\documentclass[12pt,a4paper]{book}
\newcommand{\distversion}{1.0.3}
\newcommand{\distdate}{10 May 2004}
\newcommand{\licence}{
This document is part of the AFFS Grant Project.
Copyright \copyright\ 2004 Marc Eberhard, Association for Free Software (AFFS),
c/o Turo Technology LLP, 79 Sir Lewis Street, King's Lynn, GB-PE30 2AL.
The copyright holder grants permission without fee and in perpetuity for all
acts restricted by the copyright of this work provided that the above
copyright notice and this notice are included. This work is offered on an
``as is'' basis without warranties and with all liabilities excluded, except
as required by applicable law.
}
\usepackage[latin1]{inputenc}
\usepackage[pdftex]{color}
\usepackage[pdftex]{graphicx}
\usepackage[pdftex]{hyperref}
\parskip 0.5\baselineskip plus 0.5\baselineskip
\parindent 0mm
\hypersetup{
  pdftitle={The AFFS Grant Project},
  pdfsubject={The AFFS Grant Project},
  pdfauthor={Marc Eberhard <marc@affs.org.uk>},
  pdfkeywords={AFFS grants},
  colorlinks=true,
  linkcolor=blue
}
\begin{document}
\title{The AFFS Grant Project}
\author{Marc Eberhard\\[5mm]
 {\small Association for Free Software (AFFS)}\\
 {\small c/o Turo Technology LLP}\\
 {\small 79 Sir Lewis Street}\\
 {\small King's Lynn PE30 2AL}\\
 {\small UK}}
\date{\vfill\ \\[20mm]
  {\small This document refers to the AFFS Grant Project, \\
  version \distversion, released \distdate.\\
  The latest version is available from
  the project home page \url{http://www.affs.org.uk/~marc/grants/}}}
\maketitle
\clearpage
~
\vfill
\licence
Typeset with \LaTeX\ on a Debian GNU/Linux system using only free software.
\clearpage
\tableofcontents
\cleardoublepage
\chapter*{Preamble}
\addcontentsline{toc}{chapter}{Preamble}
\markboth{PREAMBLE}{}
When Jason Clifford first mentioned his idea of using an ISP to fund free
software projects, I was immediately enthusiastic about it. To ensure a fair
distribution of these funds, he suggest to pay any surplus from this ISP to
AFFS. Without thinking too much, I agreed, that AFFS would be happy to do
this. We didn't know in the beginning if the idea would work at all and if
the income from such an ISP would ever offset the initial investment made by
Jason.
Quite some time has passed since then and the
\href{http://www.ukfsn.org.uk}{UK Free Software Network (UKFSN)} 
turned out to be a real success story. In October
2003 Jason wrote a first cheque over £500 to AFFS. This took us all a
bit by surprise at the time and since then we are trying to solve the
question on how to distribute the funds generated. We obviously want to do
this in a way which is considered fair and open by at least a majority of
AFFS's members.
The simplest model would have been a jury which would decide how to spend
the money. However, this approach does usually not result in decisions
which are considered to be objective and free of personal bias. For small
amounts of money this might nevertheless be a viable alternative, but we
expect to grow the income continuously over the years. So it is desirable to
put a system in place, which will scale well to large amounts of money right
from the beginning.
The goal is to distribute funds purely based on objective criteria without a
personal component. We don't want this to turn into a popularity contest or
something similar. This way we stand the highest chance of a fair process
with equal opportunities to all applicants. We don't want applicants to be
the target of flame wars on mailing lists or similar. This implies that we
do need to create an environment in which we encourage applicants to submit
brilliant ideas without fear of being publically attacked. Further, it
should give members the opportunity to submit supportive statements for an
application if they so wish.
Projects ought to be goal directed with clear targets, milestones, and
deliverables. Aims need to be essentially specific, measurable, attainable,
reasonable and timed. At the end of the project a formal report must be
submitted. Only then can AFFS be held accountable and prove that the money
was spent wisely. It also helps to assess the efficiency of different types
of projects to further AFFS's own aims and objectives.
It is clear that the decision making will never be perfect: you may have
the ideal set of criteria, but quantifying the value of a project against
criteria is subjective. It's likely that any two groups are going to come to
reasonably similar decisions a lot of the time, though. However, like the
offside rule, it's the boundary cases which will cause arguments. Each set
of people is going to be subjective in different ways.
As most of you know, my background as a researcher has brought me in contact
with several UK and international organisations funding research. Instead of
reinventing the wheel and repeating all the mistakes others have already
made for us, we should rather copy existing best practice and adapt it to
our particular needs. The approach taken here is such a mixture, which I
believe to be a good starting point for grants to further the development
and adoption of free software here in the UK.
The whole selection process is a tradeoff and should be as open and
transparent as possible, both to give the initiators of proposals enough
feedback and to allow third parties to assess the fairness of the allocation
process as a whole. However, making it entirely open and publishing all
proposals when they are submitted is not the way to go. I wouldn't like to
see the initial stages of the process to be public. None of the funding
bodies in the UK does that for several very good reasons.
The first reason is that it is obviously embarrassing to apply for something
and end up being turned down in public. Nobody wants to be in that
situation, thus it would act as a strong deterrent to many applicants.
Second, there is always the fear that someone might steal the idea. And
third, if the identity of the applicant is known to the people judging the
value of a proposal, it is not clear if their recommendation would be purely
based on a set of objective criteria or personal attitude towards the
applicant. I thus propose a double blind scheme as it has been used
successfully by many funding bodies including the
\href{http://www.epsrc.ac.uk/}{Engineering and Physical Sciences Research
Council (EPSRC)}. It hides both ways the identity of the applicant from
referees judging the proposal as well as the identity of referees to the
applicant.
This way, nobody knows who has been turned down or has unsuccessfully
applied. I believe this is a strong encouragement to apply. You don't loose
anything if you don't get the application through. On the other hand any
personal component is eliminated as the identities of the different parties
involved are unknown to each other. 
So how does it work in detail? The application is forwarded to referees, who
fill in a questionnaire about the proposal. These assessments are then
forwarded to the applicant to comment on. Then a board meets at a published
date and establishes a ranking order for the applications received based on
the reviews and comments. The treasurer then announces funding for projects
starting from the top of the list and moving downwards as long as there is
money available for the particular funding round. At this stage the
successful projects are announced publically.
Apart from AFFS offering grants in such a way, we should at the same time
try to attract funds from elsewhere to increase the amount of money
available for free software projects. That way we might both be able to fund
projects from money donated to us or help projects to get bids through with
other agencies or organisations. A project in need of help could get free
advice and support from us about which route might be the most effective
one to win support.
Apart from pure software development there are many other interesting bits
and pieces of information, which would be useful for existing free software
packages and which would greatly enhance their attractiveness for private
users and enterprises. Thus the money should be spent not only to get
involved in the creation of or adaption of free software to the UK, but also
for UK specific data needed for these. That allows us to contribute in a
well targeted manner and on whatever front it appears to be most fruitful
or necessary.
Grants are small at the moment, but will hopefully evolve into larger sums
later. Even though it is not worth do this at the moment, it might later be
a good idea to split the grant handling off into another organisation, which
can be a charity. This would give us all the advantages in terms of tax
savings and gift aid. Small funding requests should always go to the
committee of AFFS directly to be decided at their discretion, like quick
``expenses'' type payments for conferences and similar.
The remainder of this proposal details the points made here and describes
the whole process in detail in a way it can be implemented. I like to
stress that I consider nothing here as written in stone and I expect that
we will have to make amendments as time goes by. However, I strongly
believe that this is a good starting point, which should bring us onto the
right track and in the right direction to fulfil the promise we made when
we accepted the donations from UKFSN and --- by now --- others with
hopefully many more to come!
\chapter{About the Association for Free Software}
%Include text from pages/about.xhtml
The Association for Free Software (AFFS) is a membership organisation which
promotes free software to users, business and government. Our aims work
towards continuing free software for the UK and increasing public
understanding of free software.
The term refers to four kinds of freedom for the users of a program:
\begin{itemize}
 \item The freedom to run the program, for any purpose;
 \item The freedom to study how the program works, and adapt it to your needs
       (access to the source code is a precondition for this);
 \item The freedom to redistribute copies so you can help your neighbour;
 \item The freedom to improve the program, and release your improvements to
       the public, so that the whole community benefits (access to the source code
       is a precondition for this).
\end{itemize}
AFFS recognises a program as free software if its users have all of these
freedoms.
The constitution of the Association for Free Software sets out five aims,
\begin{enumerate}
 \item to promote and advance the knowledge, development, use and application
       of free software pioneered by the Free Software Foundation and
       others;
 \item to facilitate the exchange of information and views on the use and
       development of free software;
 \item to inform upon the subject of free software;
 \item to encourage internationalisation and localisation of free software;
 \item to ensure the continued legal existence of free software within the
       United Kingdom.
\end{enumerate}
The Association for free software is an unincorporated association at
pre\-sent. It is controlled by its members and only acts within the terms of
its constitution. The day-to-day running is coordinated by the executive
officers of the AFFS committee, who are elected by the members.
\chapter{Free software grants}
AFFS is able to offer direct funding for free software projects. This
includes the production of source code as well as other relevant tasks, such
as the preparation or creation of data for existing free software
applications or activities furthering the adoption of free software in the
UK. In general, any activities which further the aims of AFFS are eligible
for funding. Collaborations of different groups are encouraged and possible.
They can submit one application jointly.
As with every other organisation, AFFS's resources are unfortunately limited
and it will never be able to fund all suitable projects proposed. Thus
priority is given to projects that have the potential to make a real
difference for the adaption of free software. Which kind of projects could
this be?
Let's assume that 95\% of personal computers in the UK would use the same
sound card and this sound card wasn't supported by the Linux kernel at
present. Lack of support for this particular piece of hardware could be
considered as a ``show stopper'' for the adoption of free software in the UK
and thus the development of such a driver would make a real difference.
Another example would be the preparation of relevant tax (PAYE) and national
insurance tables in a form suitable for use in free software accountancy
applications. This could include paying an accredited high profile
accountancy company to examine and approve these tables and modules. Being
able to advertise free software as being certified in the same strict sense
as proprietary counterparts would surely be huge step forward.
Last but not least, a project that would proposed the adaption or
development of missing software components to fulfil the specifications of
the national curriculum for schools, colleges or universities would most
certainly make it easier to promote such free software solutions to
education bodies. With learners being exposed not only to proprietary
software, the whole software landscape could be more balanced in a few years
time.
These examples all have a few very important things in common. They have
clearly defined objectives, targets and outcomes of the projects against
which the success of the project can be measured. The main purpose of the
projects is not just the technical advancement of free software, but also
the further adoption of it as a result. They make it easier to ``sell'' free
software solutions to existing and new users. 
Thus they not only further one single aim of AFFS at a time, but a
combination of several simultaneously. That's what is meant with the phrase
``making a real difference''. Projects are prioritised in terms of the
expected impact they make for further adoption of free software solutions in
areas with low exposure at the moment, with a particular emphasis on the UK.
There is no limit on the total amount an application can bid for. But it is
obviously clear that AFFS will not be able to fund projects beyond its
financial capabilities or award the entire money for one funding round to a
single project. As a rule of thumb, it is very unlikely that any particular
project would be awarded more than 25\% of the total amount available for
one round. The amount of money available for a particular round of funding
will be made known along with the announcement of a prioritisation panel
meeting to give applicants an idea of realistic amounts they can bid for.
AFFS might issue specific calls for outline proposals from time to time, if
it decides that projects around a certain topic are highly desirable. Such
calls will include further details as to which proposals are sought exactly.
Specific calls will always be in addition to the normal funding process.
AFFS will further make the total amount of money available for each specific
call known in advance.
Statistics of completed and current projects are published as well as the
total number of applications, the success rate, average amount bid for and
average amount granted to name a few. This should allow potential applicants
to assess their chances for a successful application and to provide them
with further guidance in relation to realistic amounts and resources to bid
for.
AFFS is held accountable for the money awarded and must be able to provide
details of how effective and efficient it was spent. It will publish a
summary for each successful application detailing its scope, milestones and
targets. Depending on the project, interim reports might be requested and
published, but a final report after completion of the project will be
required and published in any case. Further applications from the same
project will not be considered until a satisfying final report has been
received.
AFFS would greatly appreciate that its support in form of a grant is properly
acknowledged in the documentation and other relevant documents of the
project including the grant reference number. This also applies to
presentations of the project to third parties or in public and other methods
of dissemination of information about the project.
\chapter{The application process}
To successfully submit a grant application to AFFS it is important to follow
the procedure explained in this chapter. AFFS reserves the right to reject
any application if these procedures are not followed properly. The main
objectives of this process are to ensure a fair and reliable way to
determine the projects to receive funding from AFFS.
The process --- including all forms --- is run electronically through the
AFFS website to the maximum possible degree. This is to ensure proper
logging of all stages of an application and to provide a record both
for the applicant and for AFFS members auditing the process. It allows
automatic statistic collection and keeps the applicant informed of the
application status.
\section{Outline proposal}
The first step in applying for a grant is to submit a completed outline
proposal form. The form is intentionally kept very short, so that it won't
take long to complete. Outline proposal forms can be submitted at any time.
However, it should be taken into account that there will be a considerable
amount of time needed for the whole process to complete from this stage. If
an outline proposal is submitted too close to a prioritisation panel
meeting, there might not be enough time to submit a full proposal and to go
through the review process before the panel meeting. In this case the
proposal will not be decided upon before the next panel meeting after the
current one.
The main purpose of the outline proposal is to provide informal feedback to
the applicant about the project idea and the resources requested in
comparison to the available funds. Members of the prioritisation panel will
contact the applicant and advise of potential problems with the project or
if it fails to meet any formal criteria set out in this document. It is
hoped that it will be possible to provide applicants with a rough
idea of the chances for the project to actually receive funding.
The applicant can then decide if it makes sense to submit a full proposal
based on the advice received from members of the prioritisation panel. AFFS
will not turn applicants down at this stage for other than formal reasons.
It is entirely the decision of the applicant to continue with the
application at this stage or to withdraw it.
\section{Full proposals}
For the next stage of the application process a full proposal form needs to
be submitted. The proposal must list all resources needed for the completion
of the project in detail. This includes costs for staffing, equipment and
other expenses. Should a grant be awarded, it will be limited to this total
amount in the project proposal. No corrections to the total amount will be
made for inflation or other reasons.
The proposal must further outline the work to be undertaken, milestones and
estimated completion dates. It must also provide criteria to assess the
success or failure of a project. A brief summary of the project needs to be
included. If the application is successful, this summary will be published
on the AFFS website. It is the responsibility of the applicant to ensure
that no confidential information is included in this summary.
No letters of support from third parties are required for a grant
application. However, such letters will be taken into account and will be
considered by the prioritisation panel along with the application itself.
\section{Peer reviews}
Once a full proposal has been received, it will be sent to referees for an
assessment. If the application includes details of up to three referees,
these will be contacted. In addition, AFFS will ask up to three members of
AFFS or external referees to assess the application. 
It is important to note that the identity of the applicant will not be
communicated to the referees. These will only see the body of the proposal,
but no information about the identity of the applicant. This is to ensure
that the assessment is purely based on the project idea and not on personal
preferences towards or against any individual involved in the application.
Referees are obliged to treat any information given in the proposal
as strictly confidential.
As soon as the completed questionnaires are received from the referees, they
will be send to the applicant for comments. The identity of the referees
will not be communicated to the applicant. They remain anonymous. The
applicant can choose to comment on individual reviews, on the reviews in
general or not at all. 
Any additional information provided by the applicant will be forwarded with
the application and taken into account in the next step of the process. As
soon as a reply is received from the applicant, the proposal will be
included in the next meeting of the prioritisation panel.
\section{Prioritisation panel meetings}
AFFS will from time to time establish a group of referees --- the
prioritisation panel --- to decide on the order of importance for funding
projects based on the reviews provided by the referees, the comments from
the applicant and any letters of support provided with the application. The
dates of these meetings will be publically announced well in advance. The
frequency of these boards depends on the actual funds available for
distribution.
Should the majority of the referees suggest to reject a project, the panel
will consider this option and if it decides to follow the advice from the
referees, it will communicate this to the applicant and exclude the proposal
from the ranking process for this funding round. Depending on the advice
from the referees, the panel might ask the applicant to submit a modified
version of the proposal, which addresses the issues raised in the comments
from the referees.
All panel members must be AFFS members and will be elected according to the
same rules as the executive committee. Elections can be held at the AGM or
by postal ballot depending on the circumstances and should normally be held
alongside elections for the executive committee. However, should need arise,
the executive committee can call for elections of the prioritisation panel
members at any time. Panel members must not submit applications to the AFFS
free software grant scheme themselves.
The decision of the board is final. There is no possibility to appeal
against the decision of the board. Panel meetings and decisions must be
minuted and archived. To preserve the confidentiality of the process and the
persons involved, these minutes will not be made publically available. In
case of a severe dispute, the executive committee can decide to appoint a
commission to investigate the matter and this commission will have full
access to all information held.
\section{Official announcement of grants awarded}
After a prioritisation panel meeting has established the order of funding
for projects, the treasurer of AFFS will award grants starting from the top
of the list and moving further down until he runs out of money for this
particular funding round. All applicants will be informed, if their project
is going to be funded or not, but no information will be provided on the
ranking order or the place of any individual project in the list.
The whole application process is considered to be strictly confidential up
to the point a grant is awarded. At that stage, the summary provided in the
successful applications and the total amount awarded for each grant are
published on the AFFS website and projects are officially announced by the
treasurer. The money will be made available as soon as possible to allow the
projects to commence. Should a project fail to commence within six months
of the grant being awarded, it will be declared void and allocated funds for
this project will be made available again for future funding rounds.
\section{Interim and final reports}
A full report of the work actually undertaken and the achieved targets is
required after the completion of the project. Depending on the scope and
size of the project, interim reports might be requested when the grant is
awarded. These reports will be published on the AFFS website and it is again
the responsibility of the grant holder to ensure that no sensitive or
confidential information is contained in these reports.
This way it should be possible for everyone to verify how the money was
spent and if it was a successful project. Although the details of the
selection process are for good reasons confidential, the actual projects
funded and the amount of money spent on them is all published. The final
report shows what the project has achieved in absolute terms and relative to
its original specifications. Thus AFFS gives full account how the money was
spent.
Failure to submit an acceptable final report for a project will result in
no further proposals from the same applicant or project group being
considered by the prioritisation panel until these are received.
\section{Resubmission of proposals}
Unfunded or rejected proposals can choose to be included again in the next
round, withdrawn or resubmitted. In the last case a new peer review process
is initiated. In any case the board is obliged to not take this fact into
account in any way when ranking proposals. Past rejections of an application
does not influence future decisions of the board concerning this
application. It should be noted that selection or rejection of a project
proposal does not constitute a statement of the quality or value of the
project by AFFS in any way.
\cleardoublepage
\chapter*{Outline proposal form}
\addcontentsline{toc}{chapter}{Outline proposal form}
\markboth{OUTLINE PROPOSAL FORM}{}
This forms should be implemented on the AFFS website for electronic
submission.
\begin{enumerate}
 \item Date of application.
 \item Applicant's details (joint applicants should be listed) consisting
  of full name, full email and postal address, telephone and fax number,
  date of birth.
 \item Title of proposed project (not to exceed 100 characters).
 \item Please give a summary of your proposed project describing its scope,
  aims and targets paying particular attention to its novelty and
  significance. Why is this project important? What is the current state of
  the project if it already exists? What is or will be your relation to the
  project? Please also justify the main resources. This summary should not
  exceed 1,000 words.
 \item Amount of money requested and brief description what the money would
  be spend on.
 \item Proposed starting and completion date.
 \item Please explain why you believe the AFFS to be best suited for the
  support of your proposal. This explanation should not exceed 200 words.
\end{enumerate}
\cleardoublepage
\chapter*{Full proposal form}
\addcontentsline{toc}{chapter}{Full proposal form}
\markboth{FULL PROPOSAL FORM}{}
This forms should be implemented on the AFFS website for electronic
submission.
\begin{enumerate}
 \item Date of application.
 \item Applicant's details (joint applicants should be listed) consisting
  of full name, full email and postal address, telephone and fax number,
  date of birth.
 \item Title of proposed project (not to exceed 100 characters).
 \item Summary of the project for publication on the AFFS website (not to
  exceed 1,000 words).
 \item Your personal track record. Please list past applications and their
  outcome. Also include any work done by you in the past or experience you
  have, which might be relevant to this application.  Explain how you want
  to achieve your goals and which methods you plan to use.  Which guarantee
  are given to deliver at the end? AFFS must be convinced that you can
  actually achieve the set goals with the funding provided and within the
  duration of the project.
 \item Will the funding be used for matched funding to attract further money?
  Will it trigger extra money to become available? Highest efficiency and
  effect is important.
 \item Describe the project in detail, including its aims, objectives,
  milestones, targets and time scales. Provide justification for all
  resources requested. This part will be the most important one to 
  determine the importance of funding for your project (not to exceed
  10,000 words). Make a good case of support for your application!
 \item Provide details of resources requested including exact costs
  with and without applicable taxes (VAT and others). For staffing costs
  list tax and NI contributions.
 \item Please name up to three referees to be contacted to assess this
  proposal including full contact details. We assume that you have
  contacted these referees and that they are happy to be contacted by
  AFFS.
\end{enumerate}
\cleardoublepage
\chapter*{Referee report form}
\addcontentsline{toc}{chapter}{Referee report form}
\markboth{REFEREE REPORT FORM}{}
This forms should be implemented on the AFFS website for electronic
submission.
\begin{enumerate}
 \item Application reference.
 \item Your name.
 \item Comment on the quality of the proposed project.
 \item Comment on the feasibility of the proposed methodology. How likely is
  this project to succeed in your opinion (in percent from 0\% to 100\%)?
 \item How high is your confidence level in assessing this (choose either
  low, medium, high)?
 \item How important is this project for the future development and adoption
   of free software in the UK and to further AFFS's aims in your opinion (in
   percent from 0\% to 100\%)?
 \item How high is your confidence level in assessing this (choose either
  low, medium, high)?
 \item Have appropriate project management plans been specified, including
  realistic time scales and milestones?
 \item How high is your confidence level in assessing this (choose either
  low, medium, high)?
 \item Comment on the overall management arrangements of this project in
  order to fulfil its objectives.
 \item How high is your confidence level in assessing this (choose either
  low, medium, high)?
 \item Do the project objectives justify the requested resources?
 \item How high is your confidence level in assessing this (choose either
  low, medium, high)?
 \item How crucial is funding from AFFS for this project to achieve its
  goals in your opinion (in percent from 0\% to 100\%)?
 \item How high is your confidence level in assessing this (choose either
  low, medium, high)?
 \item Your conclusions. Please use this space to make any comments you may
  wish, relating to this proposal, not made elsewhere. If you believe this
  proposal should be rejected, please state so here and explain why. If you
  believe this project should be resubmitted in amended form, please state
  so here and give details of what you would like to see changed for a
  resubmission.
\end{enumerate}
\cleardoublepage
\chapter*{Acknowledgements}
\addcontentsline{toc}{chapter}{Acknowledgements}
\markboth{ACKNOWLEDGEMENTS}{}
The author wishes to express his greatest appreciation for the support from
other members of the Association for Free Software (AFFS). Special thanks go
to Jason Clifford for bringing up the idea in the first place and for
starting the
\href{http://www.ukfsn.org.uk}{UK Free Software Network (UKFSN)} 
as a method to generate income for this fund.
\cleardoublepage
\chapter*{Contact info}
\addcontentsline{toc}{chapter}{Contact info}
\markboth{CONTACT INFO}{}
Please report any problems with this document to 
\href{mailto:marc@affs.org.uk}{marc@affs.org.uk}. Comments
and suggestions are of course also very welcome!
See \url{http://www.affs.org.uk/~marc/grants/} for the latest info.
You may also contact the author at:
\hbox{\vbox to 4cm{\vfill\hbox to 3cm{%
 \hfill\includegraphics[scale=0.15]{marc.jpg}\hfill}\vfill}
\vbox to 4cm{\vfill
\begin{tabular}{l}
Dr Marc Eberhard\\
 Association for Free Software (AFFS)\\
c/o Turo Technology LLP\\
79 Sir Lewis Street\\
King's Lynn\\
PE30 2AL\\
UK
\end{tabular}
\vfill}}
\cleardoublepage
\end{document}