The proposed concept is for there to be a charitable organisation dedicated to promoting the development of Open Source software for RISC OS. Being a charity would set it apart from any of the organisations which have gone before it, most notably because of the extent to which it would be required to be both transparent and accountable. In return for adopting these disciplines it would gain certain tax advantages and, it is hoped, the trust of those on whose support it would rely.
In order for an organisation to achieve charitable status the principal requirement is that its objectives must be exclusively charitable in nature. Software development is not a traditional type of charity work, but there is good reason to believe that it can qualify provided it is be shown to be for the benefit of the public as opposed to any particular group of individuals or companies.
Open Source software meets this criterion in a way that most other software does not. It also has the virtue of having a well understood and widely accepted definition — the Open Source Definition (OSD) — making it a natural place at which to draw the boundary of the organisation's activities.
(This is not to say that the charity would be opposed to proprietary software, or that it could not cooperate for mutual benefit with proprietary software developers. It would however be necessary for any software developed using charity funds to be released as Open Source.)
There would need to be a group of trustees who would be ultimately responsible for the actions of the organisation. Normal practice is for trustees to be unpaid, and it will simplify matters greatly if they are not eligible to receive any financial or material benefit from the organisation.
One of the major benefits of charitable status is the ability to recaim income tax on donations. This potentially includes membership fees, subject to some limitations which are discussed below.
If these limitations are found to be too restrictive then it would be possible to engage in more commercially-focussed activities through a wholly-owned but non-charitable subsidiary. (What would not be allowed is to take charitable donations and use them to fund such activities.)
The main activities of the organisation would be:
It is envisaged that the main type of financial incentive would take the form of a bounty attached to a particular task, available to be claimed by the first person or group to complete that task. This has the important advantage that there is little or no risk to the organisation's funds. Issues which would have to be addressed are:
These issues will be more tractable if the individual tasks are small and well-defined. It is proposed:
One way to avoid duplication would be to allow a provisional claim to be made when a task has only been partially completed. The developer in question could then be given first refusal on the bounty provided that development continued to an agreed timetable, but no money would change hands until the task was complete.
In principle bounties could be of any size, but they are likely to be most manageable in the range £100 to £1000.
A second possible type of incentive could be the granting of awards to recognise achievements which may not have been specifically commissioned by the organisation but which nevertheless advance its aims and objects.
In addition to the above, there may be cases where existing closed-source software can simply be bought and then licensed as Open Source.
There are numerous ways in which an organisation could provide practical support to RISC OS software developers, in the form of:
The first and most basic aim should be to ensure that nobody who is willing and able to develop Open Source RISC OS software is prevented from doing so by lack of equipment.
Information is tricky because, while it is very easy to start a new directory, information portal or file collection, it is much more difficult to populate it and keep it up to date. Where the organisation might have a role is to re-invigorate and perhaps consolidate some of the information sources which already exist.
One of the challenges we face is finding developers who have the software development skills that are needed. Such skills are not a fixed resource: they can be developed, and in some cases this will be a more efficient way to proceed than for the more experienced programmers to attempt to write everything themselves. This also has the potential to be a beneficial and rewarding experience for all concerned, and would do much to bolster the charitable credentials of the organisation.
Finally, one of the objectives of the organisation could be to bring together teams of developers with complementary skills so that software can be developed more efficiently and effectively. (The emphasis here would be not so much on actual programming, but rather the peripheral activities which could largely be done by non-programmers.)
The main source of funding would be donations in the form of membership fees. As an example of how this might be organised, there could be three levels of membership:
The direct benefits of membership would have to be largely symbolic in nature in order to obtain any tax benefit. They could perhaps include:
A secondary source of funding could be from sale of publications (probably beginning with the two titles written by myself and currently published by the RISC OS Packaging Project).
Fixed overheads will be avoided wherever possible. This will ensure that the business is viable whether or not fundraising targets are met. In particular, meetings will be conducted electronically to avoid travel costs. Unavoidable or highly desirable fixed overheads that have been identified so far include:
Points to be checked include whether there is a requirement for accounts to be professionally audited, or to retain any form of legal representation.
Most other overheads (such any costs related to benefits of membership) should broadly scale with the size of the organisation.
Graham Shaw (Issue 2, 29th April 2007)