Skip to the Main Content

Note:These pages make extensive use of the latest XHTML and CSS Standards. They ought to look great in any standards-compliant modern browser. Unfortunately, they will probably look horrible in older browsers, like Netscape 4.x and IE 4.x. Moreover, many posts use MathML, which is, currently only supported in Mozilla. My best suggestion (and you will thank me when surfing an ever-increasing number of sites on the web which have been crafted to use the new standards) is to upgrade to the latest version of your browser. If that's not possible, consider moving to the Standards-compliant and open-source Mozilla browser.

September 28, 2007

Talk: Orientifold and Twisted KR Theory

I’ve alluded, before, to a seemingly interminable project I’ve been working on with Dan Freed and Greg Moore. The (original) modest goal of the project was to write down a formula for the tadpole in an orientifold background, not just rationally (as has been done before), but over the integers (i.e., including torsion).

Since the project seems to be dragging on (shifting goalposts 'n all), and since, periodically, I get asked about our result, I thought I’d provide a link to a conference talk I gave recently, on the subject. Since the talks for the conference are online, you probably could have found this lecture with a little help from Google, anyway. But here’s the direct link. (As always, you need a MathML-capable browser. And, yes, this is the boring default theme.)

Posted by distler at 12:26 AM | Permalink | Followups (1)

September 27, 2007

Twistor Yang Mills

In a previous post, I promised I would say something about Boels et al. Aside from generally being busy with other matters, I’ve been rather confused about their paper.

The idea that they want to sell is that there’s a certain (nonlocal) 6-dimensional field theory, living on projective twistor space, 𝕋\mathbb{P}\mathbb{T}'. Performing a partial gauge-fixing, and integrating over the fibers of 𝕋 4\mathbb{P}\mathbb{T}'\to \mathbb{R}^4, we obtain either conventional 4D Yang Mills, or Mansfield’s Lagrangian, depending on which gauge choice we make.

This sounds very plausible; it’s the details that I’m hazy about.

Posted by distler at 12:22 AM | Permalink | Followups (4)

September 26, 2007

svn+ssh:// and svnX

As I mentioned earlier, I recently got myself a RubyForge account, so that I could commit some critical security fixes to the main branch of Instiki.

RubyForge uses svn+ssh:// for developer access to their SVN repositories. So I went off in search of instructions for using svn+ssh:// with my favourite GUI SVN client, svnX. Unfortunately, all of the advice I found on the web was either overly-complicated or wrong (or both). So, herewith, are my notes for setting up svn+ssh:// access to RubyForge or other, similar, services.

First, let’s create a new SSH key-pair for ourselves, one which we will use exclusively with RubyForge.

  % ssh-keygen -t rsa1 -f /Users/yourname/.ssh/rubyforge

and hit return 3 times, creating a key-pair without passphrase protection. “No passphrase?!” I hear you cry. Yes, I know. This is necessary. The secret key is, nominally, only readable by you and we minimize the possible fallout should someone nonetheless manage to compromise it by using this key-pair only with this one service.

Note, also, the “-t rsa1”. RubyForge uses the SSH level-1 protocol, so we need to create a key-pair in a compatible format. For most other services, this option (and the “Protocol=...” line, below) can be omitted.

Now we upload the file ~/.ssh/ on their web form, and wait a while.

In the meantime, let’s edit ~/.ssh/config and add the lines

  Host =
  Protocol = 1
  IdentityFile = /Users/yourname/.ssh/rubyforge

to the end of the file.

We should be able to come back later and do an

  % ssh

and login (and be immediately logged out) without being prompted for a password. Good.


 % svn list svn+ssh://

should also work without being prompted for a password.

Finally, in svnX, we add an new repository with the above URL as its path, and with the User and Password fields left blank.

And. It. Just. Works.

(Here’s where it was important to create the key-pair without a passphrase. svnX has no way to prompt you for a passphrase, so the public key authentication would fail, if we actually needed to enter one.)

The same technique (modulo the bit about the level-1 Protocol) should work with other svn+ssh:// services.

Posted by distler at 11:20 PM | Permalink | Followups (3)

September 6, 2007

S5 Themes in Instiki

Ever since I added S5 slideshow support to Instiki, I’ve been itching to add support for slideshow themes. I happen to like the default theme I’ve supplied with Instiki. But choice is good, and making switching themes as simple as adding a line

slide_theme: mytheme

is even better.

So I did.

Posted by distler at 10:57 AM | Permalink | Followups (3)

September 2, 2007


How embarrassing!

Volker asked some pointed questions about the security of Instiki. I gamely tried to respond, and proffered that, among other things, Instiki has a pretty darned good XSS Sanitizer.

Well, that got me to thinking. A sanitizer is only effective on those things which it … well … sanitizes. Which prompted me to wonder: are there things which Instiki should be sanitizing, but isn’t?

Posted by distler at 1:55 AM | Permalink | Followups (1)