wordtin

DataMergingWithRDF

An unknown author — 2007-04-27T16:09:04Z

An unknown author created this page on Fri Apr 27 16:09:04 GMT 2007:

This page contains draft notes towards an essay on how data merging is carried out with RDF and Semantic Web technologies

Everyone (including myself) always says that data merging is "easier" with RDF and Semantic Web technologies. But there's very little documentation that explicitly backs up that argument. Most of the discussion involves hand waving such as "it all magically happens when we use the same identifiers"

But RDF, along with a smattering of OWL and a rules language, supports a range of declarative approaches to data integration.

Laying out the options will hopefully highlight where the "easier" data merging happens. It'll also highlight where there's still room for improvement. And perhaps also suggest where people need to focus on just the useful bits of OWL.

Where We Share the Same Identifiers

Where We Have Different Identifiers

...but have a common "primary key"

...but our resources are a "primary key" themselves

...where we can infer a relationship

Main

An unknown author — 2007-04-27T15:48:44Z

An unknown author created this page on Fri Apr 27 15:48:44 GMT 2007:

My online research notebook.

DataMergingWithRDF

LinkAnnotations

26th August

Some related notes on an essay/article I'm formulating. Will probably end up in the 'blog.

APIsAndEmergence

OpenWorldModel

13th June

ExtractingMetadataFromOfficeDocuments

30th April

CommunityTopicBuilding

17th March

Bookmarklet; Bookmarklet Bootloader; Bookmarklet As Intermediary

26th January

Some notes on Little Languages. Here's a little language called Spooky for making project structures using Ant

20th December

First beta release of the FOAF-a-Matic Mark 2 is available. Currently it only handles creation of FOAF data so the import options aren't available. Thats next!

The intention for this beta is to gain feedback on the user interface and basic application. Should be better than looking at screen shots anyway!

16th December

Added FOAF-a-matic Mark 2 Change Log.

13th December

More notes in the FOAFaMaticMark2 page indicating my progress, couple of extra thoughts prompted by excellent feedback from rdfweb-dev amongst others, and some new screen shots. Beta code will be available Real Soon Now -- I really want to get the Friends data entry stuff completed first. Won't take long I promise...

Growing to like Thinlets more and more, and will probably be able to produce some helper classes for others work with the API, e.g. for dialog management, table data entry, etc. Thinking about writing a tutorial as well.

11th December

Added page for the FOAFaMaticMark2. A GUI FOAF editing tools built around the Thinlet framework.

9th December

I've been working with Java user interfaces a bit recently, taking a more detailed look at Swing than I have previously, and finding it a lot more pleasant experience than I remembered.

I've been building a scriptable user interface and application framework. ostensibly as a jumping off point for some day job utilities I'd like to build, but I'm thinking about open sourcing it instead. I'm quite pleased with how its progressing.

I've also been taking a closer look at the Thinlet framework which I quite like, but it has some quirks which I'd love to see ironed out. Here are some NotesOnThinlets (thanks to Danny Ayers for prompting me to write these up)

Note to self: should I switch to using SnipSnap?

12th September

Here's some work I did a while back on automatically Generating JSP Tag Library Documentation.

It bugs me slightly that none of the Java web application configuration formats (web.xml, JSP tag library descriptors, ejb-jar.xml, etc) use namespaces. If Sun caught up a bit with the XML world then we could freely mix together configuration options with documentation, etc.

RDF Data Mining

10th September

A list of XML Wikis

5th September

Finished tinkering with the CSS and reworking the pages. Thanks to Andrew May for helping me out with various CSS tips and particularly a hack using @import that stopped the site crashing Netscape 4.7 (it still looks ugly, but you can read it). And thanks to the Dulux paint catalogue for helping me choose a background colour. We're not painting the kitchen this colour after all, so I decided to use its (web safe) equivalent here.

Now to get on with some work, as well as rolling out the same page design across the rest of the site. Will probably use a combination of Ant and XSLT to manage the static portions.

29th August

A while back I posted a brief summary of an Architectural Forms discussion to XML-DEV. I've pulled that summary into this site, so I can expand on it and make it into a more fully-fledged article.

See: ArchitecturalForms

Nearly finished the page redesign for this site. Basically I've just refactored the JSPWiki JSP pages into chunks that I find more manageable, made them (hopefully) XHTML compliant, and altered the CSS styling. It's the latter I'm still tinkering with mostly.

btw, only I can post here, it's not a public Wiki, its a private notebook.

23rd August

I downloaded the Topologi Collaborative Markup Editor. Here are my notes about Hacking TME to add custom plugins.

21st August

Migrating content from personal Wiki into here shortly. Not sure whether to leave this as complete open access or restrict things in some way. I may try activating the RCSFileProvider.

Also need to tweak the layout of these pages so they look a bit nicer.

I did gather together a large amount of notes about Cocoon which I've actually donated to the CocoDocoWiki as a means to bootstrap that community. Seems to have worked.

Added the notes I took whilst at the XML Europe 2002 conference.

Start of a comparison of XML Pipeline Frameworks

LinkAnnotations

An unknown author — 2007-03-12T21:56:13Z

An unknown author created this page on Mon Mar 12 21:56:13 GMT 2007:

In what different ways can links be annotated to convey useful information to a user?
What different systems and/or techniques are available to build on?
How might a document be annotated with additional links?

Adaptive Navigation

From Adaptive Navigation we get:

directional guidance, e.g. "next", "previous". How is direction determined? Must it only be by the original author or publisher?
sorting of links, based on some measure of relevance? Its easy to extra links, but how does one specify their relevance, or the goal that one is working towards?
changing anchors, using symbols or colours to indicate their target (e.g. in-site, off-site) or again based on a measure of relevance
hiding or disabling or removing "non-relevant" links

Direct Link Annotation

Adding title attributes to anchors during authoring to add description/comments
Dynamically fetching information to populate title attributes, e.g. mattb's script on hackdiary fetches del.icio.us top tags for that page
popup menus triggered by mouse overs, hover, or link selection offer a richer means of annotating links

Previews

Snap Preview Anywhere, note: backlash

WebSnapr and Link Preview

See also Matt Webb on "taste", here, sampling the ends of a link.

Discussion of web previews on readwriteweb

Cooliris (browser plugin)

iReader from Syntactica uses text analysis to summarise a site. Review on readwriteweb

Statistics

Click Tagging

UniqueNamesAssumption

An unknown author — 2005-10-12T09:37:45Z

An unknown author created this page on Wed Oct 12 09:37:45 GMT 2005:

What is the Unique Names Assumption?

And how does it apply to RDF and SPARQL?

Or: why can't I have aggregate functions in SPARQL? E.g. COUNT?

Context

Comment from Andrew Newman (showed up on planetrdf, but disappeared from blog?):

' ...purists would probably say that it shouldn't be possible to write (SQL) queries that are semantically inconsistent.

To me this would be similar to extending SPARQL with COUNT for example with out resolving the semantic inconsistencies between the RDF model and one that assumes unique names.

Comment from Dan Connolloy in evaluating SPARQL w.r.t an RDF query language survey (5th April 2005):

Count the number of authors of a publication.

nope.

unique names assumption.

Whereas Jeen Broekstra responded by saying:

I have trouble with this. I think that defining a counting operation that does not count entities, but simply labels (URIs, bNodes, literals), would be useful. This does not make any unique names assumption, AFAICS. Counting would simply be a way of retrieving the number of results a query would give, without giving the actual result.

Additional Notes

This presentation on Equality in logic (Warning: PDF!) describes the UNA as:

In many applications, one makes the assumption that every object has a unique name. This is called the unique names assumption (UNA). The upshot is that a difference in name implies a difference in referent.

σ= τ ⇔ σi = τi

The unique names assumption is not true in general!!!

This document, Ontologies and Formal Statements for ISO/IEC 11179 (19th Sept 2005) says:

...it is necessary to specify whether the unique name assumption holds for a particular ontology.

The unique names assumption (UNA) asserts that different names for objects implies that they are different individuals, i.e., individuals do not have aliases. Such an assumption is used in equality testing. Unique names are called "keys" in database parlance.

The absence of the unique names assumption implies that individuals may have many names and that no inference of distinctness may be drawn by the observation of differing names. On the World Wide Web (WWW) unique names assumptions are not valid, because hosts may have more than one name, and files may have multiple links to them. In many file systems which support multiple links to a file the UNA does not hold. Finally, in email systems a user may have multiple email addresses (identities).

OWL does not have a unique names assumption (see this resolved issue description)

See also Andrew Newmans posting: Why Different Things Are the Same, in which he notes:

So using cardinality restrictions, systems can infer that if you have two values for a property when the ontology defines you should have one, it's not that you've broken cardinality, it's that those values are actually the same/equal. In Paul's example sameAs isn't required - firstPropertyOne and firstPropertyTwo are the same. This seems weird, completely backwards and very non-intuitive.

Conclusions

I'm with Jeen, a mechanism to just count labels would be a useful one. I can (just about!) understand the wider issues for ontologies, reasoning, etc. But for a query language it would just be useful to be able to perform some aggregate operations on the results.

E.g. count number of solutions.

OnlineAccountToBookmarkletGenerator

An unknown author — 2005-06-17T08:58:13Z

An unknown author created this page on Fri Jun 17 08:58:13 GMT 2005:

Introduction

FOAF has classes and properties that can be used to relate a person to an online account that they have with a given service.

Many services provide their users with bookmarklets that allow them to interact with their functionality. And increasingly these services are RESTful, meaning the URLs typically include the users account name.

Many people write additional, improved bookmarklets that supplement or extend the basic functionality of these services, sometimes in conjunction with other services. It's often difficult to track all these extra facilities.

Wouldn't it be nice if one could use a browser extension that serves you with all the latest bookmarklets for your services, without your having to track them all down?

Implementation Notes

Craft an RDF document to describe the bookmarklets.

Write some code to process a FOAF description, containing OnlineAccount information, merging that with the above to generate usable Javascript.

Output these links as an RSS feed, so that they can be subscribed to using the Firefox Live Bookmark facility. Firefox doesn't allow this, presumably because of potential security breaches

Output these links as a bookmarks file so they can be easily imported into Firefox.

Describing the Bookmarklets

<rdf:RDF
      xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
      xmlns:rdfs="http://www.w3.org/2000/01/rdf-schema#"
      xmlns:foaf="http://xmlns.com/foaf/0.1/"
      xmlns:dc="http://purl.org/dc/elements/1.1/"
      xmlns:js="http://purl.org/net/schemas/js/">

  <rdf:Description rdf:about="http://del.icio.us">
     <dc:title>del.icio.us</dc:title>
     <js:hasBookmark rdf:nodeID="#del-popup"/>
     <js:hasBookmark rdf:nodeID="#expt"/>
  </rdf:Description>

  <js:Bookmarklet rdf:nodeID="#del-popup">
     <rdf:label>Popup Post to Del.icio.us<rdf:label>
     <js:code>javascript:......$accountName....</js:code>
  </js:Bookmarklet>

  <js:Bookmark rdf:nodeID="#expt">
     <rdf:label>extisp.icio.us</rdf:label>
     <js:url>http://kevan.org/extispicious.cgi?name=$accountName</js:url>
  </js:Bookmark>

</rdf:RDF>

Generating the bookmarks is just a matter of merging this file with the users FOAF description. The accountServiceHomepage property will link the user's OnlineAccount to the bookmarklet data. The code need only replace occurences of $accountName in the bookmark(let) data to generate the working Javascript code.

The value of the js:url and js:code properties will obviously need XML escaping.

As a nicety the service should sort the bookmark(let)s by the dc:title of the service name.

TwinkleNotes

An unknown author — 2005-02-02T20:48:21Z

An unknown author created this page on Wed Feb 02 20:48:21 GMT 2005:

Feature Requests

Unzip package into subdirectory, not current directory

Bugs

Problem with opening files with spaces in path:

<jeen> com.hp.hpl.jena.shared.JenaException: Can't open: file:/C:/Documents%20and%20Set
<jeen> tings/Jeen/My%20Documents/rdf/foaf.rdf
<jeen>         at com.hp.hpl.jena.util.FileManager.readModel(FileManager.java:280)
<jeen> (sorry for the flood)

Note: can't reproduce this

Problem with optionals

<danbri> PREFIX dc:      <http://purl.org/dc/elements/1.1/>
<danbri> PREFIX foaf:    <http://xmlns.com/foaf/0.1/>
<danbri> SELECT ?name ?url
<danbri> FROM <http://danbri.org/danbri.rdf>
<danbri> WHERE (?p foaf:name ?name) [ (?p foaf:homepage ?url) ]
...
<danbri> com.hp.hpl.jena.shared.ClosedException: already closed
<danbri>         at com.hp.hpl.jena.graph.impl.GraphBase.checkOpen(GraphBase.java:62)
<danbri>         at com.hp.hpl.jena.graph.impl.GraphBase.find(GraphBase.java:223)
<danbri>         at com.hp.hpl.jena.graph.query.PatternStage.nest(PatternStage.java:131)

LeftMenu

An unknown author — 2005-01-06T21:15:50Z

An unknown author created this page on Thu Jan 06 21:15:50 GMT 2005:

Home
About
Search
Recent Changes

Unused pages
Undefined pages

Nearby

ldodds.com
Lost Boy

CocoonHTMLGenerator

An unknown author — 2004-02-24T13:59:44Z

An unknown author created this page on Tue Feb 24 13:59:44 GMT 2004:

Some notes on the Cocoon HTML generator.

Features

Can generate XHTML/XML from a local file or remote URL
Configured as generator in sitemap, including reference to properties file for configuring JTidy:

<map:generator type="html" src="org.apache.cocoon.generation.HTMLGenerator">
<jtidy-config>jtidy.properties</jtidy-config>
</map:generator>

multiple JTidy configs supported by using multiple generator definitions
caches results based on last modified date of document

Options

Useful options when screen-scraping

Preserves original case of elements and attributes, unless you switch it off with uppercase-tags, uppercase-attributes (HTML/XHTML only)

add-xml-space=true|false -- generate xml:space="preserve" attributes for pre, style, script. Ensures that whitespace is preserved correctly (as there's no DTD to check it against)
clean=true|false -- remove presentational markup, and use styles instead. Useful for reducing amount of markup you need to wade through
numeric-entities -- use numeric entities rather than named entities. Useful as you don't have to worry about declarations.
enclose-text/enclose-block-text
char-encoding
doctype
output-xhtml -- sets doctype and default namespace. Case is alway preserved, other options ignored
output-xml

InstantMessagingClients

An unknown author — 2003-09-18T13:55:43Z

An unknown author created this page on Thu Sep 18 13:55:43 GMT 2003:

A list of Java instant messaging client libraries. Not intended to be exhaustive, just some I've come across while Googling.

AIM

Joust -- low level OSCAR api
jaimlib -- supports TOC protocol
jaimbot -- Java AIM bot, includes a library
SimpleAIM -- console AIM client

MSN

JMSN -- MSN messenger clone, not actually a library, but there's likely to be stuff to reuse
MSNj -- another MSN library, looks quite old
jmml

Yahoo

Jabber

Smack
jabberc
...and probably a million more...

MusicBrainz

An unknown author — 2003-09-12T15:28:20Z

An unknown author created this page on Fri Sep 12 15:28:20 GMT 2003:

Query url: http://mm.musicbrainz.org/cgi-bin/mq_2_1.pl

Query Examples: http://www.musicbrainz.org/MM/mq_examples.html

JavaAndPGP

An unknown author — 2003-09-09T21:07:17Z

An unknown author created this page on Tue Sep 09 21:07:17 GMT 2003:

Random links collated whilst surfing for a PGP API in Java...

JCE

Java Cryptography Extension -- doesn't seem to support PGP out of the box, basically for legal (export) reasons, but there may be other service provider implementations lurking out there...

The Cryptix JCE implementation has some form of "openpgp" support. This is a clean-room implementation of JCE. There's also Cryptix PGP. This is still in alpha and has a good feature list, although key servers aren't currently supported. Last Alpha was released on 5th Feb 2003.

SlimPGP seems to offer the desired functionality as it provides a simple command line interface that can be used to create OpenPGP keys, encrypt/decrypt files and digital sign/verify texts.

BouncyCastle.org (redirected there from openjce.org). Another clean-room JCE implementation. No sign of PGP. Although the Hush Encryption Engine includes an OpenPGP implementation designed to work with this API.

pgpjava hooks into PGP using JNI. I'd rather avoid this if possible. I'd really like a pure Java API as deployment will be less of a PITA.

PGP Message Exchange formats (RFC 1991)
Open PGP Message Format (RFC 2440)
Nice description of PGP (hey I don't use this stuff everyday you know! :)

FMMark2Features

An unknown author — 2003-09-09T20:15:50Z

An unknown author created this page on Tue Sep 09 20:15:50 GMT 2003:

A list of features for the FOAFaMaticMark 2

Implemented

FOAF data entry
1. Basic FOAF properties (similar to FOAF-a-matic Mark 1)
Add Friends -- DONE
FOAF file export to file, screen, clipboard
1. export as RDF/XML -- Generates FOAF to window, file, or copies to clipboard
2. export as N3
FOAF file import from file only currently.
1. Load and edit FOAF file
2. Supports foaf:maker
3. Add friend from another FOAF file
Other Import Options
1. Very simple vCard import
Web Browser Integration
Simple HTTP GET Response module, including ability to add friends from an Autodiscovery bookmarklet

TODO

Add Documents
Add Projects
Internationalisation support -- allow the text/instructions to be easily translated
"Advanced" export options
1. Export as VCard
2. Export as Yahoo Mail format
"Advanced" import options
1. Import from Yahoo Mail format
2. Import from Advogato profile
3. Import from RSS file (with embedded FOAF)
4. Import (friends) from OPML, etc.
Publishing of FOAF files
1. Adding entries to FOAFBulletinBoard
2. FTP to website
3. FOAFnaut/Explorer pinging
Plugin API to allow pluggable import/export options (amongst other things) will be useful to allow metadata harvesting from lots of different sources.
Signing and web-of-trust stuff

Suggestions

Suggestions from the rest of the FOAF community

nearestAirport property (suggested by Jim Ley)
Libby Miller suggested image annotation as a useful feature. I could add extraction of metadata from an annotated image for stuffing into a FOAF description.

Random Ideas

Random stuff that might be cool to add

Publishing direct to a weblog? -- can't find any XML-RPC/SOAP APIs that would allow this though...Should probably look at the Blogger API 2 discussions.
Code generation direct from the FOAF schema -- shouldn't be too difficult as I've had this one in mind whilst building the framework
FOAF Finger support.

FOAFaMaticChangelog

An unknown author — 2003-09-09T20:15:26Z

An unknown author created this page on Tue Sep 09 20:15:26 GMT 2003:

If you want to follow my progress developing the FOAF-a-Matic Mark 2 then this is my change log...

9th September

Er, where did the year go? Easily distracted that's me. Basically the code is much cleaner, it now uses Jena 2 and I've created a simple Java Bean abstraction layer between the RDF data and the GUI. Whether this is a good idea or not remains to be seen. It hides the RDF details for ordinary Java developers, but at the expense of not exposing all that data.

Considering applying code generation techniques to automatically generate this bean layer to limit potential 'drift' between the API and the RDF vocabulary...

Aiming for beta-2, which will wow nobody, before end of September...

13th January

Todays mutterings on my new blog.

In code terms I've now got the friends data export working through Jena, so I'm basically on a par with beta-1 again (yawn), but with a better model and a good idea of how to add all the remaining functionality I want to have in for beta-2 (yay!).

11th January

Following the mention in the Guardian there have been a few downloads of beta-1. I've added a note next to the download link warning folk of the bug in the foaf:mbox property serialization. Imperative I get the next release out quickly so I don't befoul the FOAF waters with dodgy RDF.

I've set up a yahoo groups mailing list for reporting new releases of the software, reporting bugs, etc.

Back to hacking...

27th December

Have thrown away custom internal model in favour of using Jena, which I probably should have looked at to start with. Quite pleased that the basics only took a couple of hours (although I've broken a chunk of functionality). Makes me think that the code refactoring I've done as I've gone along is a Good Thing.

com.ldodds.foaf.thinlet.model package now largely redundant apart from a few helpers. Have extended the Jena Model interface to provide a few helpers, and have extended the in-memory implementation to support these.
Basic FOAF generation (i.e. personal info) now works entirely off of an in-memory Jena model, Friend data is currently thrown away.
N3 export now works (comes free with Jena)
Loading of data from file also basically works, again this is provided by Jena. Grabbing from URLs, etc should be similarly easy. Grabbing data from a Friends FOAF file will involve just creating a model for their data, and grabbing the requisite statements...
Deactivated HTTP interface until finished Jena-related refactoring.
Planning beta-2 for sometime early January...
Final note to self: check out this proposal. I was planning on adding an "export" feature to the FOAF-a-Matic which would generate Dublin Core descriptions for embedding in web pages. This could be an alternative option if it takes off (and similarly is another location to grab FOAF data for the FOAF Autodiscovery feature).
Further final note to self (and the LazyWeb): someone ought to make an 'Autodiscovery Bookmarklet Generator' application. It would take the details of the tag (e.g. LINK/META), info on appropriate attributes, and the end-point of look-up service, and generate a bookmarklet. Would ensure people are following 'the rules' of how to link to/embed extra content in (X)HTML...A meta-tool basically.

22nd December

Grabbed some sample FOAF data, ranging in complexity, for testing.
Added a simple HTTP server. First step in providing "FOAF Autodiscovery" feature. The latter is awaiting the FOAF loading code (just started this), but there is a feature to dump your FOAF data into the browser.
Corrected stupid bug in which mbox property was being created as a literal not a resource. mbox_sha1sum works correctly
Also added a clear button to allow friend entry fields to be reset
Corrected bug reported by Libby Miller, friend entry fields weren't being cleared after entering a new Friend

20th December

beta-1
Made the FOAFPropertyBuilder aware of the "hide email addresses" preference.
Altered code to export friend data.
Added support for entering/updating friend information
Improved table handling code with extra utility methods

16th December

Think I might use the Attribution license for licensing the FOAF-a-Matic. It'll be open source, I'd just like acknowledgement for having written it in the first place.
Now that I'm happy with the infrastructure I can finally add the support for foaf:knows.
Added helper class for working with the Thinlet table component.
Added a Services class for mediating access to useful components (e.g. UI, preferences, etc). This paves the way for scripted plugins as I can register this object (and hence its contents) with the Bean Scripting Framework.

15th December

Aiming for first beta this week. Need to add ant scripts, etc.
Decided to add support for importing LDIF format (seems to be what Mozilla use for address books). Fairly simple format.
This heuristic evaluation might be worth a look.
Added empty implementation for N3 exporter.
Added Tables class for dynamically adding/removing rows in a table
Abstracted the main Thinlet API (plus some helpers) into a separate interface: should help manage API changes, as that API is still changing
Divided processing for saving data into generation and export. The later deals with actually turning the model into RDF (and later N3) and stuffing it into a stream. The generation aspect wraps this to ensure the model is valid and then orchestrate the export so the data goes to the right place (e.g. file, clipboard, screen).
More package and class level refactorings. This might seem like wheel spinning but the code design is improving quite a bit. Ended up losing helper package as the new classes got split up and distributed elsewhere.

12th December

Uploaded new screenshots
Stole another of Ians FOAF graphics. Note to self: create CREDITS.txt and LICENSES.txt
Added helper package, and created a number of helpers by refactoring code out of the main Thinlet class. These include a DialogManager and an ExportCoordinater. Might well be able to spin out some Thinlet based utility classes along the way.
Prototyped code to add/remove friends from table on Friends tabs. Working with multiple, non-contiguous selections. Note to self: could we end up with a lot of data here?
More re-organisation of the packages

11th December

Reworked package structure to make things a bit clearer.
Altered screen generation code to use a new pop-up window.
Added new face icons for fields. Yellow for ordinary fields, green for required, and red as a validation cues.
Moved icons and XML files for GUI into separate packages
Added new menu items (Load, Save, Help, Tools) for forthcoming features, and rejigged display of tabs. Must update screenshots.
Added dialog management code to FOAFThinlet making it easier to show/close dialogs. Some of this could probably be factored out into a base class and/or utility classes.

FMRoadmap

An unknown author — 2003-09-09T20:11:42Z

An unknown author created this page on Tue Sep 09 20:11:42 GMT 2003:

Where I attempt to prioritise my feature list, alongside implementation notes...

Beta 2

Stabilise current code to reach beta-2.

Use picocontainer

At the moment the framework has some "shared services", e.g. access to the GUI, preferences, etc. This handled using a Services class and a bunch of static accessor methods to get hold of the shared state.

I don't like this, so am going to switch to using picocontainer. This will allow more flexibility, and will provide the basics of a plugin API

Build plugin API

I envisage implementing a plugin system as follows:

Plugins are packaged as separate jars, e.g. with a "plugin.xml" file.
At startup a picocontainer is created for plugins. This will give them access to the shared services
Populate that container by using ClassLoader.getResources to load all "plugin.xml" files (which may just be a format that nanocontainer can load). Picocontainer than then resolve their dependencies.
Beef up shared services, including access to the internal web server.

Implement foaf:Document, foaf:Image, etc support

Try implementing these as plugins. Encourage creation of new plugins to allow addition of metadata about books, films, music, etc, including ratings.

Signing and Encryption

This is a must. I should be able to indicate which information I want public and private, and which communities the private information is published for.

Requires a decent Java PGP library. See Java and PGP for research notes

FOAFaMaticMark2

An unknown author — 2003-09-09T20:04:58Z

An unknown author created this page on Tue Sep 09 20:04:58 GMT 2003:

This is a quick and dirty project page for the FOAF-a-Matic Mark 2: a reworking of my original Javascript FOAF-a-matic as a desktop GUI application.

The application will be released as open source, using the Attribution license from the Creative Commons

I've included some screen shots of the work in progress and my current feature list. Comments are very welcome. Please send comments directly to me and/or the rdfweb-dev mailing list. (You can't add comments directly to this Wiki).

Features and Roadmap

See FMMark2Features for a list of features that I want to implement.

See FMRoadmap for a rough plan on how I intend to get there.

Progress

See FOAFaMaticChangelog for the dirty details.

Random Design Notes

The code is written in Java and uses the Thinlet framework for building the user interface (I've got some separate NotesOnThinlets). I'm aiming to have several versions of the application for use in different environments, e.g. background "desktop FOAF collector", authoring tool, import/export tool, applet, etc.

I'm building it in Eclipse and using Junit to write the tests. Building is done via Ant.

RDF parsing is handled using the Jena 2 toolkit.

I'm trying to make the UI as friendly as possible, having recently read About Face. I've borrowed an idea from there which means that validation errors are signalled to the user without (too many) annoying dialogs -- the UI is updated to include visual cues (change of icons and tooltips) if there were problems.

Screen Shots

Startup Screen, showing initial help text. This could probably look at little better.

Personal data entry tab, entering my details...

Entering friends will look something like this...

Oops forgot to enter my email address, so the software flags an error. Note the red face indicating the problem field, and the tooltip text explaining the problem in a grammatically incorrect way :)

Aah, finally we've got some FOAF data to look at...

Downloads

IMPORTANT: There is a bug in beta-1 which causes the foaf:mbox property to be incorrectly serialized, so the generated results aren't correct unless' you chose to encrypt the email addresses, in which case the FOAF is correct. As this is the default I hope it hasn't bitten too many people. This will be fixed in the next release.

beta-1

wordtin

DataMergingWithRDF

Where We Share the Same Identifiers

Where We Have Different Identifiers

...but have a common "primary key"

...but our resources are a "primary key" themselves

...where we can infer a relationship

Main

12th October 2005

2nd February 2005

6th January 2005

24th February

18th September

26th August

13th June

30th April

17th March

26th January

20th December

16th December

13th December

11th December

9th December

12th September

10th September

5th September

29th August

23rd August

21st August

LinkAnnotations

Adaptive Navigation

Direct Link Annotation

Previews

Statistics

UniqueNamesAssumption

Context

Additional Notes

Conclusions

OnlineAccountToBookmarkletGenerator

Introduction

Implementation Notes

Describing the Bookmarklets

TwinkleNotes

Feature Requests

Bugs

LeftMenu

CocoonHTMLGenerator

Links

Features

Options

InstantMessagingClients

AIM

MSN

Yahoo

Jabber

MusicBrainz

JavaAndPGP

JCE

Related

FMMark2Features

Implemented

TODO

Suggestions

Random Ideas

FOAFaMaticChangelog

FMRoadmap

Beta 2

Use picocontainer

Build plugin API

Implement foaf:Document, foaf:Image, etc support

Signing and Encryption

FOAFaMaticMark2

Features and Roadmap

Progress

Random Design Notes

Screen Shots

Downloads

Links