A very interesting “Blogwatch” posting at Computerworld links out to an interview with Richard Stallman wherein he posits that Cloud Computing is a trap to entice users to give up control and privacy and become subject to closed, proprietary platforms. Since RMS is a professional provocateur, I wouldn’t consider all of his pronouncements newsworthy. But the thoughtful responses linked in this blog roundup were interesting, and I believe the issue of convenience vs control vis a vis Cloud Computing is a very timely and important debate to be having at this point in IT history.
… ever agree whole heartedly with RMS on something.
People are to ready to jump on some new fad without thinking about its implications. I don’t mind using a webmail service for personal email. Non of that data is overly important or sensitive. But some of this stuff is just ridiculous, like the “WebOS” sites that store everything you do on their servers. Why would you want this?
My generation (I’m 21) is WAY to comfortable with having their lives completely open and in someone else’s possession.
Edited 2008-09-30 18:22 UTC
Exactly my thoughts. Android sounds great, except for the part where Google gets to watch everything I do. Cloud computing is fine as long as you trust the dude running the cloud.
Yep. Having realised just how much information is stored on various companies servers, I’ve started ringing some of them up, and demanding they a) turn over a copy of the information and b) destroy the records if I no longer have dealings with them. (The Privacy Act of 1992 in New Zealand lets you do this). It is turning out to be surprisingly difficult.
It’s not so much that I think Big (or little) brother is out to get me, is that I’m not comfortable with Joe Random employee having access to my personal information.
I’m 23 and, yeah, I agree too. 100%. I like to be in control of my computer, including how it works and my data. If I don’t like how it’s set up to begin with, I want to be able to change it. I don’t like the idea of everything I do being up there, in some corporate “cloud” that these companies will sell to other companies as soon as they see dollar signs.
I reject making it so that I can’t play my music and video files and access my documents when I want, how I want (ie., practically any time I want). Electricity is quite reliable as far as service goes; it rarely goes out. My Internet connection sure as hell is not! Both of my poisons… er, local choices, Time Warner Cable and AT&T DSL have absolutely CRAP service. Whether it’s constant outages (both) or throttled connections (cable), it’s as unreliable as can be. And don’t get me started on dial-up.
It’s bad enough the Web is moving that direction, changing links to from actual audio and video files. Now it’s some browser-crashing, CPU-hogging Flash monstrosity that tries its damnedest to keep you from “saving” the file by hopping right over the standards that worked so well back in ’97, when I first started going online. The Internet sure has become a messy place, with the advent of DRM, Flash, and advertising companies…
I agree with RMS. I think that they should put their effort in creating more intelligent printers and more OSS compatible printers. Cloud Computing is a trap and limits freedom. Say no.
Your generation suck
disagree with so many.
I am in the generation right behind you by half a decade and I have to say that we do put a lot of information out there. I dont understand though, why this causes so much more concern now when it has ALWAYS been this way. I don’t know if cloud computing has just made us suddenly aware of the implications and caused us to wrongfully blame it for a terrifying scenario that we could be in if we wholefully adopt it when we are already in it. I know it sounds a bit alarmist and we should be more careful with our data in the real world and on the internet (the distinction is moot)but come on, let progress occur and let us dim the chimes of the populist voices that seek to hinder the small evolutionary steps of technological advancement
When I screwed up in highschool, the whole small town knew and living that down didn’t happen over night.. but it happened.
Today, when a highschool kid does something stupid, they document it on facebook or a dozen other internet hosted locations. They don’t control that information, it belongs too the server provider the moment they hit the post button. It is perminently documented, can and will be found and used for unintended purposes. These previously highschool students are now being declined for job opertunities or graduate school applications. “we would like to have accepted your application but your background check turned up X, Y.. and W. We do not feel you are the best selection for the possition we are offering..”
At one time, it was completely normal to allow children out to play in the neibourhood and after dark; not many folk in the city that still feel that way.. unattended kids playing outside, or after dark.
Back in my day, the principal would threaten to put the stupid $#!% we did on our ‘permanent record’, a mythical document I don’t believe exists.
These days, the kids do it themselves!
We’ve got more to worry about from our own stupidity than from any conspiracy. Besides, (working) conspiracies take foresight, discipline, and cooperation, all traits I’d love to see in more people.
Even on a purely strategically level, I can see where he is coming from. The GPL (in contrast to the Affero GPL) has a ‘network loophole’. By moving applications to the web (or “cloud”, whatever it means), companies can leverage copylefted software without releasing the source to the users of the software. The Affero GPL protects against the network loophole, but is used by relatively few projects.
Aside from source code availability, there are, of course, other problems. The primary problem is that in many web applications you don’t own your data anymore. It is not exportable to a standard format, let alone migrate it to another service. Besides that, data can be misused in ways that violate privacy.
another word for “Mainframe Computing”.
I dont see any difference except marketing kids trying to extract added revenue and line their pockets. No thanks!
yep, mainframes with glossy interfaces. what x windows could have been of one could fire up a random up by entering a url into some box…
Yup. I agree entirely with the above posters. Trusting your data to ‘the cloud’ is trusting someone else to know how to store it properly, be honest enough not to exploit it for their own use, and be smart enough to store it in such a way that you can get it back and use it properly should their service ever drop dead.
Not a chance.
I don’t have *anything* important stored in any kind of ‘cloud’, thanks very much.
Perhaps this is just an age-gap issue. Older folks (like me) have a hard time imagining our data “out there” floating around. But this same attitude was held towards computers in general a few years back. My father refused to have a computer in the house because it might “suck information” out of our house.
It kind of reminds be of banks. There are a few people who like to keep their money stuffed in a mattress. Most people have no problem keeping their money in a bank “out there”. They know their money is safe… oh wait!!
This is a real interesting thought.
But I don’t see it like that. I think it’s not only about the data. Many young ones claim they don’t really care about their data, ok. Still, the data is used for vendor lock-in.
Cloud-computing is just the next step after the format war. Nowadays coming up with proprietary formats as a protection against competition is wildly discouraged. Web services are the new trick.
This concept isn’t even anything new. Think of Hotmail. Nothing’s _really_ for free, but people tend to forget that fact instead of questioning the business models they are confronted with.
Not really an age-gap issue. It’s more of an experience gap. Keep in mind that many of the people who place their money in a mattress lived during the depression when banks collapsed and people who trusted banks lost all they had. You and I never had that experience and likely never will, but if it did (knock on wood WRT current banking crisis), they’d be the ones laughing.
Similarly, people like Buffet also stayed out of the dot-com boom. Our generation looked down on him, but he saw more than a few unfounded booms in his life and wasn’t able to understand the foundation for that one. When things went bust, he was safe to buy companies of value while many of our generation were in trouble.
WRT to privacy and confidentiality, anyone who lived through the cold war or currently live in oppressive regimes or was even slightly affected by books like 1984 knows the value of privacy and confidentiality. In many cases, it’s not just an issue of getting a bit more spam, it’s an issue of personal safety itself, and in the case of identity theft there may be severe financial consequences. It’s also an issue of reputation. A few google searches allowed Palin’s email to be cracked ( http://news.cnet.com/8301-1009_3-10045969-83.html ). Employers and universities are doing google searches on potential applicants and turning people away based on all the free information people willingly give to the world. This doesn’t even get into the situation when your data is trapped on a proprietary server and you’re locked in if they decided to pull a bait and switch.
But people without these experiences, won’t see the fuss but see the deprivation. Things like gmail and facebook are incredibly convenient and they allow you to not be tied to one computer or have to worry about syncing your data to your local computers. Your data is always there within each and because it’s all there and can be connected to other people’s data seamlessly across the world and automatically, you have a very powerful global social networking tool. The benefits are powerful enough to seduce most of my peers, even though they lived through they’re in my generation and know the risks. People of the younger generation just don’t stand a chance.
Well, I’m on Facebook, too, but I mostly use it for posting pictures I want other people to see, and for finding people’s phone numbers so I can contact them (and meet old friends again, in person).
I don’t put anything on there I wouldn’t want everyone to see. No drunken pics, no random angry tirades…
It is called getting old and wise and having experience.
This is not not about computers or technology, it’s about trust. Who do you trust with your sensitive data (be it corporate or private)? Random, profit driven company X? Not so much. Of course, this may change in time if/when there’s proper regulations and such but right now there’s nothing like that.
Not everything new is good, not everything old is bad and vice versa.
Edited 2008-10-01 07:58 UTC
I admin servers for a living. Say I have a netbook with a 100 gigabyte solid state drive. I also have a quad-core xeon with 32gigs of RAM and a 2 terabyte raid10 setup. Provided my global broadband access, my netbook would offer great mobile performance and low energy consumption, while not sacrificing any of my data or raw performance I have from my server.
Point being, say we have a “housecloud” of sorts which serves us our own data and software while crunching numbers for us wherever we are. That is the kind of cloud even Stallman wouldn’t object to.
It’s time to register OpenCloud.org
Don’t you mean freecloud.org?
that one’s already taken
totally unfree!
didnt microsoft recently make something like this?
basically we are talking about a nas with a web interface here, and maybe rsync integrated with a push based notification system.
Back up your data would be be the only sensible use i could thnk of.
Why on eart would i want to use some server comp to run programs i can just as easly, if not easier, run on my desktop.
I am thinking in terms of mobile devices. Our bandwidth is at the point where we can have modest devices assign tasks to a number-crunching workhorse. For instance, it would make it possible for smartphones and netbooks to handle tasks exclusive to a desktop workstation. Or even a tablet with an atom processor could conserve resources while on the go by assigning the workhorse tasks, functioning as a mobile display.
A personal cloud would offer convenience and power to its mobile user that isn’t available today.
I have two old 2U servers in a rack in my home. I bought them dirt-cheap, they perform well enough to host my own email, website, database, file, ldap, and other servers.
It’s not an enterprise setup by any means, but it works well enough for me and my extended family. None of the services I have running accept plaintext passwords. I run as many things over SSL as possible, and so do the rest of the family.
I personally don’t trust the clouds. I’d rather take on the task of backing things up myself. I don’t put anything on my website (or any other website) that isn’t already easily accessible publicly, or that I have a problem with other people knowing about me.
I ask myself this question before I post things: If I ever run for public office, is it likely that this will bite me in the ass? Sometimes the answer is yes but I feel it’s worth saying and post it anyhow.
My point is that the expectation of privacy is really, really low these days. This is where I’m supposed to mutter something about people in glass houses…
It wouldn’t be Stallman-compatible unless all the servers were running Coreboot. The FSF unsolders the BIOS and solders in a new chip with a custom compiled version of Coreboot on every machine that comes through the door. I kid you not. I think about that whenever someone claims that the FSF is shorthanded on HR.
Edited 2008-10-01 16:33 UTC
That’s called a fileshare
Is a Privacy, and Corporate SysAdmin’s Nightmare To maintain Compliance. This Rebranding of Internet Hosted Services is NOT going to fly as long as the current enterprise security models and technologies are still present.
I can’t believe it, but I’m actually agreeing with rms on something. I never thought I’d say that, or even come close to saying that, but well there you are.
As I see it, though, the problem is bigger than this new buzz word, i.e. cloud computing. Gmail is the only cloud I use, and the only reason is that I’ve switched ISPs enough that I don’t want to go through the email switch hassle anymore. Yes, clouds have a very good possibility of establishing vendor lock-in and yes, they may be an extreme security risk for confidential data. But how many of us put our confidential data out there every day? Amazon.com has our credit card numbers and/or bank accounts. If we pay bills online those companies have them. If you use any online tax services guess what? They have more than just your employer information. Paypal has information on as many bank accounts and credit/debit cards as you give it. What about the various instant messaging services? They keep logs of your conversations. Your phone company keeps phone records, though thankfully they usually don’t keep recorded conversations. But every text message you send through that slick new cel phone is logged. While all of these services do have a privacy policy, there’s usually this little part at the end that no one bothers to read or if they do they ignore it. It says something like this: this policy is subject to change without notice at any time.
I certainly won’t be putting any confidential data on a cloud deliberately but I can’t help but wonder how much is already on one of them, simply due to purchases I’ve made online at one point or another. The closest I’ll come to deliberately putting things on a cloud is my own house cloud, of sorts. Very few people know how to set something like that up, though, unless they’ve worked with computers more than simply browsing the web.
So, how much is already on a cloud? Because whether you think of it like that or not, all of the online purchasing centers are clouds of a sort, as is your cel phone, and even Paypal.
The problem with policies is that it is not law or a legally binding contract.
So though you could mention that in your policy statement, it does not stop the disagreeing party to reject future changes in policy.
Also though you may see statements that effectively state in your employment contract that require you to accept to their policies even if it changes, it doesn’t stand a chance in court. This however does not mean lawyers wouldn’t come up with ingenious ways of hoodwinking the unsuspecting public.
Why the summary is so provocative about Stallman ?
Because Stallman’s views are very radical, and he’s very provocative when presenting them?
Because he’s an idealist with a very clear view about how the world should be, and with the time, money and boundless energy to espouse it.
Put it a different way, he IS a GPL flamewar.
Edited 2008-10-01 02:00 UTC
I think it is unfair to brand RMS as such. Though he may seem like a staunch idealist. I find that a many of his views are towards the greater good. Though this may make him seem to be a socialist.
If you honestly think that capitalism reigns supreme, you haven’t been paying much attention to the financial crisis lately. It is not acceptable to preach capitalism when the fat cats on wall street rake in profits but demand socialism when they need bailouts!
IMHO, capitalism = greed, thus it should be regulated for the greater good of everyone else.
Edited 2008-10-01 05:02 UTC
1. We had a government and legal standard for data centers. This would include privacy, theft, fire, data backup… protections and legal consequences for negligence.
2. Your data remains your own and can be exported to your local computer easily.
3. No ability for the government or any other agency to see your online files. This could be done by just relying on legal matters. In addition, it could be designed such that your files are always encrypted and only decrypted locally when you enter your password. This would probably require a different paradigm that webapps today, but it could be done.
Given those conditions, I would gladly embrace ‘cloud’ computing for all my needs.
Ah, but who says that today’s regulations wouldn’t become tomorrow’s garbage? It’s sort of the same issue of the privacy policy–it may be good now, but it could be changed at any time and without your notice. Face it, at least in this wonderful USA (sarcasm fully intended), we don’t have privacy. We have the illusion of it and nothing more. Yes, the government could put regulations such as you suggest in place. But guess what? How long would it take either them or some other lawyer to find loopholes in them? Those in power can basically do what we want as long as we bend over, and most people bend over far too much these days. Privacy is subjective, and this excuse for a government has already shown us that it means about as much to them as the piece of toilet paper they used in the bathroom today.
Edited 2008-10-01 00:44 UTC
I disagree with at least the idea that *all* cloud computing is bad. Generally, the data coming from the servers are in an open format. Further, all the data you yourself created are available from these services. Want to the see emails in gmail that you sent before? No problem. They’re available. Facebook lets me access all my photo, status, blog postings, in JPG and (formatted) text formats. And so forth with any “cloud” computing service I can think of. All my data are copyable directly from the browser. There’s no lock in. Saying so simply confuses the issue.
What it basically comes down to is that sometimes you don’t have complete control of things in life, and you have to trust that others aren’t going to abuse you. RMS’s problem is that he’s paranoid that somehow he’ll be coerced or exploited if he can’t see the source code to any project himself. But why stop at source code? Surely there’s levels of opacity to the outside world in business operation. Would RMS demand that any “cloud” computing company he does business with publish its official data use policies, or its handbook of official procedures, install cameras and keyloggers, give us full read access to their systems, and allow us to monitor all internal network traffic so we can see what they really do with our data? I know this is a bit on the ridiculous side, but it follow quite cleanly from what RMS is arguing.
While I agree that source code availability is a *very* good thing, RMS’s moral arguments are really quite bunk when applied to inappropriate problems. It’s my take that RMS’s opinions on cloud computing stem from a general paranoia when it comes to dealing with any self-interested body. It’s good be able to verify the operation of your own electronics, and I agree with him that if I buy something, *I* better be allowed to do with it as I please (and so I’m against binding EULAs, most DRM, etc.,) but part of interacting with a third party is being able to trust them. If you don’t trust them, then don’t do business with them. It’s really quite simple. We (US citizens at least) are not coerced into doing business with anyone.
Demanding transparency from business is a good thing IMHO, but RMS confuses complete ownership over things you purchase with a desire for transparency in business operations. The difference here is one of self-determination. Allowing complete ownership (including source code) of things you purchase for private use allows you to use the object as you see fit, explicitly disallowing another private entity say in the use of the product. However, purchase of a service is a different question. Hiring someone to work on your behalf (as in the case of “cloud” computing,) you give up the complete right to self-determination. As long as the contractual demands are fulfilled by both parties, there should be no coercion involved, as contracts are entered into voluntarily.
Now, there is a need to keep public instutions (which ideally have a monopoly on coercion) accountable for their actions, and transparency is one part of accountability, but RMS is demanding that private instutions follow suit. But as I mentioned above, private instutions can not force you to patronize their services. It would be a nice thing, and may gain a business more customers, as well as general good will, to have said business open up its practices to public scruitiny, but then again, any business that is too open is likely to put itself at a severe competitive disadvantage.
I would like to see laws put in place to allow a customer access to all of the information on file about them (including metadata which are mined from the user’s usage info,) as well as laws forcing a company to remove all user-submitted content, and to remove all of the user’s personally identifiable information from the record. This seems like a reasonable compromise, and a good exit strategy from contractual obligations should they be found undesirable.
Yamin: I absolutely agree with everything you said. ABSOLUTELY 100%. WRT your comment on encryption: It could be done with the technology we have now. Check out public key encryption. I actually would *love* to see a push for this, as well as a push for a single user key usable throughout the internet, because not only would it mean my data transactions with “cloud” computing services would be encrypted, it would also push people I know to adopt the technology for private emails, which in turn would facilitate the adoption for further use (phone calls, etc.)
I don’t trust Fair Isaac Corp. but do I have a choice?
The problem here is that corporations successfully lobby for legislations to fit in with their business models. Just as I do not relish the spam I receive in my mail for unsolicited credit offers, leaving my private information to an industry that promotes data mining and data warehousing as the future of computing with similar marketing models in mind is equally distasteful.
You have to leave your comfort zone to see the big picture here. Doing that after Stallman’s “I told you so” would be a little too late.
Yes. This. Thank you.
I’m the EIC for an online webzine, and IRL I work at an academic library. Google Docs & Spreadsheet, for example, have enabled me to collaborate with multiple people using a variety of OSes and programs without a single compatablity issue. Whereas before I had to worry about making sure everybody had the right version of the doc, or wonky formatting that did things like turn ” in to ? and ‘ into superscript 1. (Yes, I know about global find and replace, but not having to do that at all is even better.)
Google’s servers are robust and reliable. Unless it’s a supermegamajor catastrophe, I don’t have to worry about a power outage (happened to my campus about a month ago and lasted several hours) crippling me and my partners’ ability to work. (Nothing like coming to work and discovering that you can’t access a key document because that particular server isn’t “mission critical” and gets no power from the backup generator.)
Neither do I want to set up a server at my house and deal with securing it. (I do have personal, sensitive documents on my home computers and don’t want to take *any* risks with that data.) Plus, why keep a computer running 24/7, using electricity if I’m only going to need it about 3-4 times a week — that’s a waste of money and an inefficient use of resources.
However, I don’t use Google Docs or Spreadsheet for anything that’s sensitive or not intended (ultimately) for public consumption. Yes, I’ll be annoyed if there is a security breech that allows anybody to come waltzing into my Google Docs account (because it means that somebody else was asleep at the switch), but there’s no harm done if people get the draft versions of my thoughts on yaoi, a film review, or the caveats involved with doing reference via SMS.
Cloud computing is the wave of the future. When used appropriately with an eye to its shortcomings, it’s convenient, cross platform, and makes efficent use of resources.
1) There are some who are talking about open standards and open formats. It is inevitable that a company is going to set up at least some sort of hurdle to protect themselves from the possibility of their customers leaving. It could include opt-out clauses, contracts with penalties built in, ‘procedures’ in which a person must go through to get their data or ‘open standard’ but patented algorithms so any attempt to create an export/import utility for the service is pinged with patent payment demands.
2) This same question was asked with the portability of information between the various different ‘social networking sites’ and here we are a few years later and not a single thing has been done. The issue has been quietly avoided by the industry in favour of ‘business as usual’. The same thing will occur with cloud computing – business as usual.
3) Those of us outside the US are on metered internet – we don’t have the luxury of a flat rate internet connection, meaning, this ‘cloud computing’ would become very expensive very quickly and it would have no more benefits to it than I would if I had a large thumb drive which I hauled around with me where ever I go.
4) What is the ultimate benefit of this over existing ways of hauling data and information with you? I keep hearing computer Geeks hyperventilating over cloud computing and other such stuff – and yet, no tangible reason has been given to me as to why I should dump all my data into this ‘cloud’. What this seems to me is yet a rebranding of centralised computing, but rather than using dinky little 9600 dial up models, its over the internet using ADSL/Cable connections. In other words, same bollocks – different name.
I agree with everything except:
Here we go again with “my country is the world”. Or in this case, “every country other than the US is exactly like mine”. I have flat rate internet here (Philippines) and so does my folks back in Sweden.
Ditto here in South Korea – flat rate which has actually been reduced by almost half since I signed up in 2004 because I’m still here!
Personally I dont see the attraction behind cloud computing. While its a sound idea the bandwidth is not consistently available throughout the world to really make everything 100% cloud feasible. Along with the security concerns others have raised I probably wont use the services.
I hope Mr. Stallman sees the chance of free software here.
All his concerns raise and fall with the owner, the center of the cloud. If this center is in your own posession, everything is fine and your data is safe.
So he should concentrate on offering free cloud computing solutions, making it possible for everybody to run his own cloud. Running e.g. a word processing server doesn’t require _that_ much processing power, it just requires some bandwidth. Internet-savy poeple have this already.
The problem is that the solution isn’t going ot be half as effective as using Google apps. You don’t get the uptime or all the fancy features like having your calendar send you an sms. And it is expensive to ste yourself up like that.
Software is easy. Hardware is hard. Unless you manage to beat Google at their game, you are going ot have a hard time creating a completely free and open alternative.
“Store your private data on computers that I own,agree to usage policies that I write and screw you if it all goes titanic!”It is easy to frame the idea this way. It sounds so stupid and unreasonable that I would have to be stupid and unreasonable to agree to it. I find missing from the debate,any discussion on WHY cloud computing is now relevant. The discussion has become about the storage of private data rather than a discussion of modern-day computing environments. There is a shift in computing that moves away from the single node in a network that transports data to mobile applications that have an identity associated to real world entities such as people, organisations and communties. It is a mistake to overlook the problem that Cloud Computing poorly addresses and miss the greater issue by making straw men arguments.
Well, I’ve always been a bit skeptical about cloud computing – but if Richard Stallman is decrying something, you can pretty much guarantee that it’s going to be a success.
Invest? You could manage a whole mutual fund by that economic barometer.
Darn toot’n. That crazy computer hippie FOSS crap is sure not a success. Heck, that Linux kernel doesn’t get used anywhere outside of crazy entusiast’s basements.
But sadly, there’s a large amount of budget and marketing expertise behind the newest rebranding of centralized computing. More often, people believe what Ads tell them rather than considering all the implications; just look at Microsoft success over many better but now defunct technologies.
I don’t think Stallman has ever decried Linux. Although the term “Linux,” there’s another story/Quixotic crusade that is difficult to take seriously. Which is the point that I original jested-at: no one takes Richard Stallman as seriously as Richard Stallman does.
yeah, I know the “GNU/Linux” thing where RMS will only give an interview if all references are stated that way. Very true too.. RMS takes Mr Stallman most seriously.
I most really just poking fun at the idea that what RMS decries becomes popular in contrast to how popular the Linux kernel and the GNU apps wrapped around it without having the propellant of RMS decrying it.
Nope, I think you’re taking Richard Stallman more seriously than he does. Arguing about GNU/Linux is simply a strategy to make people pay more attention to the ideals that the GNU project stands for. Just like in that famous Zen story, you mistake the finger pointing at the moon for the moon.
Richard Stallman’s insistence on asking people to call the operating system GNU/Linux doesn’t really indicate he that he would take himself seriously. Instead, it clearly indicates that Stallman takes his free software ideals very seriously. Would a man who takes himself too seriously put a computer disk on his head and call himself Saint IGNUcius?
http://www.stallman.org/saint.html
I think you missed the following note at the bottom of his page:
Warning: taking the Church of Emacs (or any church) too seriously may be hazardous to your health.
No, I didn’t miss it. Richard Stallman can be more jocular than most people like to give him credit for, but that doesn’t mean he’s not dead serious about his free software ideals.
But since we’re discussing the facetious side of Stallman’s character, have you read the “Free as in Freedom” book?[1] I was very much caught by surprise in the ending scene of chapter 5, where Stallman suddenly pulls a prank on the waiter when they leave a Chinese restaurant where the interview took place. (And the same fifth chapter also mentions Stallman’s jest that “giving the Linus Torvalds Award to the Free Software Foundation is a bit like giving the Han Solo Award to the Rebel Alliance”.) It’s a funny incident, check it out if you haven’t yet read it.
[1]
http://oreilly.com/openbook/freedom/
Where in the world did you get that notion? Richard Stallman is the computing equivalent of that crazy uncle who the whole family is slightly embarrassed of, and who always gets drunk at Thanksgiving dinner, then goes off on rants that end with “…and next thing you know, they’ll be trying to take my guns!”
What does it have to do with free software ideals?
Nothing, really. At least not directly. It’s a matter of political strategy. He’s pretty explicit on the matter of his fearing that if it gets called “Linux” instead of “The GNU System” then he loses political clout. He states the matter with a little more sugar coating. But that is essentially what he says.
Edited 2008-10-02 20:41 UTC
A lot. There’s also a more detailed answer available here:
http://www.gnu.org/gnu/why-gnu-linux.html
For those without a masochistic streak, I’ll summarize: 1,300 words that essentially boils down to “Linux has benefited GNU’s ideology, so everyone is obligated to help us advocate that ideology by using the term ‘GNU/Linux’.”
That’s a rather tenuous link between free software ideals & the Linux vs. GNU/Linux “controversy” – and part of an argument that’s wholly unconvincing to begin with. The same argument could just as easily be used to assert that Linux should actually be called “Ken Thompson/Bell Labs/Berkley/AT&T/GNU/Linux”.
That’s certainly an unusual interpretation. Perhaps you should also read another essay that the essay I mentioned also links to. This second essay is considerably shorter and it expresses even more directly why Richard Stallman thinks that “calling the system ‘GNU/Linux’ spreads awareness of the ideals of freedom”.
http://www.gnu.org/gnu/gnu-users-never-heard-of-gnu.html
In case you find even this second short essay too long for your abilities of concentration, I’ll try to summarize the essay’s main idea:
Richard Stallman argues that most people who use GNU/Linux on daily basis have never heard of the GNU project and its ideals of free software because the system is commonly called “Linux”. If more people started calling the system GNU/Linux, that might arouse GNU/Linux users’ curiosity about the GNU project and help to spread awareness of the free software movement and its ideals of free software. Stallman also argues that “the ‘open source’ rhetoric tends to lead people’s attention away from issues of users’ freedom” because the term “open source” usually just refers to a particular model of software development, with no explicit connection to users’ freedom.
Stallman ends his essay with these words: “When we ask you to call the system ‘GNU/Linux’, we do so because awareness of GNU slowly but surely brings with it awareness of the free software ideals of freedom and community.”
Hopefully this summary helps you to figure out what Stallman’s insistence of calling the system “GNU/Linux” has to do with free software ideals.
Edited 2008-10-04 00:30 UTC
Not particularly unusual. In fact, it’s not an uncommon one at all. I agree with Stephen’s interpretation, which makes at least two people right in this thread. And I’m pretty sure we’re not the only ones here who view it that way.
Which clearly demonstrates how flawed this “argument” is to anyone who has spent any time at all in any sort of distro support forum, or any other sort of Linux related forum, or even broader-scoped ones like OSAlert. GNU and the FSF’s vaunted ideals are discussed, ad nauseum, anywhere that Linux is discussed. It is, in fact, virtually impossible to avoid that relatively small, but extremely vocal cabal of Stallmanite missionaries who do their utmost to convert the masses of those whom they seem to view as “the unenlightened” to their “One True Path”… whether their intended targets want to be converted or not.
Sorry. You’ll get no “Gnu/Linux from me. There is already *too much* Stallmanism plaguing our Internet forums.
Edited 2008-10-04 01:24 UTC
As a matter of fact, quite a lot of people in this thread have expressed their amazement that they actually agree with Richard Stallman’s view that users should have the right (or freedom) to control their own computing and their own data.
And yet it’s not some new theme that Stallman has just introduced — on the contrary, it’s the essential message that Stallman has been telling about all along. But people just don’t seem to be aware of this message. Perhaps that’s because some people are very eager to shun the ethical aspect of free software as “impractical” and “foolish” when compared to the more “pragmatic” open source ideal of an efficient software development model.
Ethics is about values — it’s about how things should be. Stallman’s argument that users should have the right to control their own computing and their own data is shunned as “political” and “religious” in the open source ideology and rhetoric, but I’d expect that a lot of people would find this ideal of freedom very persuasive if they had the chance to hear about it. In contrast, I suspect that the “practical” ideal of an efficient software development model won’t persuade many ordinary computer users to switch from the proprietary operating systems to free ones, although some software developers may find this development model attractive.
And yet many people have never actually heard of the free software ideal that users should have the freedom to control their own computing and their own data. Maybe that’s because whenever a “Stallmanite”, who has actually paid attention to Stallman’s essential message, tries to discuss these important issues, there are suddenly dozens of “anti-Stallmanites” spewing insults and shouting at the top of their voice that they are fools and should be ignored?
Stallman has spent years on his high horse being ignored by everyone that matters in the industry. Linus used the GPL for practical, not religious purposes. Linux was started for practical, not religious purposes. If you buy into the whole Freedom thing, have fun with that. Plenty of people, including the founder/leader of the Linux project have no time for that sillyness, and resent RMSs attempt to ride the coattails of an actual success.
Well, the free operating systems (GNU/Linux, *BSD, …) have so far had relatively little success when compared to MS Windows, or even Mac OSX. Also, the GNU project has produced many times as much actual code to GNU/Linux than the Linux project. And, historically, there’s no way to deny that the GNU project started to work on a free operating system long before the Linux project added a kernel to the OS. So they’re not really just attempting to “ride the coattails”.
What I see less than ingenious is that GNU and Linux (or, at least, the advocates of these projects) spend too much time insulting each other when they could join their forces and fight together the real opponents: the proprietary operating systems.
The argument that open source is an effective software development model is likely to appeal to developers and corporations. On the other hand, the “ethical” free software ideal that users should have the freedom to control their own computing and their own data might be more persuasive to ordinary computer users. So why not apply both of these arguments instead of trying to prove that only one of them is correct and the other one is wrong?
Linux has had far more success then GNU HURD, which has been going since 1984 and has yet to produce a workable kernel. Stallman has never been able to penetrate anything but the universities, while Linux has been gaining more and more traction in the UNIX space. Gnu was chosen because it was free, and it was better then the alternative.
The FSF has opponents, they are trying to spread a religion. The open source guys don’t have opponents, they have the goal of creating great software.
Cause that is a line of bull. Users dont care about source code, they care about free as in cost and quality.
With a year’s head start on Linux if you only count HURD, a 5 year head start if you count the previous failed TRIX-based attempts, and a 7 year head start overall. While there are likely many reasons for this, I would like to point out one which is relevant to this thread. This link is currently only good for LWN subscribers, but will become available to all on Oct 16:
http://lwn.net/Articles/301135/
The article is about the continued lack of a plugin architecture in GCC, despite that being a long-standing and popular feature request by GCC users, solely because the FSF (AKA Richard Stallman) is afraid that it could result in what Richard feels is a politically undesirable outcome. In a nutshell, important technical decisions are being made for purely political reasons. (For more information, see http://www.dilbert.com ) I ask you, is that the most effective way to conduct our FOSS development? Is it the way we want it to be conducted?
Edited 2008-10-04 17:34 UTC
None of those points are particularly relevant.
It also begs the question: between GNU and Linux, who has benefited more from their association with each other? I think it’s pretty clear that, in as much as there can be said to be a “debt,” GNU owes significantly more to Linux than vice-versa.
If it weren’t for prominence of being associated with Linux, who would know of / give a damn about GNU? No one outside of hardcore academic CS circles, I suspect.
And how’s that coming along?
– First, it’s all marketing the same old crap under a new name. Saas, The Cloud.. this is nothing new; centralized computing through dumb/thin-client terminals. It’s all marketing, smoke and mirrors for what has been done since the first dynamic content was presented within html tag coding. Web 1.0, Web 2.0.. BS. Adding java and php to html can calling it Web 2.0 is BS. It’s still just 1s/0s over http protocol; nothing new except the spin that marketing and advertising now puts on it and people are eating up the kraft dinner like it’s cuts of gourme steak. The only difference I can see is that we have more layers of bloat and more advertising on the screen with less and less actual content visible.
– Read Google’s EULA. Everything you put through there servers belongs to them. If they choose to sell the information or analysis done on it; they can.. and will. If not now, when times get hard and money has to be found. Facebook and the other websites are no different. The user blindly giveth, the server happily taketh possetion. So much effort is put into lulling users into blindly handing over personal information and it’s working.
– If your ISP, their ISP any step inbetween or the service provider’s webserver, database server or infrastructure goes down; your screwed. You have access to their (you gave it to them) data only when connected with the ad boxes visible. If they choose to close or change the terms of service; your screwed.
– Insecure by default. Are al those lovely webapps only accessible through SSL encrypted connections? Does that start before or after you login? Is there a cookie involved? Is sidjacking a risk? Do you know and trust everyone on the other side who has complete access to your “private” data? Do you know and trust everyone who may be somewhere inbetween the two locations listening on the wire?
– webapps are never, ever as good as locally installed software. My work productivity was cut in half simply be replacing a locally installed tool with a webapp. I can’t tell you the number of times a bug in the webapp code or glitch somewhere along the network wire chewed a task half way through completion at a point where “save” is not an option. There is reason to present some software as a webapp but that is not true for every case, regardless of what MS Live or Google Office tells you. Most tasks are best served by a locally installed program; more efficient, more secure, less ways to crash.
The only place I see “The Cloud” aka SaaS aka Centralized Computing of benefit is within a company remaining inside it’s own network. The IT department can manage software updates and distribution from a central location. The wire has controllable steps inbetween client and server. The data already belongs to the company providing the shared storage. The Info Sec people can maintain proper controls and recovery practices over one central location. Webapps still suck but at least there are some benefits woth considering beyond what the marketing and ad guys claim to be Christ 2.0.
I agree with RMS.
However, he doesn’t expunge on his comments with any subtly.
First of all, what we are talking about is not just the idea of data or services people use without knowledge of their implementation. I think that is bad, but as people point out, businesses on a daily basis have no idea what the software loaded onto their machines are actually doing beyond what they tell them.
I think that is bad in and of itself.
But my biggest concern is the concentration of not just data, but the side effect of having so many of the internet services, depend on the whole concept of cloud computing.
For example, right now the internet has services, that if lost, you could technically get around. When slashdot goes down I hardly notice because my proxy server has all of the latest up to date content.
But, that is because I have my own services and data locally.
Now, if a firm or organization can design proprietary systems that have far and over reaching effects on the transport AND manipulation and processing of that data, you no longer have local control.
All it takes is mismanagement of that system, to make it incredibly corrupt very quickly.
I think the WORST possible cloud computing model for example would be anything that has to do with banking. Socially, and economically it would just represent too big of a target on the inside and outside.
To keep things honest, any system that is centralized around a cloud computing or utility computing initiative, should be totally open.
It is the only way to track and audit abuses in the system.
Fundamentally, open source is a social contract which is very simple. If things stay open, software tends to work very well, and society as a whole can reap its benefits to make everyone prosper. This just makes better software anyway.
Make it closed, and few prosper and there is abuse. Abuse of peoples privacy, software that only is fixed if a small amount of people say so, abuse of whole markets in fact. (i.e. monopolies) and as a result it curbs investment when there is unfair competition as the rule.
So if we are to talk about Cloud Computing it must be decentralized number one, and must be open hardware and software wise.
I mean can you imagine a single company owning TCP/IP and making it illegal for anyone to understand what TCP/IP is? Cloud computing imagines a world where a large amount of services are somewhere else to get a job done.
TCP/IP connects a large amount of services in our world today. It is not proprietary and a good thing too, because it has lots of design issues.
Issues that because the protocol is open, most people can design software to correct or protect themselves.
Last thing I would want is a cloud computing model that is closed, covers something as large as the internet, and is closed.
So, I think RMS has a very good understanding of those issues even if I had to point out more of the nuances, I totally agree.
-Hack
From a privacy POV, I think it’s useful to divide this concept of “cloud computing” (or SAAS, if you prefer) in two different kinds of online service, let’s call them Gmail-style and Wikipedia-style.
A Gmail-style service consists of storing your personal data online, just as you would do it locally. Your personal e-mails are not supposed to be read by anyone else. Of course that *could* happen, but it’s not something intrinsic to the service (quite the contrary!).
A Wikipedia-style service has the particularity that, apart from your personal data (if any), there are lots of shared and relational data, so that it’s not at all clear what part of it belongs to you. This in qualitatively different from local computing. Another prominent example of social manipulation of data structures is Second Life.
This distinction may be useful when deciding what someone who is concerned about privacy, open standards and (avoidance of) vendor lock-in may demand of a SAAS/cloud-computing model.
So, for instance, instead of having Google bots read all your mail, you may decide to pay for encrypted online storage. Privacy laws and policies should treat this storage as they do when you rent a house. You would be able to install whatever you want and download everything to your local storage whenever you want. What you gain is safety against data corruption, and access to your data from anywhere.Your ISP could provide this service or mediate in its acquisition.
In Wikipedia-style services, privacy is not the main concern; vendor lock-in is. So, you want to minimize the role of the service provider. Users should be able to download and run the entire wiki (or equivalent) locally. So there should be a FOSS implementation of the wiki engine. You wouldn’t contribute to a wiki which doesn’t let you do this.
Regarding the supposed “loophole” that the GPL leaves for SAAS, I don’t think there’s one. I’d go as far as to say that a license which treats SAAS as something other than a particular case of private use is, IMO, non-free in spirit, if not legally. After all, freedom 0 is that I can use the program for any (legal) purpose, in particular I can install the program in my server and let other people access this server and use the program. Does this scenario defeat the purpose of the GPL? I don’t think so. The reason is that only released software has a license. If I keep the software to myself, and other people manage to reproduce its functionality, I can’t accuse them of violating the license, because I didn’t release it under any license. There’s no “you can’t help your neighbor” dilemma.
OTOH, I could write a specification of the user interface (and the data model) of my service, and release them with an EULA, and then assert patents on them. If the data model is somehow a derivative of the GPL’d program’s data model, the GPL should be enough to prevent this action; prior art also applies. Otherwise, I’m not hijacking the code of the GPL’d program in the first place, but merely using the program. Of course, users who want to promote FOSS should avoid online services with such EULAs as they avoid proprietary sofware.
IMO copyleft licenses like the GPL shine when they restrict their scopes to licensing and patent conditions. When they go a step further and try to defend user freedom against technical restrictions and other “threats” beyond intellectual property issues they end up stepping on some of the same freedoms they are supposed to protect. In this I’d say I’m a bit closer to Linus’s position than to Stallman’s.
Cloud computing is nothing more than a euphemism for outsourcing. It’s just basic two tier server/client architecture with the server part being outsourced to third parties.
Google Docs & Spreadsheet, for example, have enabled me to collaborate with multiple people using a variety of OSes and programs
We were doing that 20 years ago before there was a WWW. Heck I still recall using 123/m… M was for mainframe. CC is and old practice using mainframes but marketed as new idea to drive new revenues. Under the hood its still a Mainframe idea.
A mainframe idea with commodity hardware.
RMS is right on. If I lost my gmail, I don’t care, but if a business loses its data, bad things can happen. Microsoft’s marketing of “cloud” computing is simply a way for them to generate revenue, nothing more, and at the expense of the security of your data and apps.
What happens when they lose interest in this project? Cloud computing may seem like a cool, trendy idea, but regardless of what generation you comprise, handing your precious data over to an untrusted source is a monumentally bad idea.
In the United States at least, your rights to a thing are based on ownership. And where this ownership is of fuzzy thinky things, like stories and ideas, the government tries its damnedest to treat them like physical things, and gives you the rights directly through Trademark, Patent, and Copyright law.
Now, how many rights do you have if you do not OWN anything? How can you ever own anything if nobody is willing to do anything but sell you a lease or license? How can you own source code without a government backing it with copyright laws?
Luckily, the government and constitution take precedence over what corporations and individuals claim you have a right to own and do.
On a practical level, I’m with Linus in that I just want to store my data on (in?) the cloud, and fine with Google keeping even my slightly more confidential data on its servers (though I do have limits). And while it may not be as secure, it is FAR far more reliable and accessible.
Keeping everything on your private server is like managing your finances exclusively with a mattress. It’s great when you need to survive a network|financial holocaust, but horrible if you want to work with a larger group, or to keep folks from stealing said assets (it is a physical server after all).
Regardless of the resource being managed, a wise investment strategy would be to diversify, but sadly I still see many people after that ‘One Answer’ to their information|financial management woes.
It really is about how much you’re willing to trust someone you personally don’t know with power. And no human society on Earth works without its members doing this to some level.