Toby Murray | 3 Jun 02:48 2008
Picon
Picon

Wiki Spam?

Following the RSS feed for the wiki, this page has just appeared

http://wiki.erights.org/wiki/Walnut/Secure_Distributed_Computing/w/index.php

Is this the product of a bot (that has broken reCAPTCHA, or what?

I can't seem to figure out how to remove this page. I was hoping someone
here could attend to it.

Cheers

Toby
Mark Miller | 3 Jun 03:45 2008
Picon

Re: Wiki Spam?

Thanks. I killed it. However, when I clicked on block, in order to
block the account, I was asked instead whether I wanted to block an IP
address. Don't we require accounts to edit?

On Mon, Jun 2, 2008 at 5:48 PM, Toby Murray <toby.murray@...> wrote:
> Following the RSS feed for the wiki, this page has just appeared
>
> http://wiki.erights.org/wiki/Walnut/Secure_Distributed_Computing/w/index.php
>
> Is this the product of a bot (that has broken reCAPTCHA, or what?
>
> I can't seem to figure out how to remove this page. I was hoping someone
> here could attend to it.
>
> Cheers
>
> Toby
>
>
> _______________________________________________
> e-lang mailing list
> e-lang@...
> http://www.eros-os.org/mailman/listinfo/e-lang
>

--

-- 
Text by me above is hereby placed in the public domain

 Cheers,
 --MarkM
(Continue reading)

James Graves | 3 Jun 04:23 2008

Re: Wiki Spam?

On Mon, Jun 02, 2008 at 06:45:55PM -0700, Mark Miller wrote:
> Thanks. I killed it. However, when I clicked on block, in order to
> block the account, I was asked instead whether I wanted to block an IP
> address. Don't we require accounts to edit?

Since installing reCaptcha we have allowed anonymous edits.  You have to
do the word recognition if you want to add a new link to a page or
things like that.

For the last 5 months or so, there hasn't been any wiki spam.  I'd
prefer to take a little bit of a wait-and-see approach to see if this is
going to be anywhere near as bad as it was before we implemented
reCaptcha.

James
Toby Murray | 3 Jun 22:22 2008
Picon
Picon

More wiki spam

http://wiki.erights.org/wiki/Talk:Main_Page/

Note that this is a new page -- MediaWiki distinguishes this page from
Talk:Main_Page
James Graves | 3 Jun 22:52 2008

Re: More wiki spam

Toby Murray wrote:
> http://wiki.erights.org/wiki/Talk:Main_Page/
> 
> Note that this is a new page -- MediaWiki distinguishes this page from
> Talk:Main_Page

Deleted and the IP was blocked.  Did you want sysop rights Toby?

James
Toby Murray | 3 Jun 23:42 2008
Picon
Picon

Re: More wiki spam

I don't mind -- it would only be useful to use to delete the occasional
bit of spam I s'pose.

On Tue, 2008-06-03 at 15:52 -0500, James Graves wrote:
> Toby Murray wrote:
> > http://wiki.erights.org/wiki/Talk:Main_Page/
> > 
> > Note that this is a new page -- MediaWiki distinguishes this page from
> > Talk:Main_Page
> 
> Deleted and the IP was blocked.  Did you want sysop rights Toby?
> 
> James
> _______________________________________________
> e-lang mailing list
> e-lang@...
> http://www.eros-os.org/mailman/listinfo/e-lang
Toby Murray | 4 Jun 02:47 2008
Picon
Picon

Re: More wiki spam

On Tue, 2008-06-03 at 15:52 -0500, James Graves wrote:
> Toby Murray wrote:
> > http://wiki.erights.org/wiki/Talk:Main_Page/
> > 
> > Note that this is a new page -- MediaWiki distinguishes this page from
> > Talk:Main_Page
> 
> Deleted and the IP was blocked.  Did you want sysop rights Toby?
> 

Looks like a bot running on an ISP-connected host with addresses
assigned by DHCP or similar. (Or multiple hosts of the same ISP..) At
any rate, blocking individual IPs is futile:

http://wiki.erights.org/wiki/W/index.php

created by 92.112.212.88

All of the IPs that created these "666 terror" pages belong to
pool.ukrtel.net, which is presumably addresses assigned by a DHCP
service at ukrtel.net, a European ISP.

I'm impressed if this is a bot that has broken reCaptcha, despite the
obvious annoyance.

Cheers

Tob
Baldur Johannsson | 4 Jun 07:12 2008
Picon

Re: More wiki spam

>
> I'm impressed if this is a bot that has broken reCaptcha, despite the
> obvious annoyance.
>

More likely someone just bored enaugh to do it manualy.

Without wax
-Baldur
Adrian Mettler | 4 Jun 08:00 2008
Picon

Comments requested on paper: functional purity in Joe-E

Resending with URL instead of attachment, as it appears moderator 
approval isn't happening:
http://www.cs.berkeley.edu/~amettler/purecomp.pdf

-

David Wagner and I have recently submitted our first paper relating to
our work on Joe-E, an object-capability subset of Java.  It focuses on a
simple pattern that ensures that a method is side-effect free and
behaves as a deterministic function of its arguments (i.e. invocations
with equivalent arguments will yield equivalent return values) and how
this property can be used to verify security properties.  We would be
interested in comments from more people in order to improve the paper.
There's a fair amount of time before any camera ready would be due (it's
still a while before committee decision comes back).

Thanks,
Adrian
Toby Murray | 4 Jun 15:19 2008
Picon
Picon

Re: Comments requested on paper: functional purity in Joe-E

That's a seriously interesting paper. It never occurred to me how useful
the Joe-E overlay type system could be for program verification until
now. 

(It looks like this work leverages some of the "best of both worlds"
between work on languages and process calculi. Lots of work on process
calculi, such as those derived from the pi-calculus, has focused on
developing type systems for verifying similar properties. This work
appears to "lift" that to a real-world language, leveraging properties
of the object-capability model (e.g. references can only be passed on
other references) to ensure the type system is sound. This is comparable
with the process calculi work which often leverages properties of the
underlying calculus (e.g. names can only be communicated via other names
in pi-calculus) to ensure their type systems are sound. However, your is
inherently more useful since people actually happen to write real code
in Java.)

Some comments as I'm reading, mostly small nitpicks:

Sect 3:

"In particular, references serve as capabilities, and capabilities can
be granted only be explicitly passing references. For instance, the
global scope must not cantain any capabilities."

That second sentence sounds problematic. Surely the global scope does
contain capabilities, e.g. capabilities to constructors to safe classes
and those to immutable objects, right?

Sect 4.3:
(Continue reading)


Gmane