Differences between software and other forms of creative work

I’m obviously just getting up to speed on the whole Creative Commons/free content debate. So this post may well be repeating stuff that’s been said elsewhere, and even rebutted elsewhere. But here goes.

Writing software, while still an inherently creative activity, is also different to other forms of creative work, and so generalising definitions of freedom from the software publishing context (Richard Stallman’s Free Software Definitions) is not automatically correct. While I respect the urge to generalise principles of freedom across all forms of human endeavour, and indeed this kind of generalisation is a direct product of the general nature of human intelligence, the fact is one of the major ways that harm is done by humans is by misapplication of principles which applied in one context to another where they do not.

The key differences that I see arise from two things: historical context and technical nature.

Historical context

(Before I begin: yes, I know that “it’s been this way” is not the same as “it should continue this way”. But drastic change is understandably met with wariness by people who have a reasonable degree of humility about their own ability to predict the future. I’m identifying obstacles to applying Stallman’s freedoms to artistic content, not necessarily endorsing them.)

Copyright has substantially shifted the balance of the monetisation of fiction, visual art, and music away from being solely about production of the saleable copy (the copy of the book, the copy of the artwork, the performance of the tune) to the creation of the new work which is then reproduced. Prior to copyright, an author of a major work was entitled to no share of the profit her work afforded others, and might well starve while her creation was selling in the thousands. (So it’s worth remembering, even while trying to fix the problems with copyright, that it was originally introduced to remedy a substantial injustice, even if that injustice mattered more than it should have because of the grosser injustices of the system and society within which the artists were creating.)

The only way an artist would be rewarded for the composition of a new work was if it was commissioned by someone rich – a business model which somewhat rewards withholding creative efforts, and certainly reduces the incentive to create on spec – which is part of why, relatively speaking, the majority of substantial works in the pre-industrial period constituted propaganda on behalf of religions, governments, and other powerful people and/or institutions. So not only was copyright created to correct an injustice, but one of the effects that it had was to shift power away from the ruling class to individual creators. This was reinforced by other social shifts taking place at the time, and as always the powerful have found ways to shift it back to their advantage, but the development of critical intellectual culture was tremendously enabled by the ability of creators to demand both a place at the table and a slice of the pie in the distribution of their work.

None of this mitigates the clear need to remedy some of the drastic abuses of copyright that we are seeing in the present day. But it does explain why some of those whom one might most expect to support such reform are coming out strongly in favour of traditional models of copyright. Any reform is far less likely to succeed without that historical understanding, and an acknowledgment that the philosophy of copyright has some roots in justice, even if you hold that some of the assumptions that pinned that concern to notions of property are now outdated.

Technical nature

Software is different to all other creative forms – including some which sit on a platform of software – because it is rarely an end in itself to the vast majority of its consumers.

People who run an operating system do not do so because they find the experience pleasant or enlightening. They find it enabling.

The same goes for the vast majority of other software. Even reader or media player software is a means to the end of the book, song, image or movie. The only software which springs to mind as being something which people run as an end in itself is gaming software. And I do not believe it to be a coincidence that there has yet to be a successfully-produced free software game that rivals commercially available games in the same way that Linux rivals – or surpasses – commercial OS software, or OpenOffice can substitute for Microsoft Office. (In fact, a disturbingly large number of the games listed on the Free Software Foundation’s site are ports or other adaptations of games which previously appeared elsewhere.)

A key result of this difference is that the majority of operating system and productivity software – software which enables the user to do something else – is subject to objective assessment. Either it does what it is required to do reliably and efficiently or it does not. There are degrees of efficiency, and even some room for subjectivity as to preferable coding technique and user interface, but nobody other than fellow creators would give any time to flawed-but-interesting code. By contrast, in traditional artforms and even (again, not coincidentally) in games software, successful execution (comprehensible writing in fiction, functional games software, etc) is only the first step.

Yes, there are electronic games out there which come with editing tools that allow the end user to create and share levels or modules; I doubt many people, including the developers who publish them as “toolsets”, would regard these programs as games in themselves. Similarly, there have been Choose Your Own Adventure books which allow users to assemble their own narrative by selecting from the material provided, though that material is fixed. The closest comparison to most software would be the ruleset for a roleplaying game, though even there there is a range of balances between fixed content and open procedure, from the fully generic like GURPS and Mythic, to genre-based games like D&D and Mutants & Masterminds, to games designed to simulate specific fictional settings like Call of Cthulhu and A Song of Ice & Fire Roleplaying, to games that really only allow for a single story to be told through to a limited range of conclusions (the most extreme of which might be something like the quasi-RPG Werewolf).

Not coincidentally, these sets of procedures are likely to share another distinctive quality of software: multiple iterations. Very few traditional creative works will be substantially revised once released to the public, though some reference material might (and, not coincidentally, it is precisely this sort of reference material which lends itself best to being shared and open-sourced in similar ways to software, as wikis everywhere prove). There may be sequels, whose names may even include similar numeric coding to software’s version notations; but these are not only additive but may even require the first work to make sense, whereas version 2.0 of an application will generally replace 1.0 wholesale.

This, I think, points to the key difference which needs to be reconciled with Stallman’s freedoms. A traditional creative work is intended not as a process or a tool to be used by others, which makes personalisation and adaptation highly desirable, but as a definitive statement or expression which is offered as-is for others to react to on its own terms, which makes modification by others undesirable, as it is likely to lead to confusion and/or misrepresentation.

A person creating a tool is less likely to be held responsible for its effect on other people, though depending on the nature of the tool (i.e. the things it enables, the side-effects of its design etc) this may be appropriate – the leather-worker in Se7en is an example of this line being crossed. Tools generally act on things, whether concrete physical objects (even the human body) or abstract entities like words, and do so under human direction. By contrast, the author of a traditional creative work, whether fiction or non-fiction, is held responsible for its effect on other people, because such works (as opposed to their media) act directly on consciousness, not on things. They inform, educate, entertain, illuminate, advise – and/or bore, deceive, confuse, manipulate, baffle – and all those verbs have people as their objects.

In other words, publishing a creative work is an act of communication between people, whereas software is a tool used by people (possibly to facilitate communication, but VOIP is not the same as conversation) and need not be interpersonal in any way. The sharing of speech is therefore necessarily different from the sharing of software. Attribution, modification, publication and sharing take on different meanings and have different effects when applied to speech as opposed to software; being exactly correct about what was said how and when by whom matters more for speech than it does for code, where the concern is more simply whether something works or not. For this reason, certain freedoms in relation to traditional creative acts carry with them responsibilities that do not apply to the same extent in code. And as always, personal freedoms must sometimes be protected by restricting the freedom to behave in ways that reduce freedom overall.

Whether copyright is the best implementation of such protection is highly debatable, but at least both copyright and Creative Commons recognise those differences.