Re: [Pearg] Research Group Last Call for "A Survey of Worldwide Censorship Techniques"

[David Oliver <[e-mail]>]

Chelsea, thanks for this.

> First, I recommend improving the document organization to clearly
differentiate
>  between a specific censorship technique and corresponding methods to
circumvent that technique.

I'm still not understanding why - in a document defined to talk about
censorship techniques - we're talking about circumvention methods. I
believe it's particularly important to understand because, at least for the
techniques I'm familiar with, these methods work "above the stack", not "in
the stack" because we implementors can't have impact on "the stack"
itself.  IETF addresses "the stack" so, in theory at least, IETF could use
the understanding of censorship techniques to implement solutions there.
Why would we (implementors) be engaging with IETF if we do not seek new "in
the stack" methods of preventing censorship?

Thanks for any clarification that can be provided.

David Oliver
David M. Oliver | david@g <[e-mail]>uardianproject.info |
http://g <http://olivercoady.com>uardianproject.info | @davidmoliver | +1
970 368 2366


On Tue, Jun 2, 2020 at 10:31 AM Chelsea Komlo <[e-mail]>
wrote:

> Hi Chris,
>
> I have a couple meta points and a few specific points.
>
> Overall, my strongest recommendation is threefold. First, I recommend
> improving the document organization to clearly differentiate between a
> specific censorship technique and corresponding methods to circumvent that
> technique. Second, I suggest strengthening the review of current
> circumvention techniques and their effectiveness to give an improved view
> of the landscape and to prevent re-inventing the wheel, as the intended
> audience is protocol designers.. Third, I recommend differentiating between
> the capabilities of mature censors like China, and weaker censors such as
> those with only off-the-shelf tools.
>
> Here is my discussion of these points in more detail.
>
> === Meta Points ===
> - While I understand this draft to be purely informational, understanding
> censorship today is incomplete without understanding existing censorship
> circumvention techniques, and how effective these techniques are..
> Critically, some of the most effective and safe censorship techniques are
> "hanging by a thread" in terms of how much longer they will be available
> (such as domain fronting), so understanding these weak areas is important
> to understanding the sustainability of the current circumvention landscape.
> I suggest adding at minimum a discussion of "Where are we today" regarding
> circumvention.
> - To go along with the above point,  providing a better review of existing
> censorship circumvention techniques will help encourage building on
> existing work, as opposed to re-inventing the wheel from first principles.
> While perhaps this review should be a follow-up document, I strongly
> encourage providing such a review, especially since the intended audience
> is protocol designs.
> - Within the draft itself, the discussion of censorship techniques is
> often interwoven with circumvention methods. I suggest separating these
> concepts out within each section. Instead of a "Tradeoffs" section, perhaps
> have sections pertaining to "Cost to Implement to Censor", and "Techniques
> to Circumvent", for improved clarity.
> - One important point is that while China is an extremely powerful censor,
> they are often in a class of their own. I encourage including a discussion
> of something like "censor maturity" or the technical resources required to
> implement different techniques. There is a bit of this discussion, but it
> can be better standardized and applied to each technique. For example, IP
> blacklisting is trivial and does not require significant infrastructure
> (and many censors do this), but performing active probing to fingerprint
> protocols and block them on the fly requires much more infrastructure and
> planning (and is essentially only China, as I understand).
>
> === Specific Points ===
> - DPI (deep packet inspection) is technically any kind of packet analysis
> beyond IP address and port number- this concept can be better clarified.
> Further, this technique is not specific to
> - Clearly highlighting techniques that are thwarted by the use of TLS
> versus techniques which can be performed even in spite of TLS usage would
> also likely be helpful to readers.
>
> Thanks,
> Chelsea
>
> On Wed, May 20, 2020 at 11:00 AM Christopher Wood <[e-mail]>
> wrote:
>
>> This is the research group last call for the "A Survey of Worldwide
>> Censorship Techniques" (draft-irtf-pearg-censorship) draft available here:
>>
>>    https://datatracker.ietf.org/doc/draft-irtf-pearg-censorship/
>>
>> Please review the document and send your comments to the list by June 5,
>> 2020. Feedback may also be sent to the GitHub repository located here:
>>
>>    https://github.com/IRTF-PEARG/rfc-censorship-tech
>>
>> Thanks,
>> Chris, on behalf of the chairs
>>
>> --
>> Pearg mailing list
>> [e-mail]
>> https://www.irtf.org/mailman/listinfo/pearg
>>
>
>
> --
> Chelsea H. Komlo
> --
> Pearg mailing list
> [e-mail]
> https://www.irtf.org/mailman/listinfo/pearg
>