Slashdot Mirror
A Decade of OSS, 10 Years After the Summit
Jacob's ladder writes "Ten years ago this week, the Free Software Summit arguably marked the beginning of today's OSS movement. Ars Technica interviews many of those in attendance when the revolution began. John Ousterhout, creator of the Tcl scripting language and Tk toolkit and founder of Electric Cloud was there, and notes how much the landscape has changed. 'When I made my first open-source release in the early 1980s (VLSI chip design tools from Berkeley), there were probably less than five open-source projects in the world. By the time of the first O'Reilly conference, there were dozens; now there are probably thousands. Also, open-source software has received substantial mainstream acceptance. 10 years ago, people were suspicious or afraid of it; now it is widely embraced.'"
Even without the acceptance of Linux on the desktop, there's no doubt that open source has been a ridiculously huge success since then. Equal acceptance (at least) as a server OS, it runs the majority of web servers and web scripting languages. Overall, a very successful life so far. I'm excited to see where it ends up ten years from now.
I am very pleased to see the progress of the last ten years-- and there is more yet coming.
Besides, I am guessing that Microsoft's "Vista" was a cleverly positioned 'failure' that is now allowing their 'damaged' brand image to give them enough room to buy out Yahoo, whereas before they would've been stopped by the Gov because of their 'monopoly'.
But.. not now.
I do not respond to cowards. Especially anonymous ones.
So, now that OSI is not an idea, but a group, and that everyone is happily coding... Are the people in the group going to come together? Someone, somewhere has said "United we stand, divided we fall" .. That goes for any association. I feel this future has potential if OSI can develop into a united power without losing the original sight (GPL2 vs. GPL3, anyone?)
You are in a maze of little twisting passages, all different.
I'm a huge fan of OSS, but what I love even more are the spin-off movements, namely the open content projects. Of those, the two I love most are Wikipedia (of course) and the just ramping-up Metagovernment project. Together, these are in the process of completely transforming how human society operates.
There was also freeware, trialware, crippleware, shareware, talk of varying types of licenses, and anything you didn't pay for normally came with caveats that fall into the 'you get what you pay for' category. So, yes, there was a lot of suspicion about OSS because of all that it was competing with.
That was even before MS had killed off all of its serious competitors.
Then there was just MS and Windows developers. There were a few areas of competition but Windows was just a far cry above what DOS programs were doing at the time. Do you remember paradox? Qbase? WordPerfect? WordStar? Novell? 10Base5 ethernet?
I'm quite glad that OSS has made it this far and one so much.
Support NYCountryLawyer RIAA vs People
Think it would be cool if every Slashdot reader listed the open source project they have released along with the Sourceforge, Freshmeat or or repo address. I for one haven't updated my project, PHPulse (a highly scalable lightweight MVC framework for PHP) or about a year even though I have code updates on my machine at home. Get busy helping corporate customers and forget the main project. http://code.google.com/p/phpulse/
This is my sig. There are many like it but this one is mine.
They might have given it a name but there was a great deal of free software around 10 years ago. My impression from those times (and it was only 10 years, we're not talking a lifetime here) is that the primary driver for free software was the internet - not a bunch of people at a conference, even if they call it a summit.
politicians are like babies' nappies: they should both be changed regularly and for the same reasons
Less than 5 that he personally knew of perhaps, but in universities it was normal to share the code to your software. Just because people hadn't created a "movement" and a label for it, doesn't mean it wasn't there. Ignoring the people who came before you so you look like a visionary is pretty pathetic.
The one quote that really bugged me is the following one from Ousterhout:
> The third thing that has negatively impressed me is
> that open source is often used as a desperate last-ditch
> effort for loser software. If a product is doing poorly
> in the marketplace, sometimes companies release it as
> open source, hoping that will somehow magically revive
> it and make it widely used. This almost never works.
Does this guy not realize that Firefox was born from Netscape going south? I'm sure there are other examples out there, but how obvious does one need to be?
Isn't a wikipedia article on open content sort of a vicious circle? Like when Jimmy Wales got caught for editing the article on himself? :)
now it is widely embraced
Er, no.
I still, on a daily basis, run into people who would rather buy software than use OSS alternatives because they firmly believe "you get what you pay for". And this in the "Joe Sixpack" crowd, not even talking about fellow IT professionals.
Among them, I get much more polarized attitudes - They either embrace it, or shun it (with reasons ranging from the "viral" licensing BS, to (yes, seriously) tirades about damned hippies trying to buck the system).
Me, I'll just use what works. Sometimes that means paying for software, but I can usually find something comparable and Free (and with a price tag of "free", I give "comparable" quite a bit of leeway).
"It does not do to leave a live dragon out of your calculations, if you live near him." - Tolkien
I may be trolling or flaming you, but that doesn't change the fact that you're dead wrong and missed the meat of my post.
Something that's given away for free has become popular?
Wow, that's a surprise.
The open source movement was already well under way before the Open Source summit. It was already well on its way before the GNU manifesto and the founding of the FSF. There's a perception that it's big events like these that "created" the open source movement. That's not so, it's the open source movement that's created the possibility of big exciting events.
Even people talking about f/l/oss before these events seem to miss it. for example, Ousterhout's comment "When I made my first open-source release in the early 1980s (VLSI chip design tools from Berkeley), there were probably less than five open-source projects in the world."
The Software Tools applications and libraries date back to the '70s. So does Emacs. So do the enormous collection of software published in Dr Dobbs' journal. So do the DECUS and other user group tapes. Much of this was game software, but it also included free compilers and interpreters (Forth, Small C, Tiny C, Tiny Basic, Tiny Pascal), editors (including emacs), operating system monitors (and early attempts at UNIX workalikes), and networks. Usenet was an open source project, and there were soon open source gateways between Usenet and networks like Fidonet... and one of the earliest Usenet groups was "net.sources".
I would say the first open source decade was the '70s, though in a way it's as old as the computer industry: "in 1971, I became part of a software-sharing community that had existed for many years" -- Richard Stallman. It's been argued that it wasn't really until the '70s that closed source really got under way, so one might say that it was the creation of a binary, non-shared, closed-source software industry that created what became the open-source movement (under whatever name you like).
So depending on whether you include the '60s, we're coming up on the end of the 4th or 5th "open source decade", the '00s. Not the first.
Another recent interview-based take on the State of Open Source/a., with discussion from Bruce Perens, Eric Raymond, Google's Chris DiBona, IBM's Bob Sutor, Microsoft's Sam Ramji, and a host of other business-minded open source folks from Alfresco, EnterpriseDB, Mulesource , Hyperic, Asterisk, and MySQL.
I realize that a creator is not responsible in any way for the various ways in which is creation is used. But I have to wrestle with Tcl code every day because it was packaged with a large commercial application my team supports. Its strength is also its weakness: almost anyone can learn to use it (and frequently badly).
And why is the Tcl interpreter so brain-dead? Consider the complaints from the interpreter when encountering "unbalanced grouping symbols" that are contained within a comment. Most parsers throw out all contents of a comment as soon as it's identified. But if you have an expression like
set foo "bar"; # (oops forgot a closing paren
it will refuse to work. WTF?
I only post comments when someone on the internet is wrong.
The primary driver for free software was the post office. DECUS tapes, Dr Dobbs' Journal, Software Tools, all the user group floppy collections. Then the primary driver for free software became the phone company. Usenet, Fidonet, BBSes. The Internet didn't get into the game, really, until the '90s... free/open/watchamacallit software was decades old by then. :)
When I made my first open-source release in the early 1980s ..., there were probably less than five open-source projects in the world.
You've missed a lot of computing history. Maybe the capitalized phrase "Open Source" was new, but the practice wasn't. For instance, before the mini/micro-computer "revolution", I worked on a number of IBM mainframes, all of which used VM as their main OS. VM originated in academia, and its source was always available to anyone interested. Of course, not too many people wanted it unless they had an IBM mainframe. Most such installations had a VM guru on the staff, and the VM gurus I knew were quite open with their source.
Around the same time, on one such machines, the engineering staff brought in Amdahl's unix system, which ran on VM of course. When we asked about source, the reply was "That's not an option; you get it whether you want it or not." "Open Source" may not have been a catch phrase yet, but Amdahl was happy to have customers with employees who could read the source, since that made their support job a lot easier. In fact, I sent them a kernel bug fix about a month after we got the system installed; I got back a nice "Thanks!" letter and was added to their published list of code contributors.
A more accurate history would be that open source was quite common before the mid-1980s, but it didn't need a name. Software vendors routinely gave source to customers who wanted it, with the expectation that customers would find and fix bugs and maybe add new features. One of Microsoft's innovations was to hold their source as proprietary, so as not to allow customers to improve the software. A lot of people were amazed that customers actually accepted this. You heard a lot of questions like "Would they buy a truck or car that couldn't be worked on by any mechanics except the manufacturer's?" But then, when it became clear that Microsoft had gotten away with such a dodgy scheme, it was quickly adopted by others, so that customers would have to pay them for patching up the bugs.
It still sorta amazes me that customers can be so dense as to pay money for products that can't be repaired by anyone but the manufacturer (and usually now not even by them). So much for the economists' idea of a rational marketplace.
Those who do study history are doomed to stand helplessly by while everyone else repeats it.
I had been interested in the metagovernment project, but I noticed that their site had been around a while with no real openness: no way to contribute to it. Nor did I see any input on the related metascore project on sourceforge.
I wrote to them late last year, and--- I was just contacted last week. They said they are starting the next major phase of initialization this week. So I suspect there is going to be much more happening there soon.
I sure hope so. The more I pay attention to the US presidential elections, the more I want to see us get rid of this ridiculous process of so-called democracy, which seems to be anything but.
Depending who you ask, the open source concept has been working in practice for 20, 30, 40, or even 52 years: IBM SHARE was founded in 1955.
By anyone's definition it's at least as old as the Free Software Foundation, which makes this article's premise ridiculous.
Holy fact checking Batman.
you had me at #!
And why is the Tcl interpreter so brain-dead?
Because it's simple. Deliberately so. It's inspired by Lisp.
There's 11 rules that define the complete syntax for Tcl, everything else including control structures is built up on top of that.
I'm responsible for some of the complexity that IS in there, originally there wasn't a distinction between {...} lists and "..." strings at all: I'm the one who suggested that variable substitution be allowed inside "...".
But if you have an expression like
set foo "bar"; # (oops forgot a closing paren
it will refuse to work.
No, that one's OK, but if you have
set foo "bar"; # {oops forgot a closing brace
it may not work.
The reason is that the parsing of comments happens at the block level, but the parsing of blocks happens at the list level. So {# this list happens to start with "pound sign"} is a list. The fact that that list might happen to be code as well doesn't make a difference when it's parsing lists.
The development I had hoped to see in Open Source but never did:
Exploit the fact that Open Source projects (potentially) have a lot more ears then closed source.
Does anybody remember back when computer magazines used to print programs in source code form for you to type in? That was certainly one way of making the code available.
Coder's Stone: The programming language quick ref for iPad
The Floss Weekly Podcast (24) had an interview with the founder of POVRay. That's a project whose source was available in 1987! I remember using it on my Atari ST back in the mid 90s.
Coder's Stone: The programming language quick ref for iPad
No part of this paragraph is true. The OSI had existed for two months when this summit convened. The term Open Source was concieved in a meeting at VA Linux Systems by Christine Petersen.
Bruce
Bruce Perens.
I think 10Base5 was pretty much on its deathbed when Microsoft appeared on the scene. The cable was thick and unwieldy to install. It was costly, as you needed active devices to connect to the cable. 10Base2 was a lot cheaper, and it offered the flexibility to re-wire a network. 10BaseT was cheaper still, and much more fault tolerant.
When our name is on the back of your car, we're behind you all the way!
...when you count "open source" projects.
"Free" software was a concept that I think really emerged with the release of GNU Emacs. Gosling EMACS and Stallman's PDP EMACS may have been "open source" but they weren't truly "Free software". Being that it's 10 years after the FREE software summit that the discussion is around that rather than simply computer programs with widely distributed source code.
In the loosest sense of the definition, open source has existed since at least the 1960s. Computers were not commodities as they are today; they were large, expensive and complex. There were less computers around, and the user community was smaller and more knowledgeable. The industry was driven by hardware and services provided by mainframe vendors (IBM, AKA "Snow white" and the "seven dwarfs", IBM's main competitors). A good deal of software at that time was open source if not Free, because there was little if nothing at all lost by the vendors if the code was freely shared (after all, the systems this software ran on typically were leased with lucrative service agreements and so on).
The 1970s really WERE the decade of decline for open source. When minis like the PDP series and UNIX systems emerged to become a significant part of the market most systems were then bought outright and these smaller systems didn't always require vendor service agreements, so closed software started becoming more important as a revenue stream.
Then, in 1975 the Altair 8800 came out and, for the most part, revolutionised computing by making individual ownership and control of computing resources feasible. I think, however, that it very nearly killed open source software. Thankfully it lived on in the PC space in the form of magazine type-in programs and the like.
Perhaps it was the near-demise of a significant open source community that gave Free software advocates the resolve to succeed. The near-extinction of open source tools in UNIX systems certainly was the prime motivator in the creation of the GNU license (to keep open source code from being co-opted by closed software vendors).
I think the damage done by the rise of closed source in the 1970s and early 80s still shows to this day. For a time, open source sometimes meant you paid dearly monetarily for the privilege to see the code and modify it for your own purposes, and free software was "gratis", not truly Free (or "libre"). And since there is no "gratis" lunchyou paid in some way, whether it was timed lockouts, feature reductions or the like in shareware or through advertising revenue streams.
When I started using Linux in around 1996 and I first got paid to work on Linux systems used commercially in 1997, so I've seen first hand that though it was embraced by academia there was a fair bit of suspicion to overcome with management and nearly total ignorance from end users. I think today business has fully embraced Free software where it is sponsored and backed by "expert" corporations, such as is the case with Linux and Apache by IBM, Red Hat and Novell.
However, the stigma of closed freeware, particularly adware and spyware, remains a major challenge for Free software acceptance amongst naive Windows users. For example, when Firefox 1.0 was released just before 2005 I started advocating it as a safer alternative to IE and FF was viewed with great deal of doubt. I encountered a few people who steadfastly believed that FF was spyware. IE was part of Windows, and most of the offerings out there as an alternative weren't really browsers but "skins" over top the IE engine, and such skins, tool bars, "helper objects" and the like were most often pop-up factories or spyware. Furthermore, Free software apps were behind the curve in terms of creating signed distribution packages, so the Microsoft warning about an unsigned app just reinforced the FUD.
I think that the first decade since the Free software summit was the decade of commercial acceptance (and understanding that Free software isn't the opposite of comme
It's hard to find successful examples of higher-level applications that are open source. I think this is fundamental in the nature of open-source software and will never change: open-source software comes about when developers build things that they themselves want to use. Things that aren't used by developers won't be implemented in open-source fashion; there is no incentive for anyone to do this.
I partly agree with this, but there are also end-user applications where there is virtually no commercial software and a lot of good open source software. I'm not sure what the mechanism is that creates these open source niches, but (for some example) virtually all the software available for doctors and technicians to deal with DICOM images from CAT scans and the like seems to be open source... and some (like OsiriX) is tremendously good.
set foo [list {#valid list item} {bar} {baz}]; # { unmatched brace in a comment produces error here
There's no distinction in Tcl between {#valid list item} and {#a code block starting with a comment}.
Here's another example that might help you understand: These two "if" statements are the same, as far as Tcl is concerned. A Tcl block consists of Tcl statements separated by newlines and semicolons. Each tcl statement is a list, with the first element of the list being a command, and the rest being the arguments. So that "if" statement is the command "if" with two arguments, or it's a three element list containing "if", {[lindex $separators $token] >= 0} and $block.
Whether the block is the result of a variable substitution or not isn't relevant. So the parser operates on lists, one list at a time. If a block is used, it gets JIT interpreted as that code and the resulting code is stored in the object alongside the list format, so the next time around it doesn't get recompiled and execution stays in the bytecode interpreter... but that's a side effect of the implementation. As far as the language is concerned it's all just strings that may be lists, code, or plain text.
I really can't see any expanding role for FOSS in the long-term. Let's say, for example, that a FOSS project actually manages to become a market leader (doubtful, but for the sake of discussion). What can they possibly do from there? All of FOSS's good ideas come from other people, like Microsoft or Apple. Where do they go when there is nobody left to steal ideas from?
Also, groupthink is notoriously suppressive of creative solutions, so it's hard to imagine a FOSS project which could handle being a market leader. Probably the most glaring example is in the server space: at one time Linux held a clear (and even dominant) lead in web servers. Today? Not so much, with people ditching Linux left and right for far easier to manage, and far more secure, Windows servers. Linux just couldn't improve itself enough to hold on to it's lead- their only hope was that standing still long enough would help, and that a constant flow of anti-MS FUD would keep companies away. Obviously, it has not been a winning strategy.
So yeah, it's nice that there's a source of free software out there, but then again there always was. Now it's just more formalized, since the internet is a great distribution vehicle which didn't exist for the mainstream computer user. But despite that formalization, it still doesn't really have much to offer the average non-programmer computer user, who is going to expect a far more 'finished' product than the majority of FOSS can offer. But hey, I loves me some 7zip, so that's something at least.
A lot of end-users use OSS without realizing it. A prime example would be Mac OS X and iPhone OS. They both are built around Mach, BSD, and the host of other OSS projects that is called Darwin. They run Safari (built on KHTML), PHP (and other scripting languages), MySQL (and other databases), and so on.
For that matter, I cannot imagine how many end-users use MySQL every day in all kinds of different ways. I've never looked into what runs /. but I imagine that the moment I click this submit button, I will be "using" MySQL and Perl.
Sure, back in the old days (1970s, 1980s) there was a lot of freeware, crippleware and abandonware but that's true today too. Buy a new Windows PC and it is packed with 30 day trial stuff like anti-virus, winzip etc.
Back in the old days there was quite a bit of OSS too. But what was lacking was an effective mechanism to share the code and keep the changes together. Sure there were BBSs and uucp, but that's a far cry from having cvs servers etc over the internet.
Engineering is the art of compromise.
I've been coding off and on since the early 1980's. The idea that open source coincided with a big conference that supposedly kicked everything off is nuts. Conferences require people interested in the same topic to be profitable. Conferences therefore are thought followers, not leaders.
When I first got on the Internet some 19 years ago, there was already a healthy community of free and semi-free projects. There was a lot of code sharing, particularly in Unix sources on newsgroups.
I ported more than my fair share of things to run on A/UX, and I helped GNU for a short while with a port of stty back in the day when they wanted their own operating system (this is back circa 1990).
I was writing device drivers for Matrox cards back in 1996, and by then the Linux kernel was a heady and healthy open source project with thousands of contributors.
Honestly, either this guy was incredibly insular or his Internet connection was broken. O'Reilly didn't create open source any more than the Muppets did.
Andrew van der Stock
Comment removed based on user account deletion
Comment removed based on user account deletion
The reference you refer to uses the words "open source" in a sense closer to the sense of "open source military intelligence", which was a well-known usage at that time and still continues to be used. It means something that has value but wasn't taken from a secret source. In early February 1998, the phrase gained a new usage which was promoted by the Open Source Initiative.
I will not, however, take any credit for the usage of "Open Source" in a series of articles by one "Violet Blue". This seems to be closer to the military sense "not a secret" than the sense I have promoted :-)
Bruce Perens.
I work as a sysadmin in a pretty large mobile network operator. This is a business packed with obscure and higly proprietary systems (tho most protocols are free and open), yet I use OSS every single day. All of our machines, from the most mission critical call handling clusetrs to the most insignificant terminal pc, are in some way depending on OSS tools. Even an ancient Sinix/Reliand UNIX cluster from like... 1996 runs openssh and some other gnu apps.
I could not for my life imagine a world without OSS tools readily available for almost any platform I use. So, a big thank you to everyone who is making it possible!
When in danger, whewn in doubt! Run in circles, scream and shout!
Even Linux is more than 15 years old. But the term "open source" as applied to source code can be dated very precisely to the free software summit 10 years ago.
And the birth of that phrase, together with ESR's "Cathedral&Bazaar" and the Netscape source code release, did represent the beginning of the popularization of the idea of free software as a business opportunity. If you read the interviews, you will see that they almost all deal with exactly that aspect, namely the acceptance of free software in the commercial world.
Of course commercial free software didn't start with the summit, we had Russ Nelson's Crynwr, and Peter Deutch' Aladdin, and of course the poster child, Michael Tiemann and John Gilmore's Cygnus. Michael Tiemann said is best, he read the GNU Manifesto but saw a business plan. But the Cygnus web site also exemplified the problem. It started with centered around a friendly child drawing and information about the free software philosophy and how it could help your company, and ended up (at the time of the summit) as a boring business-like page stripped of all mentions of free software.
Free software was not a sales argument. The open source movement was about changing that, and it was wildly successful. It is cause for celebration.
That the FOSS movement hasn't yet learned the personal computing paradigm is a crucial insight.
Its first problem is that most FOSS developers are coding for themselves or their programmer/sysadmin peers, not end-users (and usually not even application developers, since a "Linux" desktop platform remains poorly defined / non-existent).
The second problem is going overboard with Unix culture, which leads FOSS advocates to stump for thin client / server architecture and centralized control often without even realizing. This results in an internal contradiction that is getting bigger each day. The FOSS crowd loves its LAMP platform, which tellingly doesn't even include 'Browser' in the acronym, so grudging is the acceptance of PC software! The community has no other platform that even approaches user-facing status, nor can it even articulate the need for such a thing, so the reflexes and the subtext are aimed at "replacing the desktop" with the web.
So naturally, FOSS continues to flounder with end-users not because the latter are 'dumb', but because they sense the lack of any substantial intent to serve them.
You MUST offer an OS that has both a well-defined API and a well-defined UI in order to have a ploatform that is attractive to end-users and app developers alike (and esp. the end-users who will start to explore coding). It must also offer a comprehensive default IDE that a child can use-- but there is no FOSS analog to Xcode + ADC or Visual Studio + MSDN.
So, dear reader, THAT is why so few people are ever inspired to follow their passion on a FOSS operating system to create the next Skype, the next Paint Shop Pro, the next Quicken or Sketchup -- the applications that sell the systems. You can have less than 5% of the installed userbase like Apple did and still attract considerable talent with the right structures (a platform) and tools.
Define a real platform - from the kernel on up to the IDE and file dialogs - and they will come!
In the second paragraph, elide "In Tcl, this is a single comment:", I missed that when I removed an example.
Free Software was the first campaign to clearly associate rights with source code. Publicly distributed source code existed even before then, and sometimes had rights that complied with the OSD. The OSD was written to fit existing licenses, primarily BSD, GPL, and Artistic. Although Richard had published an article about the four freedoms in GNUs Bulletin number 4, he didn't maintain any publication about them after that, online or elsewhere, until after the OSD existed. The references to "open source" before 1998 don't clearly associate any rights with the fact that source code is distributed.
So, what I am claiming credit for is getting "Open Source" to be one thing that very many people ask for. It is essentially the same thing that Richard was (and is) promoting, but he was unable to reach the masses nearly as well, simply because of his emphasis that the audience must place its a priori appreciation of freedom above all else. I agee with Richard, but it wasn't the best way to convert the unconverted - at least those who didn't think very similarly to Richard.
Bruce Perens.
And the birth of that phrase, together with ESR's "Cathedral&Bazaar" and the Netscape source code release, did represent the beginning of the popularization of the idea of free software as a business opportunity.
It recognised the fact that whatever-you-call-it software was already viable in business. All the examples you cite, and many more, already existed.
Free software was not a sales argument.
Free software, in the GNU sense, was also late on the scene. The people who were already writing and using free software, tangible software, all the other names freely redistributable source code had from the '70s and early '80s, we were pretty ticked off at the way the FSF co-opted the term. Because their kind of "free software" wasn't a sales argument. But the people who made free software, open source software, whatever you call it software... the people who made it work... we were already creating it and using it and sharing it long before any of the "big names" made their splashes by taking things that were already happening and convincing people they invented it.
It's great to have a name for the movement that already existed, and I'll celebrate that, but the movement wasn't created by the summit, any more than it was started by the GNU Manifesto.
The name is *useful*, but it's just a name, it's not the same as the thing it refers to.
The first published lament about non-free software that I'm aware of is a letter to Communications of the ACM from Professor Bernie Galler of the University of Michigan (an ol' prof of mine).
In it he gently flamed a couple programmers who were charging more than card/tape reproduction costs for a copy of a subroutine they had written in the course of their employment. He continued by predicting software-as-proprietary-product marketplace and its chilling effect on software development as a worst-case "if this goes on" scenario.
Don't recall the issue. But it's the one that also has Dijkstra's "GOTO statement considered harmful" letter, which puts it in 1968 or so.
Bantam Dominique roosters crow a four-note song. Once you've heard it as "Happy BIRTHday" you can't NOT hear it that way