path: root/ideology/fscons-2010.mdwn

# Syndicated E-mail Service Against Software as a Service

Denna sida innehåller dels allmänna anteckningar om Friposts bidrag
till FSCONS 2010 genom Gustav Eek och Stefan Kangas, och dels
innehåller en slutgiltig artikeltext. Rekommendationen är att direkt
läsa artikeltexten.

This page contains general notes that has to do with the preparation
to Fripost's contribution to FSCONS 2010 through Gustav Eek and Stefan
Kangas. It also contains a final article text. The recommendation is
to directly read the article text.

[[!toc levels=2]]

# Anteckningar om föreningens bidrag till konferensen FSCONS 2010 

 * *Authors* – Gustav Eek and Stefan Kangas

 * *Contact email* – skangas@skangas.se

 * *Preferred timeslot* – 30 min

 * *Proposal title* – Olika förslag...

   * "[http://en.wiktionary.org/wiki/syndicate Syndicated] E-mail
     Service Against Software as a Service"
   * "Serve our self the server service"
   * "Servicing our own software as a service"
   * "Defeating SaaS by socializing email services"
   * "Collaborative email servers against SaaS"

## Proposal summary

It is becoming increasingly common for persons and organizations to
hire external services for e-mail handling, and even for the use of
spreadsheets and word processing. This phenomenon, known as *Software
as a Service* (SaaS), implies restrictions in the users freedom that
is not only tangential.

In our ambition to counteract this tendency, we would like to present
a project of constructing an independent and autonomous e-mail hosting
service, run by a syndicate, *The Free E-mail Association*, for the
associations members. The syndicate is driven by a simple framework
that guarantees that it is

 * of independent means,
 * democratically structured, 
 * built on trust between its members, and
 * ready to be far-reaching with regards to avoiding traffic logging and to protect the members privacy.

The service that we are hoping to create is a full featured e-mail
solution with high reliability and accessibility.

This project exemplifies that the creation of a freedom preserving
network service is possible, and it is also an example of what that
can look like.

The obvious question to raise is whether this service actually is not
just another SaaS? Obviously, all individual members can not at all
times reach absolute freedom and computational control. However, these
limitations must be weighed against that investments in a more
reliable technical infrastructure are possible only through
collaboration. Also, we believe that the association's democratic
structure and numerical and geographical boundaries are enough to
mitigate the limitations; the limited user participation is still
sufficient for the service to bee regarded as free.

We hope that this project can raise curiosity enough to tantalise
users away from private suppliers of e-mail and other services, to
either join our association or even better start more similar
projects.

## Other information

It is preferred from our point of view that the talk is scheduled in
an early stage of the conference. This since the hope is to raise
questions for further discussion during the conference.

# Antagen till konferensen

Den 24 juli blev föredraget "Syndicated E-mail Service Against
Software as a Service" antagen till temat "Infrastructure" på FSCONS
2010. Jeremiah Foster är koordinator för temat som har beskrivningen

> Wikipedia defines infrastructure as "the basic physical and
> organizational structures needed for the operation of a society."
> The theme focuses on the infrastructure needs for tomorrows future
> society, as well as the basic need for privacy and secure
> communication, together with monitoring, both as a threat and as a
> tool to manage our ever increasing networks.

Det är ett antal saker som vi från Jonas Öberg blivit ombedda att
utföra och återkomma till Jonas och Jeremiah angående.

 1. Granska den föreslagna sammanfattningen av föredraget och revidera
 det med det övergripande temat i åtanke och om något tillkommit i
 föredraget.
 2. Om vi har några begränsningar vad gäller schemaläggningen. 
 3. Frågan om när vi kommer till Göteborg och om vi ska vara med på de
 sociala tillställningarna.

Frågan är också om konferensen vill ha någon längre artikel för
föredraget. I så fall vore det bra att sätta igång med det nu,
snart. En idé skulle vara att översätta principförklaringen om den kan
skrivas på ett tillräckligt vetenskapligt sätt.

# Second article for the conference

This is the second proposal, which is longer and more adjusted to the
infrastructure theme description. The proposal follows declaration of
principals.

## Introduction

User freedom is the most important property of tomorrows
infrastructure. This property is necessary to safeguard the relative
freedom of speech provided by the Internet against increasingly
aggressive attacks by preying commercial and opportune state
interests.

In this abstract of a talk on FSCONS 2010 we first try to define
freedom in computer work. Then the problem of increased centralization
of the Internet is discussed and a number of ongoing threats to this
freedom are identified. We then present the *Free Email Association*,
what built up infrastructure we have, and our principles. Finally we
try to sketch what we think the future paths might look like.

## Background

The centralisation of influence and rectification of decision-making
is not unique for the Internet. This is written in a wider context of
general social criticism of economic and cultural globalisation and
the current forms of the globalisation of information flows. Many
decisions are made in multi-lateral arenas where the democratic
control is limited.^1 A decreasing number of transnational operators,
not only controls the means of production and the production of goods,
but also have great influence in markets demands.

The Internet is, or will soon be, the most important communication
medium in the majority of the industrialized world. The story of its
development from its conception as a highly distributed network
through the establishment of the free and open web, towards the
increasingly privatized web we see today is indeed saddening for those
who takes user freedom seriously. A few strong parties control major
segments of important infrastructure that millions of users depend on
every day. Those who control the technology and its infrastructure
also have power over its users.

## Computing, computer labour, and power over the infrastructure

The *Free Software Foundation* suggest definition of *free software*
consisting of four requirements: the "freedom to run, copy,
distribute, study, change and improve the software". (Free Software
Foundation "The Free Software Definition") As a complement to those we
define the more general *freedom with regard to own computer labour*
^2 as requiring that

 1. the work is performed exclusively with free software,
 2. the work is performed with computer hardware that entirely is at
 one's possession and control,
 3. the information worked with is information that one possess, and
 4. the result from the computer work also is at one's possession and
 control.

We here use "information" to denote data and documents that are the
object to computer work. In this context *computer labour* is all use
of computers, and *own computer labour* is computer work performed for
one's own part. ^3

## Internet and its servers

The *Internet* seen as an infrastructure scheme was constructed as a
distributed *peer-to-peer* non-hierarchical network of independent and
self-determined parts. ^4 Despite this immanent property, the Internet
today, seen from a regular user's point of view, is structured in a
hierarchical manner around a decreasing number of server clouds, which
are continuously growing in size and power. Historically the meaning
of *servers* was to gather and publish information provided from its
clients. However, in many applications today, publishing is not
performed as a separate process, and the clients are no longer always
independent. For example social networking sites, like Facebook, often
require their users to perform their work directly on the company's
servers. (Moglen 2010)

## Software as a service

A concept introduced in the spirit of centralisation is *Software as a
Service* (SaaS). Shortly SaaS is that users are invited to perform
their computer work on or through a network server on Internet or a
local network. The main purpose of SaaS is to separate *possession*
and *ownership* of software from its *usage*. This software is said to
be *licenced on demand*. (Turner 2003)

In this text *Software as a Service* is used in a more narrow sense in
accordance with Stallman (Stallman 2010), to mean one's own computer
work on hardware that the user do not control. Popular Internet
services that are examples of SaaS are Google Docs and Facebook, but
the concept is widely used. Computer work performed with this software
is non-free in a double sense; using SaaS also leads to *vendor
lock-in*. However, the complex of lock-in problems reach far outside
SaaS.

## The infrastructure of email

The email infrastructure is not an exception from the tendency towards
centralisation and rectification of Internet's services and usage. We
now see even large institutions being lured in by the economic
benefits offered by these large scale solutions.

Email communication through the Internet involves several computers
and servers, among those a *mail user agent*, a computer program
controlled by the sending person; several *mail transfer agents*,
Internet servers responsible for getting the mail though using the
SMTP protocol; *domain name system servers*, servers keeping track of
the addresses to all those servers; and finally another *mail user
agent*, used by the receiving person to read the emails. It is also
common to make use of extra inbox handling services like *IMAP access*
or *webmail*, which usually involve separate servers.

What about email and freedom? We here need to distinguish what part of
emailing that is one's own computer labour. Editing email definitely
is, along with all sort of contact management. The transfer process,
however, is not. Whether the email arrives is of course of great
concern to the sender, but there is generally no human activity
(directly) involved and thus no actual work performed. In principle
the same holds also for the process of receiving email. However, most
popular email services are not content with that. Stallman writes,
<blockquote>Some sites whose main service is publication and
communication extend it with *contact management*: keeping track of
people you have relationships with. Sending mail to those people for
you is not SaaS, but keeping track of your dealings with them, if
substantial, is SaaS. (Stallman 2010)</blockquote> And using SaaS is
not free computer labour. Furthermore, whether hiring a company for
handling one's email imply usage of SaaS or non-free computer labour
might not be the only matter of importance.

## Privacy and survelliance

With large clientele comes a lot of power. Google is currently not the
largest email service provider; both Windows Live Hotmail and Yahoo
Mail has more customers. (Brownlow 2010) We nonetheless believe it is
important to single Google out as a company, because of how
effectively they utilize privacy invading schemes that are integrated
between their services. Together with Google's e-mail service one also
get services that probably was not asked for: advertisement, semantic
analyses of email contents, and spying. (Moglen 2010) The data
resulting from Google's analysis and espionage is later used
indirectly in marketing campaigns with Googles customers or sold
directly to third party. Google is profiting on their email users with
the means of the users private information provided by themselves.

Google link that data to individuals' surf activities using Google's
search engine, Google accounts and cookies. Additionally, many
websites utilize Google's JavaScript APIs, web site statistics or
reCAPTCHA service, which indirectly exposes individuals to Google's
supervision.

## Organisation for a change

Technology has never been neutral. Behind technology changes and
innovations today lies commercial interests and social factors. On the
one hand, the ownership structure of technology companies and their
endeavour for profit ensures that usage of their services for work and
communication will never be free. On the other hand, only a fraction
of the everyday Internet users have knowledge and resources enough to
create free alternatives on his or hers own hand. The question is also
what we can reasonably expect from an average user in terms of
engagement in their privacy and freedom.

Our idea is to form an association and together take back a small but
important part of our Internet life, namely the email communication
infrastructure. We gather around several servers that receives and
stores the members' email. The purpose of the association is to render
it possible for individuals to bring their computer usage under their
own control. And to show that user freedom is possible, even if only
through active, prolonged and collective struggle.

In more detail, we start with five or six servers spread out in Sweden
and Europe connected through the Internet: at the very least two mail
exchange and one IMAP server, separate backup and log servers, and
hopefully a webmail server. In the beginning we will not have an SMTP
send mail service, and we will hire the DNS service from third
party. Though we will have tight economic boundaries, our focus on
reliability and stability forces us to try hard to maintain a
sufficiently high level of redundancy.

The email service provided through the Free Email Association is free
as in free speech, not as in free beer. This because of the democratic
structure and of non-profit and commonly owned organisations like the
Free Email Association. The free email association promises to

 1. work hard to receive email for the sake of its members in a reliable manner
 2. protect its member's integrity, that is to never
    * read or analyse its members emails, either manually or
      automatically, (possibly with exception for voluntary spam
      filter services and alike)
    * gather statistics about or analyse member's traffic through the association's servers, and
    * under no circumstances hand out information about members, their
      emailing and other activities, or any other information to third
      party, and
 3. provide for transparency regarding administration, economy and the
 decision-making process.

Transparency and continuous information about the work of the board is
of cause extra important, and the democratic regime requires constant
attention. A declaration of principles serve as a founding document
for the association, but still a high level of trust is laid on the
elected officers.

## On the long term

We recommend and encourage everyone to perform their computer labour
with free software at machines that are in their own possession. But
many solutions on Internet servers, like Google Docs and Facebook, are
quite usable and practical and adopted to a modern way to meet and
work computer-aided. The Free Software Community has a great challenge
in the creation of free and distributed alternatives, where free means
that their usage imply only free computing.

Despite this, we must not deceive ourselves into thinking that the
final and greatest challenge is about building physical infrastructure
or programming advanced distributed social networking solutions. The
real challenge is to get people engaged in the issue of the Internet's
power structures and in their own freedom and privacy.

No one in their right mind would put their blind trust in a government
that was dabbling in surveillance on a scale anywhere near what Google
is doing. But when it comes to Google, their marketing strategy has
been so successful that many people need no additional guarantees that
Google will behave &ndash; people are willing to take Google's word
for it.

The reasons behind the described changeover of the Internet are
structural. The structural tendency, as described, is that already
powerful operators get even more power when computer labour is
generally becoming less free. Structural problems need structural
change, but projects like the Free Email Association serve as an
alternative to structural change for those who are eager to establish
free alternatives. In the long-term, however, we realize that a larger
social change on a structural level is necessary. The best we can hope
for is to give a small contribution to bringing about that change.

## Footnotes

^1 Good, recent examples are the European Unions IPRED directive or
the ACTA agreement.

^2 Stallman uses *your own computing* to denote what we call *one's
own computer labour*. [2] We use the latter because of its broader
associations.

^3 Stallman points out that work performed as employee in some company
or in a cooperation project as Wikipedia is not one's own computer
work, but a part of that company's or project's work. In that case it
is not one's own freedom that is threatened, but the company's or
project's. (stallman 2010):

^4 Technically the parts of a network are *nodes*, which in the case
of Internet are servers, switches, and personal computers, and
*edges*, which are interconnecting wires.

## References

 * [1] Turner, M. et.al. (2003). "Turning Software into a Service",
   Computer vol. 36, IEEE Computer Society 2003.
 * [2] Stallman, R. M (2010). "Who does that server realy serve?",
   Boston review, only web version
   http://bostonreview.net/BR35.2/stallman.php. Revised version på
   http://www.gnu.org/philosophy/who-does-that-server-really-serve.html. Fetched
   2010-08-30.
 * [3] Moglen E. (2010). "Freedom in the Cloud", talk given to the New
   York chapter of ISOC February 2nd 2010. Video available at
   http://new.law.columbia.edu/isoc/eben_moglen_freedom_in_the_cloud.ogv,
   and transcription at
   http://www.softwarefreedom.org/events/2010/isoc-ny/FreedomInTheCloud-transcript.html. Checked
   2010-09-10.
 * [4] Free Software Foundation. "The Free Software Definition",
   http://www.gnu.org/philosophy/free-sw.html. Fetched 2010-08-30.
 * [5] Brownlow M. (2010). "Email and webmail statistics",
   http://www.email-marketing-reports.com/metrics/email-statistics.htm. Updated
   May 2010. First published April 2008. Fetched 2010-09-21.