Sunday, December 29, 2002


JAlbum A free web photo album generator.

Eclipsing All

ApectJ's Eclipse Home "PARC has decided to transfer AspectJ to an openly-developed project. This project will include documentation, web site, mailing lists, bug database, and sources for the AspectJ compiler (the primary tool produced by the project)." Just another part of the growing Eclipse ecosystem.

This follows the Koi Project: "The goal of the Koi project is to develop an infrastructure that can be used to create collaborative features. The addition of specific collaborate features to the Eclipse Workbench are beyond the current scope of the project. However, as a test of the infrastructure, we have created a set of sample collaborative functions that support activities such as messaging, metadata storage, and a shared calendar. These are available as part of the Koi download."

They're using a client/server architecture because "A server-based architecture provides a centralized place to store any collaborative metadata. It also permits expensive computations to be performed without impact on the responsiveness of client machines." most peer-to-peer systems exist to extend both computation and storage capacity of client/server systems. Although they do say, "Thus using Koi, there is the potential for every “client” to also be a “server”. This could form the basis of a Koi peer-to-peer architecture. We think this could be an interesting application of the Koi infrastructure and encourage experimentation in this regard."

Saturday, December 28, 2002

The Nonsense of KM

The Google cache of "The Nonsense of Knowledge Management". He likens KM (or DKM - Distributed Knowledge Management) to TQM, ERP, CRM, BPR and others (love TLAs).

"These have sometimes been called management fads and fashions, but it would be wrong to assume that, for that reason, they all lacked effectiveness when applied in organizations. Some, however, have been [downsizing] could be a recipe for industrial fact, two thirds of BPR efforts are said to have failed...Happily, it is quite easy to distinguish between 'knowledge' and 'information' in such a way as to remove ambiguity and, at the same time, demonstrate the fundamental nonsense of 'knowledge management'." Most of the papers he found used "information" as a synonym for "knowledge". He also pokes fun at major consulting firms like Accenture, Ernst and Young and KPMG and their use of "search and replace" marketing.

In the end, we are left with: "'If getting promotion, or holding your job, or finding a new one is based on the knowledge you possess - what incentive is there to reveal that knowledge and share it?...Organizations need to learn to think about problems, rather than grab at proffered 'solutions' - which often turn out to be expensive side-tracks away from the main issues.'".

Open Source Visualizer

OpenDX - "As its name implies, OpenDX is Open Source. The license allows you to freely create, distribute, and develop visualization solutions as you see fit.

OpenDX is a uniquely powerful, full-featured software package for the visualization of scientific, engineering and analytical data: Its open system design is built on a standard interface environments. And its sophisticated data model provides users with great flexibility in creating visualizations...The GUI is built on the standard interface environment: OSF/Motif(tm) and X Window System(tm). The current version supports software-rendered images on 8-, 12-, 16-, 24-, and 32-bit windows." Supports Windows, Linux, Solaris, OS X and others. A Java version is planned. The gallery of animations is pretty cool.

Friday, December 27, 2002

It's Trying to be Easy to Use it's just Failing

I'm not the best drivers, I'm very tall, and have fairly large feet. I drove a manual 1925 Chevy. It's top speed is 25 miles and cruising speed is about 15 miles. The accelerator is between the clutch and the brake. I could not just depress the accelerator, I had angle my foot between the clutch and the brake, trapping it between them. This meant I could not brake quickly. The clutch and the brake are merely levers - no hydraulic assistance. Only the two rear wheels brake, it's braking distance is fairly long, even when you push it down hard. The steering is awfully unresponsive and as the car is top heavy, turning, even at its modest speeds, it felt very dodgy. At first I could not change the gears because my legs did not fit properly; nothing is adjustable except for the driver. No one else was game to drive it. In every way this car was slower, more dangerous, less stable and convienent than modern cars.

For a comparison, here's Aaron Swartz's review of a Segway.

MS to Carry Java

Main point - IE must install Java Plugin " After reading the judges decision from the PDF helpfully linked elsewhere, I found what I was looking for - the judge does not just generically demand that any .Net implmentation must also ship with Java, but also that in particular IE must ship with the Java Plug-In and Windows Update must notify MS users of Java to make it availiable for install."

Microsoft ordered to carry Java links to the PDF of the ruling.

JPEG 2000 Interview

Interview with Dr. Daniel Lee, who currently convenes ISO’s JPEG group, "Compared to current JPEG and other common image file formats, we have seen better compression efficiency, anywhere from 30% to 60%, bearing in mind that compression efficiency is image dependent...The committee has obtained the generous offer of royalty and license-fee free conditions by its technology contributors. While one can never be certain that there will never be any assertion of patent rights that may cause problems in the future". Mozilla will support it around 1.4 according to bug 36351.

Saturday, December 21, 2002

End of On-To-Knowledge

EU funded project delivers cutting edge knowledge management tool " of the aims of the on-to-knowledge project was to identify an ontology language standard for the Internet that could be adopted by the Worldwide Web consortium, bringing the next generation web - the semantic web a step closer." The deliverables included Sesame, OntoExtract, OntoEdit and more.

Other deliverables included: visualization (the cluster map is a pretty standard ball and stick type interface), EnerSearch case study, Analysis and Requirements of Ontology Middleware (includes security (which TKS can do of course), versioning and metainformation) and an implementation of a DAML+OIL reasoner. They (BT) developed a text mining tool called ViewSum.

Burning Records

Librarians Receive Advice on Law and Reader Privacy - "The speakers all agreed that proper federal requests for data should be dutifully complied with, but only when an official court order is presented, and not just because an FBI agent asks for information. There was also a general consensus that fewer records should be kept by libraries and that even necessary records should be destroyed after use." I wonder how it will work in the future when libraries are web services.

Death of Swing

Two articles from Martin LaMonica "Open-source group broadens its reach" and "Oracle plan exposes Java rift" both focus on the continuing momentum of Eclipse and SWT. The issue does seemed to be confused around the IDE and the GUI framework.

You could say that SWT does not compete with Swing because it does not provide a platform independent widget set and it requires management of resources (why use Java then?):
"The advantages of using SWT come at a price. You now have to manage allocated resources such as fonts and colors instead of relying on automatic garbage collection to free resources. Whatever your requirements may be, knowledge of the trade-offs can help you decide which Java GUI is right for your application."

Friday, December 20, 2002

JDK 1.4.1 DP8 for OS X

"This is just a quick note to alert you guys that DP8 is available for download now! This is an important release, because it essentially represents our Beta milestone. There are still a few features not implemented, but is mostly complete. It is very important for those of you who've been staying off the bandwagon, to get on and try your stuff out and submit bugs. We're almost there! "

How to Compare URIs

"Software is commonly required to compare two URIs. Such comparisons can have two outcomes, in this document labeled "equivalent" and "different". Since URIs exist to identify resources, presumably they should be considered equivalent when they identify the same resource. This definition of equivalence is not of much practical use for reasons which include:

* Resources may have many different identifiers.
* Web architecture defines how resources are named and how their representations are interchanged, but doesn't define resource equivalence."

IEEE Article on Jena

Jena: A Semantic Web Toolkit: " The Jena download contains three implementations of the API: One stores its data in main memory, another stores its data in a relational database, and a third uses Sleepycat Software's open-source embedded database, Berkeley DB ( A system programming interface makes it easy to introduce new stores...The Berkeley DB store, like the relational store, is persistent. While it lacks the transaction support of the relational database, it is up to an order of magnitude faster than some relational databases." Also notes they're working on a RDF-S and DAML APIs.

Thursday, December 19, 2002

The Shrinking Mac

Following the Motley's McDonald's Cuts Fat where "The company announced today that it'll shutter around 175 locations and cut 400 to 600 jobs worldwide. It will also give up capital investments in four markets and completely leave three markets, all located in the Middle East and Latin America." A local story says that they're cutting staff and that " would post its first quarterly loss in its 47-year history."

Aplix Releases MIDP 2.0 and more

Aplix Ships JBlend Platform Based on J2ME MIDP 2.0...: With JBlend they're offering "...complete support for MIDP 2.0..." as well as "...optional extensions, the J2ME Wireless Messaging (JSR 120) and Mobile Media (JSR 135) APIs." MIDP 2.0 covers things like games and security. The other vendor that seems to have much of MIDP 2.0 and the optional extensions is Insignia's Mobile Fondation which seems more suited to Pocket PC, Smartphone and others.

Big designs for small devices covers four design patterns for UIs under MIDP.

Wednesday, December 18, 2002

Unstructured Data Management

the451 has released a report called "Unstructured Data Management" it divides the sector into 4 fields: document management, search and retrieval, categorization and XML databases and tools. The winners would seem to be Documentum, Verity (or Google), Inxight and amazingly Sleepycat. They do believe that in the future it will be Microsoft, Oracle and IBM.

Inxight's profile from the451 report and the full report is here. Interestingly, the companies to watch are: 80-20 Software, Octave Software, Recommind and Semagix. From what I've read no mention of RDF or the Semantic Web (although they do note people like Semagix).

Less than a Week

U.S. says no to Aussie libel lawsuit "The 4th Circuit Court of Appeals said two Connecticut newspapers could not be sued for libel in a Virginia court on the basis of allegedly defamatory articles posted on their Web sites."

Same Tune

The Microsoft/Linux Connection "There are precedents for a Microsoft about-face. "There have been other examples of Microsoft vociferously challenging technologies, only to turn on a dime and start doing it themselves," says Dwight Davis, an analyst for Summit Strategies in Seattle. Davis points to examples such as thin-client computing and Java." Yup, from stupid idea to the TCO CDs that they sent out last time.

They call me Dr Mem

"This is an attempt to dig into a typical Java programmers problem: A Java program is running for hours and weeks, finally stopping with an OutOfMemoryException. A typical profiler like JProbe or OptimizeIt can be of help, if the problem comes up within a short time, but they slow down the operation too much."

Nope, doesn't sound familar. Dr Mem.

Tuesday, December 17, 2002

Get Real, get the Semantic Web

The Semantic Web gets real, "And yet the Semantic Web may ultimately hold the answer to many of the knotty problems raised by Web services and other machine-to-machine communications schemes. A handful of major players including Adobe, Hewlett-Packard, IBM, Nokia, and even the Department of Defense have thrown resources at it--the Defense Advanced Research Projects Agency (DARPA), which founded the Internet, is a backer..."We fundamentally believe that the Semantic Web is the only way we can solve the very difficult issues of the growing complexity of information systems," says Celcorp's CEO Jas Dhillon. He acknowledges that building ontologies can be "very time consuming and painful." But, claims Dhillon, "one of our breakthroughs is that we can automatically extract the ontology of the system and the process" as the first step in developing its Celware applications. Another milestone in easing ontology development was the W3C's release last summer of a working draft for a Web Ontology Language (abbreviated OWL)."


I don't know advanced trees (as in binary, red/black, etc) but I know what I like. And I think I like this applet as an example of R-Trees:

This seems very graph/model like to me. I think you could use RTrees to do datatyping, lists and bags.

From a synopsis of the paper:
* Allows searches for contained in/overlapped with some rectangle.
* Like B+-Trees except each child key bounds all the objects below in
the hierarchy.
* Searches may now follow down multiple children because the keys can
cover overlapping area.
* Insert chooses the sub-tree that minimizes the expansion of the
bounding rectangles
* Under-full nodes are entirely re-inserted
* Nodes are split with the goal of minimizing the area of the two new
bounding rectangles

Apart from the basic RTree implementation, some other functions have
been implemented in the applet. In particular the RTree currently
supports the following operations:

* Traverse By Level: Returns a list of all nodes sorted by their level.
* Traverse Post Order: A post order traversal of the RTree nodes.
* Traverse Pre Order: A pre order traversal of the RTree nodes.
* Enclosure: Returns an enumeration of all rectangles that contain the
query rectangle or point.
* Intersection: Returns an Enumeration with all rectangles that
intersect with the query rectangle.
* Nearest Neighbor: Returns the rectangle nearest to the query point.

It looks to me like when to split R Trees is important. It talks about
when to split the overlapping of the boxes into to separate branches.
Again, playing with the applet it seems quite smart as to when to split
and group them.

I think traversing R-Trees is similar to log(n) but it looks like each
sub-tree may be visited more than once.

The paper I've been trying to finish reading is:

Friday, December 13, 2002


There's a lot of rubbish in Australian court's upside-down Internet ruling article, like implying that he could be sent to jail in Australia for opinions published in his article:
"Under the court's perverse logic, every person or organization that posts something on the Internet will need to understand and comply with the libel laws of 190 nations and who knows how many sub-national jurisdictions. That's absurd, of course.

It's also dangerous, because it encourages powerful and paranoid people to use local laws, some of which will be designed for such purposes, to stamp out unwelcome news or opinions. If we all have to temper our speech to fit the restrictions of the worst abusers of liberty, no one will say anything worth hearing."

In the physical world companies operate in jurisdictions that are most supportive of their business (like Nevada or Delaware). It's the same for publishers.

New Google Tools

It begins (already) with Froogle. "This is a smart move by Google. Segmented search is the future of search," said Ken Cassar, senior analyst at Jupiter Research. "The universe of information that the Web has to offer is simply too vast to be able to effectively handle very specific inquiries efficiently without segmentation." Google Feels a Little Froogle

"The Google Viewer displays the pages found as a result of your Google search as a continuous scrolling slide show."

"Google WebQuotes annotates the results of your Google search with comments from other websites."

Thursday, December 12, 2002

Search Engine Technologies

"A report issued by Forrester Research in September 2002 concluded soberly, "Most companies already own a search engine—one that doesn't work.""

"Questions about accessing information from different locations and devices have proved to be among the most vexing, forcing employees to waste time bouncing between company intranets and browsers, with no interface. Among companies making inroads in that realm is Divine (, which in June 2002 began offering SinglePoint Search, a tool with an open architecture that enables users to search all of their resources, in any format, simultaneously."

Goes through different approaches to searching including: clustering, linguistic analysis, natural language processing, ontology, probablistic, taxonomy and vector based. Also talks briefly about image searching. The software that does the linking in Infoworld articles, RichLink, is highlighted and even Zoe gets a mention (go email search engines).

On page 3 there's a side bar at the bottom talking about who owns the metadata. Using trademarked or copyrighted materials as metadata is fair use. Similar to the the article on Feist Publications Inc. vs. Rural Telephone Service Co. Inc. where it was ruled by the Supreme court that copyright does not protect the facts themselves, even though the compiler of the collection may have invested substantial funds, labor or both in collecting and compiling them.

Apparently, MS is in the probablistic camp and Autonomy is in both probablistic and clustering.

Creating Applications with Mozilla

The O'Reilly book, "Creating Applications with Mozilla" is available online in HTML for free. Chapters include XUL, XBL, XPCOM and RDF.

Tuesday, December 10, 2002

Novosoft Tools

A while ago, I was looking for a C to Java converter there's C2J. They also have FL and Zebra which allows you to construct business logic using UML (use case, class, state and collaboration diagrams).

Making his mark on the Internet map

This interview and background article profiles Tim Bray, "This map is very interesting," Bray begins, positioning his tall, imposing frame beside the chart. "It's from 1790 or so, and it shows part of Eastern Europe. Why is this map full of writing and not illustration? Because there's no common agreement on measurement. More than half of the map shows the scale of distance in all the different units of the day: Polish miles, Prussian miles, Ukrainian miles."

Yes we've come so far, still haven't decided on a standard measure of distance and now we have Word.

Semantic XML

Age of Discovery: "R.V. Guha, who has been thinking about these problems for longer than XML has been around, has a suggestion that I think will be helpful, a simple usable convention for identifying real objects and referencing them in any XML document. He proposes two magic attributes with a special namespace. The first would suggest that an element was a real object and would provide a way to find it—a URL or URI whose value would be in the attribute."

The paper by Guha has already changed the names from about/resource to key/ref. Already people are changing these names. I don't like the fact that all the examples include numeric keys but as it says any URI is fine.

The Thinking Web

"This fundamental shift in web publishing will have far-reaching repercussions for web search engines. Rather than visit a search engine and trawl through a flat listing of possible matches, users will be able to issue high-level information requests and receive a distilled answer.

The semantic web is intended to complement humans in areas in which they do not perform well, such as processing large volumes of information quickly or analysing large texts for certain pieces of information."

"The upshot is that the semantic web may act as a 'collective memory', augmenting individual brain power and accelerating the pace of human learning and discovery. But we will need to careful about controlling its development and our dependence on it if we wish to avoid the emergence of a dystopian digital dictator."

Monday, December 09, 2002

Opencola Pro 1.1

"Opencola has selected seven search engines that can be searched with a single query: Google, AltaVista, Yahoo!, AllTheWeb, NorthernLight, Teoma, and Dmoz. Users can add their own news sources or blog sites to include in the search and can also choose to search more than 3,000 news sources provided by Moreover. A folder is automatically created for each search.

Users can also add text-based documents (HTML, text files, Adobe Acrobat PDF, Microsoft Outlook, Microsoft Word, Microsoft PowerPoint, or Microsoft Excel) to a folder to start a search. Opencola will extract the key phrases from the documents and use them to find relevant information or relevant peers. Users then tag results items that they feel to be relevant, and the Semantic Relevance Engine scores other documents (scale of 1 to 100) on how relevant they are to the tagged documents. It then alters the query based on a user’s preferences. New search results are automatically delivered directly to folders at frequency intervals that are defined by the user (daily, weekly, or hourly).

Opencola PRO customers may access other registered users on the global Opencola Knowledge Network. Opencola analyzes similarities between peers on the Knowledge Network and can suggest a user in search results as a “relevant peer.” Opencola allows customers to set permissions for every item in every folder residing within Opencola. Users can browse folders designated as “shared” and download files."

Available for most versions of Windows. There's also an Enterprise version of the product.

Friday, December 06, 2002

Semantic Web Services

"For Semantic Web services to become a reality, a markup language must be descriptive enough that a computer can automatically determine meaning." Goes through the steps of Discovery, Invocation, Composition and Monitoring.

"the Web Services Modeling Framework (WSMF) has been proposed to facilitate the creation of Semantic Web services. It is based in part on IBM's Web Services Flow Language (WSFL), an XML language for the description of business process-driven compositions of multiple Web services."

Written by the VP of Business and Technology Development for Semaview, Inc.

The Amicalola Conference, which I have talked about before, has a paper called "A Conceptual Architecture for Semantic Web Enabled Web Services" which covers the WSMF. Also of interest is an article by Uche Ogbuji called "Supercharging WSDL with RDF" which talks about replacing WDSL and UDDI with RDF and how this would simplify web services.

Wednesday, December 04, 2002

Semantic Patent

Well this was linked to recently on the RDF Interest scratchpad about Talee's (now Semagix) patent. It is a rather broad (that's about the nicest thing I can say about it) patent on metadata storage it's number 6,311,194:

"A system and method for creating a database of metadata (metabase) of a variety of digital media content, including TV and radio content delivered on Internet. This semantic-based method captures and enhances domain or subject specific metadata of digital media content, including the specific meaning and intended use of original content. To support semantics, a WorldModel is provided that includes specific domain knowledge, ontologies as well as a set of rules relevant to the original content. The metabase may also be dynamic in that it may track changes to the any variety of accessible content, including live and archival TV and radio programming.".

I have a sense of deja vu with patent 5,684,985 which was used to claim ownership of RDF.

Why or how this is possible escapes me. I'd rather think about things like integrating Slide (a WedDAV framework) into TKS. It makes a lot of sense to me by allowing you to combine your metadata and freetext searching in the one store, stupid patents aside.

REST, SOAP and Java

The AXIS 1.1 beta is now available for download. Now I've always thought abstraction was good and AXIS pretty much takes the cake with its SOAP over JMS. JMS allowing you to use HTTP, RMI, or whatever as the protocol but giving you things like guaranteed delivery too. I also appreciate the fact that there's a WSDL2Java rather than the other way around (language to WSDL).

AXIS has a recommended reading section that includes a link to the Architectural Styles and the Design of Network-based Software Architectures. The debate over the SOAP and REST is well covered in Roots of the REST/SOAP Debate, for example. SOAP vs REST is one of the few topics where I don't have a firm opinion one way or another. If you write applications with SOAP they're just not like the Web it's that simple. The web is all about simplicity and scalability just looks after itself.