More about Meaning

cl-http-70-23 virtual machine

drewvid — Tue, 30 Nov 2010 16:50:12 +0000

Image by dpicker via Flickr

Here is a version of cl-http (the Common Lisp Hypermedia Server) for downloading (link), that runs on Mandriva 2006 with cumcl-18b.linux.glibc2 installed. As you probably won’t want to install this version of Mandriva on your computer I’ve created a Virtual Appliance (link), that you can download and run in VMware Player.

Mandriva 2006, is an ideal platform for small lisp projects and integrates nicely into Windows via the Openssh interface which allows applications to log into a remote machine and execute commands. I always install andLinux on my dos box because then the Xming X server is set up automatically and can be accessed from the Mandriva Virtual Machine. I just set my DISPLAY variable to the address of the Xming X server which is usually 192.168.11.1:0.0. Then when a command is executed on the Mandriva Virtual Machine using ssh (plink on windows), the Xming server is chosen for graphics output by default.

Once you’ve booted Mandiva and started iceWM and the cl-http server, just start Firefox and you’ll be presented with a start page from which you can access the documentation. To start the iceWM you’ll need to login as “plateau” with password “toor” and type startx at the shell prompt. When iceWM has started then open a shell by clicking on the terminal button in the bottom menu bar, and type in cl-http. Firefox can be started by clicking on the browser button.

The root directory for cl-http is “/usr/local/cl-http-70-23”. The patches I downloaded with this version have already been applied. One patch was rejected so I fixed the lisp code by hand. To compile cl-http just do the following:

    > cd /usr/local/cl-http-70-23
    > lisp < cmucl/start.lisp

At the end of this post you’ll find some example code. With luck, I’ll post some more examples about creating dynamic web pages within the next few weeks.

If you want to launch applications using ssh (plink) then you will need to log into the Mandriva Virtual Machine as “root”, password “toor”, and run ifconfig to find out what the IP address of the Virtual Machine is. Then commands can be executed as follows:

plink 192.168.86.128 -l plateau -pw toor xterm -fn 12×24

Where 192.168.86.128 is the IP address I obtained by executing ifconfig, plateau is the name of the remote account and toor is the password. Don’t forget to set the DISPLAY variable for user “plateau” to point to the Xming server.

This version of cl-http has not given me any problems but then I only created an interface to the Wordnet Lexicon which works fine.

Just for reference here are the important links, user names and passwords mentioned in this post:

user: “root”, password “toor”
user: “plateau”, password “toor”
cl-http-79-23 source code: (link)
virtual machine: (link)

Mandriva forks into Mageia Linux (linuxfordevices.com)
Developers Fork Mandriva Linux, Creating Mageia (linux.slashdot.org)
Welcome Mageia! (nowwhatthe.blogspot.com)
Common LISP Hypermedia Server (cl-http.org:8001)
Practical: An HTML Generation Library, the Compiler (gigamonkeys.com)
WCF Web APIs, HTTP your way (codebetter.com)
How I Can do web programming with Lisp or Scheme? – Stack Overflow (stackoverflow.com)
History: Applications that were developed on the Symbolics Lisp Machine (lispm.dyndns.org)

Example 1

;;; exporting computed URLs. A. Ritz, 17 April 2001.  Example 1
;;; This example shows how to write a computed url that displays text from a file.
;;; This is probably the simplist web page that you will ever write.
;;; Two functions are provided for displaying text from a file. The first
;;; function calls lisp functions to open and iteratively read and display the
;;; contents of the file. The second function uses the run-program feature and the unix
;;; command cat.

;;; example of the preformated text, , tag.
(defun Display-File1 (filename stream)
(setq a (open filename))
(loop
(setq str (read-line a nil))
(when (not str) (return))
(write-string "" stream)
(write-string str stream)
(write-string "<\pre>" stream)
)
)

;;; example of the preformatted text, , tag.
(defun Display-File2 (str stream)
(write-string "" stream)
(run-program "cat" (list str) :output stream)
(write-string "<\pre>" stream)
)

;;; example of the paragraph, , tag.
(defun write-paragraph (str stream)
(write-string "
" stream)
(write-string str stream)
(write-string "" stream)
)

;;; a horizontal rule
(defun hr (stream)
(write-string "")
)

;;; write the HTML for the document body.
(defun Document-Body (str stream)
(html:with-document-body (:stream stream)
(write-paragraph str stream)
(hr stream);
(Display-File1 "test.txt" stream)
(hr stream);
(Display-File2 "test.txt" stream)
(hr stream);
)
)

;;; write the HTML for the whole document.
(defun Document (str stream)
(html:with-html-document (:stream stream)
(html:with-document-preamble (:stream stream)
(html:declare-title str :stream stream)
)
(Document-Body "Testing export of computed url" stream)
)
)

;;; Function Document outputs the following HTML.
;;; 
;;; 
;;; 
;;; 
;;; 
;;; The HTML output by function Document-Body.
;;; 
;;; 

;;; Write the web page
(defmethod write-test-page ((url url:http-url) stream)
(http:with-successful-response
(stream
:html
:cache-control       (url:response-cache-control-directives url)
:content-language (url:languages url)
)
(Document "Test Page" stream)
)
)

;;; export the url
(http:export-url #u"/cl-http/test.html"
  :computed
  :response-function #'write-test-page
  :public t
  :language :en)

;;;                 Function index
;;;
;;;    http:with-successful-response
;;;    html:with-html-document
;;;    html:with-document-preamble
;;;    html:declare-title
;;;    html:with-document-body
;;;    http:export-url

SliTaz Live CD: Blog Post Enrichment

drewvid — Thu, 18 Mar 2010 13:06:41 +0000

You can download a Live CD dedicated to wordpress plugins for Blog Post Enrichment (similarity search) from here. Once you have booted the CD just start Firefox and go to this address: http://localhost/~tux/enrichment. The address is bookmarked. You will also find POPLOG and Dokuwiki installed.

Out of the available plugins for blog post enrichment, the following have been installed because they are the most stable and work on a SliTaz platform.

All of the above plugins allow you to create short posts rich in content. Now, when someone accesses your blog, they will essentially create their own post to read by interacting with the dynamic content you provide them with. Your role as an author has now turned into one of creating a framework for structuring the dynamic content added to a post so that the reader can explore it effortlessly. Because what someone reads unfolds as they interact with your post, no two visits are the same. As a result your posts will have a longer life and relevance. Also, empowering the reader in this way makes surfing the web more personal and enjoyable.

Apture and Zemanta are two products for Blog Post Enrichment that are gaining visibility on the Internet. The Apture plugin, allows bloggers to add an extra dimension to their web pages containing third party content. This dynamic layer increases the appeal of blog posts which often seem short and terse. A nice tutorial on Apture can be found here. Zemanta, a Semantic Web company, provides a plugin

to assist with online content production but this time links to related content are listed at the end of a post. In addition, unlike the other plugins mentioned here, Zemanta’s plugin analyses the text you type in automatically. Zemata’s blog contains tutorials and lots of other information. Both the Apture and Zemanta plugins are included in the Read Write Web’s top 10 Semantic Web products of 2009.

The Insights plugin is similar to the former plugins but unlike Zemanta does not analyse your text automatically and unlike Apture does not create a dynamic layer of content above a post. The Insights plugin is very good for searching the Web, or your own blog, for potential content and as a result compliments the plugins already mentioned.

Another plugin for blog post entichment that I came across just recently is wp-hover. So far I haven’t found this plugin as useful as the rest but it’s early days.

All of these plugins are superb because they allow you to create posts rich in content on most of the main blogging platforms. Unfotunately, with the exception of Zemanta, these plugins are not available for desktop blogging clients. Zemanta is available for Windows Live Writer.

There are also some authoring environments available that provide the minimum tools necessary for writing blog posts. Writerfocus is one such tool. I like this tool, because its about as close to a typewriter as you can get – just you and the printed page! Another no frills product for creative writing is PageFour. The latter editor is ideal for organising your ideas before the enrichment phase.

Finally, One of my favourite websites for writers is suite101. On this site there is a wealth of information about fundamental writing techniques.

Zemanta is coming of age (onlineobservations.net)
Blogging is fun again with Zemanta and Apture! (72suited.com)
A Chance To Engage Readers: The Zemanta Plug-In (sweetbusinesses.com)
My List of Indispensible WordPress Plugins… (lanebailey.info)
Apture.com and WordPress ROCK (frontofficebox.com)
Zemanta (securewebs.com)
How to Use Apture (problogger.net)

wordpress: gap-fill plugin

drewvid — Sun, 10 Jan 2010 23:16:30 +0000

I’ve just installed a new plugin I developed for generating gap-fill exercises on my wordpress development blog. The plugin connects with the opencalais interface to generate the forsaid exercises. When you access the page you’ll be presented with a text widget in which you can cut and paste any text you like. The default text is the transcript of a talk by Bonnie Bassler on how bacteria communicate. After cutting and pasting your text, just click on the “submit” button to generate your exercise. The words tagged by opencalais will then be displayed followed by the gap-fill exercise from which these words have been removed. A list of TED talks can be found here.

Some of the gap-fill exercises created by this plugin are pretty boring to say the least. However, I have managed to create quite a lot of exercises that fit the bill.

wp-arc-1.1: new version

drewvid — Mon, 10 Aug 2009 19:49:22 +0000

wp-arc bug fixes and improvements

HTML entities are now handled correctly and some extra code was included to remove the slashes automatically added to queries during form processing. The plugin can be viewed on my new blog. Just go to the “wp-arc demo” page and click on the submit button. Download wp-arc from here.

As an example, the following query will work when input as is, or after being run through an html entity encoder. You might like to try the entity encoder pointed to by this url.

PREFIX owl: 
PREFIX xsd: 
PREFIX rdfs: 
PREFIX rdf: 
PREFIX foaf: 
PREFIX dc: 
PREFIX : 
PREFIX dbpedia2: 
PREFIX dbpedia: 
PREFIX skos: 
PREFIX dbo: 

SELECT ?name ?birth ?death ?person WHERE {
?person dbpedia2:birthPlace
            .
?person dbo:birthdate ?birth .
?person foaf:name ?name .
?person dbo:deathdate ?death
FILTER (?birth < "1900-01-01"^^xsd:date) .
}

Here are some example shortcode commands:

1) Shortcode command for form processing:

[sparql input='form']
[/sparql]

2) Shortcode command to query a SPARQL endpoint

[sparql input='inline' endpoint='http://dbpedia.org/sparql'
      display='name']
PREFIX owl: 
PREFIX xsd: 
PREFIX rdfs: 
PREFIX rdf: 
PREFIX foaf: 
PREFIX dc: 
PREFIX : 
PREFIX dbpedia2: 
PREFIX dbpedia: 
PREFIX skos: 
PREFIX dbo: 

SELECT ?name ?birth ?death ?person WHERE {
?person dbpedia2:birthPlace
             .
?person dbo:birthdate ?birth .
?person foaf:name ?name .
?person dbo:deathdate ?death
FILTER (?birth < "1900-01-01"^^xsd:date) .
}
[/sparql]

3) shortcode command to query an RDF file

[sparql input='file'
rdf='http://foafbuilder.qdos.com/people/
andrew.ritz.myopenid.com/foaf.rdf']
[/sparql]

WordPress plugin: wp-arc

drewvid — Fri, 29 May 2009 10:18:01 +0000

While reading about SPARQL I developed a wordpress plugin (wp-arc) which I’ve decided to release into the public domain. The plugin implements a shortcode interface to the ARC RDF system. To use the plugin you’ll need to install my forms library first. Both plugins can be found here: wp-forms, wp-arc. The lastest verison, wp-arc-latest.zip, contains a few improvements. There was an extra line at the end of the old version of wp-arc.php, which may have caused compilation problems. Also the processing of HTML entities and backslashes has been improved. If you have any problems installing them please let me know because I’ve only tested them on my local installation running wordpress 2.7.1. The interface is minimal but OK for building on.

The shortcode command has four options:

input = ”form”, “inline” or “file”
endpoint = ”the SPARQL endpoint”
rdf = “an rdf filename”
display = “the query variable to display” – only one variable can be input at present

To display a form for inputing SPARQL queries just create a page containing the following shotcode with the input option set to ‘form’:

    [sparql input='form']
    [/sparql]

When the form loads you will be presented with a “submit” button and three text fields for inputting the “endpoint”, the “sparql query” and the “variable” to display.

If you don’t want to use the form interface then just input the same shortcode with the options shown below. The SPARQL query should be enclosed between the opening and closing shortcode tags.

[sparql input='inline' endpoint='http://dbpedia.org/sparql'
          display='name']
PREFIX owl: 
PREFIX xsd: 
PREFIX rdfs: 
PREFIX rdf: 
PREFIX foaf: 
PREFIX dc: 
PREFIX : 
PREFIX dbpedia2: 
PREFIX dbpedia: 
PREFIX skos: 
PREFIX dbo: 

SELECT ?name ?birth ?death ?person WHERE {
     ?person dbpedia2:birthPlace  .
     ?person dbo:birthdate ?birth .
     ?person foaf:name ?name .

     ?person dbo:deathdate ?death
     FILTER (?birth < '1900-01-01'^^xsd:date) .
}
[/sparql]

The triples in an RDF file can be displayed by inputting the following shortcode and options.

    [sparql input='file' rdf='http://foafbuilder.qdos.com/people/andrew.ritz.myopenid.com/foaf.rdf']
    [/sparql]

Sorry about the lack of documentation for these plugins. I promise to rectify the situation when time permits. I also intend to increase the number of options.

One last thing, you may need to encode the HTML entities within your sparql queries that you quote with shorcode tags. You can do that here.

WordPress & Dokuwiki on a Slitaz LiveCD

drewvid — Tue, 05 May 2009 12:23:56 +0000

You can download a Slitaz LiveCD with Dokuwiki and WordPress already installed from here. To get started just boot Slitaz and start Firefox. Once you have started Firefox you will find links to three of my favourite plugins for blog post enrichment.

Slitaz 2.0, Protégé and Prolog

drewvid — Fri, 01 May 2009 20:06:17 +0000

This month’s LiveCD contains Protégé, Prolog and the PrologTab. Protégé is a nice lightweight tool for creating customised knowledge-based applications. The LiveCD contains one such application – a rule based system (RBS) for training assessment.

Get started by booting Slitaz from the LiveCD and logging in as ‘tux’ with no password. A brief description of the demo RBS can be found by viewing Dokuwiki’s start page. The fasted way to start Dokuwiki is by clicking on the Firefox icon at the bottom of the screen. The description of the system is rather terse but should suffice. To run the Protégé ontology editor and knowledge acquisition system just double click on the relevant desktop icon.

POPLOG V15.62 is also installed. These days I tend to use POLOG as a scripting language (POPSH) so the version I have shipped with the LiveCD is compiled without the XWindows system. To start POPLOG just click on the desktop item.

In summary the Slitaz LiveCD contains the following:

Protégé
PrologTab
Demo RBS
Dokuwiki
POPLOG
POPSH

April’s flavour was delayed slightly because I decided to switch to Slitaz 2.0 at the last minute.

SliTaz and the Semantic Web

drewvid — Sun, 15 Mar 2009 19:00:37 +0000

Recently I’ve started to become interested in small linux distributions such as SliTaz, which boots from a CDROM or pendrive and resides in computer memory.

So far I’ve managed to port the following semantic web toolkits to SliTaz:

Twinkle
Protege
RDF123
ThManager
JRDF
Graphl
GrOWL/OntoBrowser
Kaon
Smore
Swoop
Mulgara
Swi Prolog
POPLOG

Each month Langtech plans to release a LiveCD with a selection of these tools on. This month you can download a LiveCD containing the Redland RDF Libraries and Twinkle. The LiveCD can be downloaded from ifile

LiveCD: the Redland RDF libraries with a little twinkle

drewvid — Wed, 04 Mar 2009 18:44:52 +0000

This month’s flavour is the SPARQL query language for searching RDF-based websites for semantic content. Information on the Semantic Web is stored in RDF files which can be converted into triples – SUBJECT PREDICATE OBJECT. With the tools on the LiveCD, you can extract all the triples in an RDF file and write SPARQL queries asking for specific information. For instance, “find me the names of all the elements in the periodic table”.

Two query tools are introduced for accessing the data in an RDF file. The first tool is called, ROQET, which is distributed with the Redland libraries. The second tool is called, TWINKLE, and is a GUI interface that wraps the ARQ sparql query engine. The tool for finding all the triples in an RDF file is called RAPPER. In addtion you will also find a tool called LQRAPS, for reconstructing RDF from tabular data.

The LiveCD can be downloaded from here. Once you have booted the LiveCD and logged in (username: hacker, no password), you should start firefox and read the contents of the start page.

In summary the LiveCD contains the following:

The Redland RDF libraries
TWINKLE
LQRAPS
Dokuwiki

If you have any queries do not hesitate to contact me (aemr@tlen.pl).

The Semantic Web and Change

drewvid — Mon, 16 Feb 2009 13:13:28 +0000

Recently I’ve been reading a lot of articles about the failure of the semantic web. What intrigues me about such discussions is that not one of them addresses these issues:

* how/who transforms a large dynamic information network like the WWW into something else?
* how does one measure change on the web?
* over what timescale does/should change take place?

I would not like to answer these questions now, but just make a few comments.

A few years ago I was interested in the semantic web but did not spend too much time trying to use any of the emerging technologies myself. The main reason for this was that I did not have the computing power at my disposal. Over the years, as I read more about the web, I eventually decided to upgrade my computing facilities so that I could experiment with some of the available technologies. Now I’m in the position where I can use many of the available tools when building my own applications and also port them to, by today’s standards, moderate sized computer systems. This possibility has been brought about by the availability of tools such as VMware and small LINUX distributions which enable developers to explore the robustness and applicability of available software in a reasonable amount of time. Therefore from my point of view the Semantic Web is already here and I’m using it. In fact I’m making more use of web 3.0 technologies in my applications than I ever did of web 2.0. That’s because web 3.0 technologies allow me to seamlessly integrate the storage, transformation and display of data.

Before I could change the way I worked there had to be alternatives. Surely that is what the web is about. Unless alternatives exist change will never happen. At present the semantic web is the only real alternative for future development. If there are others I’m not aware of them. I chose to start using Semantic Web Technologies and I’m sure that lots of others will follow. Individual users have a lot of power on the web but it takes a long time for users to mobilize their resources and effect change. Also without a well defined set of development alternatives change is not possible. Therefore the Semantic Web is a success because it does present people like myself with a real alternative which makes computational sense.

One result of my explorations into Semantic Web Technologies are the LiveCDs that I will be releasing every month. These CDs are designed to demonstrate a particular aspect of Semantic Web Technologies. This month you can download SLiTaz/POPLOG, a multi language programming environment which I’ve been using for years and quite like. While, POPLOG, does not come with any Semantic Web Tools, it does come with PROLOG, POP11 and a foreign function interface. I plan to use POPLOG to develop a high level language interface to the Redland libraries.

The semantic web may not be the future of the web but at present it is the only coherent development path that I’m aware of and for that I’m very grateful.