Two Second Memory

24 July, 2007

Note to Self — caps lock -> escape

printf "! Esc on caps Lock\n \
remove Lock = Caps_Lock\n \
keysym Caps_Lock = Escape\n"  > $HOME/.Xmodmap

Filed by rgh at 12:56 under Notes to Self
No Comments

20 July, 2007

Project Honey Pot

Well after being hounded by Mr Palmer (well he suggested that I join a couple of times) to join Project Honey Pot I finally joined and donated some MX records and put it on this website. I want to put it on Limes & Lycopene but there is no script for Ruby as of yet.

There are a number of good introductions to Project Honey Pot: here & here but it essentially boils down to laying traps for spammers (or more specifically harvesters) and then tracking them when they send spam. This information is then made available via their http:BL service which allows you to do a DNS lookup on an ip address to determine if it is a known spammer.

Given that I’ve just finished the new blog and Kathryn is still getting some spam even though she is using akismet (albeit a very small amount) I thought that I use http:BL there. Unfortunately there was no ruby code written that I could find (this is an unusual thing in it’s own right) so I’ve knocked up a small class that interfaces to the http:BL.

The code in it’s entirety is:

require 'resolv'
    require 'ipaddr'

class HttpBl

TYPE = { "0" => [:search_engine],
                 "1" => [:suspicious],
                 "2" => [:harvester],
                 "3" => [:suspicious, :harvester],
                 "4" => [:comment_spammer],
                 "5" => [:suspicious, :comment_spammer],
                 "6" => [:harvester, :comment_spammer],
                 "7" => [:suspicious, :harvester, :comment_spammer] }

# Create a new HttpBL with the key as required to access this service.
        # See http://www.projecthoneypot.org/httpbl_configure.php to get a key.
        def initialize(key)
            @key = key
        end

# Perform a lookup on the provided ip address.
        def check(ip_addr)
            ip = IPAddr.new(ip_addr)
            query = "#{@key}.#{ip.reverse.gsub("in-addr.arpa", "dnsbl.httpbl.org")}"

begin
                lookup = Resolv.getaddress(query)
                values = lookup.split(".")
                { :last_activity => values[1], :threat => values[2], :type => TYPE[values[3]], :raw_data => lookup }
            rescue Resolv::ResolvError => error
                return {}
            end
        end
    end

And that’s it. You can download a tarball here containing the above code and some tests.

Any comments welcome.

Filed by rgh at 17:27 under General
2 Comments

Brittany Spears and Semiconductor Physics

Not to be outdone by Mr Palmers post Britney Spears explains Garbage Collection here is Britney’s Guide to Semiconductor Physics.

But the thing I really love about this site is that if you do a google search for britney spears this site appears ninth. Go Britney!

Filed by rgh at 11:24 under General
No Comments

Inernet Explorer

This article posted on slashdot stating that IE is down to 70% market share in Europe is very encouraging. Microsoft has been playing silly buggers with respect to web browsers for so long it’s nice to finally see some comeuppance. IE has a very poor track record with regards to security, is wantonly non-standards compliant (something that Microsoft has excelled at for years) and is significantly lacking in useful functionality. Here are a few articles to whet your appetite: why you should dump internet explorer, Developers gripe about IE standards inaction, The Web Standards Project and there are plenty more.

So in short if you are using IE change! The most obvious choice is Mozilla but if you are using a Mac then Safari is probably a good choice. Safari is now available under windows but is still in beta. If you use a number of different platforms then again mozilla is probably the best choice.

Do it and the web developers of the world will thank you!

Filed by rgh at 10:20 under Internet
No Comments

16 July, 2007

New Blogging Engine

For the past few months I’ve been writing a new blogging engine for Limes & Lycopene. It is written in Ruby on Rails and is designed to provide the functionality need for a blogging engine but nothing more. Although i’ve tried to make it fairly minimal I’ve put some effort into doing the Right Thing™, for example the search functionality is quite sophisticated. If you want to search for ‘lycopene’ in the title you would enter title:lycopene.

Now I know most people at this stage say why on earth would you write you own blogging engine there are loads of them out there. And that would be a fair question! There are a number of reasons:

The blog is going to part of a much bigger system and I didn’t want to have to maintain and integrate a third party blog.
The existing blog (Wordpress) is extremely unreliable. To give you an example of what I mean I have a script that runs every minute and makes a five request to the blog, if the requests time out after five seconds I restart the server. The server is sometimes restarted 20 times a day. Yes I know it is widely used and therefore “must be reliable for me it isn’t.
Google Analytics would not work on Kathryn’s site but would work on mine. This seems to be a product of the theme.
The page used to write posts can only be described as evil (yes I know you can change it).
Wordpress is written using php which is being run using fcgi under lightty and is not a particularly elegant solution and uses a fair amount of memory (see this post for more details). The machine that runs Limes & Lycopene is a small memory constrained Xen VM and the less running on that machine the better.
I wanted to gain more experience using ruby on rails.

So on Saturday afternoon I released the new engine, it wasn’t a particularly pleasant experience, mainly due to exporting the existing data from Wordpress, but it is now done (and the export only needs to be done once). There are a number of defects that need to be fixed but are not serious enough to delay the release. I have to say that I’m very pleased with the results and my admiration for rails has increased dramatically. Let me know what you think.

I would also like to thank Styleshout for there fantastic website template and an even bigger thank you for releasing it under the Creative Commons Attribution 2.5 License. I had been looking for an open source site design for quite a while before I found pixel green so thank you, thank you, thank you.

As I mentioned above there are still a few defect to fix up and some more functionality that really does need to be added but once I’ve done that I will be releasing it under the GPL license. If anyone has any bright ideas for a name please let me know!

Filed by rgh at 18:27 under Internet
3 Comments

12 July, 2007

Spam

I’ve just done a brief analysis of the amount of spam that Kathryn gets on her blog. The number is … wait for it … 94.13% spam!

Kathryn had a comment from a Troll recently ranting (he’s apparently famed for being a troll) about Limes & Lycopene being broken because she uses Spam Karma. The troll proudly announced that he moderates all his comments. Go for your life matey, if you want to moderate 16998 posts then you can only be described as a fool.
So the moral of this story is make sure you use a good anti-spam tool.

Filed by rgh at 14:27 under General
No Comments