Open Source
Content
Camford
info vol.2 no.1 february.2000
the journal of policy, regulation and strategy for telecommunications information and media
005
© 2000 Camford Publishing Ltd 1463-6697/00/010005-03
inform...
and software as a political statement
Freeware or open source software, even called brown bag software, seems to have no value. Because why would anybody give away anything that does have value? In fact the community in the largest sense, of the economy and the society that depends on it, have benefited enormously and tangibly from such open development software: the internet and the World Wide Web are just some of its children. Open source software heralds a new kind of economic cooperation, rather than combat. Moreover its donation is a political statement, affecting billions of people. Its growing power now makes the mightiest of software publishers take it seriously.
Software has attained incredible economic power. So much so that we now need ways to break out of the current hold that its creators have over us. But how do we escape its hold? Over the past few years free software has been developing at an accelerating rate. It represents a new and exciting path for software development, challenging the autonomy held by certain major software publishing companies. Is free software, therefore, set to become a key feature in the future of a network economy, or will its advance result in an overall decrease in its quality? Are we, indeed, on the verge of a new age in software? Much will depend upon people’s readiness to exchange their trusted, yet expensive, brand names for a product which, at present, is considerably less polished. But what exactly does free software mean? Unfortunately, there seem to be no `free lunches’ .
Simon Forge is a Principal at OSI (Email: [email protected]).
open away stuff, source: the economics of giving
Simon Forge
Thus, it is comforting to find that: The term `free software’ has nothing to do with price ± it is about freedom. So writes Richard Stallman, one of the contributors to a book charting the growth in the open source revolution.1 Today, the list of free software,`freeware’ or open source code is already impressive and constantly growing. It ranges from the basic origins of the internet itself, and the World Wide Web with the first browser (Mosaic), to operating systems such as Linux, GNU and Berkeley UNIX. It also encompasses the scripting language Perl, the web server Apache, as well as the GNU C and C++ compilers. But open source means far more than free, in the sense of `no price’. Fundamentally, it implies no owners. Anyone can improve on the original, not just a chosen set of academics and researchers. So its code is constantly cared for and improved upon by some of the most
006
inform
open source: the economics of giving away stuff, and software as a political statement
able talent in the world. This worries the world’s largest software publishers considerably, rather understandably. W hat does such an apparently fringe, hacker-led, even nerd-dry subject hold which is of interest for those outside a circle sometimes viewed as consisting of introspective ageing hippies (be they young or old in years) who are also software freaks? The reply is that it is clearly already an essential for our future economy and prosperity, even our freedom. Because open source software is really about free speech. Software is speech, in the sense of creativity. Thus, freely sharing ideas in software is about communicating without hindrance. Open source software could almost be seen as a return to the political ideas of syndicalism and shared benefit through freedom from ownership. And the net is the key component. It brings the ultimate freedom of publishing globally, at a keystroke. The net turns open source into a new paradigm for software creation and network sales ± a new economic mechanism for free global trade. However, open source’s significance for any future world economy is still misunderstood and certainly unappreciated, except within its own wired planet. Free software with free care and improvement is already a key, but
underestimated, driver for and feature of a network economy. It established the W WW and electronic comm erce through its free distribution. N o comm ercial or governm ent project could do this. From telecommunications analysis of revenues and margins, the value of a network economy appears to increase exponentially with the number of consumers attached, while the sunk cost in its infrastructure tends to go up linearly. The main enabler for consumption, packaged intellectual capital in the form of software utilities that enable and support connection, declines towards zero with open source code (see Figure 1). The conclusion ± for conventional sales, it becomes increasingly attractive to sell software over as large a network as possible. For open source software, the same network mechanism works differently ± it leverages its power as an economic factor in accelerating productivity gains for its users. Free software makes the net affordable not to tens of millions, but to billions. Its freeness wields massive economic power for a network-based society. The interesting thing is that freeware ± opensourcing ± will increase the quality of software as its availability increases. The network effect assures a constant quorum of quality, in the numbers of competent critics and workers who
F ig u re 1: Network economics v software economics.
Relative cost
Software cost/unit Network cost to build and operate
Network value
Volume of users
1. Richard Stallman, ‘The GNU operating system and the free software movement’, in Chris DiBona, Sam Ockman and Mark Stone, eds, Opensources: Voices From the Open Source Revolution, O’Reilly & Associates, Sebastopol, CA, 1999, USA.
open source: the economics of giving away stuff, and software as a political statement
inform 007
provide the enhancements, improve the robustness and fix the bugs. Furthermore, giving away software is highly attractive in commercial terms; it is taken up more widely and in the open source world it then improves, free. Free software can then act as the catalyst to sell other products. So, IBM supports Linux for the RS6000 server and its AS400 platform, previously renowned for being highly proprietary (the AS400 operating system even bundled in its own database). Giveaways can shape, then w in markets, as giving away a platform triggers sales for products based on that platform. It’s not a new idea ± Sun, with NFS (network file system , a U N IX utility) and Netscape, with its browser, have been doing it for years. Moreover, free software trains the next generation of designers ± freeware may often be all that their centres of learning can afford. But this is not new. Such was the rise of UNIX, as AT&T intended, by giving UNIX to universities. The University of California at B erkeley subsequently seeded an early academically free vehicle, Berkeley UNIX, and Sun was formed in its embers. Its technical directors, such as Bill Joy, created the Berkeley version and supported its distribution throughout academia. The free UNIX world (and UNIX’s place on the ARPA internet project of the 1970s) created its even more successful networking protocol, TCP/IP, the basis of internet connectivity with IP version 6 aimed at 21st century e-commerce.Today, the economic model has moved to the point where com mercial suppliers build complementary products for a successful freeware offering. In the case of Linux, for instance, many of its new device drivers come from the suppliers of peripherals themselves, to jump on Linux system sales for disks, printers, networking gear, etc. The image of hackerware from am ateurs has passed, although m ajor software publishers may often encourage this. The history of pure open sources begins with the GNU project at M IT in 1984 when Richard Stallman released GNU and its utilities into the public domain to create the first such `product’ . Stallman and GNU were seen as part of the establishment, yet now open source is the way forward: Stallman has even received the Grace Hopper award for achievement in the software industry. Open source represents a new path for software development. It is not like proprietary freeware directions that deg raded into proprietary products. For example, UNIX has 30 different marketed versions, each slowly failing as competing gets tougher, and all hiding differences in their various advances which other suppliers
2. Linus Torvalds, ‘The Linux edge’, ibid.
cannot use. In contrast, an open source operating system such as Linux is only available as one version, and in the most valuable form of any software, the source (and not just the rather indecipherable run-time binary) is available free. So, any computer supplier can see and understand any innovations. Open source thus creates its own unifying pressure to conform to a common reference point ± leveraged by the network affect of millions of users and thousands of improvers. W hile the major model of software success so far has been to sell proprietary binary only (one reason why enormous CDs are needed for software now ± the literal is not the codified version, and the code is itself increasingly inefficient), the new model is to deliver free software, but also to make money out of that action. That’s easier than it seems. Most people prefer to buy a branded water at reasonable cost, as it is a life-giving essential, than to trust the water company. The same goes for open source. But the brand may be Red Hat for Linux, rather than Evian or Badoit for water, with an operating system and support at a low $50 to $70, rather than the thousands for the alternatives. However, this may change. As the web changes from the Wild, Wild Web to more trusted forms (of `street’ , `mall’ , `home’ , `office’ , etc) so the raw open source software may gain preference. Today we only have the `street’ web so it can be pretty frightening. Opensources is a book which espouses free software in ter ms of the freedom from restrictions of ownership. It covers this `politics of software enabled by the network’ as a collection of papers from some of the major software authors in the open source world. It is worth looking into just for its insights into the software creation process, the quality of what goes on behind the curtain, showing up the reality of what we have to use everyday. Improvers of freeware tend to offer stark comments on what they see as the faults and gaps in competitive (commercial) offerings and in their own freeware creations.We see the design flaws in operating systems close up. But what is most riveting is the passion. Criticism of other freeware also abounds: listen to Linus Torvalds’ views on an editor called Emacs (nb a serious software detraction is size):2 The Emacs editor is horrible for instance, while Linux is larger than Emacs, at least Linux has the excuse that it needs to be. Criticism that counts ± to break our software chains and bring us quality, not monopoly, in our code.
info vol.2 no.1 february.2000
the journal of policy, regulation and strategy for telecommunications information and media
005
© 2000 Camford Publishing Ltd 1463-6697/00/010005-03
inform...
and software as a political statement
Freeware or open source software, even called brown bag software, seems to have no value. Because why would anybody give away anything that does have value? In fact the community in the largest sense, of the economy and the society that depends on it, have benefited enormously and tangibly from such open development software: the internet and the World Wide Web are just some of its children. Open source software heralds a new kind of economic cooperation, rather than combat. Moreover its donation is a political statement, affecting billions of people. Its growing power now makes the mightiest of software publishers take it seriously.
Software has attained incredible economic power. So much so that we now need ways to break out of the current hold that its creators have over us. But how do we escape its hold? Over the past few years free software has been developing at an accelerating rate. It represents a new and exciting path for software development, challenging the autonomy held by certain major software publishing companies. Is free software, therefore, set to become a key feature in the future of a network economy, or will its advance result in an overall decrease in its quality? Are we, indeed, on the verge of a new age in software? Much will depend upon people’s readiness to exchange their trusted, yet expensive, brand names for a product which, at present, is considerably less polished. But what exactly does free software mean? Unfortunately, there seem to be no `free lunches’ .
Simon Forge is a Principal at OSI (Email: [email protected]).
open away stuff, source: the economics of giving
Simon Forge
Thus, it is comforting to find that: The term `free software’ has nothing to do with price ± it is about freedom. So writes Richard Stallman, one of the contributors to a book charting the growth in the open source revolution.1 Today, the list of free software,`freeware’ or open source code is already impressive and constantly growing. It ranges from the basic origins of the internet itself, and the World Wide Web with the first browser (Mosaic), to operating systems such as Linux, GNU and Berkeley UNIX. It also encompasses the scripting language Perl, the web server Apache, as well as the GNU C and C++ compilers. But open source means far more than free, in the sense of `no price’. Fundamentally, it implies no owners. Anyone can improve on the original, not just a chosen set of academics and researchers. So its code is constantly cared for and improved upon by some of the most
006
inform
open source: the economics of giving away stuff, and software as a political statement
able talent in the world. This worries the world’s largest software publishers considerably, rather understandably. W hat does such an apparently fringe, hacker-led, even nerd-dry subject hold which is of interest for those outside a circle sometimes viewed as consisting of introspective ageing hippies (be they young or old in years) who are also software freaks? The reply is that it is clearly already an essential for our future economy and prosperity, even our freedom. Because open source software is really about free speech. Software is speech, in the sense of creativity. Thus, freely sharing ideas in software is about communicating without hindrance. Open source software could almost be seen as a return to the political ideas of syndicalism and shared benefit through freedom from ownership. And the net is the key component. It brings the ultimate freedom of publishing globally, at a keystroke. The net turns open source into a new paradigm for software creation and network sales ± a new economic mechanism for free global trade. However, open source’s significance for any future world economy is still misunderstood and certainly unappreciated, except within its own wired planet. Free software with free care and improvement is already a key, but
underestimated, driver for and feature of a network economy. It established the W WW and electronic comm erce through its free distribution. N o comm ercial or governm ent project could do this. From telecommunications analysis of revenues and margins, the value of a network economy appears to increase exponentially with the number of consumers attached, while the sunk cost in its infrastructure tends to go up linearly. The main enabler for consumption, packaged intellectual capital in the form of software utilities that enable and support connection, declines towards zero with open source code (see Figure 1). The conclusion ± for conventional sales, it becomes increasingly attractive to sell software over as large a network as possible. For open source software, the same network mechanism works differently ± it leverages its power as an economic factor in accelerating productivity gains for its users. Free software makes the net affordable not to tens of millions, but to billions. Its freeness wields massive economic power for a network-based society. The interesting thing is that freeware ± opensourcing ± will increase the quality of software as its availability increases. The network effect assures a constant quorum of quality, in the numbers of competent critics and workers who
F ig u re 1: Network economics v software economics.
Relative cost
Software cost/unit Network cost to build and operate
Network value
Volume of users
1. Richard Stallman, ‘The GNU operating system and the free software movement’, in Chris DiBona, Sam Ockman and Mark Stone, eds, Opensources: Voices From the Open Source Revolution, O’Reilly & Associates, Sebastopol, CA, 1999, USA.
open source: the economics of giving away stuff, and software as a political statement
inform 007
provide the enhancements, improve the robustness and fix the bugs. Furthermore, giving away software is highly attractive in commercial terms; it is taken up more widely and in the open source world it then improves, free. Free software can then act as the catalyst to sell other products. So, IBM supports Linux for the RS6000 server and its AS400 platform, previously renowned for being highly proprietary (the AS400 operating system even bundled in its own database). Giveaways can shape, then w in markets, as giving away a platform triggers sales for products based on that platform. It’s not a new idea ± Sun, with NFS (network file system , a U N IX utility) and Netscape, with its browser, have been doing it for years. Moreover, free software trains the next generation of designers ± freeware may often be all that their centres of learning can afford. But this is not new. Such was the rise of UNIX, as AT&T intended, by giving UNIX to universities. The University of California at B erkeley subsequently seeded an early academically free vehicle, Berkeley UNIX, and Sun was formed in its embers. Its technical directors, such as Bill Joy, created the Berkeley version and supported its distribution throughout academia. The free UNIX world (and UNIX’s place on the ARPA internet project of the 1970s) created its even more successful networking protocol, TCP/IP, the basis of internet connectivity with IP version 6 aimed at 21st century e-commerce.Today, the economic model has moved to the point where com mercial suppliers build complementary products for a successful freeware offering. In the case of Linux, for instance, many of its new device drivers come from the suppliers of peripherals themselves, to jump on Linux system sales for disks, printers, networking gear, etc. The image of hackerware from am ateurs has passed, although m ajor software publishers may often encourage this. The history of pure open sources begins with the GNU project at M IT in 1984 when Richard Stallman released GNU and its utilities into the public domain to create the first such `product’ . Stallman and GNU were seen as part of the establishment, yet now open source is the way forward: Stallman has even received the Grace Hopper award for achievement in the software industry. Open source represents a new path for software development. It is not like proprietary freeware directions that deg raded into proprietary products. For example, UNIX has 30 different marketed versions, each slowly failing as competing gets tougher, and all hiding differences in their various advances which other suppliers
2. Linus Torvalds, ‘The Linux edge’, ibid.
cannot use. In contrast, an open source operating system such as Linux is only available as one version, and in the most valuable form of any software, the source (and not just the rather indecipherable run-time binary) is available free. So, any computer supplier can see and understand any innovations. Open source thus creates its own unifying pressure to conform to a common reference point ± leveraged by the network affect of millions of users and thousands of improvers. W hile the major model of software success so far has been to sell proprietary binary only (one reason why enormous CDs are needed for software now ± the literal is not the codified version, and the code is itself increasingly inefficient), the new model is to deliver free software, but also to make money out of that action. That’s easier than it seems. Most people prefer to buy a branded water at reasonable cost, as it is a life-giving essential, than to trust the water company. The same goes for open source. But the brand may be Red Hat for Linux, rather than Evian or Badoit for water, with an operating system and support at a low $50 to $70, rather than the thousands for the alternatives. However, this may change. As the web changes from the Wild, Wild Web to more trusted forms (of `street’ , `mall’ , `home’ , `office’ , etc) so the raw open source software may gain preference. Today we only have the `street’ web so it can be pretty frightening. Opensources is a book which espouses free software in ter ms of the freedom from restrictions of ownership. It covers this `politics of software enabled by the network’ as a collection of papers from some of the major software authors in the open source world. It is worth looking into just for its insights into the software creation process, the quality of what goes on behind the curtain, showing up the reality of what we have to use everyday. Improvers of freeware tend to offer stark comments on what they see as the faults and gaps in competitive (commercial) offerings and in their own freeware creations.We see the design flaws in operating systems close up. But what is most riveting is the passion. Criticism of other freeware also abounds: listen to Linus Torvalds’ views on an editor called Emacs (nb a serious software detraction is size):2 The Emacs editor is horrible for instance, while Linux is larger than Emacs, at least Linux has the excuse that it needs to be. Criticism that counts ± to break our software chains and bring us quality, not monopoly, in our code.
