Re: [GNU/consensus] [RFC][SH] User Data Manifesto

[Frank Karlitschek]

On 07.01.2013, at 03:20, Richard Stallman <address@hidden> wrote:

>    I'm not a native speaker so sorry I I choose the wrong words. What words 
> do you suggest?
> 
> I suggest talking about which _actions_ should be under who's control.

Good idea.


>    Sure. This documents is targeting every data that is not a program. I 
> think freedom in programs are perfectly covered by the FSF principals and the 
> GNU licenses.
>    This is about holiday pictures, microblogging messages, blog posts, 
> personal documents, emails and so on.
> 
> Some of these works should be free/libre too.  For instance, if they
> are educational resources.
> 
> However, if you don't use the word "own", and instead talk about
> access to the data, this issue goes away.  It is not a real
> substantive issue, just a spurious issue of language.

O.K. Thanks for the help.


> 
>>> I guess so, but in the long term, this is aiming low.  The real goal
>>> should be that everyone has a server and keeps her data there.
> 
>    Yes. It would be the perfect solution if every user would have a
>    personal server. But in the midterm this is not realistic so
>    people store their data on server that are owned and run by other
>    people. And this is not necessary a problem if the principals of
>    this document are respected. Data is encrypted, can be migrated,
>    ...
> 
> We need to focus on the long term!  To ask for less, because in the
> short term we can only get less, is to risk making a temporary
> compromise permanent.
> 
> Storing data on servers run by someone else is a bad idea!  We should
> teach people to worry when they do this.  The company that gets the
> data may store it in a server in the US and the NSA will copy all the
> data on its way to and from the US.
> 
> It's not so bad, if the user encrypts the data before uploading it
> and the server has no access to unencrypted mail.  But we need
> to talk about this as a compromise.

Yes. We could add that it's the the preferred way to have an own server.


> 
>>> "Invulnerability" is too strong.  Nobody can achieve that.
> 
>    Yes. You are right. This is impossible to achieve. The idea is
>    that this is a principal where we should aim for but propbalby
>    can't be reached. Do you know what I mean?
> 
> Yes, I understand.  I suggest saying it differently; perhaps
> "Protect the data from loss".

Great. Thanks


> 
>>>   8. Server software transparency
>>>   Server software should be free and open source software so that the
>>>   source code of the software can be inspected to confirm that it works as
>>>   specified.
>>> 
>>> Please don't use the term "open source" here.  This is part of the
>>> free software movement.  "Open source" is the slogan of people who
>>> disagree with our ethical ideals.
> 
>    I'm sorry. My mistake. I will change this to free software.
> 
> Thanks.  However, there is an issue of substance here too.
> 
> If the server does the users' own computing, that is SaaS,
> and it always tramples the user's freedom as explained in
> http://www.gnu.org/philosophy/who-does-that-server-really-serve.html.
> 
> Not all services are SaaS.  If it doesn't do _the user's own computing_
> then it isn't SaaS.

I read you document and I see what you mean.
Should we add a paragraph that SaaS is not recommended?



>    Do you like the general direction of this document?
> 
> Not yet, but with changes maybe I would.

Great. Let's do it.

Thanks a lot.
Frank


> 
> -- 
> Dr Richard Stallman
> President, Free Software Foundation
> 51 Franklin St
> Boston MA 02110
> USA
> www.fsf.org  www.gnu.org
> Skype: No way! That's nonfree (freedom-denying) software.
>  Use Ekiga or an ordinary phone call
>