-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On 9/27/09 6:34 AM, Winfried Tilanus wrote:
> On 09/27/2009 02:32 AM, David Banes wrote:
> 
> Hi,
> 
> First of all, my apologies. I just finished reading the chapter on
> humour in the book "Native English for Nederlanders" and tried to put it
> to practice in my original posting. I think I should suggest the author
> politely that his remarks on irony were inadequate.
> 
>> I've been in IT security R&D for 20 years and can say that the crims use
>> whatever protocol will get the job done,
> 
> And that is a big compliment for XMPP: it gets its job done, whatever
> the job is...
> 
> On 27/09/2009, at 6:59 AM, Peter Saint-Andre wrote:
> 
>> http://www.rsa.com/blog/blog_entry.aspx?id=1525
>> http://www.rsa.com/blog/blog_entry.aspx?id=1515
> 
> In matter of facts, it is this one:
> http://www.rsa.com/blog/blog_entry.aspx?id=1520

Thanks.

BTW one of those blog entries said something like "download the [sic]
Jabber server", indicating that the RSA folks have not done a lot of
research about our technology.

>> Hold it. Now that the unsuspecting user has visited a fraudulent
>> website, anything is possible! The criminals could use XMPP, SMTP, HTTP,
>> IRC, or whatever they want at that point.
> 
> Correct, and as I stated above, it is a compliment for XMPP that it is
> the technology of choice for both creating a webchat and for real-time
> distribution of data.
> 
> But what struck me most, was the explicit mentioning of jabber (and not
> XMPP!) as the technology behind the scenes. In most cases the news
> articles don't care about the technologies behind the scenes of a trojan
> and just talk about its functionality. (Like some other articles on this
> case did).

Good point. I will reach out to the RSA folks soon to educate them a bit
about XMPP.

>> There is nothing particular to XMPP in these attacks, other than the
>> fact that the criminals are using a chat interface and then sending
>> information through an XMPP server that they have installed on their own
>> machines. These attacks are not being perpetrated against Jabber users,
>> but against regular older Internet users via fraudulent websites, with
>> XMPP as an information transport.
>>
>> There is nothing we can do about this kind of attack.
> 
> No, we can't prevent these attacks. But IMHO they are a concern for the
> XMPP standards foundation. One of the factors that gave IRC a bad name,
> was its use by botnet herders. Nowadays any IRC-server is suspect.
> 
> For me the news was not XMPP doing something (illegal or not), but
> Jabber (and not XMPP...) getting 'bad press' like this.

Yes.

> If we have any possibility to avoid XMPP/Jabber getting a bad name like
> IRC did, we should do so. One way of doing so, might be generating
> 'good' press about XMPP's security features. An other way might be to
> differentiate clearly between the protocol, the public Jabber network
> and private applications of the protocol. A case like this, would then
> have a much smaller impact on the public image of XMPP/Jabber.

Agreed. A post at blog.xmpp.org is in order...

>> However, we might want to look more seriously at XEP-0165 so that we can
>> help prevent similar attacks over the real XMPP network.
> 
> Proactive working on things like XEP-0165 can indeed help with
> controlling the possible damage of illegal use of XMPP. It might lessen
> the damage when the day comes that the public jabber network is a big
> time target for things like identity theft. It might also help by
> showing that insecurity is not a feature of the protocol but that the
> insecurity is a result the way the protocol is applied.

Yes, we need to make an effort to communicate about this more publicly.
More soon.

Thanks for bringing this up, Winfried!

Peter

- --
Peter Saint-Andre
https://stpeter.im/


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.8 (Darwin)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iEYEARECAAYFAkq/ZN0ACgkQNL8k5A2w/vxOpwCeIfIvYjeJSIO0Na6GruqtdoBF
8hcAoI9W9xGGpVVzkh0BrZd1vbBmZNZ8
=yGq1
-----END PGP SIGNATURE-----

Reply via email to