Early JMI implementation

As part of the Netbeans project Sun has released its Metadata Repository.

Homepage:
http://mdr.netbeans.org/

Standards used:
http://mdr.netbeans.org/standards.html

What's XUP?

The world has gone SOAP mad. I've been working with XWT and I recently came across XUP from MartSoft.

"XUP is a protocol for delivering events and user interface updates. It is independent of the actual UI or event model. It places no restriction on the UI component set, or the attributes or events associated with each component. Furthermore, it supports both delegation and capturing/bubbling event models."

"XUP provides a model that bridges between the traditional desktop based and web page based user interface paradigms. With XUP, events are delivered from user agent to server as SOAP messages. Programmers implement event handlers on the server side. They no longer need to process form data as URL-encoded strings. User interface changes are delivered from server to user agent as incremental updates, so end users will no longer experience slow page refreshes, and network bandwidth is conserved."

While most of this can be achieved without SOAP I guess the industry's will must prevail. I also think that relying on the network for user input/output for responsiveness and consistency is mad.

As the note says:
"Because events are dispatched over the network, to conserve network bandwidth, the user agent should not send certain types of high-frequency events, such as mouse movement and key press."

While the specification doesn't dictacte the UI model they do mention XUL an awful lot.

W3C note:
http://www.w3.org/TR/2002/NOTE-xup-20020528/

MartSoft:
http://www.martsoft.com/

XWT:
http://www.xwt.org/

Submit

This is a bit of fun. Games written in Javascript on submit buttons.

Asteroids:
http://kisrael.com/viewblog.cgi?date=2002.05.25

50 yard dash (literally):
http://kisrael.com/viewblog.cgi?date=2002.05.27

1D Tetris:
http://kisrael.com/viewblog.cgi?date=2002.05.29

More button fun:
http://kisrael.com/viewblog.cgi?date=2002.05.22

Link Annotation

Metalinker takes a very simple approach - every time you put a link into your website, metalinker uses javascript to add a link to the page on blogdex that lists other people talking who have mentioned that link. And so with just a couple of mouse-clicks people reading your site can read a webful of opinion and debate..

http://www.thinkblank.com/metalinker/

Kartoo

Take the worst from metasearch engines like Vivisimo, the presentation aspects from something like RDFViz, implement it in an appallingly slow Flash interface (or a broken HTML one) and you have Kartoo. Initially, I was pretty impressed with this little flash bestie but it's slower and more cumbersome to use than Google and provides less information per screen than any search engine. If you are going to show what's related you need more information displayed than a tradition search engine not less. It also does literal words and not concepts. Also, the size of the balls and some of the other graphical qualities seem to be meaningless.

http://www.kartoo.com/flash.php3

Suckers aren't we all?

The scam is best known as the "Nigerian email scam" or "419 scam," a reference to Nigerian penal code against fraud. Authorities now call it "advance fee fraud."

According to an FBI report, about 2,600 Americans said they were victims of the scams in 2001. Sixteen reported losses totaling $US345,000 ($A626,000); two individuals lost over $US70,000 ($A127,000) each.

http://www.smh.com.au/articles/2002/05/24/1022038469314.html

An online examples of KIF ontology browser:

http://virtual.cvut.cz/cgi-bin/kifb

Standford spec on KIF (interesting links to Knowledge Sharing at the bottom):
http://www-ksl.stanford.edu/knowledge-sharing/kif/

The NZDIS project (associated with DSTC too), SableCC grammars (KIF, FIPA and OQL), AgentCities services and UML based visualization tools:

http://nzdis.otago.ac.nz/servlet/Dis/
http://sourceforge.net/projects/sablecc/
http://nzdis.otago.ac.nz/resources/allprojects.xml

Btw, I like SableCC. We use it for work (ITQL) and I use it in RCOSjava (SimpleC).

AgentCities is a website I look at from time to time trying to understand what it is. I know it has something to do with FIPA and agents and there's some sort of federation of clients/servers but I don't really understand what it's actually doing.

http://www.agentcities.org/

RMIT has a project associated with this (there's a link to JACK an Australian made intelligent agent - struth!):
http://www.cs.rmit.edu.au/agents/protocols/

Finally Kevin and Chris can stop regailing me with their chance to become millionaires...

"Six people were arrested in South Africa over the weekend on suspicion of being involved in the infamous "Nigerian" e-mail and letter fraud.

Four of those detained were Nigerian, one was Cameroonian and the sixth was South African. Police in South Africa believe that the six are part of an international fraud and drug-dealing cartel, sending out thousands of e-mail and letters in an attempt to defraud."

http://netscape.com.com/2100-1105-918960.html

Bitkeeper and Branding

" Linux, the kernel, is often thought of as the flagship of free software, yet its current version is partially non-free. How did this happen? This problem, like the decision to use Bitkeeper, reflects the attitude of the original developer of Linux, a person who thinks that "technically better" is more important than freedom.

Value your freedom, or you will lose it, teaches history. "Don't bother us with politics," respond those who don't want to learn."

http://www.linuxworld.com/site-stories/2002/0520.rms.html

Every artist is a cannibal, every poet is a thief.

"Unless you earn your living as an intellectual property lawyer, you probably don't know that the Supreme Court has granted certiorari in Eldred v. Ashcroft, a case that will test the limits of Congress's power to extend the term of copyrights."

"Back in 1998, representatives of the Walt Disney Company came to Washington looking for help. Disney's copyright on Mickey Mouse, who made his screen debut in the 1928 cartoon short "Steamboat Willie," was due to expire in 2003, and Disney's rights to Pluto, Goofy and Donald Duck were to expire a few years later. "

"Cultural giants borrow, and so do corporate giants. Ironically, many of Disney's animated films are based on Nineteenth Century public domain works, including Snow White and the Seven Dwarfs, Cinderella, Pinocchio, The Hunchback of Notre Dame, Alice in Wonderland, and The Jungle Book (released exactly one year after Kipling's copyrights expired)."

http://writ.news.findlaw.com/commentary/20020305_sprigman.html

EDevelopment Platform Paper

About a month ago I linked to the ERights page. They presented a paper at the O'Reilly Emerging Technology conference. After spending a frustrating week signing applets and ActiveX controls this is sounding more and more attractive.

What's not to like about providing "...inter-process capability-secure distributed programming. Its cryptographic capability protocol enables mutually suspicious Java processes to cooperate safely, and its event-loop concurrency and message pipelining enable high performance deadlock free distributed pure-object computing."

The paper:
http://www.oreillynet.com/cs/weblog/view/wlg/1422

Link to someone who saw the presentation:
http://www.oreillynet.com/pub/wlg/1422

Nothing like attending cool conferences through other people.

Quake in Mozilla

Monkeyglobs is a Javascript API for OpenGL and GLUT. Javascript based Quake anyone?

Screenshots:
http://monkeyglobs.mozdev.org/screenshots.html

K-Log - Knowledge Logs

As blogging is giving consciousness to the web (or so say some) K-Logs seems to enable knowledge. Lotus Notes that works? Or extending things like Zoe farther.

"Within a corporate context, K-Logs make it possible for any employee to add knowledge to an Intranet. It's easy enough to use (start-up in less than five minutes) that it overcomes resistance. Further, K-Logs provide people that use them two immediate benefits: 1) it is a highly visible way to enhance personal brand and 2) it is a great organizing tool that you can share with co-workers (it organizes your most important information over time). There is no other better way to get employee knowledge off the desktop and out of their heads and onto an Intranet where it can be archived, browsed, and searched."

An article with John Robb:

http://writetheweb.com/read.php?item=123

Microsoft Business Intelligence

MS released SQL Server Accelerator for BI and it sounds like a couple of spreadsheets, an access database, SQL Server. I love MS material. After reading the material I'm still not sure what it is but I know that it will increase my ROI and 4 other things that all sound like the same thing (sales, customer trends, marketing and ordering and shipping).

http://www.microsoft.com/solutions/BI/evaluation/overview/

RDF and Java

Using rdf2java (of course) you can convert RDF to Java objects and back again:
http://www.dfki.uni-kl.de/frodo/rdf2java/

To validate against a schema you could use:
http://sourceforge.net/projects/sesame/
http://www.ics.forth.gr/proj/isst/RDF/
http://www.daml.org/validator/

Also, the KOAN people have some recent publications about SEAL (Semantic Portal):

http://www.aifb.uni-karlsruhe.de/WBS/Publ/pub2002.html

Video killed the Jini

Jini was pretty cool when it was announced. However, together with Sun unable to place it in a product niche combined with the heavy requirement of a JVM meant it never really took off. It looks like Sun is repositioning it as a "web service". JXTA is much better and looks like it's succeeding. Dropping the requirement of a JVM and being XML based like UPnP (copying Microsoft for a change).

It looks like UPnP will succeed in the embedded device market where Jini failed:
"Since UPnP was announced, responsibility for UPnP has moved to the UPnP Forum (www.upnp.org), which boast a large membership list of participating companies (Sony, Philips, Siemens, IBM and Pioneer to name a few - Sun was a latecomer but also joined the Forum).

Ultimately, to be successful, UPnP faces the same challenge Jini did: it needs a critical mass of devices shipping, which requires the industry to come together to create device and service specifications. This seems to be happening: On Friday the specifications by the AV working group (covering the specifications for media servers, media sinks and the transports between the two) were posted for 45 day review for forum members. Once approved, we need to start seeing AV products ship, and hopefully we will."

Author of a Jini book and almost writing a eulogy for Jini:
http://www.ericfreeman.com/stories/2002/04/28/isUpnpPluggingJinisBottle.html

Whatever happened to Jini?
http://softwaredev.earthweb.com/prog/article/0,,12064_1015771,00.html

A paper written for the WWW2002 conference about the JXTA project Edutella:
http://www2002.org/CDROM/refereed/597/index.html

Do humans do it better?

Tap looks a lot like Alpiri. The reason it turns out, is that Alpiri is Tap. The Alpiri.org redirects you to tap.stanford.edu.

http://tap.stanford.edu/ss/

http://tap.stanford.edu/index.html

http://search.alpiri.com/tap

Paper on Tap presented at W3C:
http://www.w3.org/2002/Talks/www2002-w3ct-swdemo-em/

W3C demo:
http://tap.stanford.edu/w3c.html (I typed Tim into the search bar)

After the stirring debates caused by the humourous Metacrap paper. I went looking for some human generated metadata repositories. Sure there's dmoz and such but one based on files (as the Metacrap's examples gave). It seems that Bitzi is such a site. Funnily enough it's music, text and DivX files. They have a program that you mark-up and describe the files on your hard drive.

Bitzi Top 100 (Memento, Jurassic Park 3, lots of novels):
http://bitzi.com/stat/bitprints_ranked_100

Bitzi RDF:
http://bitzi.com/openbits/datadump

Microsoft Tax

Here in QLD and in Australia the government pretty much pays a fixed amount per PC for various MS software. This is bad enough for all the Novell, Linux and other alternative PC based opertaing systems as well as Client Access Licences (like database servers).

However, the US Education department goes one better. They include Macs. Now Office is fine but you also pay for upgrades to the OS - the Microsoft OS that is. So you pay for software you either can't use as well as stuff you might not want to use.

Want more? The UK Campus agreement count licences based on the number of people employed.

The Register story has all the facts:
http://www.theregister.co.uk/content/4/25179.html

I'm moving to Peru

If there's anything I got out of "Free as in Freedom" is that you should be pedantic about open vs free.

http://www.gnu.org.pe/resmseng.html

http://www.oreilly.com/openbook/freedom/

Current State of RDF Storage and Querying

Query Languages: ICS-FORTH RQL, ILRT SquishQL, Intellidimension RDFQL, RDFPath, VERSA RDF Query Language, TRIPLE, DAML+OIL Query Language, TMQL, Ontopia Tolog.

Interesting TRIPLE is the only that supports reification. Versa is the only one that supports nested queries. Most of them do not support grouping, sorting and aggregate functions.

RDF Stores: ICS-FORTH RDF Suite, Sesame, Inkling, RDFdb, RDFStore, EOR, Redland, Jena, RDF Gateway, Triple, KAON, Cereba, Emplois K24, Ontopia Knowledge Storage.

Around half of these are Java based. Cereba and Emplois look like the only commercial offerings. One of them uses OODB, most of them use PostgreSQL. The two commercial ones look like they have their own persistance. They also give a list of the performance and scalability characteristics where available.

Cereba is an RDF Inference Engine (an automated reasoner) developed by Network Inference. Its supporting facilities are similar to the University of Manchester’s FaCT Description Logics engine, with the difference that it supports instances and concrete datatypes. Cerebra allows for collaboration and has been reported to collaborate successfully with OilEd (version 2.0 and above) and Protegé-2000 with the OIL plug-in. Demonstration versions of Cerebra® for all major platforms (Windows and Linux) are available from the download section of the Network Inference website.

http://www.networkinference.com/

The full PDF is here:
http://139.91.183.30:9090/RDF/publications/tr308.pdf