grarpamp | 18 Apr 05:59 2013
Picon

CloudFlare

Noticed a recent surge of sites using CloudFlare.
CloudFlare seems to block Tor exits by default.
So that legitimate users of these sites can't patronize/use
them in a manner that allows the user to maintain control
over their privacy.
They're based in San Francisco, along with Craigslist (which
is another misguided arbitrary blocker of Tor exits).
Any other SF based companies that could benefit from
a visit or hackerspace talk about why they should not
be blocking Tor?
It would make for a good 'all in one' trip or conference :)
Matt Pagan | 18 Apr 11:01 2013
Picon

Re: CloudFlare

> They're based in San Francisco, along with Craigslist (which
> is another misguided arbitrary blocker of Tor exits).
> Any other SF based companies that could benefit from
> a visit or hackerspace talk about why they should not
> be blocking Tor?

Yelp is based in San Francisco. So is Pinterest. Getting the Wikimedia
Foundation (also based in San Francisco) to come over would be a huge
victory, IMO.

--

-- 
Matt Pagan
matthew.a.pagan <at> gmail.com
PGP key ID: 0xA521D36F
Matthew Finkel | 18 Apr 22:01 2013
Picon

Re: CloudFlare

On Thu, Apr 18, 2013 at 09:01:21AM +0000, Matt Pagan wrote:
> > They're based in San Francisco, along with Craigslist (which
> > is another misguided arbitrary blocker of Tor exits).
> > Any other SF based companies that could benefit from
> > a visit or hackerspace talk about why they should not
> > be blocking Tor?
> 
> Yelp is based in San Francisco. So is Pinterest. Getting the Wikimedia
> Foundation (also based in San Francisco) to come over would be a huge
> victory, IMO.
> 

Wikimedia is actually willing to discuss an alternative setup if a
usable one is found. Their current implementation is not really
acceptable, but there also isn't really a working/implemented alternative
solution, at this point (and it's not exactly at the top of their list
to implement their own).

- Matt
Gregory Maxwell | 18 Apr 22:45 2013
Picon

Re: CloudFlare

On Thu, Apr 18, 2013 at 1:01 PM, Matthew Finkel
<matthew.finkel <at> gmail.com> wrote:
> Wikimedia is actually willing to discuss an alternative setup if a
> usable one is found. Their current implementation is not really
> acceptable, but there also isn't really a working/implemented alternative
> solution, at this point (and it's not exactly at the top of their list
> to implement their own).

It's the same old story:  There are persistent highly annoying trouble
makers— not even many of them— who are effectively deterred by
blocking whatever proxies they use. Eventually they hit tor, and thus
tor must be blocked from editing.  This abuse isn't imaginary.

The various magical nymtoken ideas would probably be acceptable— they
just need to make it so that an unbounded supply of identities is not
any cheaper than it already is— but they need to be implemented and
not have a high deployment or operating cost.

There are some people who hold the position that instant doubling of
identities (w/ and w/o tor) that attackers would get is not acceptable
but with things like
http://en.wikipedia.org/wiki/Wikipedia:Wikipedia_Signpost/2013-04-08/News_and_notes
 and Tor's effectiveness at evading censorship I expect that most can
be convinced that it's worth it.  Harder would be the fact that
English Wikipedia (and many other larger Wikipedias) blocks most data
centers and VPS services with large rangeblocks as they get used as
account multipliers by socks and an obvious nym implementation would
partially defeat that.
_______________________________________________
tor-talk mailing list
(Continue reading)

grarpamp | 18 Apr 23:51 2013
Picon

Re: CloudFlare

> It's the same old story:  There are persistent highly annoying trouble
> makers— not even many of them— who are effectively deterred by
> blocking whatever proxies they use. Eventually they hit tor, and thus
> tor must be blocked from editing.  This abuse isn't imaginary.

Of course it isn't imaginary. However this is where kneejerkers are
just being dumb... 2^8 exits will *never* ever be anything in comparison
to the 2^30 IP's reasonably estimated to be actually in use. Ten
square kilo's of your favorite big city has more abusable open IP's
than 2^8. Know how many big cities there are? Know how many
laptops and wifi and open wallplates there are? Lots.

Though sure, I do suggest and accept that Tor may present a
different *class* of abuse than other categories of abusable
IP's.

> not have a high deployment or operating cost

I think cost is large what they think about. Just a...  'Really? You mean we
can turn a flag and whack 2^8 at zero cost, sweet, we just eliminated a
help desk drone's worth of salary from our costs'. That's pretty cold.

> The various magical nymtoken ideas would probably be acceptable— they
> just need to make it so that an unbounded supply of identities is not
> any cheaper than it already is

Nyms wouldn't be usable by legitimate anons unless they are
free from linkable properties. Whether it's usage and cookies across
sites, or back to the anon themself. Even then you must trust the third
party nym provider with the nym logs. History proves that trust is always
(Continue reading)

Gregory Maxwell | 19 Apr 03:20 2013
Picon

Re: CloudFlare

On Thu, Apr 18, 2013 at 2:51 PM, grarpamp <grarpamp <at> gmail.com> wrote:
> Though sure, I do suggest and accept that Tor may present a
> different *class* of abuse than other categories of abusable
> IP's.

Tor exits were not banned prior to their use for abuse. At the point
automated exitlist banning was performed a substantial portion were
manually blocked. (Which had the three way bad effect of not
completely blocking the trolls, while blocking most use by non-free
users, while also blocking ex-exits and punishing people for even
trying out being an exit).

There is no particular blocking efficiency gain that comes from using
exitlists relative to other kinds of abuse sources. The site can and
does block /16's all by itself.  (
http://en.wikipedia.org/wiki/Special:BlockList?wpTarget=&wpOptions[]=addressblocks&limit=5000
)

>> not have a high deployment or operating cost
> I think cost is large what they think about. Just a...  'Really? You mean we
> can turn a flag and whack 2^8 at zero cost, sweet, we just eliminated a
> help desk drone's worth of salary from our costs'. That's pretty cold.

Your approach is why the tor community will make absolutely no
progress on this subject.  Telling me that you don't think the problem
is imaginary doesn't help when everything else you say shows that you
believe it is.

You might think you're being only slightly insensitive to other
people's needs, but I am here to tell you that I am inside the both
(Continue reading)

grarpamp | 19 Apr 06:35 2013
Picon

Re: CloudFlare

>> Tor may present a different *class* of abuse than other categories
>> of abusable IP's.

> There is no particular blocking efficiency gain that comes from using
> exitlists relative to other kinds of abuse sources.

The skill needed for the masses to download and use Tor for personal
style abuse is far less than for them to have formerly attempted old
school proxy chains. Would not an analysis of the type of misdeeds
carried out via all the various abuseable sources yield at least measurable,
if not substantial, weightings for each class of source towards particular
classes of misdeeds. Finance/deep crackers use some subsets, packet
scanners and flooders some others, data leeches others.

>> 'Really? You mean we
>> can turn a flag and whack 2^8 at zero cost, sweet, we just eliminated a
>> help desk drone's worth of salary from our costs'

> You might think you're being only slightly insensitive to other
> people's needs, but I am here to tell you that I am inside the both
> communities

Many of us have hats in both places. I've seen places where
IP based whack a mole was phrased pretty much that exact way.
Or as 'sweet, now we don't even have to perform the balancing act.'

> you are coming off as a clueless jerk.

We're all free to debate each other, point out this or that, supply
links and so forth. That's all good :) And I even expressed some
(Continue reading)

Andreas Krey | 19 Apr 07:27 2013
Picon
Picon

Re: CloudFlare

On Fri, 19 Apr 2013 00:35:58 +0000, grarpamp wrote:
...
> > you are coming off as a clueless jerk.
...
> However the recent direct name calling and abuse amongst
> people on this list needs to stop right now.

You failed to demonstrate an understanding of the
specific situation of wikipedia which makes the
response in question not entirely unappropriate.

> Yes, it's a hard human problem. One for which I think there
> are better solutions than just IP based blocking.

Now, what would they be in this case? As I understand
the problem we have very few people that are so toxic
to the wikipedia community that they simply need to
get blocked. If you don't block tor just like any
other IP range those jerks appear from then you need
either a way to identify those jerks (obviously
impossible given tor's properties), or identify
the good guys coming in via tor (which is something
we might not want to, re anonymity).

The only idea that comes to my mind is a reputation
system, but how do you tell apart the cooperative
newbie from the fivehundert accounts of the jerk
initially? (Reputation needs to be built into
a site, see stackoverflow.com, i guess. Reputation
is also 'I already have an account', but the question
(Continue reading)

grarpamp | 19 Apr 09:34 2013
Picon

Re: CloudFlare

> You failed to demonstrate an understanding of the
> specific situation of wikipedia

Though the wikipedia problem model does apply to other usage scenarios
elsewhere, I did not intend or wish my posts as addressing the
wikipedia subthread but rather the practice of performing just IP
blocking in general.

With the growing knowledge of users regarding their online privacy,
and even interests in having anonymity and communication freedom
options, more are turning to tools like Tor. When their good intents,
participation in, and contributions to services are bluntly blocked
by IP filters, that's a problem.

We're all here because we believe and understand the same basic
things.

>> Yes, it's a hard human problem. One for which I think there
>> are better solutions than just IP based blocking.

> Now, what would they be in this case?

If there's one thing this thread illustrates it's that one size
doesn't fit all. Different services have different possible solutions.

As you've noted, reputation (and learning systems in general) are
a powerful tool over time. With the same noted exception regarding
new users... the real world tells us that the first driveby from a
new user bent on trouble may not ever be truly preventable. Yet
input from other users and help desks into a good system can stop
(Continue reading)

Andreas Krey | 19 Apr 12:21 2013
Picon
Picon

Re: CloudFlare

On Fri, 19 Apr 2013 03:34:15 +0000, grarpamp wrote:
...
> As you've noted, reputation (and learning systems in general) are
> a powerful tool over time. With the same noted exception regarding
> new users... the real world tells us that the first driveby from a
> new user bent on trouble may not ever be truly preventable. Yet
> input from other users and help desks into a good system can stop
> that user rather fast.

The problem, as I understand it, is that wikipedia has: A single
hellbent user needs 10 minutes to cause trouble that takes an
hour to fix. And unfortunately, that user won't go away but
instead appear newly anonymous, and can continue to consume
a factor of five to one in maintainer's manpower. The question
is whether a) they can afford that at all and b) whether the
potential loss of input outweighs the binding of maintainer's
man power (in a way that won't keep them as maintainers for long).

...
> In English some people say 'shit happens'
> and continue on with their day :)

Problem is, this kind of shits keeps happening if you don't
want to block the shitters.

Andreas

--

-- 
"Totally trivial. Famous last words."
From: Linus Torvalds <torvalds <at> *.org>
(Continue reading)

grarpamp | 19 Apr 13:35 2013
Picon

Re: CloudFlare

> The problem, as I understand it, is that wikipedia has: A single
> hellbent user needs 10 minutes to cause trouble that takes an
> hour to fix.

Not knowing the specific case of wikipedia I might suggest
in general...
If you do not require accounts, and are open to the net at large,
you are fighting a largely unsolvable problem [1]. Learning systems
and community and staff moderation can help there. Yet if your
tolerance is still exceeded, you are best to move to accounts
thereby enabling more powerful solutions.
If you require accounts, invest time in better rollback systems
so that a single click makes the user and their contributions
disappear. Because of the N-way de-merges it may be necessary
as staff to roll it all back to before the bad epoch and notify
your community contributors that 'hey, we had to rollback abuse,
please reapply your bits on such and such pages'. If that community
has since died, others will eventually fill in the loss. The quicker
you are notified the less to roll. Definitely not ideal I know. A wiki is a
rare case of built-upon contribs, whereas other types of services
really can just yank the offending user and their stream out of the
system. Sorry I really don't know WP enough to comment in this
subthread. I like WP though and have never noticed anything offensive
on it so something is working there for that category of abuse :)

[1] For instance, email spam. It is not sucessfully fought with
just IP blocks at all. Only when bayes, markov and other adaptive
intelligience and combined systems came online has spamfighting
kept pace. Maybe now they're enough along that if they turned off
the IP parts they'd still fare acceptably well.
(Continue reading)

Andreas Krey | 19 Apr 18:14 2013
Picon
Picon

Re: CloudFlare

On Fri, 19 Apr 2013 07:35:27 +0000, grarpamp wrote:
...
> If you require accounts, invest time in better rollback systems
> so that a single click makes the user and their contributions
> disappear.

You keep ignoring that determined troublemakers can't be forced
to use a single account. Who is determined to creatively
deface user generated content will have no problem solving
many captchas or similar. (And I guess you wouldn't want
to tie accounts to mobile phone numbers or similar.)

Thus, there *is* no one-click way to undo all the changes
of a troublemaker, as they don't come from a single account.

...
> [1] For instance, email spam. It is not sucessfully fought with
> just IP blocks at all.

Email spam has a completely different goal. Unlike the WP jerk,
no spammer spams just to annoy/damage people.

Andreas

--

-- 
"Totally trivial. Famous last words."
From: Linus Torvalds <torvalds <at> *.org>
Date: Fri, 22 Jan 2010 07:29:21 -0800
grarpamp | 20 Apr 04:49 2013
Picon

Re: CloudFlare

>> If you require accounts, invest time in better rollback systems
>> so that a single click makes the user and their contributions
>> disappear.
>>
>> You keep ignoring that determined troublemakers can't be forced
>> to use a single account. Who is determined to creatively
>> deface user generated content will have no problem solving
>> many captchas or similar.
>>
>> Thus, there *is* no one-click way to undo all the changes
>> of a troublemaker, as they don't come from a single account.

The use of multiple accounts is known. Again, enhance your systems
to make your administrative life easier... add a rollback and delete
function. Integrate it with your user feedback, rating and alert
systems. Be creative.

I don't know what's available on the WP backend as I'm not in that
space... but if life with WP means that, as a WP site admin, you're
stuck with using the article history, contributions and user pages
to manage abuse manually, then yeah, WP admins are at a disadvantage
and WP should meta up and enhance that part of the software for
them.

> (And I guess you wouldn't want
> to tie accounts to mobile phone numbers or similar.)

I would provide that as an option to my users, not as a mandate for
all accounts on my site.

(Continue reading)

NoName | 19 Apr 23:08 2013
Picon

Re: CloudFlare

On 19.04.2013 13:35, grarpamp wrote:
> [1] For instance, email spam. It is not sucessfully fought with
> just IP blocks at all. Only when bayes, markov and other adaptive
> intelligience and combined systems came online has spamfighting
> kept pace. Maybe now they're enough along that if they turned off
> the IP parts they'd still fare acceptably well.

Actually blacklists work better. Bayes filters are puzzled by whole 
paragraphs taken from Project Gutenberg. Actually, they can and will put 
gibberish and the Bayes filter won't know s*it. But blacklists work wonders.
mirimir | 19 Apr 23:52 2013
Picon

Re: CloudFlare

On 04/19/2013 05:27 AM, Andreas Krey wrote:

> On Fri, 19 Apr 2013 00:35:58 +0000, grarpamp wrote:

<snip>

>> My problem is with simple IP blocking, especially when you take
>> out an entire shared access system such as Tor with it. It's
>> crude and takes the ham with the spam.
> 
> Wikipedia obviously thinks different - that spam is too
> poisonous to accept with the ham.

The problem with IP blocking is that the "ham" will give up, but the
"spam" will just tunnel a proxy or VPN through Tor.
NoName | 19 Apr 21:05 2013
Picon

Re: CloudFlare

On 18.04.2013 23:51, grarpamp wrote:
> Of course it isn't imaginary. However this is where kneejerkers are
> just being dumb... 2^8 exits will *never* ever be anything in comparison
> to the 2^30 IP's reasonably estimated to be actually in use. Ten
> square kilo's of your favorite big city has more abusable open IP's
> than 2^8. Know how many big cities there are? Know how many
> laptops and wifi and open wallplates there are? Lots.

The only problem you guys are ignoring is the human factor. It might 
sound harsh, but it's plain reality: most system admins are too narrow 
minded to figgure out most things beyond where and how to ask for a loaf 
of bread. They are just complex drones in a complex system. As long as 
computers are still not bright enough to autoconfigure themselves, these 
guys are going to earn a living. So the 15 day crash course says make a 
rule, block an IP. And they are going to do it as it is the only thing 
they know.

Throw in the mix the fact that building a community is much harder than 
policing it. Ever wondered why nice places like England are day by day 
turning into Airstrip One?

And throw in the mix that this is how the tools are shaped. Because, 
even if you like it or not, software developers are regular people. 
Meaning most have an ability to grasp finer concepts way below average 
in order to give the global average of today. Looking at some of the 
things done and said by a client to a server and you can believe most of 
the current conspiracy theorists. Only there is no conspiracy. Only 
plain stupidity. Take Tails for example: once upon a time they used to 
default to check.torproject.org. Only that somebody decided it would be 
cool to have some statistics. Now it defaults to the tails homepage. 
(Continue reading)

Lunar | 20 Apr 16:52 2013

Re: CloudFlare

NoName:
> Only there is no conspiracy. Only plain stupidity. Take Tails for
> example: once upon a time they used to default to
> check.torproject.org. Only that somebody decided it would be cool to
> have some statistics. Now it defaults to the tails homepage.

According to the discussion that lead Tails to change its homepage,
statistics were never mentioned:
<https://tails.boum.org/todo/decide_what_web_homepage_to_use/>

What is your source?

--

-- 
Lunar                                             <lunar <at> torproject.org>
_______________________________________________
tor-talk mailing list
tor-talk <at> lists.torproject.org
https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-talk
NoName | 22 Apr 08:00 2013
Picon

Re: CloudFlare

On 20.04.2013 16:52, Lunar wrote:
> NoName:
>> Only there is no conspiracy. Only plain stupidity. Take Tails for
>> example: once upon a time they used to default to
>> check.torproject.org. Only that somebody decided it would be cool to
>> have some statistics. Now it defaults to the tails homepage.
>
> According to the discussion that lead Tails to change its homepage,
> statistics were never mentioned:
> <https://tails.boum.org/todo/decide_what_web_homepage_to_use/>
>
> What is your source?

http://www.mail-archive.com/tails-dev <at> boum.org/msg02810.html
Lunar | 22 Apr 18:19 2013

Re: CloudFlare

NoName:
> On 20.04.2013 16:52, Lunar wrote:
> >NoName:
> >>Only there is no conspiracy. Only plain stupidity. Take Tails for
> >>example: once upon a time they used to default to
> >>check.torproject.org. Only that somebody decided it would be cool to
> >>have some statistics. Now it defaults to the tails homepage.
> >
> >According to the discussion that lead Tails to change its homepage,
> >statistics were never mentioned:
> ><https://tails.boum.org/todo/decide_what_web_homepage_to_use/>
> >
> >What is your source?
> 
> http://www.mail-archive.com/tails-dev <at> boum.org/msg02810.html

Could you point us where on this page it is said that the homepage was
changed to "have some statistics"?

--

-- 
Lunar                                             <lunar <at> torproject.org>
_______________________________________________
tor-talk mailing list
tor-talk <at> lists.torproject.org
https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-talk
NoName | 22 Apr 19:04 2013
Picon

Re: CloudFlare

On 22.04.2013 18:19, Lunar wrote:
> NoName:
>> On 20.04.2013 16:52, Lunar wrote:
>>> NoName:
>>>> Only there is no conspiracy. Only plain stupidity. Take Tails for
>>>> example: once upon a time they used to default to
>>>> check.torproject.org. Only that somebody decided it would be cool to
>>>> have some statistics. Now it defaults to the tails homepage.
>>>
>>> According to the discussion that lead Tails to change its homepage,
>>> statistics were never mentioned:
>>> <https://tails.boum.org/todo/decide_what_web_homepage_to_use/>
>>>
>>> What is your source?
>>
>> http://www.mail-archive.com/tails-dev <at> boum.org/msg02810.html
>
> Could you point us where on this page it is said that the homepage was
> changed to "have some statistics"?

Well, Google Translate or some other service can change the text into a 
language you could read.
bvvq | 23 Apr 13:57 2013

Re: CloudFlare

On 23/04/2013 3:04 AM, NoName wrote:
> On 22.04.2013 18:19, Lunar wrote:
>> NoName:
>>> On 20.04.2013 16:52, Lunar wrote:
>>>> NoName:
>>>>> Only there is no conspiracy. Only plain stupidity. Take Tails for
>>>>> example: once upon a time they used to default to
>>>>> check.torproject.org. Only that somebody decided it would be cool to
>>>>> have some statistics. Now it defaults to the tails homepage.
>>>>
>>>> According to the discussion that lead Tails to change its homepage,
>>>> statistics were never mentioned:
>>>> <https://tails.boum.org/todo/decide_what_web_homepage_to_use/>
>>>>
>>>> What is your source?
>>>
>>> http://www.mail-archive.com/tails-dev <at> boum.org/msg02810.html
>>
>> Could you point us where on this page it is said that the homepage was
>> changed to "have some statistics"?
> 
> Well, Google Translate or some other service can change the text into a
> language you could read.

There's nothing on the page that explicitly states it. Put up or shut
up, please.
Andrew Lewman | 23 Apr 15:28 2013
Picon

Re: CloudFlare

On Tue, 23 Apr 2013 21:57:26 +1000
bvvq <beveryveryquiet <at> lavabit.com> wrote:

> There's nothing on the page that explicitly states it. Put up or shut
> up, please.

Play nice, children.

--

-- 
Andrew
http://tpo.is/contact
pgp 0x6B4D6475
intrigeri | 30 Apr 12:47 2013

Re: Tails statistics / browser homepage [Was: CloudFlare]

Hi,

NoName wrote (22 Apr 2013 06:00:48 GMT) :
> Only there is no conspiracy. Only plain stupidity.

Regardless of the actual wording of the emitted judgment, I suggest
you check your facts, and make sure you know what you're talking of,
before judging other people's action. Just a friendly advice.

> Take Tails for example: once upon a time they used to default to
> check.torproject.org. Only that somebody decided it would be cool to
> have some statistics. Now it defaults to the tails homepage.

For the record, this is an entirely incorrect description of the
decision process we had.

Facts:

  * the homepage was changed in Tails 0.16, released on January 11
    this year
  * the "Tails report for August, 2012" has boot statistics

=> We were publishing boot statistics _months before_ the browser home
page was changed... no magic involved: the data from which these stats
are computed simply does *not* depend on the web browser homepage.

I think I've guessed what may have confused you in the first place:
the Tails report you're pointing to reads "this number is an
approximation from the requests made to the security announcements
feed". I'm sorry that "security announcement feed" is vague enough for
(Continue reading)

Matthew Finkel | 19 Apr 05:12 2013
Picon

Re: CloudFlare

On Thu, Apr 18, 2013 at 01:45:12PM -0700, Gregory Maxwell wrote:
> On Thu, Apr 18, 2013 at 1:01 PM, Matthew Finkel
> <matthew.finkel <at> gmail.com> wrote:
> > Wikimedia is actually willing to discuss an alternative setup if a
> > usable one is found. Their current implementation is not really
> > acceptable, but there also isn't really a working/implemented alternative
> > solution, at this point (and it's not exactly at the top of their list
> > to implement their own).
> 
> It's the same old story:  There are persistent highly annoying trouble
> makers— not even many of them— who are effectively deterred by
> blocking whatever proxies they use. Eventually they hit tor, and thus
> tor must be blocked from editing.  This abuse isn't imaginary.
> 
> The various magical nymtoken ideas would probably be acceptable— they
> just need to make it so that an unbounded supply of identities is not
> any cheaper than it already is— but they need to be implemented and
> not have a high deployment or operating cost.
> 

Yeah, the various ideas for nym systems was what I was implying and the
"limited resource" aspect of them is definitely hard to specify.

> There are some people who hold the position that instant doubling of
> identities (w/ and w/o tor) that attackers would get is not acceptable
> but with things like
> http://en.wikipedia.org/wiki/Wikipedia:Wikipedia_Signpost/2013-04-08/News_and_notes
>  and Tor's effectiveness at evading censorship I expect that most can
> be convinced that it's worth it.  Harder would be the fact that
> English Wikipedia (and many other larger Wikipedias) blocks most data
(Continue reading)

Jacob Appelbaum | 18 Apr 23:57 2013
Picon

Re: CloudFlare

Matthew Finkel:
> On Thu, Apr 18, 2013 at 09:01:21AM +0000, Matt Pagan wrote:
>>> They're based in San Francisco, along with Craigslist (which
>>> is another misguided arbitrary blocker of Tor exits).
>>> Any other SF based companies that could benefit from
>>> a visit or hackerspace talk about why they should not
>>> be blocking Tor?
>>
>> Yelp is based in San Francisco. So is Pinterest. Getting the Wikimedia
>> Foundation (also based in San Francisco) to come over would be a huge
>> victory, IMO.
>>
> 
> Wikimedia is actually willing to discuss an alternative setup if a
> usable one is found. Their current implementation is not really
> acceptable, but there also isn't really a working/implemented alternative
> solution, at this point (and it's not exactly at the top of their list
> to implement their own).

I was involved in writing the DNSBulkExitList program specifically for
Wikipedia at the request of Tim S. At the time, I believe that it was
better than simply blocking every Tor node - it only blocks exit nodes
that allow exiting to Wikipedia.

It is possible to request a special flag on a Wikipedia account that is
granted by way of some special handshake. It is possible to take an
already created account and use it for edits as the flag overrides the
Tor block.

A workable solution would be to continue to use such a list to detect
(Continue reading)

Gregory Maxwell | 19 Apr 04:34 2013
Picon

Re: CloudFlare

On Thu, Apr 18, 2013 at 2:57 PM, Jacob Appelbaum <jacob <at> appelbaum.net> wrote:
> It is possible to request a special flag on a Wikipedia account that is
> granted by way of some special handshake. It is possible to take an
> already created account and use it for edits as the flag overrides the
> Tor block.

The flag is called ipblock-exempt

You can see the the list of uses on english wikipedia that have it here:

http://en.wikipedia.org/w/index.php?title=Special%3AListUsers&username=&group=ipblock-exempt&limit=500
(bot accounts and administrators also inherit this ability without the
ipblock-exempt flag)

(As an aside, your own account was previously flagged this way, (by
Wikimedia's chairman of the board), but the flag has since been
removed because your account has been inactive:
http://en.wikipedia.org/w/index.php?title=Special%3ALog&type=&user=&page=User%3AIoerror&year=&month=-1&tagfilter=
)

[snip]
> I think we should ensure that Wikipedia understands that the account was
> created with Tor and that the user may be using this to circumvent
> censorship, to protect what they are reading or editing from their local
> network censors or surveillance regime as well as to protect IP address
> information that the US currently doesn't really protect (see USA vs.
> Appelbaum; re: my Twitter case). Since the US can see a lot of the
> traffic to Wikipedia, I'd guess that this is important worldwide.

I've been generally unable to convince people that surveillance of
(Continue reading)

Jacob Appelbaum | 19 Apr 07:07 2013
Picon

Re: CloudFlare

Gregory Maxwell:
> On Thu, Apr 18, 2013 at 2:57 PM, Jacob Appelbaum <jacob <at> appelbaum.net> wrote:
>> It is possible to request a special flag on a Wikipedia account that is
>> granted by way of some special handshake. It is possible to take an
>> already created account and use it for edits as the flag overrides the
>> Tor block.
> 
> The flag is called ipblock-exempt
> 

Right - it might make sense to make a second flag - anonymity-allowed
and set it to true for everyone until they abuse it.

> You can see the the list of uses on english wikipedia that have it here:
> 
> http://en.wikipedia.org/w/index.php?title=Special%3AListUsers&username=&group=ipblock-exempt&limit=500
> (bot accounts and administrators also inherit this ability without the
> ipblock-exempt flag)

That page is a very predictable side effect of having a flag for people
with strong need for privacy. I guess we know which Wikipedia users are
valuable or doing something interesting, right? o_0

> 
> (As an aside, your own account was previously flagged this way, (by
> Wikimedia's chairman of the board), but the flag has since been
> removed because your account has been inactive:
> http://en.wikipedia.org/w/index.php?title=Special%3ALog&type=&user=&page=User%3AIoerror&year=&month=-1&tagfilter=
> )
> 
(Continue reading)

Griffin Boyce | 19 Apr 20:18 2013
Picon

Re: CloudFlare

Jacob Appelbaum <jacob <at> appelbaum.net> wrote:

> That page is a very predictable side effect of having a flag for people
> with strong need for privacy. I guess we know which Wikipedia users are
> valuable or doing something interesting, right? o_0

  Disabling account creation by Tor users also doesn't prevent the more
insidious conflict-of-interest editing problem on Wikipedia.

  Why not allow account creation via Tor, but limit each node to X per day,
require email validation, and build in a time delay before they can be used
to edit?  Maybe block page creation and edits on semi-protected pages
entirely?  It would be annoying to spammers and vandals, but still allow
real users to get involved.

  It would be cool if accounts could be anonymity-enabled in such a way
that they could submit "anonymous" edits (where their real username is only
visible to admins or high-level moderators).  That's doable with Mediawiki
installations, so why not Wikipedia itself?

Jacob Appelbaum <jacob <at> appelbaum.net> wrote:

> That page is a very predictable side effect of having a flag for people
> with strong need for privacy. I guess we know which Wikipedia users are
> valuable or doing something interesting, right? o_0

Yeah, that's a bit problematic, especially in light of the recent French
police / Wikipedia kerfuffle.

--

-- 
(Continue reading)

Matthew Finkel | 19 Apr 04:53 2013
Picon

Re: CloudFlare

On Thu, Apr 18, 2013 at 09:57:06PM +0000, Jacob Appelbaum wrote:
> Matthew Finkel:
> > On Thu, Apr 18, 2013 at 09:01:21AM +0000, Matt Pagan wrote:
> >>> They're based in San Francisco, along with Craigslist (which
> >>> is another misguided arbitrary blocker of Tor exits).
> >>> Any other SF based companies that could benefit from
> >>> a visit or hackerspace talk about why they should not
> >>> be blocking Tor?
> >>
> >> Yelp is based in San Francisco. So is Pinterest. Getting the Wikimedia
> >> Foundation (also based in San Francisco) to come over would be a huge
> >> victory, IMO.
> >>
> > 
> > Wikimedia is actually willing to discuss an alternative setup if a
> > usable one is found. Their current implementation is not really
> > acceptable, but there also isn't really a working/implemented alternative
> > solution, at this point (and it's not exactly at the top of their list
> > to implement their own).
> 
> I was involved in writing the DNSBulkExitList program specifically for
> Wikipedia at the request of Tim S. At the time, I believe that it was
> better than simply blocking every Tor node - it only blocks exit nodes
> that allow exiting to Wikipedia.
> 

Interesting, I assume this was before Onionoo was around. I understand
why it was/is necessary.

> It is possible to request a special flag on a Wikipedia account that is
(Continue reading)

Jacob Appelbaum | 19 Apr 07:12 2013
Picon

Re: CloudFlare

Matthew Finkel:
> On Thu, Apr 18, 2013 at 09:57:06PM +0000, Jacob Appelbaum wrote:
>> Matthew Finkel:
>>> On Thu, Apr 18, 2013 at 09:01:21AM +0000, Matt Pagan wrote:
>>>>> They're based in San Francisco, along with Craigslist (which
>>>>> is another misguided arbitrary blocker of Tor exits).
>>>>> Any other SF based companies that could benefit from
>>>>> a visit or hackerspace talk about why they should not
>>>>> be blocking Tor?
>>>>
>>>> Yelp is based in San Francisco. So is Pinterest. Getting the Wikimedia
>>>> Foundation (also based in San Francisco) to come over would be a huge
>>>> victory, IMO.
>>>>
>>>
>>> Wikimedia is actually willing to discuss an alternative setup if a
>>> usable one is found. Their current implementation is not really
>>> acceptable, but there also isn't really a working/implemented alternative
>>> solution, at this point (and it's not exactly at the top of their list
>>> to implement their own).
>>
>> I was involved in writing the DNSBulkExitList program specifically for
>> Wikipedia at the request of Tim S. At the time, I believe that it was
>> better than simply blocking every Tor node - it only blocks exit nodes
>> that allow exiting to Wikipedia.
>>
> 
> Interesting, I assume this was before Onionoo was around. I understand
> why it was/is necessary.

(Continue reading)

Andrew Lewman | 18 Apr 17:32 2013
Picon

Re: CloudFlare

On Wed, 17 Apr 2013 23:59:45 -0400
grarpamp <grarpamp <at> gmail.com> wrote:

> Noticed a recent surge of sites using CloudFlare.

Actually, I've talked to cloudflare in the recent past. They don't
block Tor per se, they rate limit connections/request per IP address.

While I don't agree with this model, it seems consistent with how they
treat Tor. I can connect to cloudflare sites by forcing circuits to
exit through non-busy exit relays just fine.

--

-- 
Andrew
http://tpo.is/contact
pgp 0x6B4D6475
grarpamp | 18 Apr 21:34 2013
Picon

Re: CloudFlare

> Yelp is based in San Francisco. So is Pinterest. Getting the Wikimedia
> Foundation (also based in San Francisco) to come over would be a huge
> victory, IMO.

I'd bet there is room for some one day presentations in the big
corporate cities, sf, nyc, chi and so on. Find the businesses
and invite them.

>> Noticed a recent surge of sites using CloudFlare.
> They don't
> block Tor per se, they rate limit connections/request per IP address.

I seem to be denied with them, but that could be the apparrent
effect of a stern rate limit. So I need to do better and copy the
message/screens to the list. I often give up though.

> While I don't agree with this model, it seems consistent with how they
> treat Tor. I can connect to cloudflare sites by forcing circuits to
> exit through non-busy exit relays just fine.

Tor is already slow. So though silly, a rate limit is probably better
than a universal Tor block.

My main issue with sites that are Tor aware and then take action
against Tor nodes specifically, is that most seem to say
they get attacks, spam, illegal stuff from Tor. While true, that
is a drop in the pond when compared to from the internet at large.
Yet they don't block the internet, the coffee shops, the cable
ranges, Romania, etc. It's the being dumb about the net and the
kneejerk and the push to privacy destroying phone based auth.
(Continue reading)

Andrew Lewman | 18 Apr 22:17 2013
Picon

Re: CloudFlare

On Thu, 18 Apr 2013 15:34:21 -0400
grarpamp <grarpamp <at> gmail.com> wrote:

> My main issue with sites that are Tor aware and then take action
> against Tor nodes specifically, is that most seem to say
> they get attacks, spam, illegal stuff from Tor. While true, that
> is a drop in the pond when compared to from the internet at large.
> Yet they don't block the internet, the coffee shops, the cable
> ranges, Romania, etc. It's the being dumb about the net and the
> kneejerk and the push to privacy destroying phone based auth.

Right. They only look for the attacks and not the 99% of traffic which
is non-attack. If they looked at total traffic from Tor, they'd likely
find the normal usage vastly overwhelms the attack traffic.

Getting some real data here would be interesting. I'm constantly told
by "network security" people with more letters after their names than
years of experience that "everyone knows tor is bad traffic", but when
you push them on it, they have no idea why or even where their traffic
originates at all.  And "everyone" turns out to be companies selling
products or their certification instructor. 

Protecting networks or hosts based on rumors and hearsay is a pretty
poor way to protect anything. Empirical data should rule the decisions.

Cloudflare, google, akamai, and others would have a pretty good view of
how much traffic from Tor exits can be classified as good or bad. If
only they'd share the data or summarized results. I'm interested in the
answer, no matter what it is.

(Continue reading)

Griffin Boyce | 18 Apr 23:01 2013
Picon

Re: CloudFlare

Andrew Lewman <andrew <at> torproject.is> wrote:

> Protecting networks or hosts based on rumors and hearsay is a pretty
> poor way to protect anything. Empirical data should rule the decisions.

  Some people also make decisions based on perception of how likely traffic
is to "convert" into sales or signups.  Which is also a very problematic
way to look at things.

~Griffin

--

-- 
Please note that I do not have PGP access at this time.
OTR: saint <at> jabber.ccc.de / fontaine <at> jabber.ccc.de
grarpamp | 19 Apr 00:28 2013
Picon

Re: CloudFlare

>> Protecting networks or hosts based on rumors and hearsay is a pretty
>> poor way to protect anything. Empirical data should rule the decisions.
>
>   Some people also make decisions based on perception of how likely traffic
> is to "convert" into sales or signups.  Which is also a very problematic
> way to look at things.

Let's take bitcoin for a sales example. There are lots of people who would
like to find a site that would actually take the sale. Some of those people
have needs to trade from work, or school, or to keep their ISP from
giving away their profile... Tor does that. So these sites are LOSING
business, lots of it, because they're being stupid about where someone
*appears* to be coming from and/or whatever else they see from there.
I don't know what business school they went to, but in my book you take
the sale and stay out of the privates of your customer.

And right now there is no law anywhere in the world that says you
cannot do business with customer merely for being via Tor. And only a
couple situations with banned countries and known agents lists. All
of which are addressible by KYC, if you are so regulated, to again
permit you to accept business from wherever. With bitcoin, or even
any other type of business, actual regulation which you MUST
follow by law often doesn't exist... only 'guidance' and broadly applied
best practice.

Within that realm of always being legal... there is thin and rich with
happy customers, all the way to broad and poor with a reputation
that minus1's you.

I think Amazon takes the sale via Tor :) Hats off to Jeff.
(Continue reading)

Micah Lee | 19 Apr 23:25 2013
Picon

Re: CloudFlare

On 04/18/2013 08:32 AM, Andrew Lewman wrote:
> On Wed, 17 Apr 2013 23:59:45 -0400
> grarpamp <grarpamp <at> gmail.com> wrote:
> 
>> Noticed a recent surge of sites using CloudFlare.
> 
> Actually, I've talked to cloudflare in the recent past. They don't
> block Tor per se, they rate limit connections/request per IP address.
> 
> While I don't agree with this model, it seems consistent with how they
> treat Tor. I can connect to cloudflare sites by forcing circuits to
> exit through non-busy exit relays just fine.

Just want to jump in and say that Freedom of the Press Foundation
starting using CloudFlare to fend off a DDoS attack a couple months ago
and we've never had a single complaint from a Tor user not being able to
access it.

I've tried visiting https://pressfreedomfoundation.org over Tor myself
several times and have always gotten through.

--

-- 
https://twitter.com/micahflee

_______________________________________________
tor-talk mailing list
tor-talk <at> lists.torproject.org
https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-talk
(Continue reading)


Gmane