Stallman Video: If You Want Freedom, Don’t Use Proprietary Software
mashable.com
mashable.com
My personal choice is slightly more fine-grained - I use proprietary software, but will not invest in it, so as to avoid lock-in. I won't use anything that requires proprietary file formats. I use a proprietary text editor, but only because it offers a familiar vi-mode interface. I do a brief mental audit every few months - If I expect it to take more than a day to migrate to a fully-open environment, I know I've ceded too much control to software companies.
I think Stallman is a fundamentalist, but an important one. I think he prompts debate and exemplifies practice in a way that is only possible because he simply will not touch proprietary software. Seth Godin recently argued that one of the most underrated properties of the internet is its ability to show us the limits of what is possible and in that sense Stallman is vital. I see him as a figure like Heston Blumenthal or Ray Jardine - there aren't many people with a sous-vide setup in their kitchen or hiking the Appalachian Trail with an eight pound pack, but their example broadens all our horizons.
If a piece of software is useful, use it. If it's useful enough to pay money for, pay money for it. It's just not an area that requires ideology to enter into your decision-making process in any way.
As you've noted, they've got a guy on the internet saying every possible thing there is to say. Do yourself a favor and ignore pretty much all of them.
Two major upheavals shafted a lot of audio engineers.
One of the most popular DAW suites is a package called Logic. After nearly twenty years of development, Emagic sold out to Apple and Apple decided to abandon the PC version. Suddenly guys who had used the same piece of software all day every day for a decade or more were faced with either buying at least one top-end Mac, having to learn to use OS X and potentially a whole bunch of other software, or abandon their preferred DAW and have to relearn how to do their work in a different DAW. Either option would prove enormously expensive for a lot of engineers, both in direct costs and lost work.
When Apple shifted over to Intel, there was a huge lag from many developers in getting updates released. For engineers, this is a serious problem - having the fastest hardware and the latest software is a significant competitive advantage. A lot of people lost out on work, or had to buy PCs or new audio hardware.
The experience of that world has led to my current position. Professionals in all sorts of disciplines have huge personal investments in particular software and are often completely betrayed by the developers. EULAs are a one-way street and only the most rarified of enterprise software has any sort of reciprocal commitment. What Stallman talks about isn't purely abstract. Relying on proprietary software is dangerous, for very straightforward practical reasons.
In other words, what happened to make the software that you were running for 10 years suddenly cease to be useful?
Others sleep better knowing there's a company with a financial incentive to keep the product alive.
Honestly, either way, I don't see where morality enters the equation. It's not wrong to release software that's closed-source, it's your choice as a developer. It's not wrong to give up some of your freedom, it's your choice as a consumer.
In other words, going open source in this type of sense means that you won't have the latest and greatest anyway. So your particular case example doesn't change things.
http://en.wikipedia.org/wiki/Source_code_escrowAccording to Wikipedia (http://en.wikipedia.org/wiki/Logic_Pro), Emagic and Logic weren't around until the 90's, even though Logic was based on ideas from previous software. It's stated there that Emagic and Logic, as software named "Logic" didn't exist until '93. Then Emagic was acquired by Apple in 2002 at which point the Windows version was discontinued. The information there, if factual, would mean that Logic was only being developed by Emagic for about 10 years before they were acquired by Apple.
Information here implies (http://www.tweakheadz.com/history_of_notator_and_logic3.html) that the Mac version came in '93 followed by the PC/Windows version in '95 or '96 which would mean that Logic wasn't being used on PC/Windows for more than 6 or 7 years or so before the Windows version was discontinued by Apple. Logic and all the previous software whose ideas it was based on were also originally on Atari.
The extent to which those skills accumulate is highlighted by the fact that "Pro Tools Operator" is a job title in the industry, and that their pay varies more widely than that of session musicians. Using a DAW isn't just a secondary task, like a writer uses a word processor - at the high end, operating a DAW is a job in itself. People have built whole careers simply off the back of being very quick at piloting Pro Tools. Being told that your DAW is being discontinued is a bit like being told that all your old electric guitars are obsolete and won't work with new amplifiers. It's vaguely tolerable if you're a hobbyist or an artist who can afford to be fussy about their kit, but if you're a professional who needs to get the job done every time, it's a catastrophe waiting to happen.
The alternative is that Visicalc, or Deluxe Paint for Amiga continue to be updated forever. I wouldn't invest in that business, and I'd hate to be a developer there.
Trying to get into the commercial software packages as an amateur, one gets this sinking feeling of, "oh my god, there's too much stuff to maintain." Installers for every little plugin you come across, some of them using dongles, others using serials, some a tiny sub-megabyte download, others being dozens of gigabytes in samples or loops. And then you have to think about the handy presets and effects chains, hardware setups for different situations...and then a learning curve for everything, even in mostly-similar situations. It essentially becomes impossible to restore a project from an earlier setup, other than to salvage raw audio and MIDI.
I can see why a lot of people still put their stock in hardware even today; the product cycles are longer and the hardware isn't going to "disappear into the ether" so easily.
However, I also think that open source is going to gradually gain an advantage here because, once developed, it can remain stable, well-integrated, and easy to install in ways that the commercial packages can't. It hasn't hit the point where you can adequately point to equivalents for all studio functions, and it probably won't do so for some years to come, because of the extremely slow development of UI polish, new features in existing software, and API standardization. The Linux audio distributions today are still embarrassingly cumbersome to work with. But it's possible to envision a day where it all works together and has some form of healthy relationship with the commercial market.
Money doesn't have anything to do with it. His point is about freedom, not cost.
This is a key point I think. His approach (i.e. not touching proprietary) is clearly silly as a general attitude (both from the point of view of the software ecosystem and as a user). But by staying true to an ideal he believes in helps counterpoint the closed-tight ecosystem of some companies.
Somewhere in the middle there is a happy medium (possibly where we sit right now).
There seem to be a couple of implicit arguments in this that I find interesting. First, that others aren't to be trusted. Second, that software is in no way an invention.
Regarding the first, I think that for most people they have to rely on the developer of the software they use. If the developer decides to stop working on it they can get screwed. With open source software they may be able to find another developer, but then they are simply at the mercy of the other developer. Their lack of freedom isn't shaped by whether the software is open source or not, it's shaped by their lack of knowledge.
Regarding the second, it seems that the argument for free software is an argument for free speech. This in essence means that software is exlusively a work of creativity (an thus copyrightable) and is not a work of invention (thus not patentable). I don't completely buy this argument. I think software is unique in that it has aspects of both creativity and invention. On one end of the spectrum _why would definitely epitomize the creative aspect of software. On the other end, software paired with super computers and embedded systems certainly come closer to machines.
Overall, I these are tricky things to deal with. I don't think that all software should be out in the open, nor do I think that every piece of software should be cloesd away. I think the free software community needs to have a more nuanced view of the world and that having one will bear more fruit.
With regard to programs that I don't develop, I'm very much a user who enjoys that almost everything I use comes nicely in packages with Ubuntu. Almost.
A few programs that I use need patches or modifications that aren't included in stock Ubuntu.
Because I use free software the appropriate patches do exist and are available on which ever website their authors have published them. Or they circulate around on the internet. So, the features I desire suddenly become available via side-channel distribution. Often someone keeps prebuilt binaries of the software available with the same patches compiled in. Or they're available via alternate or backported APT repositories. In some cases, being a developer myself, I download the patches and dpkg-buildpackage them into binaries myself, but generally I not.
If I had used proprietary software:
- No third party could write patches for it;
- I would have to stick with whatever the author decides to release;
- I couldn't run old binaries if my OS went through some big upgrade and the author didn't bother to support a yet another binary;
- I also wouldn't enjoy the vast repository of prepackaged and configured software that comes with Ubuntu: unless the author was agreeing to take part in an OS distribution, I would have to search, download, and install each program separately.
Free software is much more than a single developer's ability to hack on other people's code. Free software is a whole ecosystem that benefits each participant. But since all software will come down to programming, that software ecosystem comes down to individual developers' ability to hack.
I hear this again and again, and I think it's absolutely false. It's like saying
"I don't have the time to shop around for the cheapest product, so a market economy is useless to me"
In both cases the point is that even if you don't have the time or the knowledge someone else does, and only a small number of such people are required for the system to bring benefits to everyone.
That doesn't mean the freedom to elect a democratic government is not valuable to those people.
Anymore? You really think it was easier in 1980s?
Over the years Stallman has singe-handedly affected the software ecosystem in very tangible and important ways. Yes he may look like Osama preaching from a cave in these videos, but make no mistakes, software landscape would've been different and even more sad place if not him.
You can get Linux distributions that are 100% Free Software nowadays. 8 years ago that was simply impossible. You can be wholly productive on a FOSS system for a multitude of tasks.
Seriously, do you even go to FOSS conferences or meetings? If anything the number of participants has increased exponentially in the last few years. More people know and use Free Software right now than at any other point in history. And more governments and states look to Stallman for solutions on many technological issues (look it up; the Indian state of Kerala, several South American countries, etc).
FOSS has never been used as widely. Stallman's campaign is a resounding success in comparison to the way things were before.
Bingo.
I disagree. While perhaps it's wrong that many people give up freedom without really knowing that they're doing so, I think it's reasonable to on occasion to do it if you know the trade-off you're making.
If it were possible to make a list of everything I wish to do, and then arrange to do those things in the most time-efficient manner possible, there would still not be anywhere near enough time in my likely maximum lifespan to accomplish more than a small fraction of them. I'm not going to further waste my limited time using tools that aren't the best I can obtain for the task at hand.
1. Software companies like Google: Google open-sourced Android, but Android isn't their money-making business. They want to use Android to help their primary business - search and advertising. They haven't open-sourced their search code or their ads code.
2. Consulting companies: These companies focus on generating money from software consulting, not from creating software products. So promoting the open-sourcing of software-products is good for them. The software/consulting division of IBM is a good example of this category of companies
3. Hardware companies like Sun: These companies make their money from selling hardware. They stand to benefit if software is free/cheap and easily available for their hardware.
4. Non-profits: These organizations get money from wealthy individuals/foundations and their software products/websites don't need to generate revenue.
I'm curious to know of other open-source business models that have been used to build a sustainable business.
If you need to pay for support services to use a web browser, a text editor, or some other completely normal piece of consumer level software for example, that would be the software's problem and only exemplifies the areas where open source doesn't work as well as proprietary.
Companies make money from software because they devote lots of resources into making software easy to use. Relying on support services to float every random xyz piece of software will make for a very dead industry.
The funny thing is that I've never actually seen any viable business models from the open source community about how to build a business around free software (as in beer and freedom) that wasn't entirely couched in "support services". This creates the unfortunate situation that if a company were to create a consumer level piece of software that was easy to use (say, anything by Apple), they would put themselves out of business. Their incentive then is to simply make the software terrible to use so they can sell support services.
On top of this, the price of support services is often quite a bit higher than the alternative of just buying software. A thousand dollars of custom consulting will get you about a day's worth of labor hours. If I need to bring in a guy for a week to setup some software, or purchase the $3000 easier to use proprietary software, I'll go with the cheaper option. Even replacing that software at some point in the future, due to lock-in or what have you, is cheaper.
Paying a bunch of money so I can figure out how to use an OS to connect to the internet so I can send an email or make a home movie is a waste of my time and money. Ergo, support services only work when the cost of buying a proprietary solution that works out of the box (or with minimal support) is greater than the cost of the free software plus support services.
And yes, I recognize that I'm talking about beer here, but for end-users of software, that tends to be the principle consideration. The truth is, most people and companies that use software care very little if their freedoms are being impinged by using MS-Office.
Mostly what RedHat sell is server services and a lot of the original market they competed with were similar services from IBM, Sun, SGI etc.
I don't think support services work for desktop software, particularly not for non-business critical parts
I don't think Open Source is really about the cost for larger businesses. It's about having something stable, reliable, predictable and ultimately knowing that you can get it fixed if it breaks. Google, for example made $23.7bn last year, their software stack, as I understand it, is completely either open source or coded internally, so they can fix anything that goes wrong. Had they been based on proprietary software, then they would be at the whims of some company who doesn't want to fix their issue (think Microsoft for example). Facebook have found the software they used was not able to work at the scale and cost they wanted so were able to tailor it to their needs, they would have found it difficult and expensive to do that with proprietary software.
Companies around free software often sell services because that's what people want and they didn't often write most of the software themselves anyway.
Whilst Open source does get a bad rep. for being hard to use, I think it's unjustified and if your people can't work it out then their not very good TBH. Most startups are based entirely on open source software, so there is a certain amount of proof that on the server, at least, it is easy enough to use.
I think you under-estimate the cost of replacing software when large numbers of people have become used to it's intricacies and not forgetting many staff need training in order to use any software effectively. I've been involved in a few big projects and the people time involved usually outweighs the software by 10:1 even with proprietary solutions. Also bigger organisations are less risk adverse so they tend to buy support contracts on everything.
Software is a commodity and the ones making a lot from it have mostly engineered a artificial situation where it's difficult for anyone to compete with them effectively. That is not good for you as customer and is a risky strategy for them.
"The funny thing is that I've never actually seen any viable business models from the open source community about how to build a business around free software (as in beer and freedom) that wasn't entirely couched in "support services". "
When I said "build a business around free software" I meant as the producers or authors of the software.
The reason I make this distinction is that I think most of your examples involves companies building a viable business around open source software as users of the software. I'm also implying a distinction between authors of software vs. people who modify the source to do something they want -- and yes I realize that's a bit of a stretched distinction w/r to open source. Often times the authors are the same people who use and contribute small bits of modifications to the code.
What I mean is that, say Apache for example, comes from the Apache Foundation. If I make some modification to the source to make it do some particular thing I want, I'm modifying the software in the same way a car enthusiast may modify their Honda to do something they want. (If, what I made is generically useful, I can submit it back to the Apache Foundation and perhaps cross the line to become and "author", but in most cases we're talking about people who mod software in some way).
The last place I worked for does about $10bn a year. It's a technology company with their hands in projects as diverse as autonomous undersea robots to text processing systems. They've successfully built a business around a mix of proprietary software solutions and some small percentage of open source software. The default OS is some Microsoft Server OS, databases are almost exclusively Oracle or SQL Server, Web servers are almost always IIS. They also write a lot of their own software, typically in Java (even before it was opened) and/or Perl but most of the software jobs involve writing code in some highly proprietary development environment like those you find in finance, HR, or some other types of systems. Java and Perl just happen to be go-to languages because it's easy to find people who write in those. Python was starting to show up a bit as well when I left that job.
They also get ok mileage from really true open source software. Red Hat finds its way onto some fractional percentage (along with Apache, JBOSS, and a few other odds and ends -- including the aforementioned Perl). But in many cases, they could do what they wanted with completely proprietary solutions. Those systems were usually selected due to some client requirement, or because the licensing costs are free (beer). I only ever ran into one true open source advocate in the entire place, and he was usually laughed out of proposal work because his "solutions" options would have extended the development effort by a year and required hiring an entirely new staff of people raised on that ecosystem rather than the preferred and proprietary MS/Oracle/Sun/Cisco ecosystem. Finding qualified staff to work on the open source stuff was frankly a continuous and quite major problem for us -- often times, because those people counted as people with specialized skillsets vs. the average run-of-the-mill trade school "tech guy" they simply cost more to hire. In other words, we ran the numbers over and over again, and open source rarely made sense for us.
I do think that the reverse is also true though, for small startups open source can make a lot of sense in both senses of the term "freedom". For the bottom line, it's free, if you are doing something weird you can, in theory, mod the software to do something unique. It's not always easy, and I've seen many cases where the target goal was too big for the staff and the startup folded. Web startups are also interesting because for the most part, the work they do with open source results in a completely proprietary web app (something RMS has lamented on quite a bit recently).
Many big businesses likewise do great things with and for open source (IBM, Sun and Apple come to immediate mind) in terms of funding the kinds of expensive Open Source development efforts that would otherwise not exist because building world class software by part-time hobbyists is extremely difficult -- there's really only a very small handful of those projects, and most of them have been funded recently by large corporations looking to leverage the initial investment/development effort to make their solution work like they want.
But by and large, large organizations will move to whatever accomplishes their goals for the least amount of $$$. Even startups. If a startup can use a proprietary solution for $x thousands of dollars vs. modifying some existing open project to scratch some itch for $y thousands of dollars, and x<y they'll go with the solution that costs x. For example, how many startups use Windows or OSX as their primary desktop OS vs. some flavor or Linux/BSD? Probably most.
This guy gets a lot of what I'm saying here.
It's a great analysis of the problems in the community flavored with a hard dose of pragmatic realism. He gets it that "just do it open source" doesn't really make sense in many cases because the community is by and large made-up of people who don't understand what it takes to make a business.
That said, there still is a need for support around this products, so people create businesses around them an employ the people who wrote it.
I don't think creating a open source project solely to make a business out of makes a lot of sense in most cases, because profits in the early days will be low and it will take long time to scale that business.
Most businesses don't see the value of open source really and your story shows that the company you worked for didn't either. That's not to say there are aren't lots of companies that do realise the benefits. To use your example, some businesses like the Honda how it is and others want to make a race car and absolutely can benefit from modifying it. Google, Apple, Facebook and so on have all greatly benefited from modifying the Honda.
I love the idea of free software, but in practice it's left a lot to be desired for me. For others, more power to them if it works for them.
"much", not "all" by any means. Some utterly, incomparably awesome stuff has come out as OSS.
For us, OSS has worked a lot better. Having the ability to inspect the source code all the way down to the hardware level, tweak it, adjust it, and share the fixes with the world is a lot better than waiting for that phone call back or that tech support email back. -- This is the practical benefit of OSS. It also appeals to those who want to understand how the system works, and those who dislike dealing with black-boxes that constantly break.
And remember that the deciding factor in many cases is user interface - the average user does not care if they can hack it. Many many many OSS projects which could be contenders have absolutely abysmal UI and help documents.
Your argument would be significantly stronger if you actually gave examples here.
Maya / 3DS Max / Whatever - Are you really going to claim Blender is any good? Functional, sure, but that's about it.
Windows - Linux graphics drivers are often an absolute joke, so Windows really is just about the only real PC gaming option. And don't claim it's because graphics cards are closed hardware - then why hasn't an open-hardware group made anything competitive? Their only major success has been the Arduino (a BIG success, but rather lonely).
OSX - Linux OSes, even Ubuntu, have all had clunky UI and seriously lacking normal-user help documentation. Last time I used Ubuntu, I had to resort to the CLI inside of 30 minutes to correct a permissions error the UI could only inform me of. Try telling most people how to do that.
Outlook / Mail.app - Thunderbird's the biggest contender, and has many things right... but while the new UI is much nicer, I ran into many display bugs in less than an hour of use, one of which resulted in my deleting an email I did not have selected (when I didn't hit delete).
MS Office suite / iWork suite - OpenOffice has a UI lagging behind even MS Office, and in many of my uses has failed to have a working updater and lags behind my typing frequently. And then there's the preferences window... most people simply cannot understand 90% of that, and will be turned off the entire application by its mere existence.
Half Life / Quake / Unreal Tournament / PC games of any kind - ?
Console games of any kind - ?
McAfee / Norton / AVG - ? ClamAV? Brought my system to a crawl last time I used it. Not that McAfee / Norton are much better there, but...
And in the other corner, we have:
Firefox / Chrome, Pidgin / Adium (though Pidgin has a bad install process and a nasty UI for most people), Eclipse (though it's slow)... I know there are more, but I'm having trouble coming up with them right now. Just woke up.
I had generally thought the major ones were self-explanatory, and this is rather long, hence its initial omission.
Free software works for us. We ship a complete box with a free OS, built on completely free frameworks, languages and libraries. We release any contributions or updates we make to the OS software.
Just recently we had to license a close-source library. It has been very frustrating having to deal with licensing and support. Send emails back and forth each time something doesn't work instead of being able to just look at the code, apply a patch locally and send it upstream.
The problem with something like Facebook, though, is that choosing to use a platform other than Facebook for Facebook's purposes is useless, because the entire purpose of a "social network" site is lost unless a large number of people are using it. Most people aren't very technical, so they're going to use one networking site, and guess what? It's going to be what "everyone else" is using.
So, sure, you still have the choice to use it. But it's not like, say, system software, in that if you don't like the primary offering, you can go implement your own or download somebody else's.
That being said, I've still found absolutely no reason to use Facebook, and as such I don't.
I'm living fine without it. No one is forcing me to use it, and email (free! but using Mail.app and my iPhone to give it a boost I want and need–though I wouldn't mind Letters.app to appear) handles most of my needs, along with Flickr and Twitter, which are also both closed platforms.
I'm not too burdened by the software I use. I use what pleases me most. Right now that spectrum includes emacs at one end and iPhone OS at the other.
As someone pointed out above, free software fundamentalism has its place, and I respect it. But I also respect the hard work and effort that goes into software like 1Password and Things, and they both make my life a little bit better.
Those concerns are orthogonal.
I'm not sure freedom is what one gives up in this situation. It seems more like giving up control. And one has to do that in life just to be able to delegate tasks to others.
Is it really freedom being given up or freedom obtained by freeing up one's time to do things you value more than examining the code for the microprocessor that makes your clock radio work?
And its freedom that makes it possible for one person to offer the work of their minds to others without having to disclose everything about how it is that they are capable of doing it.
We are free to make such a deal and then later if we choose another alternative. There is no loss of freedom at any point, just the exercise of one's freedom.
I guess I'd say that I'll unhappily sacrifice some freedom to gain quality.
Over time the issue of freedom in the digital domain has become increasingly important. Today perhaps more than ever there are an array of forces who are trying to snatch away people's control over their own computing and put it into the hands of highly centralised technocrats, usually by dangling the carrot of convenience. The personal computer is in danger of no longer being personal.