CNN's misuse of quotes is sorta annoying.

At the time Sendo declined to give all the reasons for the split, saying only that it had not been given access to so-called 'source code' which would allow it to tailor the software.

It is source code, damnit. Not so-called "source code".

Now I remember why I don't read CNN's so-called "website" too often. Long live boortz.

I am perfect. I know everything. My opinion in infallible. I only use the good things and eschew the bad. Anything that I don't use sucks, because I don't use it. There is no way someone else can find value in it. If they do, then they suck, too. And if it sucks, not only do I not use it, but it is my job, ne, my duty, to ensure that you don't use it.

update: Folks keep asking what inspired this. It was inspired by the massive number of "Foo Sucks" headlines I've seen in the past 4 days. We could all get pretty tired pretty quickly trying to enumerate all of the sucky things in the world. Instead, why don't we just search out and promote the good non-sucky things instead?

Be like Plato: search for The Good.

mati in #werkflow pointed me to Workflow Patterns

This site serves as a repository for workflow modelling patterns. These patterns can be used to examine the expressive power of a workflow server that you intend to work with or they can serve as a set of ideas how to implement given business requirements having some workflow server already deployed in your organization

Duncan told me about ebPML.org, which seems to contain a veritable plethora of useful BPM information.

Welcome to the home of ebPML.org, a website dedicated to the architecture and technologies of Business Process Management Systems (BPMS).

For some reason, probably due to the number of fingers most of us have, we prefer nice 'round' numbers that include lots of zeros. So, here's my 100th entry.

For some reason, probably due to the number of fingers most of us have, we prefer nice 'round' numbers that include lots of zeros.

Since I've shot my mouth off about not knowing who's behind a blog, here's me, working on my 100th entry.

(yes, intentionally self-referential)

I was alerted to Jeff Schneider's Service Oriented Enterprise blog. And he even puts his name right there on the page.

Keywords: messaging, async, service, ovaltine.

Alex Chaffee's XPath Explorer is looking really cool these days. Alex is one smart cookie.

Plus, the screenshot includes mention of cheese, so it absolutely has to be good.

Includes plugins for Eclipse and NetBeans.

XPath Explorer (XPE) is a GUI application that lets you interactively experiment with XPath. Basically, you type in a URL (to an XML or HTML document) and an XPath expression, and it displays the elements or attributes from that document which match that expression. It also displays the value (string, number, or boolean) of the expression, and (in a stunning coup de grace) displays the entire XML tree of the original document, but with the matching nodes highlighted in bold. This makes it easy to play with and debug your XPath expresions.

By popular demand, we've added a reverse gear! Now when you click on a node, XPE will generate an XPath that uniquely identifies that node. If you double-click (or click the "Use This Path" button) it will close the circle and select all nodes that match that XPath -- should be the one you clicked on, huh?

The generation algorithm basically walks up the tree until it finds the document root, or a node with an id tag (since ids are supposed to be unique in XML), then walks back down, building the path by position.

Mr. Hacking Log 2.0, whoever that is, seems to think fairly highly of himself. What's with the trend of not obviously identifying yourself on your blog?

Attack me in your blog and you'll get in the triple digits on FreeRoller!

Just so you know, as others have discovered. If you don't have any actual legitimate points, just trash me and the way I express myself in your blog! It will be a sure winner and you'll reach the triple digits onfreeroller! So have at it! Flame away! (But for now on I'll ignore you...sorry ;-) )

Kibo. Kibo. Kibo. Kibo. Kibo.

Just a note about the new classworlds uberjar functionality. Pete Kazmier has written a maven plugin to make it super-easy, if you're sync'd up to maven HEAD.

classworlds allows the creation of a single standalone jar for your project which may internally include any other additional jars that are required for your application. This allows for easy java -jar myapp.jar type of execution.

Blogging Roller is an example of what's wrong with the world...

Please do not interpret the above statements as an indication that I condone Phish.

Would there be war if we all listened to Phish, the Dead, Widespread Panic, Blueground Undergrass and all of the other happy twirly music? Listen to Phish. Do it for the children. You don't hate children, do you?

Test Frameworks for W3C Technologies could really be about any specification from any standards (or recomendations) body.

Everyone involved wants to write conformant implementations, for technical or marketing reasons. In order for conformance to mean anything, it needs to be defined. Bear in mind that conformance and certification are not the same thing: "certification" seems to be what marketing people hear when technical people say "conformance". It would be much easier to allocate resources if a case could be made that the extra effort of testing and, thus, conformance is worthwhile. Without making conformance meaningful or trying to enforce it, testing will be seen as an added bonus and will not be as central as it needs to be. Again, end users and consumers are at the losing end, since they won't be able to make sure the implementation they choose is conformant.

Personally, to me there is nothing better than a well-written, clear, concise spec. Marshall Rose's RFCs for BEEP and its bindings to TCP are excellent examples. Most RFCs are pretty good. Most specs lead by individuals (such as most RFCs) tend to have focus, whereas those authored by an organization which costs thousands of dollars to join tend to be meander muddled messes.

To its credit, the BEEP community is attempting to put together cross-implementation compatibility and conformance test suites.

The XPath 1.0 recommendation is pretty good, aside from a few ambiguities and the confusion inherent in XML namespaces. OASIS recently came out with conformance tests for XPath, but alas, they are quite tied to XSLT.

JSR-94 (Java Rule Engine API) has its share of problems.

Gilgul is an extension to Java, specifically supported by an augmented JVM.

One of the key features seems to be allowing an object reference to actually be a reference-to-a-reference but still used without the programmer being aware of the double indirection.

For those of you with C/C++ backgrounds, gilgul seems to add pointer to a pointer with automagic deference semantics.

The examples on the site seem to indicate that it could be used to solve some of the same problems solved by AOP, while at other times, seems to be a distinct architectural method (see their state pattern example).

Looks cool. Guaranteed to be stuck in academia of course, since it so radically mutates the familiar Java concepts and isn't bytecode-compatible with the JVMs in the wild.

Oh well.

Object identity is usually regarded as a corner-stone of object-oriented programming. The observation that this concept in fact combines two distinct notions lies at the heart of Gilgul's model that provides the conceptual basis for the programming language Gilgul. Its central idea is a strict separation of these notions, namely the notions of reference and comparison.

Gilgul's model can be illustrated with an implementation technique called Identity Through Indirection [KC86]: Here, a reference to an object is realized as an object oriented pointer (OOP). An OOP points to an entry in an object table which holds the actual memory address of the object (see figure 1). Since Gilgul's model needs not to be restricted to the implementation technique of an object table, it is only claimed that object references point to entries which hold referents that represent the actual objects. Contrary to Java, where references themselves are immutable, it is possible to manipulate referents in Gilgul and in this way, to replace objects at runtime.

In an effort to further consolidate the software design tools market, after purchasing both Borland and Rational (much to IBM's chagrin), Microsoft is negotiating the purchase of bob's notebook and pen in a stock-swap deal estimated to be worth $35.

Big Picture of the XML Family of Specifications is pretty cool.

Workflow Management Coalition Publishes XML Process Definition Language (XPDL) Version 1.0, because of course we don't already have enough specifications regarding workflow.

XPDL 1.0 uses the popular XML language to describe a business process. A process defined in XPDL (a set of XML statements) can be imported into any workflow engine that supports XPDL. The related objects and attributes (data associated with the process) are now also included in the XPDL process definition. The XPDL process definition can be generated by workflow modeling and simulation tools, or can be manually coded, or can be exported from another XPDL-compliant workflow engine.

Introduction to XPath Processing in Java with Jaxen is a session I'll be doing at SD-West in March, 2003.

Session Title: Introduction to XPath Processing in Java with Jaxen Time/Date: Friday (March 28, 2003) 8:30 am - 10:00 am Track: XML Programming Format: Class Level: Intermediate

Description: XPath is one of the more useful and powerful specifications developed around XML. Jaxen, a free and open-source XPath implementation can assist developers in applying XPath within their own application. This siminar will address the issues surrounding XPath, such as the XML-InfoSet, XML-Namespaces,variable bindings, and how Jaxen can be used within an organization's XML-handling application.

1 comments

Somehow, amongst the random bpfoo specification releases over the last 45 days, I missed Business Process Modeling Notation from the creators of BPML.

It aims to be a graphical notation, sorta like the UML and the spec includes mappings to BPML and BPEL4WS. So, maybe we should consider it a graphical equivalent to the XML Infoset, for business processes. Maybe that's cool. I like abstract Infosetesque things.

Conversation Support from IBM seems to be yet-another addition to the fray currently occupied by bpml, bpel4ws and wsci.

I have yet to dig deeply into it, but on the outset, it does seem closer to what I've been pondering with regards to the strengths of werkflow.

The technology proposes and implements a conversational model of component interaction. In that model, a component’s functioning is divided into two broad categories: interoperability technology and decision logic. Here, “decision logic” is a catch-all term for what is usually regarded as the true function performed by the component, regardless of how or by whom it is done. In an e-business, for example "decision logic" is often called the "business process". The interoperability technology is the software the e-business uses to communicate and interact with others, especially other e-businesses.

Test Driven Development: By Example by Kent Beck has changed my life in 48 hours. While I haven't completely gone test-first, I have radically changed the way I write code.

Specifically, I'm using a lot more fine-grained interface-based designs in order to make testing with mock objects easier. My test-cases now actually demonstrate the logic instead of just blindly exercising methods after a bazillion lines of setup code.

TDD has forced me to think about coding in the small so that I focus on exactly and only the portion I'm currenlty working on. Over the past 3 days, I've noticed that when I do jump around and revert to my old ways, I end up writing myself into a corner again.

Initially, TDD seemed much slower but in the end, elapsed-time per tested functionality is actually much lower.