Difference between revisions of "CommonsCloud"

From Free Knowledge Free Technology Wiki
Jump to: navigation, search
(Visions and working groups)
 
(6 intermediate revisions by 2 users not shown)
Line 1: Line 1:
With "CommonsCloud" we refer to a set of network services that are operated by and for the commons. We consider servers and web applications that are published under free licenses (as Free Software), rigorously adopt Open Standards and the self-governance and self-management by the participating users. Our long term vision is to evolve towards decentralised, federated systems, like email.
+
With "CommonsCloud" we refer to a set of network services that are operated by and for the commons. We consider servers and web applications that are published under free licenses (as Free Software), rigorously adopt Open Standards and are self-governed by the participating users. Our long term vision is to evolve towards decentralised, federated systems, like email.
  
 
For this purpose we are doing research on existing systems that can be used, and possibly adapted, and on initiatives with similar purposes, to learn from them and possibly collaborate.
 
For this purpose we are doing research on existing systems that can be used, and possibly adapted, and on initiatives with similar purposes, to learn from them and possibly collaborate.
Line 5: Line 5:
 
While this work has a global perspective, we are also involved in setting up a concrete set of CommonsCloud services in Barcelona. This we do as an alliance of several likeminded actors.
 
While this work has a global perspective, we are also involved in setting up a concrete set of CommonsCloud services in Barcelona. This we do as an alliance of several likeminded actors.
  
 +
To contribute please join the [http://lists.freeknowledge.eu/listinfo/commonscloud mailing list].
  
 
== Systems review ==
 
== Systems review ==
Line 77: Line 78:
  
 
==Existing CommonsCloud initiatives==
 
==Existing CommonsCloud initiatives==
like Framasoft, Sandstorm.io, cozy.io  
+
 
 +
* [https://framasoft.org/ Framasoft] different FS cloud services each at a different domain; however several relevant services are limited, e.g. the NextCloud FramaDrive doesn't allow new users.
 +
* [http://Sandstorm.io Sandstorm.io] and test at [https://fki.sandcats.io beta service run by FKI]
 +
* [http://cozy.io cozy.io]
 +
* [http://cloudy.community/ Cloudy]
 +
* [https://disroot.org/ Disroot] the use a combination of FS cloud software but didn't publish (yet) how exactly
 +
 
 +
===Cozy vs Sandstorm comparison===
 +
 
 +
_(this is a collage of notes and messages exchanged by M. Fioretti, which needs a deeper study to be confirmed!)_
 +
 
 +
As far as I can see, Cozy...
 +
 
 +
unlike cloudron, has no embedded email server, and domain name support, i.e (almost) one click domain name registration, creation of SSL certificates for that domain..
 +
 
 +
no package for movim.eu client, which is IMO essential
 +
 
 +
CURRENT VERSION ONLY: too resource hungry? they themselves say it is n[https://blog.cozycloud.cc/post/2016/11/21/On-the-road-to-Cozy-version-3?lang=en not efficient enough that "hosting LARGE numbers of Cozy instances...) is economically doable"]
 +
 
 +
TOO LIMITED? From the [https://cozy.io/en/faq FAQ]: "Q: Can I install my favorite PHP, Python, Ruby, etc. application on Cozy? A: Currently Cozy supports only Node.js applications."
 +
 
 +
Also from the FAQ: "Q: Can I share a calendar with a friend? A: Not at the moment, but this feature is on our roadmap."
 +
 
 +
'''"QUANTIFIED SELF"''': One reason for Cozy's interface and certain apps being what they are, is one more philosophical difference between cozy and percloud: the former focuses a lot on the "quantified self", that is collecting and using yourself all conceivable data you may get from your fitbit device, your smart power meter, etc.. and easily combine/share those data among your apps or with
 +
third parties.
 +
 
 +
In other words, what DOES make Cozy stand out wrt cloudron etc... is that it puts a lots of effort on certain ''types'' of personal data that many people simply don't have. Even people who live in many "first world" cities could not get smart meters, or e-banking, that interface with third party services, if they wanted to. The KYou cozy app imports health data from Jawbone bracelets, a kind of product that, regardless of price, has little or no interest for many people... There's nothing wrong in Cozy offering those specific services, of course. Still, they may be simply not relevant enough, or not usable, by too many people, to point on Cozy for a platform that has real mass adoption as one of its main goals.
 +
 
 +
on the same topic, check out the owncloud and sandstorm sections of [https://cozy.io/en/comparison/ this page]:
 +
 
 +
 
 +
Localization: I may be totally wrong here, but I got the feeling somewhere that having cozy apps in other languages is something one would have to do by himself. Integrating existing apps like cloudron and others do, instead, makes it more likely to get an environment that is ALREADY, at least partially, localized in other languages.
 +
 
 +
Also relevant may be this "Why I left Cozy" post, even though several of them may be '''not valid anymore''' today:
 +
https://forum.cozy.io/t/a-review-about-cozy-and-why-im-leaving/814
  
 
==Visions and working groups==
 
==Visions and working groups==
Line 83: Line 118:
 
* [http://per-cloud.com/ PER-cloud] by [http://freeknowledge.eu/about/people#Marco_Fioretti Marco Fioretti]
 
* [http://per-cloud.com/ PER-cloud] by [http://freeknowledge.eu/about/people#Marco_Fioretti Marco Fioretti]
 
* [https://open.coop/2017/01/10/planet-open-source-operating-system-collaborative-sustainable-economy/ Planet], by [https://twitter.com/DefactoDesign Oliver Sylvester-Bradley]
 
* [https://open.coop/2017/01/10/planet-open-source-operating-system-collaborative-sustainable-economy/ Planet], by [https://twitter.com/DefactoDesign Oliver Sylvester-Bradley]
 +
* [http://ioo.coop/clouds IoO.coop/Clouds] Internet of Ownership (Nathan Schneider cs.) platform coop cloud projects directory in progress
 +
* [http://people.ac.upc.edu/leandro/docs/cloudy.pdf Cloudy in guifi.net: Establishing and sustaining a community cloud as open commons]

Latest revision as of 15:54, 23 February 2018

With "CommonsCloud" we refer to a set of network services that are operated by and for the commons. We consider servers and web applications that are published under free licenses (as Free Software), rigorously adopt Open Standards and are self-governed by the participating users. Our long term vision is to evolve towards decentralised, federated systems, like email.

For this purpose we are doing research on existing systems that can be used, and possibly adapted, and on initiatives with similar purposes, to learn from them and possibly collaborate.

While this work has a global perspective, we are also involved in setting up a concrete set of CommonsCloud services in Barcelona. This we do as an alliance of several likeminded actors.

To contribute please join the mailing list.

Systems review

User management and authentication

Collaborative writing, office and filesharing

Collabora Office/LibreOffice+nextCloud
Developer/Maintainer:
Description: collaborative editing and filesharing through LibreOffice + nextCloud
Core technologies LibreOffice + nextCloud
Software licenses:
Language/s:
Server requirements: nextCloud: MySQL/MariaDB, PHP 5.4 +, Apache 2.4 with mod_php
Server requirements: [LibreOffice]:
Installation with Docker
Standards used
Synchronisation with netCloud/ownCloud clients for mobile and desktop
Maintenance effort
To be dev/adapted
User mgt. / auth LDAP in LibreOffice; LDAP in nextCloud

Project, task and group management

Social networks and web content management systems

Calendar, contacts and notes

Radicale, Davical


Email, mailing list

mailman


Instant messaging and groupware

Miscelaneous

password management, like sandpass (using keypass format)


Existing CommonsCloud initiatives

  • Framasoft different FS cloud services each at a different domain; however several relevant services are limited, e.g. the NextCloud FramaDrive doesn't allow new users.
  • Sandstorm.io and test at beta service run by FKI
  • cozy.io
  • Cloudy
  • Disroot the use a combination of FS cloud software but didn't publish (yet) how exactly

Cozy vs Sandstorm comparison

_(this is a collage of notes and messages exchanged by M. Fioretti, which needs a deeper study to be confirmed!)_

As far as I can see, Cozy...

unlike cloudron, has no embedded email server, and domain name support, i.e (almost) one click domain name registration, creation of SSL certificates for that domain..

no package for movim.eu client, which is IMO essential

CURRENT VERSION ONLY: too resource hungry? they themselves say it is nnot efficient enough that "hosting LARGE numbers of Cozy instances...) is economically doable"

TOO LIMITED? From the FAQ: "Q: Can I install my favorite PHP, Python, Ruby, etc. application on Cozy? A: Currently Cozy supports only Node.js applications."

Also from the FAQ: "Q: Can I share a calendar with a friend? A: Not at the moment, but this feature is on our roadmap."

"QUANTIFIED SELF": One reason for Cozy's interface and certain apps being what they are, is one more philosophical difference between cozy and percloud: the former focuses a lot on the "quantified self", that is collecting and using yourself all conceivable data you may get from your fitbit device, your smart power meter, etc.. and easily combine/share those data among your apps or with third parties.

In other words, what DOES make Cozy stand out wrt cloudron etc... is that it puts a lots of effort on certain types of personal data that many people simply don't have. Even people who live in many "first world" cities could not get smart meters, or e-banking, that interface with third party services, if they wanted to. The KYou cozy app imports health data from Jawbone bracelets, a kind of product that, regardless of price, has little or no interest for many people... There's nothing wrong in Cozy offering those specific services, of course. Still, they may be simply not relevant enough, or not usable, by too many people, to point on Cozy for a platform that has real mass adoption as one of its main goals.

on the same topic, check out the owncloud and sandstorm sections of this page:


Localization: I may be totally wrong here, but I got the feeling somewhere that having cozy apps in other languages is something one would have to do by himself. Integrating existing apps like cloudron and others do, instead, makes it more likely to get an environment that is ALREADY, at least partially, localized in other languages.

Also relevant may be this "Why I left Cozy" post, even though several of them may be not valid anymore today: https://forum.cozy.io/t/a-review-about-cozy-and-why-im-leaving/814

Visions and working groups

Various people and groups are working on visions and ideas for a personal or collaborative cloud: