Re: [Freepto] Initial comments

Delete this message

Reply to this message
Autore: vinc3nt
Data:  
To: freepto
Oggetto: Re: [Freepto] Initial comments
On 02/25/2014 01:06 PM, intrigeri wrote:

>
> 0. What I've found to be the hardest thing when looking at Freepto was
>    the lack of design goals clearly documented in a language I can
>    read. 


there is now an initial draft:

https://wiki.freepto.mx/en/Design


> 22. Regarding transparent proxying through Tor, we have stopped doing
>     that a while  ago in Tails. I'm not sure if this applies to
>     Freepto too, but transparent  proxying had two drawbacks for us:
>     1. it grants access to the network to  applications bound to any
>     network interface (e.g. eth0 or ppp0), so if the application leaks
>     (e.g. in application layer packets) the local IP address it knows
>     about, it may be leaking the IP address on the LAN (which might be
>     identifying sometimes, and at least allows linking activities
>     together), or worse (in the case of ppp0) the public IP address on
>     the Internet. 2. It somewhat lets the user believe that any
>     application can be used safely with Tor, without special
>     configuration, which is clearly untrue.


Both risks you mentioned are true.
Maybe we have to clarify these risks on the TorTP UI (also a manpage is
still missing).


>
> 23. Using persistence-media=removable-usb, you'll soon enough be
>     affected by the same problem as Tails caused by vendors such as
>     Sandisk now shipping USB sticks with the removable flag set (in
>     accordance to the USB standard, but mainly to be allowed a fancy
>     Windows 8 logo): https://labs.riseup.net/code/issues/6397 We have
>     a plan to deal with that, and perhaps we could collaborate to add
>     the missing bits to live-boot, and upstream it?


We used this setting basically to avoid (during the boot) the question
about "encrypted hard driver unlocking" on computer with FDE.

Until now we never experienced this problem, but of course we can work
to fix it together.


> 24. I see you disable TCP timestamps. We're thinking of doing the same
>     in Tails. Any drawback of doing so you would be aware of?


I checked months ago and I don't remember worrying issues:

http://www.forensicswiki.org/wiki/TCP_timestamps
https://stackoverflow.com/questions/7880383/what-benefit-is-conferred-by-tcp-timestamp

>
> 25. Does anything prevent live-* to use a cleartext swap partition
>     found on an internal hard drive?


Is this a realistic scenario? I never experienced with _automatic_ swap
mount from an internal HD.
I will check this behavior. Is there any hack in Tails to prevent it?


> 26. Tails Greeter:


It is very interesting, but I think we have more important priority at
moment.
We should deal with this in the future, maybe after the next release.


> 27. Maybe install bash-completion?


why not :)


> 29. I'm part of the people who are currently bootstrapping two Debian
>     teams that might interest you:

>
>       https://wiki.debian.org/AppArmor 
>       https://wiki.debian.org/Teams/OTR


Cool!
I would be very happy to help, even if only for triage.


> 31. If there's work you do that you would like to upstream in Debian,
>     and in case you need help / mentoring / sponsoring, feel free to
>     ask me. But I'm sure you already know plenty of other Debian
>     developers who are eager to help :)


Many thanks :)



--
vinc3nt

+-------------------------------------------------------------+
|GPG Id key: 4096R/DAA26AC4 2012-12-07                        |
|GPG Fingerprint: 19D8011C 81891778 7A1986DC 28E8567F DAA26AC4|
|OTR Fingerprint: 86F3EE43 8A9340D2 FF92A828 81C84DC2 FEDD839E|
+-------------------------------------------------------------+