Wilted buttercup, grey skies, and geek

I have been looking for a set of links for all of the (Linked Open Data) LOD clouds so that I could have a pictorial description of how the LOD cloud has evolved since it was first assembled. Sadly there are no links to the old LOD clouds on from the LOD cloud’s home on the web.

So I hacked together a dirty perl script to see if I could guess the URLs for the old clouds, the code looks like so :

#!/usr/bin/perl

$rootUrl = ‘http://richard.cyganiak.de/2007/10/lod/’;

#now to brute force the URL mangling to grab all of the pictures

for ($i = 2007 ; $i < 2010; $i++) {
for ($j = 1 ; $j < 13; $j++) {
for ($k = 1; $k < 32; $k++) {
$stringToFetch = $rootUrl.”lod-datasets_$i-”.make_big($j).”-”.make_big($k).”.png”;

`wget $stringToFetch`;

}
}
}

#padding out string …
sub make_big($) {
my $string = shift;
if (length($string) == 1) {
$string = “0″.$string;
}
return $string;
}

After running the script I have found the following old instances of the LOD cloud :

• http://richard.cyganiak.de/2007/10/lod/lod-datasets_2007-10-08.png
• http://richard.cyganiak.de/2007/10/lod/lod-datasets_2007-11-07.png
• http://richard.cyganiak.de/2007/10/lod/lod-datasets_2007-11-10.png
• http://richard.cyganiak.de/2007/10/lod/lod-datasets_2008-02-28.png
• http://richard.cyganiak.de/2007/10/lod/lod-datasets_2008-03-31.png
• http://richard.cyganiak.de/2007/10/lod/lod-datasets_2009-07-14.png

Finally, I still don’t think I have the first LOD cloud, would love to know its URL if anyone out there knows it!

Meh,

So we now have more links from dbpedia to other RDF resources on the web. Dbpedia now links from dbpedia classes to freebase ones, this example SPARQL query gives some examples of the described linkage. None of these make any sense to me. Why have people decided to use owl:equivalentClass and not owl:sameAs (ducks behind computer screen). Neither of the two seem correct to me, am guessing there is a skos:broader relationship (or something similar) which would be more appropriate, but at least we have come to accept that owl:sameAs tends to get abused, do we really need to loose faith in the semantics of the other OWL classes. Does anyone know why dbpedia decided to go this way?

The OWL spec defines owl:equivalentClass to be :

“A class axiom may contain (multiple) owl:equivalentClass statements. owl:equivalentClass is a built-in property that links a class description to another class description. The meaning of such a class axiom is that the two class descriptions involved have the same class extension (i.e., both class extensions contain exactly the same set of individuals).”

I really don’t think these links are suitable for the knowledge at hand. Perhaps even rdfs:subClassOf would be more fitting.

I have long been thinking about how the interweb affects the notion of practical obscurity and how one can no longer expect to be forgiven for a crime after they have served their sentence.

An example I have used for a while now is the Georgia Sex Offenders mashup

http://www.georgia-sex-offenders.com/maps/offenders.php

IMHO sites like the above one will just end up creating ghettos of sex offenders as real-estate agents start to adopt such online resources to help sell properties to future homeowners. Eventually we will see neighbourhoods of sex offenders as no family would ever choose to live next to a rehabilitated offender. The key word in the previous sentence being “rehabilitated”, as they have been released by the judicial system into the community as reformed human beings.

Now one can install an iPhone App, which tells the phone own about sex offenders in their local area, GPS/web magic, note that this only works in the US :

http://www.telegraph.co.uk/technology/apple/5918923/iPhone-app-tracks-sex-offenders.html

I believe that practical obscurity is a dying concept. At this point in the post I should stress that I DON’T classify sex offences are petty crimes, but I believe that the advent of such data on the web will set a president for other forms of crimes to be being posted to the public domain. I can easily imagine a future where all crimes committed in some US state X are posted to the web.

For example, high-school student Bob gets arrested for shop-lifting and gets a minor punishment that could be community service or something of a similar vain. Alice a classmate of Bob’s finds this so funny that she posts it to whatever cool social network she is currently a member of, pushing it into the public domain. Now after Bob has served his sentence in pre-interweb days this information would have been practically obscure, it would have been logged in a filing cabinet in some local magistrate court, and unless you had the impetus to seek out this information you would probably never have found out about it. Alice would have been able to communicate the “funny story” to her social network, but those conversation’s would not have been in the public domain. And now they would be.

Well that is all from me, would love to know if people in the US have installed this APP, and wonder how many linch mobs are going to run around US cities taking following their trusty iPhone and taking the law into their own hands. Here is a link to an article which describes some of the vigilantism which occured in the UK after the tabloid new paper “The News of the World” published a list of sex offenders in the year 2000. Do excuse the fact that I am pointing to a document on the “world socialist web site”, but it seems to report the story well

Furthermore, Kieron O’Hara, Nigel Shadbolt and I wrote a paper touching on this a while back, it can be downloaded from ECS eprints.

I have recently been given a ASUS EEEpc 701, thanks swh, and have till now found it way too slow and jerky. The 701 I have has got 1G of RAM and 4G of disk space built in, which I added a 16G flash disk too.

Thanks to Seb (old housemate), I have now downgraded the graphics driver, and as a result it is now totally usable, yay to small computers!! I followed this howto guide when attempting to fix the EeeeeePc.

These are the steps I took :

1. Edited and added the below lines to : /etc/apt/sources.list:

deb http://ppa.launchpad.net/siretart/ppa/ubuntu jaunty main
deb-src http://ppa.launchpad.net/siretart/ppa/ubuntu jaunty main


2. Imported the needed key:

sudo apt-key adv --recv-keys --keyserver keyserver.ubuntu.com 0xce90d8983e731f79


3. Installed the older driver:

$ sudo apt-get update
$ sudo apt-get install xserver-xorg-video-intel-2.4


4. Restarted X.

sudo /etc/init.d/gdm restart


And then, like magic a usable working EEEbuntu on the 701. Notes that I tend to install Fedora OSs on my machines, this is the first time I have decided to stick with an Ubuntu flavoured OS.

In order to determine how trustworthy a digital signature of a file is, you need to grab the file, the digital signature, and you will need to import the user’s public key. This wikipedia fragment describes what is meant by a “trustworthy signature” in terms of the Web of Trust.

This is the command I run to determine the level of trust of my signed foaf file.

gpg --verify --no-tty --status-fd 2 --command-fd 0 foaf.rdf.asc foaf.rdf


Which results in the folowing output :

gpg: Signature made Wed 3 Jun 23:19:52 2009 BST using RSA key ID 51F2F7EF
[GNUPG:] SIG_ID foL1PiWCT+546VnE17UG2QvWJeE 2009-06-03 1244067592
[GNUPG:] GOODSIG 9ED0B04E51F2F7EF Mischa Tuffield (Mischa@Garlik) <mischa.tuffield@garlik.com>
gpg: Good signature from "Mischa Tuffield (Mischa@Garlik) <mischa.tuffield@garlik.com>"
gpg: aka "Mischa Tuffield (http://id.ecs.soton.ac.uk/person/6914) <mmt04r@ecs.soton.ac.uk>"
[GNUPG:] VALIDSIG 18A2AF280CA59E77AE512BB39ED0B04E51F2F7EF 2009-06-03 1244067592 0 4 0 1 2 00 18A2AF280CA59E77AE512BB39ED0B04E51F2F7EF
[GNUPG:] TRUST_ULTIMATE


This is an automatic way of evaluating how trust worthy statement at the end of a URI are.

Hello All,

I had one of my hard-drives fail miserably a couple of weeks ago, and have only now recovered all of my blog content from Google’s and Yahoo’s caches. I have recreated all of the content, and have set up a cronjob to backup my sql tables, I had copies of most of the important stuff on my machine, and only lost my sql tables … d’oh….

I have set the following 301 redirects from all of the old URLs, I don’t think I have missed any out. Do shout if you find any old broken URLs on my site.

redirect 301 /blog/2009/04/02/barcamp-09/ /blog/2009/07/07/barcamp-09/
redirect 301 /blog/2009/03/24/time-machine-to-a-linux-box/ /blog/2009/07/07/timemachine-to-a-linux-box/
redirect 301 /blog/2009/03/23/signing-a-public-key/ /blog/2009/07/08/signing-someone’s-public-key/
redirect 301 /blog/2009/03/20/making-foaf-useful/ /blog/2009/07/08/making-foaf-useful/
redirect 301 /blog/2009/03/19/yum_64bit_binaries/ /blog/2009/07/07/configuring-yum-to-only-install-64bit-binaries/
redirect 301 /blog/2009/03/19/webdav_webid/ /blog/2009/07/08/enabling-a-writable-webid-with-webdav/
redirect 301 /blog/2009/03/18/foafwot/ /blog/2009/07/08/enabling-trust-in-a-foaf-document/
redirect 301 /blog/2009/03/17/signingverifyinggpg/ /blog/2009/07/08/signing-and-verifying-documents-using-gnupg/
redirect 301 /blog/2009/02/05/ah-good-work-tobyink/ /blog/2009/07/07/ah-good-work-tobyink-…/
redirect 301 /blog/2009/01/28/duplicate-dylib-libiconv2/ /blog/2009/07/07/duplicate-dylib-libiconv-2-dylib/
redirect 301 /blog/2008/12/31/ld-duplicate-symbol-mac-osx/ /blog/2009/07/07/ld-duplicate-symbol-_g_bit_nth_lsf-mac-osx-leoparddarwin/


I have used Jules blog post to have a go at iPhone tethering and it seems to work fine. Make sure that if you attempt to tether you should use Jules’s update’d howto on 2009-06-18. All you have to do is point your iPhone at the following url http://www.jules.fm/files/uk.o2.mobileconfig and you can follow the instruction on how to run it here http://richardlai.xanga.com/704930537/enable-tethering-on-iphone-30—too-easy-worldwide-carriers/

I have made a simple webpage which makes use of the W3C Geo API. The page will prompt you for your location, given you are using FF3.5, and will subsequently ask you for a WebID and some text to describe what you are up to.

The service then generates a call to another endpoint I bashed together, that takes the following cgi arguements.

webid - lat - long with an OPTIONAL alt - datetime - doing(what I am doing now field)

e.g.,

That in turn generates a FOAF person scrobble, or a FOAF Event. I have made us of the Event, Timeline, FOAF, dc, and the Geo ontologies.

So this service can be found on my site, http://mmt.me.uk/geo. It should be noted that I DO NOT store any of the information which I output on this site. I will make it HTTPS at some point, and then I will replace using Plazes.com with my own service. I would rather a world where I was running all of my own social networking from my own machine.

The code to do this is so simple. In order to do the W3C geo stuff all you need to do is write some html and javascript, like so (sorry about the indentation)

<script src="http://maps.google.com/maps?file=api&v=2&key=YOUR_API_KEY_HERE" type="text/javascript"></script>
<script type="text/javascript">
function load() {
navigator.geolocation.getCurrentPosition(showMap);
}
function showMap(position.coords) {
// (position.coords.latitude, position.coords.longitude).
if (GBrowserIsCompatible()) {
var map = new GMap2(document.getElementById("map"));
map.setCenter(new GLatLng(position.coords.latitude, position.coords.longitude), 13);
var point = new GLatLng(position.coords.latitude, position.coords.longitude);
map.addOverlay(new GMarker(point));
}
}
</script>
<div id="map" style="width: 620px; height: 310px"></div>

and this :

<body onload="load()" onunload="GUnload()">


Here are a bunch of links which I used to find out how to do this :

• http://www.skyhookwireless.com/whoweare/privacypolicy.php
• http://dev.w3.org/geo/api/spec-source.html
• http://ajaxian.com/archives/navigatorgeolocation-using-the-w3c-geolocation-api-today
• http://www.tralfamadore.com/2008/08/w3c-geolocation-api-on-iphone-with.html
• http://almaer.com/blog/using-the-w3c-geolocation-api-specification-today-extending-whereareyou
• https://developer.mozilla.org/En/Using_geolocation

This blog post follows on from my previous one signing and verifying files with GnuPG, whereby I showed (he says), in its simplest form, how one can digitally sign and verify a document. This in turn allows anyone reading the document to verify whether or not it has been tampered with since it was signed.

In this post I will describe two methods of linking to a digital signature from a RDF document. The RDF document I will be describing in this post is a FOAF document, but is is needless to say that this approach can be used from any RDF file.

The method described below makes use of the Web of Trust ontology (WOT). WOT allows for RDF documents to be signed using Digital Signatures and Public Key Cryptography.

Whilst putting together the foaf validator, which checks the semantics of a RDF document to ensure that it is a well formed foaf:PersonalProfileDocument, I came across these two different methods of using the Web of Trust ontology.

Linking to an armored digital signature using the WOT ontology from your FOAF file:

Step 0: Declare the wot namespace in the FOAF file



@prefix wot: <http://xmlns.com/wot/0.1/> .



Step 1

Add a triple from the Document pointing to the digital signature like so:



<> wot:assurance <http://foo.com/foaf.rdf.asc> .



Step 2

Add a triples associating the public key used to sign the FOAF document to the FOAF person. This can be done in one of two ways, like so:

Style 1



_:bnode0 a <http://xmlns.com/wot/0.1/PubKey> .
_:bnode0 dc:title "Public Key Bnode" .
_:bnode0 wot:fingerprint "FW89F7WF78SD8F7SD7FG21JL213192" .
_:bnode0 wot:hex_id "12A75E9B" .
_:bnode0 wot:identity <#me> .
_:bnode0 wot:pubkeyAddress <http://foo.com/me.pubkey.asc>


This is how I sign my FOAF file

Style 2



<#me> wot:hasKey _:bnode0 .
_:bnode0 a <http://xmlns.com/wot/0.1/PubKey> .
_:bnode0 wot:pubkeyAddress <http://foo.com/me.pubkey.asc>
_:bnode0 dc:title "Public Key Bnode" .
_:bnode0 wot:fingerprint "FW89F7WF78SD8F7SD7FG21JL213192" .
_:bnode0 wot:hex_id "12A75E9B" .


This is how Kjetil signs his FOAF file

These two methods of associating a publicKey to a FOAF WebID, which is in turn can be used to digitally sign a FOAF file are both supported by Garlik's FOAF validator.

In this post I will describe how you can enable write access to a file, specially a RDF one, via Apache’s HTTP server and the Web Distributed Authoring and Versioning protocol (WebDAV) extension to the HTTP protocol.

So, why would you want to do this?

I use WebDAV on my FOAF file to enable write access via Tim Berners-Lee’s Tabulator and Garlik’s foafbuilder. This technology allows me to write updates straight through the HTTP protocol, so that I don’t have to save the file to my local machine, and scp it over.

These are the configuration settings needed in your httpd.conf file:

Setting up WebDAV on a whole directory:



<VirtualHost *:80>

ServerName www.foo.com

ServerAlias foo.com

Alias / /var/www/foo/public_html/

<Location />

DAV On

AuthType Basic

AuthName "webdav"

Header set MS-Author-Via DAV

AuthUserFile /var/www/foo/passwd.dav

<LimitExcept GET HEAD OPTIONS POST>

Require user bar

</LimitExcept>

</Location>

</VirtualHost>



Enabling WebDAV for all files ending in .rdf:



<VirtualHost *:80>

ServerName www.foo.com

ServerAlias foo.com

Alias / /var/www/foo/public_html/

<Files ~ ".*\.rdf">

DAV On

AuthType Basic

AuthName "webdav"

AuthUserFile /var/www/foo/passwd.dav

Header set MS-Author-Via DAV

ForceType application/rdf+xml

<LimitExcept GET HEAD OPTIONS POST>

Require user bar

</LimitExcept>

</Files>

</VirtualHost>



It should be noted that the methods presented above allow for the files to be read normally via HTTP, as well as catering for writing via WebDAV.

WebDAV related HTTP Headers:

The correct HTTP header used to tell a client that a file is WebDAV enabled is:

MS-Author-via: DAV

Some of this information was taken from the ESW wiki’s article “EditingData”, and I should thank everyone who helped put it together.