Slashdot Mirror
Open Source Foundations Coming of Age — What Next?
An article at The H makes the case that many open source foundations have successfully proven their worth and withstood the test of time as legitimate entities. This leads to the question: where do they go from here? The author suggests an umbrella foundation to provide consistent direction across many projects. Quoting;
"As you might expect, the main aim of most foundations is to promote their own particular project and its associated programs. For the putative [Open Source Foundation Foundation], that would generalise into promoting open source foundations as a way of supporting open source activity. In practical terms, that might translate into establishing best practice, codifying what needs to be done in order to create an open source foundation in different jurisdictions with their differing legal requirements. That would make it far easier for smaller projects – such as Krita – to draw on that body of knowledge once they have decided to take this route. It might also encourage yet more projects to do the same, encouraged by the existence of support mechanisms that will help them to navigate safely the legal requirements, and to minimise costs by drawing on the experience of others. After all, this is precisely the way open source works, and what makes it so efficient: it tries to avoid re-inventing the wheel by sharing pre-existing solutions to problems or sub-problems."
Multiple "umbrella" organizations? Sure. One single central authority? Do Not Want.
Warning: this article may contain humor, sarcasm, parody, and perhaps even irony. Read at your own risk.
These foundations should get into the online advertising business.
Maybe then -finally- their market cap would increase... or something.
*NO*.
Open source can be a "mess". But it's exactly this "mess" that makes the FOSS resilient.
Lisias@Earth.SolarSystem.OrionArm.MilkyWay.Local.Virgo.Universe.org
I think that the copyleft adoption is next. Currently corporations like non-copyleft open source licenses like Apache, eclipse so they can create commercial derivative works. The next step is don't caring about commercial software anymore and focusing 100% in services, turning on the obligatory open source on derivate works. ... nah... don't have the time to think on details.
We could call it the Free Desktop Standard, and all Open Source Desktops could implement it -- Yes, even you my little Haiku. What's that Haiku? You don't want to conform to some generic standard? It would limit your ability to differ from other desktops other than by look & feel? You don't want to implement X11 just to comply with the Standard?! Well, fine. You just won't be under the Umbrella of the Free Desktop Standard. You just won't benefit from the services we provide! What do you mean you greatest assets are user/developer hybrids not money and legal advice? Haven't you heard of the patent wars? Well, just remember this you rebel factions, when the Propriety Axis of Evil Patent-Cross-Licensors comes to call. Don't bother whining to the EFF! They're under OUR Open Source Umbrella! HA!
On second thought, maybe the Umbrella Corporation should be fought tooth and nail. Maybe open source foundations should just act as a collection of independent tools that each do one thing and do it well regardless of loyalty to any platform, standard, or ideology (CopyFree vs CopyLeft), instead of become a single insane bureaucrazy... You know, because THAT'S the Open Source Way.
Well, yes and no. As a leader/owner of a F/OSS project myself a reviewed and tested legal framework, along with some operation guidelines, would be immensely helpful in "business" situations or any time monetary issues are at hand.
That's probably what they meant with "to provide consistent direction across many projects", not some plan to force everyone to rank and file under their command.
And since no-one is born with universal knowledge of every imaginable field then that sort of thing would benefit medium (or even largish) sized minds by saving them from a boatload of research and brain-melting legalese.
The exactly quote is more specific:
A foolish consistency is the hobgoblin of little minds, adored by little statesmen and philosophers and divines.
However, this does not give a complete dismissal of the value of consistency, and if I wanted to dig up some pithy quote about the value of a regular structure, I'm sure I could do so.
And you can bet if Open Source becomes a mish-mash of inconsistency, there will be people who decide to cut that Gordian Knot.
Microsoft Mission Statement (http://www.microsoft.com/about/en/us/default.aspx) - "At Microsoft our mission and values are to help people and businesses realise their full potential." their values are more interesting "As a company, and as individuals, we value integrity, honesty, openness, personal excellence, constructive self-criticism, continual self-improvement, and mutual respect. We are committed to our customers and partners and have a passion for technology. We take on big challenges, and pride ourselves on seeing them through. We hold ourselves accountable to our customers, shareholders, partners, and employees by honoring our commitments, providing results, and striving for the highest quality."...they haven't looked at those for a while.
Apples Mission Statement (http://investor.apple.com/faq.cfm?FaqSetID=6) - "Apple designs Macs, the best personal computers in the world, along with OS X, iLife, iWork and professional software. Apple leads the digital music revolution with its iPods and iTunes online store. Apple has reinvented the mobile phone with its revolutionary iPhone and App Store, and is defining the future of mobile media and computing devices with iPad." - Seriously Apple!! that is less of a mission statement and more of a list of products.
The one great attribute (besides being free & open) about F/OSS is flexibility. Don't like this distro/program/script? Use this one. Like this distro/program/script but not feature X? Use a forked version with alternate feature Y, install alternate feature Z, or hack it up yourself. Don't like the license being BSD-like? Find one that uses the GPL, request for someone to write a new version (or for the author to change licenses), or again hack up your own.
Creating one umbrella foundation sounds good on paper, but how to make it flexible? And how do you guarantee there won't be corruption? We (humans) have spent years on philosophy, politics, etc., and there is always a little corruption. The problem is that corruption trickles down. If there's a break in the main foundation of the umbrella, then you might as well throw it away. Do we really want to risk this? And is it even possible to get everyone to agree?
It could work, and there are good intentions here, so perhaps I'm just being selfish? But for me personally, when I write open source code I don't want standards forced upon me. I want the freedom to write my code like I want, put my braces/brackets/whatever where I want to put them, name my program with f***ed up acronyms, and I don't want to be part of corporation-like practices when I "code for fun." I want individuality, and I want that from others too. I want unrestricted creativity and intelligence without being delayed and distracted by yet another "standard." And, I may be wrong, but I think that's some of the roots of the original F/OSS.
It just depends on what this umbrella foundation is supposed to do. From reading the article, it sounds like it's just to pool money and create documents about legality in different countries? This doesn't really sound like a foundation to me; you could just make a small website for this and ask people to post best-practices documents and for donations. I think there needs to more details about what this "umbrella foundation" is actually for.
One Foundation to rule them all, One Foundation to code them,
One Foundation to bring them all and in the darkness bind them.
The G
businesses have to have their lawyers go over such issues with a fine tooth comb anyway, they're not going to trust some "foundation"
How about an interface instead of a abstract base class?
If two or more organizations feel they can do better if they pool resources together, why not make an alliance or a small partnership or such? If there is no real need for a meta-bureaucracy to handle the existing bureaucracy, why make it?
s/a abstract/an abstract/
slashdot needs an edit button.
This is pure hogwash. First of all, Ubuntu, or Mint, or the dozens of distros are not recognized by GNU as legit; instead, what is endorsed are distros like Trisquel, Dyne:bolic, gNewSense and some others that nobody outside the FSF has heard of. Secondly, while it's true that Linux distros have been made that boot automatically and so on, very often, this plug & play functionality is b'cos they have binary blobs - something that's hated by the FSF guys. OSI doesn't have an issue w/ them, but FSF does. That's why one is more likely to find holes in those Linux-libre distros than in the standard Linux distros. Heck, even Debian, recognizing that there are times that un-liberated software is needed, provides those separately from its liberated software for those who are not that uptight about this purity, and guess what? Debian too doesn't have the endorsement of the FSF.
Open source is more accommodating of such deviations than Free Software, and that is why most businesses are more receptive to it.
As someone who spent the last 20 odd years coding C# NET / C++ COM among other things, I've recently become disillusioned with the MS nosedive as a dev platform.
I am exploring open source alternatives, learning Ubuntu, Apache, GNU tools, Google App Engine, Clojure, etc.
Now that MS has abandoned the desktop (& stagnated NET) with its Win8 disaster, Its no longer "developers, developers, developers" - it is now a "devices & services" company. Their problem with this is that they have completely lost to Android on the device front, and on the cloud services front, people are mainly interested in VM hosting not PAAS - so they are just another 2nd rate host provider. The only thing going for them is their massive workplace install base, which is rapidly ossifying into legacy tech.
So, what is stopping Linux penetrating the corporate market? Companies dont like paying MS license costs. here's my insider's take:
1) people (both LOB dev poohbahs / fanboys & office workers) are dumb, lazy to learn new things, resistant to change
2) people are unaware of the alternatives - joe shmoe does not know that you can use Google Drive for free instead of paying exorbitant license fees.
3) there is no hand-holding guidance for IT win admin bob to migrate his platform away from windows / office, or marketing materials to help convince his manager that this is the path to take.
4) retraining is expensive & disruptive.
5) A lot of in-house info systems are built on an underlying MS infrastructure: SQL Server, ShitePoint etc - there is no cheap migration path here.
So what is need is :
1) market awareness - an intense & prolonged marketing campaign regarding replacing windows & office
2) training material
3) migration guidance
4) migration tools
I was there when IBM went from hero to zero in 1992 with OS/2, & saw Digital VAX die in the 90s. Paradigms shift & change can happen fast.
It would be great if Free Software Foundation and OpenSource Initiative worked together to minimize the differences and together support the critical projects like a free/open source Skype replacement.
P.S. I personally would be very happy to see the term "Libre Software" to become the common denominator. Open source sounds cool and everyone knows it, but the name only says the source has to be open, that's only a part of what we all mean by it, and FSF is opposing it. Free software, on the other hand, is a little ambiguous in the meaning of the word 'free'. Also I don't like that FSF only says a system is free if it doesn't provide tools to install non-free software, and thus excludes Debian. User must be free to make the choise on his own (oh I mean her own:), while understanding the risks.
This shows me that RMS was right: open source is too lose term. 100 businesses making similar modifications to the source code of a single program on their own, fixing the same bugs, adding the same features is a waste of time. Also they may begin suing each other eventually. Libre Software is all about cooperation, and source availability is just a technical requirement for that.
Cause that would be cool.
They should all hire some hipster UI company to ruin the remaining desktop environments and UI-heavy software with some moronic tablet-wannabe interface. After all, Windows 8 is selling so well, GNOME3 is welcomed by the users, and Unity did not hurt the popularity of Ubuntu at all!
Contrary to the popular belief, there indeed is no God.
Fixing bugs is just one aspect of it, and usually something that the original vendor would want to do. But there are a lot of other things that individual customers may want, but which the vendor may not be interested in b'cos there ain't many other customers interested in it. For instance, let's say an ISV makes a program that helps a company manage its finances. One customer, which is located both domestically and offshore, would like some features that helps it transact business not only according to domestic, but also foreign tax laws. The vendor has limited employees and can't/doesn't want to explore the latter. In the model that I suggested, their customer can just go ahead w/ it, implement it in-house and use it, w/o worrying about support.
As FreeBSD has shown, when a company is making modifications that others may be interested in and that the vendor may be willing to own and maintain, they are only too happy to send those modifications to be merged upstream. That's what Juniper has done in FBSD. But other than that, when Juniper wants to add modifications to FBSD that it can then use in its routers, they do it, and the FBSD foundation has nothing to say to them, and is only too happy to let them mind their own business. OSI too doesn't care.
Cooperation is a fine idea, but forced cooperation is not cooperation. What's more, the basic question out there is whether software developers have the right to earn a living simply by selling software, and not having to worry about services and those aspects. RMS' answer to that question is practically no. Their current choices are to either go proprietary, thereby denying themselves the advantages of open source, or go open source, and be put at a disadvantage in that they can't restrict redistribution of their software. My above idea would pick a sweet spot that allows them to benefit from the advantages of open source, while not having to suffer the disadvantages. That way, people who simply want to make & sell software can do that being forced to allow their customers to become their competitors.
Aha!!! Here is the problem! Why would a new API not be backward compatible? The only reason is to accommodate the FSF's insistence that all software be liberated. Not everybody is comfortable w/ liberating their software, but why should a customer who likes their software - or hardware - be punished b'cos of that? Binary blobs are again a win-win situation for both vendor and customers, since the latter gets to use the thing, and former gets to sell it. Why should the former be forced to publish their specs if there are either business or other reasons for not doing it? If a vendor and a customer like a particular solution, who is any third party to dictate whether that solution is good enough or not?
This incompatible by design philosophy is also the reason that old software doesn't work on newer distros. On top of everything, there are several combinations of kernel versions, library versions, compiler versions and the like which make even FOSS tough to work w/, since one can't freely mix and match, due to incompatibilities that already exist. Things like drivers should be a write-once, forget about it deal. It's not that way in Linux. In Windows, the only time there has been a disruption has been when the very base of the OS has changed, like going from Windows ME to Windows 2000, or XP to Vista (since the latter was a migration from a win32 to a win64 OS). Allowing things like blogs also forces some discipline w/ OS makers, and forces them to maintain backwards compatibility.
The FSF is not a live-and-let-live organization - their whole philosophy is about disrupting the businesses of others by encouraging things like boycotts and the like. That's not the deal w/ the OSI. Otherwise, I'd have had no problems w/ them. Maybe the organization should become more independent of RMS, for starters - I wonder whether the rest of their membership are really as fanatical as he is?
From what I understand, Debian wasn't previously a member of OSI, but joined them not too long ago. I guess being tarred by the FSF for not being free enough (by providing separately the option of un-liberated software for those who wanted them) probably made them decide that they're better off w/ an open-source certification as well, in addition to the Debian Free Software Guidelines.
Seems like that would be a good role for the Open Software Initiative.
Software sucks. Open Source sucks less.