded by Tim O'Reilly as part of a memetic marketing campaign to make people stop thinking about software freedom:
It was an exacty synonym in 1998. There have been very minor deviations in very obscure licenses most people haven't heard of, so it's a minor quibble. The article you quote acknowledges twice,
The two terms describe almost the same category of
software, but they stand for views based on
fundamentally different values.
and
“Free software.” “Open source.” If it's the same
software (or nearly so), does it matter which name you
use?
OSI's intent was always to refer to the exact same thing as "free software", just use different language, and thus different arguments in favour of it.
I agree that OSI thinks the two are the same. "The article you quote" disagrees, and that's the entire point of that article. It's totally fine that you might agree with OSI, but it boggles the mind that you can't see the difference between freedom and free, or that you don't see that the GNU Manifesto was always about freedom. Whatever happened to understanding both sides of an argument?
> OSI's intent was always to refer to the exact same thing as "free software", just use different language, and thus different arguments in favour of it.
If you use "different arguments" than freedom to argue about software freedom, then it's no longer the same movement. The free software movement has always been about ethics. The open source movement has no ethical backing, it values code quality over ethical treatment of users (which is why people talk about "open source DRM implementations" -- since DRM is meant to subjugate users, it cannot be free software. But it can be "open source").
I didn't say it was the same movement. I said it was the same software.
The DRM point you make is nuanced. There can be free DRM implementations -- it would just mean that users could modify the software to remove the DRM. If the DRM is implemented in hardware, like tivoisation, then it is indeed no longer free. I think open source advocates, if they still exist (the open source movement seems dormant because it has "won"), might consider that to be open source. Stallman certainly thinks they would:
wow, I really do not want to quibble with you about something we agree on, so for the record, "hey everybody, I agree with cyphar!"
However: the GPL2 says that you can impose no restrictions on what what people can implement with free software so long as they it is only redistributed as free software, so it seems to me that while DRM'ed software is not free, DRM software itself could be free. Hey, if a DRM scheme could be designed to ensure software freedom, it might even be embraced; we already have GPL clickwrap license clicks which a narrow reading of the GPL might suggest you can't have (it's an "additional restriction" that the original software you licensed may not have had; for instance, if clickwraps are covered by law that the GPL is not covered by, you are forcing additional terms)
I know that the GPL3 had some fine tuning with regard to TiVoization and web servers, but I'm not aware that it would bar DRM software itself; but perhaps.
OK, I found it here for GPL3. You are partially correct that the GPL3 contains some anti-DRM measures but not in the sense you describe:
http://www.gnu.org/licenses/quick-guide-gplv3.en.html
"It's always possible to use GPLed code to write software that implements DRM. However, if someone does that with code protected by GPLv3, section 3 says that the system will not count as an effective technological "protection" measure. This means that if you break the DRM, you'll be free to distribute your own software that does that, and you won't be threatened by the DMCA or similar laws."
I thought one aspect of the GPL (copyleft being a clever copyright hack) is that the license applies only to the code you are licensing, and not to other copyrighted works you may have authored, but in a certain sense it does. I'm sure RMS and Moglen and FSF have thought harder about this than I have, so it's quite likely I'm missing something; however they do have broader agendas as well.
If you don't have a system that takes away your freedom with tivoisation, then yes you can have free software that implements DRM. But there's no purpose to that software because it cannot be used to restrict your users (the whole point of DRM). In addition, the software is actually not effectively free because you cannot exercise any of the freedoms because you might get DMCA'd (especially freedoms #1 and #3 -- where modification involves removing the digital handcuffs).
So, practically speaking, you can't have free DRM software. But you can have "open source" DRM software, because there's no part of the OSI that classifies software that acts like such a trap as being immoral. The GPLv3 essentially ensures that a user is not threatened by legal threats about breaking DRM in a piece of free software. It's the only software license that ensures this AFAIK, so I'm a bit sad more people don't use it for firmware and other places where defence against tivoisation matters.
Open Source includes BSD style licenses, which allow distribution in binary form only. Free Software doesn't, that's the difference and I think that difference is large enough to make them not synonyms.
You are not alone in thinking that only copyleft licenses are free. I think people get this association from GPL = free = Stallman = copyleft, since rms does keep talking about all of those things, so that's what people think he is saying (or they get it second-hand from people talking about what rms said). However, rms has never said that non-copyleft licenses are not free. He does call them weak or pushover licenses, though, but he doesn't think they're wrong.
Ok, thanks for clarifying that, you're right I was thinking that free = copyleft. This came from thinking that open source = non-copyleft + copyleft, and that it doesn't care at all about copyleft, while FS does, so FS = OS - non-copyleft = copyleft.
Isn't the parent commenter correct that BSD permits a developer to distribute a binary without offering the source? At the same time, a free software developer can incorporate and distribute BSD libraries without compromising a free license like GPL.
I spent some time looking at the webpage for this project
and the links https://www.odoo.com/page/upgrade (from the front page). seems to me to be exactly what RMS is talking about?
Hi, I would just like to applaud you (pinky07) for your even and clear responses in this thread. If the quote above is the whole message you sent RMS/FSF, I would suggest you bring in some from your reply below, specifically the part where you quote the bit about "2011", and "we used to do this, but we don't any more (and there was a Free alternative back then too)".
There are some Non-Free modules, which apparently are for use with the proprietary Enterprise Edition. From their legal page:
"Odoo 8 is released under version 3 of the GNU Affero General Public License (also known as AGPLv3)."
"Odoo 9 Community Edition is licensed under the LGPL version 3 (also known as LGPLv3)."
"Odoo 9 Enterprise Edition is licensed under the Odoo Enterprise Edition License v1.0, defined as follows: (...) can only be used ... with a valid Odoo Enterprise subscription ... You may develop Odoo modules based on the Software and distribute them under the license of your choice, provided that it is compatible with the terms of the Odoo Enterprise Edition License (For example: LGPL, MIT, or proprietary licenses similar to this one)."
Yes, but that's not the (controversial) claim in the article, the claim is that you can't upgrade without reverse-engineering the new database schema yourself. The article doesn't really address the issue of a plugin not being available as Free software. While FSF/RMS doesn't particularly like closed source software of any kind (it interferes with the four freedoms), this article deals more with Free software that is hemmed in by various "external" constraints. Data migration being the example referred to. And, which I perhaps was unclear that I was referring to, when I paraphrased: "used to do this, but we don't any more".
"upgrading to the next version of OpenERP required sending the database (full of patients' medical data) to OpenERP's server for reformatting."
"Required" is a big overstatement. OpenERP (now Odoo) is FOSS, and uses a standard Postgres database. The thing is that Odoo didn't release the migration scripts when a new version was released, but you weren't required to use their service; OpenUpgrade[1], a community project for implementing these scripts, is proof of that - we've successfully used it to migrate all our clients from version 6.1 to 7 to 8.
Disclaimer: I don't work for Odoo S.A., but I do work for a company that sells services based on the software.
Why do Odoo S.A. not release the migration scripts? Is it the core of their business model that people do not have access to migration and are thus forced to buy from Odoo?
Two years ago, it used to be our business model: selling maintenance services to bigger companies (includes unlimited technical support for bugfixes and upgrades). We tried hard to grow on services rather than selling proprietary features. (I don't think another open source software vendor succeeded to push a business model relying on services that far)
We grew up to 300 employees with this model but reached the limit of the model. (most companies don't want to pay for bugfixes and prefer to buy upgrades later) Now, we rely on 3 different offers: https://odoo.com/pricing
Since we switched to the new open core model, everything is simpler. Revenues grew a lot (monthly MRR growth is 242% higher than before) and we can invest more in the development of the open source product. Relationship with partners is easier too.
Our promise is to cover all enterprise needs in a single software, while being super easy to use. It's something nobody succeeded for now on. On the market, you have two types of players: ERP (SAP, Oracle, MS. Dynamics) who deliver a huge business benefit but a poor usability (expensive and inefficient) and one-app vendors (trello, slack, mailchimp) who have a great usability but a low business value. The first category has a huge revenue per customer, and the second category can attract a lot of customers.
We think we can crack that, and offer the best of the two worlds: all business needs, fully integrated, with a top notch usability. We are in a very good position to do it, and be the first to do it. [1] (and every competitors try to do that but they are far from succeeding)
To succeed, we need serious investments in the product and the service offer. We don't need to be very big (e.g. 10.000 people), but 300 is too small for what we do.
tl;dr we want to fix a huge problem and to do it, we need resources.
Just wanted to add, I think it is regretful that it was not possible for Odoo to grow a business on free software. I hope to be able to best Odoo at this some day, in a different domain.
At the same time, I am unapologetic for classifying pinky07's language as marketing. Evidence of marketing-speak is in "open source" instead of "free software", "open core" instead of "proprietary extensions" and other things such as "top notch usability", "best of both worlds".
When I first saw SaaSS I figured RMS had his own acronym for SaaS, and sure enough further in the article he spells it out - Service as a Software Substitute.
I think it detracts from his message when he uses these derogatory nicknames in anything other than in introduction. Like when he's talking about the "Swindle" (kindle), he will refer to it that way throughout the entire article, rather than just make a point an move on.
Services are offered in substitute for software precisely because IP has become socially, legally, and technically impossible to protect, so the only way to protect it is to put it behind an API gate, which (ironically) offers far less Freedom than any copy of Windows ever did.
After all, your copy of Windows never forced deprecation of API functionality you wanted to use.
There are cases where the service is not substituting software. E.g., something like Stripe or Twilio... versus software which is offered as a service because it's a nice form of DRM.
It seems to me that people rarely think about the software behind a service, this is a mechanism to bring that secret software into people's minds and discussions.
I recently came across this situation when attempting to hack on Keepass2Android [0], a Keepass client for Android. It was written with Xamarin, which was a closed-source framework at the time. Having to download and use software with a 30-day trial to hack on an open-source project seemed wrong.
Now that the Xamarin SDK is open-source[1] it looks like it's moving towards being freed, which I think is a good thing. That said, I'm not sure the IDE is available on Linux yet, so I guess it's still relying on non-free software.
Leaving aside the factually inaccuracy in the example...
If you want to stick with 'N', you can. And you can fix it to suit your needs. If you want to convert your database, there are no barriers in your way -- and you can distribute your results. Or maybe you want N+1 to work with the old format -- either way, nothing's stopping you.
If you're not categorically against proprietary software, the converter may be a really good purchase; better than any of the above options. But you're not 'trapped' into buying it -- no more so than, say, GIMP 'traps' you into using photoshop by lacking non-destructive editing features.
The fact that the situation would be kind of lame is because for almost everyone, in practice, It's vastly more useful for your software to be actively maintained than it is for it to be 'open' -- and the developers abandon 'n' when they move on.
Hi! I'm a developer working for a company (which happens to work with the software mentioned in the article) and which pays me to write AGPL/LGPL licensed code. And I can assure you there are many similar companies around the world, usually not behemoths like MS or Google, yet writing business critical software for very large companies, often based on the cooperation that only FOSS allows.
A company needs some kind of software that doesn't exist (or which is too expensive when they only need a subset), so we build it for them on top of Odoo. We also offer hosting, training, support, data migration, etc.
Note: while our software is FOSS, we don't publicly publish all of it, though our clients are free to do so.
"Note: while our software is FOSS, we don't publicly publish all of it, though our clients are free to do so."
That sounds like what I used to do with my trade secrets that I semi-publish. Enough detail out that hardworking specialists can build it. I'm curious what you mean by it though. Ive considered a model that sounds similar as I figured companies would be incentivized to keep paying anyway so updates/support have steady quality. Was worried about cloning, though.
Odoo is essentially a generic platform for which one writes modules, either self-contained or as extensions of others. When a client comes to us with a need, we implement a solution using a mix of new modules and existing ones. When it's finished, the client gets a FOSS license to all the modules, so they aren't locked-in to us. What we don't do is necessarily publish all modules on Github or similar; we make that decision case by case.
Interesting. Reminds me of a discussion here where so many FOSS projects were barely scraping by, overloaded with reauests from non-paying users. A few showed up to say simply: "tell them to pay you go build or support the feature if they really want it." Seems your company is playing it wiser than most. ;)
Free Software exists because someone says "hey, wouldn't it be quite useful to share this code with others and see if it improves".
It has absolutely nothing to do with corpo-cracy. Since it is a maxim that anyone who writes software can decide to publish it, at the individual level, this is as equally applicable to any corporation as it is on the individual level. However, I would say there are far, far more individual authors of free software out there, than corporate.
This is true for some of the biggest and best: majority of commits are sponsored by commercial sector. Not all and I won't say majority unless I see data.
Isn't Github a standard closed-source service? I don't see how the article would apply. It's close to the opposite; migration from git host to git host is easy and open-source.
Not when all your code, bug reports, wiki, etc. etc. etc are on a single platform, as a service.
Don't get me wrong: I'm not nearly as fanatic about this as Stallman and I'm fine using GitHub for everything, but it is a non-free dependency for every single project hosted on it.
That's still just a simple proprietary service. There's no connection to what the article focuses on. Part you can't migrate, and the part that does migrate does so with open-source tools.
For someone so passionate about security of the internet, I was surprised to find that you're loading, trusting, and executing random 3rd party javascript from Google and others.
the only thing that (should) be loading on the page is bootstrap (which IIRC, is hosted locally). Please let me know if _anything_ from google is being loaded in the background, so I can remove it promptly.
Your reply is kind of off-topic, don't you think? Nowhere on the site do I say that a desktop browser uses OpenSSL. I'm not sure where you got that from.
If you'll notice, our website does not use broken encryption (because SSL delivered by OpenSSL is a joke anyway). And we currently require you to install Tor Browser in order to have dependable encryption before purchasing.
But that's up to them, to choose a server which doesn't. Whereas currently, people are being redirected to the safe site from a completely unprotected landing page, so they can get MITM'ed and sent to a honeypot instead.
please let me know how someone can be MITM'ed when connecting to medicalcannab.is over Tor.
When you connect to the domain with torbrowser, the server detects your IP, then redirects to the appropriate .onion site if coming from an exit node. In order to be MITM'd, you'd have to either be using a Tor exit node that was compromised, or my server would have to be completely compromised beforehand.
neither of these seem like likely scenarios, so I'm curious as to what you see that I don't.
You should assume Tor exit nodes to be hostile. There have been exit nodes that injected malicious code into binaries downloaded through them, and running an exit node is an obvious choice for an attacker, hoping for people running unencrypted traffic through them.
Fair enough. someone could mirror my site's content, and we wouldn't have the ability to check where it's going.
However, that's not too much an issue. we don't offer binaries (outside of two pdf files that may have a jpg embedded), and nothing on our Onion site requires a download. The most they could really do is make someone give bitcoins to the wrong wallet. That's a pretty easy customer service issue to solve ("We are not responsible for bitcoins sent to wrong wallets").
Proprietary software companies trap you, so you utilize GPL to trap more lazy programmers? Selfish people are always selfish, lazy ones are still lazy. I don't like this model.
While I'm not a fan of the GPL and its viral nature either either, the kind of trapping you're referring to is not at all the same thing.
If GCC was the only C compiler in the world, my Apache licensed (or hell, public domain) C project wouldn't really be "trapped" by GCC being GPL licensed, because that license allows me and my users unfettered access to GCC whenever we want, so there's no one who can't access compilation of my project, now or ever.
Your example does not solve this problem: You may rely on some excellent GPLed softwares, but you want to distribute it with some close-source softwares. Then you might become the lazy programmers. I can see how GPL fans hate close-source software, but they can also choose another license, so programmers could share their works easier, also get more ideas from non-GPL fans.
ADDED: This same problem originates from close-source software, so GPL is another extreme I don't like. People from both sides still continue to develop great softwares, but what about MIT/Apache/BSD/...-licensed softwares? I choose the golden mean. I don't mean GPL is a bad thing, I just don't want to blindly support it. Times are changing.
The world of free software would not exist today without the GPL. If everyone had used lax licenses, companies would've made proprietary forks and driven free software out of existence.
Also, choosing MIT because "it's the middle ground" is a middle of the road fallacy. The GPL ensures users will always have freedom when using the software. Lax licenses provides no such assurance and proprietary software obviously doesn't provide anyone freedom.
The purpose of the GPL is to halt malicious companies from using free software to mistreat users. The middle ground isn't better in this case.
Even if the source is guarded by GPL, the idea behind the source can be copied and represented in another form. Then the freedom defined by GPL is obscure to me. MIT licenses saves me from time-consuming copycat situations. Yeah, I am lazy, but I'm more willing to share (hard-working now) if you don't push me to extreme.
I know the pattent laws in the past were somewhat ridiculous compared to what they are today (though still ridiculous sometimes), so GPL was/is really great then, just like the Creative Commons licenses.
Free software licenses have to do with copyright law, not patent law. And the only major change in copyright law in the last few decades was the Digital Millennium Copyright Act, which isn't super relevant to the issues in free software.
Oh yes, thank you. I always relate free software to patent trolls, because software patents have huge impact on software development. When I relate GPL to Creative Commons, I mean they are good role models for defending legal rights and interests.
Patents and copyright are not related. But yes, software patents are still ridiculous, and getting more so.
As for "the idea behind the source". Feel free to take the "idea behind the source" of FFmpeg or VLC or Linux. The point is that you can't make changes to the free software and then take away your users freedom. As for "but I'm a poor developer, etc" then you have two choices:
1. Use the GPL code and comply with the GPL. This is actually a good thing if you care about your users' freedom.
2. Don't use the GPL code.
Simple as that. There is no third "but I want to use the GPL without complying" option.
> The point is that you can't make changes to the free software and then take away your users freedom.
Still, you cannot exactly explain what the freedom is. Proprietary softwares can copy the ideas without contributing back. I know manpower is important, so another great softwares similar to FFmpeg/VLC/Linux with MIT/BSD/... licenses won't appear very soon.
If your freedom just means
1. Hey, just look at our GPL-licensed softwares, you propietary softwares can't use our code directly without any contribution.
2. Everyone should sell services, not code, for a more open world.
then I've already understood. And I don't believe GPL can do much better than MIT/...
The fact that it would take several man-years to replicate a large GPL project is enough of a deterrence. And the freedom is quite clearly defined by the FSF:
0. The ability to run the program as you wish.
1. The ability to modify the program and run the modified program. This requires source code.
2. The freedom to distribute exact copies of the software and source code.
3. The freedom to distribute modified copies of the software and source code.
The GPL ensures that those freedoms are upheld for any single body of work or any works derived from that work. If someone decides to completely replace your software with an alternative, that's fine. But if it's not worth their time (hint: it usually isn't) they'll begrudgingly grant their users the freedom that you wanted them to have in the first place.
The GPL does do much better than lax licenses like the MIT at both points. With an MIT license, it doesn't take any effort to make the code proprietary and thus violate both of the points you made.
> The fact that it would take several man-years to replicate a large GPL project is enough of a deterrence.
So what? Proprietary softwares have said similar words enough.
> The GPL ensures that those freedoms are upheld for any single body of work or any works derived from that work.
Okay, this is the only big difference, the "virus". I can use MIT to fight GPL.
> With an MIT license, it doesn't take any effort to make the code proprietary and thus violate both of the points you made.
My two points above are mainly for GPL. For MIT, that will be:
1. Our open-source softwares are used by more and more people, welcome to contribute if you don't want to see our softwares running everywhere with bugs in your operating system.
2. Our open-source softwares are great and free of charge.
3. Welcome to use our code for your proprietary softwares, dude. There are more and more open-source developers waiting for you.
As usual, RMS is content to sit back and complain rather than take action to try to effect positive change.
Has he written any code in the last decade, or is he still going around giving the same tired speeches over and over?
Do his anachronistic computing habits contribute to his continued irrelevance?
Ubiquitous broadband connectivity has changed the world and it seems RMS is still living in the past. I hate SaaS models just as much as the next user, but I am also a realist. There's just no reason for corporations to take RMS's "ethical" route when the profits lie squarely in the other direction. Lofty academic ideals don't buy your family groceries.
> There's just no reason for corporations to take RMS's "ethical" route when the profits lie squarely in the other direction. Lofty academic ideals don't buy your family groceries.
I agree he's a bit too extreme; I recall a video where he told a proprietary software developer that between continuing to do his job and letting his children starve, he should do the latter.
But there does have to be someone advocating for this, because free software is in general a social good and is unarguably indispensable in our industry. It's not impossible to make money off it either; look at Red Hat and Canonical, for example. But I'd never argue that all software has to be free or that everyone should do it.
I agree he's a bit too extreme; I recall a video where he told a proprietary software developer that between continuing to do his job and letting his children starve, he should do the latter.
I'd like to know the context, because that sounds like a "ask a stupid question, get a stupid answer" situation. It's a common trolling/derailing technique: you make a general point, and people reply with these very specific and extreme situations that "disprove" it.
I'd never argue that all software has to be free or that everyone should do it.
>I'd like to know the context, because that sounds like a "ask a stupid question, get a stupid answer" situation.
Agreed. The situation where you have to do an ultra-specific job or starve from lack of employment doesn't really exist. Jobs manage to appear and disappear over time without a death toll of millions.
>> I'd never argue that all software has to be free or that everyone should do it.
> Why not?
Because not all useful types of software have volunteers willing and able to write them or business models that sustain them. Expecting all software to be free is unrealistic.
I'm having trouble making any sense of your comment. Are you trying to say that writing code is the only effective way to spread free software, and that activism and awareness efforts are useless? But then you just fall back and insult RMS's computing habits, as if they somehow affect his ideals or the message he's trying to get across.
While I agree that we're not going to see companies mass-adopt free software as long as there is no legal or economic reason for them to do so, I don't see how randomly insulting one guy has anything to do with the movement.
> As usual, RMS is content to sit back and complain rather than take action to try to effect positive change.
Apart from inventing your world, RMS also has the absolutely unforgivable habit of being right about things. This is what really pisses people off about him. (see also: Theo deRaadt)
emacs, GCC, gdb, earlier GNU software were written by RMS.
He gave meaning to the term free software and propagated use of free software and free hardware. His ideas on using non-free Java, Flash, TiVo and DRM put those issues on centerstage.
If we see GPL being used so widely his dedication cannot be ignored at all.
Just like Steve Jobs didn't write all code of Apple or Next (FYI: He wrote none) but still he is "creator" it current Apple, similarly RMS is a chief figure of free and copyleft software.
Although we do sell a service to help companies upgrade, there is also an open source tool by the Odoo Community Association to upgrade OpenERP: OpenUpgrade. [1] There is not a single dependency in OpenERP/Odoo against proprietary software or services. We have always fought about any kind of lock-in.
Can you fix your post? it's probably not the intention of gnu.org to make FUD against others open source project.
I guess you have been misinformed by Luis since he has done a lot of misinformation when he decided to fork to promote his project.
Having said that, you should have a look at what we do. We replace thousands of proprietary management software by Odoo, converting whole governments [2], SMEs and big companies as well as NGOs. (60% of the top 10 NGOs are run by Odoo: doctors without borders, red cross, ...)
[1] https://doc.therp.nl/openupgrade/ [2] https://joinup.ec.europa.eu/community/osor/news/portugals-an...