[IRP] Blocking and filtering text

Tapani Tarvainen tapani.tarvainen
Wed Nov 10 15:44:08 EET 2010

On Wed, Nov 10, 2010 at 12:30:50PM +0000, Dixie Hawtin (Dixie at global-partners.c

> I just wanted to remind everyone that if you have specific
> suggestions for more concrete wording of any of the provisions, or
> specific suggestions for principles that you would like to see in
> the Charter but which are not currently there, please send them
> through to the list by Monday.

As I seem to have scheduled next full night's sleep to mid-December
or so, I've been reading the list somewhat sporadically and may
have missed something - my apologies if I repeat something that's
already been said. But a couple of comments:

> Tapani, I apologise for the absence of "freedom from religion" in
> the draft Version 1.1. I had taken on board your earlier comments and
> inserted it into the text, however it unintentionally got lost in the
> proliferation of copies I was working on! Thank you for pointing this
> out and it will be reinserted into the text. The line I used was as
> follows: "This right also includes the right to freedom from religion
> and belief". Is that enough do you think?

I don't really like that wording. I'd prefer something that makes it
clear that it's not about what one believes or not, but about rules
based on those beliefs, and that this right specifically means that
no-one should be forced to observe rules of religions they don't
believe in.

Good wording needs some thought. I'll try to think of something
before Monday.

> With regards to the very interesting discussion about filtering and
> blocking
> What I tried to do was to more stringently define the cases in which
> filtering and blocking may* (although it seems I may have been wrong
> about this) be acceptable.

Even the terms are not unambiguous. As I read the draft,
I intepreted them there as implying censorship,
rather than as, say, the Council of Europe documents Meryem
referred to do, as tools provided to users.

> I am not a technical expert but is it true that blocking and
> filtering is always necessarily prior censorship?

That depends on your definition of blocking and filtering.

For example, I have no problem in ISPs filtering viruses and spam *as
a service*; I am not comfortable with it if they wouldn't allow me to
bypass it if I like (I do, I run my own mail server and my own spam
filter), and even less comfortable with the idea of government
mandating it - but even that doesn't count as censorshop *if* the
filtering rules are of technical nature (limiting number of messages
in a timeframe, requiring strict observance of standards &c).

But if the filtering is based on the *content* being transmitted
*and* enforced by the government, then it is censorship.

In particular note that there's no universally agreed-on definition of
spam: what is spam to one recipient may be interesting message to
another. And spam-detection is never foolproof.

> I thought that if a particular piece of content could legitimately
> be limited under human rights law, and that the decision to limit
> the content was made by a court (following all trial requirements)
> and that the content was only then blocked (and the block only
> covered that specific piece of content), this would not constitute
> prior censorship.

If "all trial requirements" include openness, i.e., everybody could
see for themselves (and reporters could report in newspapers) what is
being censored and judge by themselves if the decision is right, sure
- but that'd already mean it's not being censored, wouldn't it?

Or is the idea that accessing the material would be possible
but only by means other than through Internet?
That would again be saying freedom of expression is somehow
less protected than it is elsewhere.

- I note in passing that I could accept prior censorship
in certain situations, like about reporting troop movements
during a war. The key limitation there is that it must be
for so short a duration that the information being censored
can be seen and the justification for its censorship
evaluated soon enough afterwards to bring thouse
responsible to justice as needed.
But I can't see any situation where any material should
be permanently censored from the general public.
(Note: this does not mean publishing anything should
be allowed. People should still be punished for
libel, fraud, spreading private information &c.
But not prevented from doing that by censorship.)

> I know that none of the blocking systems (which I am aware of)
> follow these requirements, and thus my intention was to make the
> majority or even all blocking systems illegal on that premise.

While I appreciate the idea, I don't think it would work,
even if we could agree on the requirements.
Those who want to censor the net just take anything
they can interpret as supporting it and ignore such
requirements as would get in the way.

> The current thinking on the list seems to be along the lines of:
> Filtering and blocking is always in contravention of Article 19 with two limited exceptions:
> a) end-user filtering; and
> b) filtering and blocking aimed at controlling spam and malware

As noted I'm not quite happy even with b), unless it is also
optional to end users (or at very least to ISPs),
but I could accept it if formulated tightly enough.
On the other hand I'd have no problem even actively
promoting filtering tools for end users to use if they like,
so perhaps we could add some such wording?
For example, we might want to state that organisations
like spamhaus should have the right to keep operating,
spammers' protests notwithstanding.

That is, say that blocking and filtering tools and services
should be made available to users but never forced on them.
(Again, precise wording needs some more thought.)

> from what I understand the reasons that filtering and blocking are
> always in contravention of Article 19 are that they cannot be
> targeted narrowly and they disproportionately undermine the capacity
> of the internet to support freedom of expression.

I guess you could say that - although I think that applies to any
permanent censorship outside the Internet as well.

> I am not completely clear on why we think that blocking for spam and
> malware is a valid exception - does this not also undermine the
> openness of the internet, and if not why not?

Two key points (sorry for repeating myself):

(1) Enforcing *technical* standards and other restrictions that are
agnostic on the information content of the messages are not generally
problematic to FoE. This category presently covers most of spam, and
arguably viruses and similar malware, too.

(2) Spam is almost by definition messages people don't *want*:
filtering it becomes a problem only when it's defined in
a way (some) people don't like and they're given no choice.
But spam filtering is (as far as I know everywhere)
ultimately optional to end users (or at least ISPs):
those who really want it can get it.

Tapani Tarvainen

More information about the IRP mailing list