[Freedombox-discuss] FreedomBox Summit

[Danny Haidar]

Hi everyone,

I want to open a discussion about the FreedomBox Foundation's upcoming
FreedomBox Summit. As we discussed on our progress call on October 13,
the FreedomBox Foundation is hosting a meeting in New York City on
Tuesday, November 13, 2018 with Sunil Mohan Adapa, James Valleroy, Eben
Moglen, and me. The aim of the meeting is to discuss the main challenges
facing the FreedomBox project and to plan for our future.

When I first announced this meeting on our progress call, I mentioned
that we want the agenda to include items recommended by the community. I
want to ask each of you on this discussion list to help us out with
this. In short, we are planning to discuss the most important issues
facing the FreedomBox project as we continue to grow, attract new users,
and expand into hardware sales. My questions for you: what do you think
are the most important challenges facing the FreedomBox project? What do
you want to see us do in the future? We want FreedomBox to become a
mainstream consumer privacy solution. What do you think we need to do
before we get to that point?

These questions are intentionally broad--please feel free to interpret
them however you wish and answer as narrowly or broadly as you'd like.
As I said, we want to give the community a voice in this meeting, so you
should feel free to raise any issue that you care about.

I will take note of all the questions I get and try my best to
incorporate your ideas into our agenda. You can feel free to email me
your ideas in this thread, or to my email address privately at
<***@freedomboxfoundation.org>. Alternatively, I also plan to bring
this up on our upcoming progress call this Saturday, November 10--so you
can bring your questions to that call if you want.

I hope to hear from many of you!

Best,

Danny

[Adrian Gropper]

Hi Danny,

I love the FreedomBox project and have been following it for years. I lead
a Free software project for patient controlled independent health records
that would benefit immensely from a consumer-friendly FreedomBox but you
don't seem to be any closer to that for our use-case than you were two
years ago. I have to admit that this is my impression as a spectator and
that I have not invested much time in actually trying to get things to work
myself so please take my comments with a large grain of salt.

The problem is that the project is trying to do way too much and in too
vertical a way. It would be much more useful to me and a lot of other Free
software product developers if FreedomBox presumed separate platform and
app components. Current hardware is so cheap that it hardly makes sense to
put two different apps on the same appliance. Combining apps from multiple
developer communities on the same FreedomBox also muddies the support
relationships. To get mass adoption, people need "one neck to choke" and
that should be the app developers as much as possible.

For example, a $10 Raspberry Pi Zero W with a $8 OLED to display an IP
address and a USB keyboard to enter a wifi password could be a reference
point as a platform for about $30. After that, everything else could be
done from a Web browser on the LAN. That, along with a preset choice among
3 alternatives of dynamic DNS services to allow almost anyone to set up a
specific server app and appliance with one page of instructions. Yes, I
know the Raspberry pi has Freedom issues so a totally Free alternative
should also be available but the $30 appliance platform needs to be our
benchmark nonetheless.

Along with this alternative strategy, I would hope for more modular support
around Debian for Free software appliance developers. I'm not sure how that
would work, but app developer communities need to be able to support their
most unsophisticated customers around platform security issues, backup and
restore, and hardware migrations. Given that partnership, every feature or
app currently offered on FreedomBox and many more could be supported by
their respective communities as separate appliances with FreedomBox
completely agnostic to what they are. This would be a much more diverse and
more scalable ecosystem.

Thank you all for this really lovely project!!

Adrian

Post by Danny Haidar
Hi everyone,
I want to open a discussion about the FreedomBox Foundation's upcoming
FreedomBox Summit. As we discussed on our progress call on October 13,
the FreedomBox Foundation is hosting a meeting in New York City on
Tuesday, November 13, 2018 with Sunil Mohan Adapa, James Valleroy, Eben
Moglen, and me. The aim of the meeting is to discuss the main challenges
facing the FreedomBox project and to plan for our future.
When I first announced this meeting on our progress call, I mentioned
that we want the agenda to include items recommended by the community. I
want to ask each of you on this discussion list to help us out with
this. In short, we are planning to discuss the most important issues
facing the FreedomBox project as we continue to grow, attract new users,
and expand into hardware sales. My questions for you: what do you think
are the most important challenges facing the FreedomBox project? What do
you want to see us do in the future? We want FreedomBox to become a
mainstream consumer privacy solution. What do you think we need to do
before we get to that point?
These questions are intentionally broad--please feel free to interpret
them however you wish and answer as narrowly or broadly as you'd like.
As I said, we want to give the community a voice in this meeting, so you
should feel free to raise any issue that you care about.
I will take note of all the questions I get and try my best to
incorporate your ideas into our agenda. You can feel free to email me
your ideas in this thread, or to my email address privately at
this up on our upcoming progress call this Saturday, November 10--so you
can bring your questions to that call if you want.
I hope to hear from many of you!
Best,
Danny
--
Danny Haidar
Vice-President for Product & Development
FreedomBox Foundation
_______________________________________________
Freedombox-discuss mailing list
https://alioth-lists.debian.net/cgi-bin/mailman/listinfo/freedombox-discuss

...

[mray]

Hi Danny,

I think the most important issue for the project right now is:

Earn trust by providing a reliable and easy user experience.

Setting up, running and backing up (and restoring!) a service needs to
be an easy successful, even joyful task. Only then can we earn the trust
of people to see us as alternative to centralized services that perform
excellent in that respect. This polished and streamlined interaction
won't be easy to provide for ALL our applications, which is why I'd
suggest to focus on selected services and care about less important ones
later.


-Robert

Post by Danny Haidar
Hi everyone,
I want to open a discussion about the FreedomBox Foundation's upcoming
FreedomBox Summit. As we discussed on our progress call on October 13,
the FreedomBox Foundation is hosting a meeting in New York City on
Tuesday, November 13, 2018 with Sunil Mohan Adapa, James Valleroy, Eben
Moglen, and me. The aim of the meeting is to discuss the main challenges
facing the FreedomBox project and to plan for our future.
When I first announced this meeting on our progress call, I mentioned
that we want the agenda to include items recommended by the community. I
want to ask each of you on this discussion list to help us out with
this. In short, we are planning to discuss the most important issues
facing the FreedomBox project as we continue to grow, attract new users,
and expand into hardware sales. My questions for you: what do you think
are the most important challenges facing the FreedomBox project? What do
you want to see us do in the future? We want FreedomBox to become a
mainstream consumer privacy solution. What do you think we need to do
before we get to that point?
These questions are intentionally broad--please feel free to interpret
them however you wish and answer as narrowly or broadly as you'd like.
As I said, we want to give the community a voice in this meeting, so you
should feel free to raise any issue that you care about.
I will take note of all the questions I get and try my best to
incorporate your ideas into our agenda. You can feel free to email me
your ideas in this thread, or to my email address privately at
this up on our upcoming progress call this Saturday, November 10--so you
can bring your questions to that call if you want.
I hope to hear from many of you!
Best,
Danny

...

[permondes - sagen]

Hi Danny,
thanks for posing this topic to the forum.
Before talking about the future of the FreedomBox project, let me first
put down my impression of the current status:I think the FreedomBox
currently has a number of interesting individual solutions running
(some are part of my daily life by now). As development is performed in
the Testing repository, one has to accept occasional issues after
updates. Thus the project is dedicated to Debian experienced people.
In order to define the next steps and the directions to take, I think
we should understand and define whom we are targeting at. As I am
working as product manager (not in IT), I like the concept of Personas,
i.e. to define one or two fictive personalities who represent the
target audience. Any development is done by asking these Personas how
they see the changes, if they can deal with them.Based on this we have
to define the applications needed, by understanding what our customers
are trying to achieve with the Box. This is in contrast to the current
approach which is rather software package centered. An application
centered approach provides a solution to a problem which may consist of
the interaction of various software packages. E.g. a wiki requires a
DNS record and ideally also Lets-Encrypt. Such an approach can often be
realized with adapting the user interface and the corresponding
documentation. Here, users need more guidance.
Classifying these needs can be done following the KANO-Analysis, so
defining - must have- exciters (wow effect)- valued requirements-
indifferent- undesired (listed just for completeness)
Based on this we should set up a roadmap with time-line and work
through this. 
Coming back to the fact that the development is done in Testing: As
users need a reliable solution but also want to benefit from the
development, we should add backport support for the Stable
distribution. I think this was already discussed. Once the Box is
running in a stable way, (and it improved a lot during recently), it is
time to spread word.
Dietmar

Post by Danny Haidar
Hi everyone,
I want to open a discussion about the FreedomBox Foundation's
upcoming
FreedomBox Summit. As we discussed on our progress call on October 13,
the FreedomBox Foundation is hosting a meeting in New York City on
Tuesday, November 13, 2018 with Sunil Mohan Adapa, James Valleroy, Eben
Moglen, and me. The aim of the meeting is to discuss the main
challenges
facing the FreedomBox project and to plan for our future.
When I first announced this meeting on our progress call, I mentioned
that we want the agenda to include items recommended by the
community. I
want to ask each of you on this discussion list to help us out with
this. In short, we are planning to discuss the most important issues
facing the FreedomBox project as we continue to grow, attract new users,
and expand into hardware sales. My questions for you: what do you think
are the most important challenges facing the FreedomBox project? What do
you want to see us do in the future? We want FreedomBox to become a
mainstream consumer privacy solution. What do you think we need to do
before we get to that point?
These questions are intentionally broad--please feel free to
interpret
them however you wish and answer as narrowly or broadly as you'd like.
As I said, we want to give the community a voice in this meeting, so you
should feel free to raise any issue that you care about.
I will take note of all the questions I get and try my best to
incorporate your ideas into our agenda. You can feel free to email me
your ideas in this thread, or to my email address privately at
this up on our upcoming progress call this Saturday, November 10--so you
can bring your questions to that call if you want.
I hope to hear from many of you!
Best,
Danny

...

-- 
FreedomBox: https://permondes.de

[Joseph Nuthalapati]

Danny,

Here are some developments I'd like to see in FreedomBox in the
near future:


Table of Contents
_________________

1 Separation of Operating System and Applications
2 FreedomBox Contrib
3 User Feedback and Analytics
4 Improving Adoption


1 Separation of Operating System and Applications
=================================================

I think the FreedomBox platform should take a lesson from the Android
platform in separating the operating system from its applications.
FreedomBox, the consumer operating system for servers, should provide
the core framework with a few server applications shipped as stock
applications, e.g. Radicale, Syncthing, Dynamic DNS, Let's Encrypt,
Tor etc. The remaining applications should be provided as apps through
an "app store" of sorts. The development of FreedomBox apps can be
taken up by interested third-parties, not the core team. We already
have 25 applications and counting, the maintenance of which is taking
up significant time and efforts from the FreedomBox team. We'll be
completely overwhelmed with application maintenance when the number of
applications reaches 50.


2 FreedomBox Contrib
====================

Debian packaging for some large applications like federated social
networks (diaspora*, mastodon etc.) has turned out to be a Sisyphean
task. It's worth noting that the one year's effort spent on packaging
diaspora* was undone when the upstream released a new major version.
Packaging the new version turned out be as much work once again. This
is likely to happen again when the next major version is released.

Most modern federated applications are being written in languages with
packaging systems that encourage too many small packages. diaspora*
alone has 250 Ruby dependencies. Mastodon needs both Ruby and NodeJS
dependencies.

The diaspora* team has been fairly successful with their contrib
package called disapora-installer, but the diaspora package itself
requires too much manpower to maintain. I propose that we should
create a freedombox-contrib package that when installed can install
applications like diaspora* through their contrib packages
(diaspora-installer in this case).


3 User Feedback and Analytics
=============================

Passive collection of anonymized user interaction data is what guided the
development and improvement of most modern commercial platforms. I
believe that we should do proper web and mobile analytics with a
self-hosted solution like Matomo. A low-hanging fruit in this category
is to do regular user surveys with a few carefully formulated
questions.


4 Improving Adoption
====================

Discussed earlier on the mailing list at
[https://alioth-lists.debian.net/pipermail/freedombox-discuss/2018-July/008451.html]

Post by Danny Haidar
Hi everyone,
I want to open a discussion about the FreedomBox Foundation's
upcoming FreedomBox Summit. As we discussed on our progress
call on October 13, the FreedomBox Foundation is hosting a
meeting in New York City on Tuesday, November 13, 2018 with
Sunil Mohan Adapa, James Valleroy, Eben Moglen, and me. The aim
of the meeting is to discuss the main challenges facing the
FreedomBox project and to plan for our future.
When I first announced this meeting on our progress call, I
mentioned that we want the agenda to include items recommended
by the community. I want to ask each of you on this discussion
list to help us out with this. In short, we are planning to
discuss the most important issues facing the FreedomBox project
as we continue to grow, attract new users, and expand into
hardware sales. My questions for you: what do you think are the
most important challenges facing the FreedomBox project? What
do you want to see us do in the future? We want FreedomBox to
become a mainstream consumer privacy solution. What do you
think we need to do before we get to that point?
These questions are intentionally broad--please feel free to
interpret them however you wish and answer as narrowly or
broadly as you'd like. As I said, we want to give the community
a voice in this meeting, so you should feel free to raise any
issue that you care about.
I will take note of all the questions I get and try my best to
incorporate your ideas into our agenda. You can feel free to
email me your ideas in this thread, or to my email address
I also plan to bring this up on our upcoming progress call this
Saturday, November 10--so you can bring your questions to that
call if you want.
I hope to hear from many of you!
Best,
Danny

...

- --
Regards,
Joseph Nuthalapati

[Petter Reinholdtsen]

Thank you for bringing up these suggestions.

[Joseph Nuthalapati]

Post by Joseph Nuthalapati
The diaspora* team has been fairly successful with their contrib
package called disapora-installer, but the diaspora package itself
requires too much manpower to maintain. I propose that we should
create a freedombox-contrib package that when installed can install
applications like diaspora* through their contrib packages
(diaspora-installer in this case).

While I agree that several program systems seem to be very hard to
package for Debian, I must admit that I am very reluctant to run on my
computers random stuff downloaded from the Internet. It is part of the
reason I like that Freedombox is based Debian, where there is hard
control over what is provided to the users. It is to regain control
over my computer equipment and avoid having to trust a bunch of random
package repositories I want to use Freedombox.

Post by Joseph Nuthalapati
3 User Feedback and Analytics
=============================
Passive collection of anonymized user interaction data is what
guided the development and improvement of most modern commercial
platforms. I believe that we should do proper web and mobile
analytics with a self-hosted solution like Matomo. A low-hanging
fruit in this category is to do regular user surveys with a few
carefully formulated questions.

Submission of person information, like IP addresses and the fact that a
program is being used, accompanied with a story about throwing away part
of the collected information to make it 'anonymized', is part of the
reason I stay away from "most modern commercial platforms" and use
Freedombox. Introducing call home functions to the Freedombox seem like
a very bad idea that is unlikely to further our goal.