David Norwood

David's Resume

2012-02-25T20:26:45Z

19 years of design and development experience, implementing and deploying software for Fortune 500 industrial clients with manufacturing and distribution business environments
Architect for JAVA and .NET SOA platforms and wireless to ERP integration solutions
14 years as a hands-on developer (still current)
JEE mentor and trainer
Instructional designer for technological course-ware
Specialist in federal and state government compliance applications
Project manager for on-and-offshore development teams for large projects
Integrator of COTS applications to custom deliverables and SOA initiatives
Team Leader and development manager with excellent motivational skills
Profound knowledge of object-oriented methodologies, database application design, web-based application development
Deep background in large-scale enterprise applications with transaction processing using JEE, .NET, XML, messaging, and other integration technologies
Wide experience with open-source cross-platform development and integration
Excellent communicator - regular speaker at technical conferences
Published on integration issues

Technical Skills

JAVA: JEE, EJB to 3.0, Servlets to 2.4, JSP to 2.3, RMI, JNDI, JDBC, SAX/DOM, J2ME, Java Mail, i18n (internationalization),JAXB, and JMS. JAAS for transaction management
.NET: Framework 1.1 to 3.5, Compact Framework to 3.5, C#
MOBILE: Android to 2.1, Objective C (iPhone), .NET
APP SERVERS: JBoss entire family to 5.1, Glassfish to 3.0.1, WebSphere to 5.1, Weblogic 7.0 to 9, iPlanet, Oracle AS to 10g
PATTERNS: MVC, Session Façade, Business Delegate, Value Object, Value List Handler, DAO, Service Activator, Aggregate Entity, Service Locator, Decorating Filter
WEB: Flex4, JSP to 2.4, STRUTS to 1.2, Eclipse to 3.5, WSAD 5.0.1,ASP.NET, PHP to 5.3, Ruby on Rails
XML: XSLT, XML Schema, XLink, XBase, XPoint, XPath, JAXP
MESSAGING: IBM MQ-Series to 5.1, JMS, CORBA, RMI
WEB SERVICES: SOAP, REST, WSDL, UDDI, SOA Architectures
RULE ENGINES: Java rules engines (ILOG), Jess, Mandarax,JBoss Rules (Drools), RETE
DATABASES: Oracle 8 to10g. DB2 to 8, PostgreSQL to 7.3.5, MySQL to 5.1, SAPDB to 7.3

Current

Employer: Hortonworks, Inc. 2012 to present

Yahoo-backed startup San Jose, CA

David is now back to teaching, and will be instructing new clients and independent students on administration and programming in Apache Hadoop.

Past

Employer: Red Hat Consulting 2009-2012

Division of Red Hat, Inc. Raleigh, NC

Returning to his roots, David again consulted with clients, delivering solutions to everyday problems. He implemented Single Sign-On (SSO) for a large Atlanta-based hotel chain. His last project was to manage a team of consultants to convert 20+ media applications from Tomcat to JBoss 5 for a cable network.

Project History and Clientele

Project: Real-time Financial Analysis 2008

$1.4BB Wall Street Financial Credit Analyst Firm via Red Hat, Inc. Atlanta, GA

This client, the world’s leading source of commercial insight on businesses, had written a large query system for its credit analysis customers, responsible for over $1.5MM in revenue per day. David was brought in by Red Hat to help determine best optimization of code and application server on JBoss 4.0.3. David presented solutions that included JBoss tuning, SQL optimization, and JVM analysis which helped to solve outages plaguing the infrastructure and development teams.

Project: Enterprise Content Management System 2006

Fox Interactive Media Los Angeles, CA

As the primary architect for this large custom-built CMS, David introduced a Service-Oriented solution (SOA) for the application. The application was comprised of text, photo,and video story management for television affiliates of this worldwide network.Launching in April 2006, the 26 station affiliate web sites driven by the CMS are to support over 1.5 million users in the first year. David was also tapped to manage part of the development team for the project, which included up to 12developers and a new offshore team (Viet Nam). The entire project from design to go-live was delivered in 4 months. Again, David introduced best-practice solutions to the development organization, including his now-famous Wiki-based design tools.

Technologies utilized:

Web Services (SOAP over HTTP)
JSP & AJAX in the web tier with web service clients
XFire for WebService management
Video, image and multimedia capture over HTTP
XML (DOM/SAX) with dom4j feeds from video streamer (www.anystream.com)
Eclipse 3.2 with custom plugins
Hibernate for Oracle 10g abstraction

Responsibilities:

Strategy,Architecture and standards
Technical design documentation(UML and use cases)
Web service definition (xsd and wsdl description)
Hands-on management and development
Autonomy in selecting tools
Timeline and deliverable decisions

Project: Worldwide Strategic Logistics Integration 2005

UPS Alpharetta,GA

The project consisted of completely redesigning an existing Visual Basic suite of applications into a J2EE enterprise-capable application deployed to BEA Weblogic 8.1. David held the position of chief architect on the project, mentoring employees on J2EE development along the way. The application has planned responsibility for managing over $1BB annual revenue, and is considered pivotal for growth based on scalability and performance improvements. David also completed over 500 hours of hands-on development for the project. As with earlier projects, David introduced several development best-practices into the group, including continuous integration and Wiki-based knowledge management.

Technologies utilized:

J2EE (Weblogic 8.1)
Eclipse 3.2
Dom4J XML
JMS to MQSeries bridge
UML
XML (DOM/SAX)
XMLBeans (BEA)

Responsibilities:

Requirements documentation
Hands-on development
Wiki evaluation,development and deployment
CruiseControl platform for continuous builds
Evaluation of procedures and tools to improve performance and quality of application

Project: SOA Strategy 2005

Kubota Manufacturing Gainesville, GA

This client, a manufacturer of small tractors based in Japan, needed a strategy for integration of their SAP-based sales and purchasing and the AS-400 based manufacturing systems used for manufacturing in Gainesville. In this short engagement, David called upon his SAP and SOA experience to prepare and present a strategy for a Service-Oriented approach.

Project: Travel Reservation System 2004-5

Datalex, Inc. Alpharetta, GA

The project consisted of massaging XML structures to facilitate fast response times from a multitude of reservation suppliers. David was called upon to create class, sequence and activity diagrams for each functional unit of development for the team. He also suggested and then implemented an open-source knowledge base (a “Wiki”) for the development teams, which included Dublin (Ireland), New Jersey, and Atlanta.

Technologies utilized:

J2EE (JBoss and Oracle 10g database)
Oracle OC4J
Eclipse 3.0M
JAXP 1.3
JBoss 3.2
UML (Poseidon)
XML (DOM/SAX)

Responsibilities:

Hands-on development
UML designs
Wiki evaluation,development and deployment
Evaluation of procedures and tools to improve performance and quality of application

Project: Employee Timesheet portal 2004

University of Massachusetts Amherst, Mass

This project was a development of a portal for all 450 employees of UMass Amherst, facilitating time tracking and approvals for management. All time entries were fed into the payroll system, based on time approvals. David accomplished the painstaking task of upgrading the Oracle application server and database from 9i to 10g.

Technologies utilized:

J2EE (Oracle 9iAS,10g server and 10g database)
Oracle OC4J
Oracle BC4J web tags
JSP 1.2
Servlets2.1
XML (DOM/SAX)
Oracle portal

Responsibilities:

Hands-on development
Upgrades of the application server and database
Evaluation of procedures and tools to improve performance and quality of application

Project: Paperless Enterprise HACCP System 2002-4

Kraft Foods Chicago, IL

The Project involved developing a JAVA and .NET-based application for the meat processing industry for USDA HACCP (Hazard Analysis and Critical Control Point) compliance. David was the architect and designer for the project, which hit deliverable #1 within3 months. The system had 3 important components: 1) collecting data from PocketPC handhelds at the shop floor level (temperatures from handhelds with attached probes) and 2)queuing the transaction via SOAP to the JAVA-based server over the corporate 802.11 wireless intranet. A 3) Java-based rules engine in the server evaluates incoming data, updates the database and compares that data with compliance to XML-based rules in the database. The main feature of the system was the ability of the administrator to change business rules on the fly without changes to the code. After development of the use cases, David designed classes for a java based rules engine. David designed the entire application and oversaw administration of the Oracle 9i database and creation of the indexes, triggers and sequences to maximize performance of the database. He implemented the integration of service requests between IIS and Apache. David also managed parts of the project and the pilot as well, which included four teams for delivery of the project, including 2 teams overseas.

Technologies utilized:

Poseidon UML
J2EE (JBoss 2.4 to 4.0), JBoss Rules
JSP 1.2
Servlets 2.1
Session façade Pattern
.NET to 3.0 (Visual Studio), Compact Framework (currently at 2.0)
XML (DOM/SAX) XSLT, XML Spy
JUnit, Log4j, Ant,Cactus, Eclipse Java IDE
Oracle 9i

Responsibilities:

Complete database model and use case development (Rational UML)
Hands-on development
Evaluation and setup of the development and testing environment, scheduling migrations between dev, fit and prod environment
Mentoring and formal classes of technical approach and best-practice to the COBOL development community to teach Java to 15 developers
Evaluation of best-of-breed procedures and tools to decrease development time and improve stability and quality of application
Code reviews
Managed a team of offshore developers (5)

Project: Web-based Offender Search Portal 2004

NC Department of Correction Raleigh, NC

This project was to implement modifications to the Offender Search capability for state agencies that needed enhanced access to the internal DOC Offender database(s). Hitting milestones within weeks of beginning, an offender search portal was implemented with enhanced search criteria including weight, gender, hair, etc. Enhancing usability, David designed a web-services based approach, which allowed for portlets that converged several business processes into a single user interface.Subsequent to David’s engagement with the DOC, a decision was made to implement portlet-based web services throughout all new applications. David also proposed implementation of a RETE-based rules engine to make the Offender Search much more powerful. The decision is pending funding.

Technologies utilized:

J2EE (JBoss 4.01):
STRUTS 2.0
Servlets2.3
XML (XML Spy) with XSLT
Junit, Log4j, Ant
Eclipse Java IDEM9
DB2 7.3 CICS

Responsibilities:

Database model onDB2
Hands-on development of the application
Evaluation and setup of the development and testing environment, scheduling migrations between dev, fit and prod environment
Mentoring of technical solutions and “best-practice” approaches to the COBOL development community to teach Java to 15 developers
Evaluation of best-of-breed procedures and tools to decrease development time and improve stability and quality of application
Code reviews

Project: Web-based timesheet data collection portal 2003

Fluor Irvine, CA

The company (world’s largest project and construction company) needed a web-based time card portal for acquiring time from remote locations to SAP. Large construction sites in different cities were not integrated, and time cards would be delayed as much as 2 weeks from time of work done to time of billing. The need was for a web solution that could consolidate timekeeping and project management hours, thereby replacing multiple PC-based “sneaker-integrated” systems. The development of the system would help reduce the license costs of SAP, allowing a large number of users access to the portal. This system used an open-source application server to create a high-availability transaction engine for messaging occasional 5,000 hourly time sheets to e*Gate, which managed integration to SAP. Also established CVS as the software management repository.

Technologies utilized:

J2EE (JBoss 3.2.3):
Servlets2.1
Eclipse Java IDE
Ant, Enhydra
Oracle 8 to 9i

Responsibilities:

Entire database model and use case development (Rational UML)
Hands-on development
Code reviews and direction to the team of 8 developers
Integration withe*Gate to SAP
Interfacing with client’s Project Manager for timelines and deliverables

Projects: SAP to WMS transaction manager for wireless data collection 2002-3

Various Clients

The project involved the creation of a mapping tool integrating a popular WMS (warehouse management system) with SAP. In real time, the system parses text or flat files to SAP IDOC transactions as well as returning IDOC transactions from SAPto the WMS system. The system allows the users to define (via an .ini parameter file) locations of inbound and outbound ETL transactions. Utilizing publish-and-subscribe JMS messaging, the project included creation an IDOC parsing routine to convert data to flat files as needed by the interfaced system (InfoScan’s D/WMS).

Technologies utilized:

J2EE (WebSphere 3.1 to 5.0)
JSP2.0
Servlets2.3
JMS
Visual Basic 5
SAP IDOCS
XML and XSLT
WSAD 3.1 IDE
Oracle 8 Windows

Responsibilities:

Database and model with UML (Rational ClearCase)
Integration model to SAP utilizing messaging to IDOCS
Hands-on development
Code reviews

Project: SAP Web 2001-2

UniLever

This worldwide client wished to web-enable an ERP application for hundreds of users. The process was to develop a user interface for the entry of shop floor data, replacing functionality in the ERP solution. The data entered was then transferred from the Web application to the back end ERP system thru messaging software (MQ-Series) for the transaction, as well as for reports. Data was then updated into the ERP application for the transaction. The JSP user-interface provided for Sales Orders, Sales Order Shipment, Purchase Orders, and Journal Vouchers. David developed the UML for the application,defined the queue manager, queues and the Channels for MQ-Series to integrate the data between the Web Server and the UNIX Server, which had the backend ERP solution. He designed the data format to be integrated to MQ-Series queue and triggering of the Queues. He also created UNIX scripts needed to start the ERP session to upload the data. Reports were also managed via the HTML front end.Report criteria entered was transferred to the ERP system via JAVA RMI methods.

Technologies utilized:

J2EE (WebSphere 3.5.1):
JSP 2.0
Servlets 2.3
MQ-Series 5
RMI
WSAD IDE
Oracle 8 Windows

Responsibilities:

Database model and use case development
Hands-on development
Evaluation of best-of-breed procedures and tools to decrease development time and improve stability and quality of application.
Code reviews
Managed 3 off site teams of developers
Project reviews with the client

Project: Strategic Distributed Procurement System (Sun) 1996-2000

General Electric Power Systems Atlanta, GA

David managed implementations for solutions ranging from 50 to 1000 users on distributed technologies for GE's EDI subsystems. He also was responsible for managing client requirements and deliverables for a team of 12. The project was so high profile that it was written up in Business Week as a case study for Jack Welch’s management achievements.

Responsibilities:

Entire redesign of the database into 4th Normal Form
Hands-on development
Evaluation and setup of the development and testing environment, scheduling migrations between dev, fit and prod environment
Code reviews
Providing to PM timelines and development plan of deliverables

Project: 911 Call Center Application 1993-6

Electronic Data Systems Raleigh, NC

David was chief architect for an EDS product, a 911 Call Center application that was implemented at many police and fire locations in the US, including Palm Beach County Sheriff, Raleigh Police Department, and Cobb County EMS and Police. This application managed open calls and police unit dispatching as well as NCIC database integration to Mobile Data Units in the field.

Technologies utilized:

Pick (Universe 3.1):
HP UNIX 10.5

Responsibilities:

Complete redesign of the database model
Hands-on development
Implementation of code management tools
Code reviews

Education

1997, BS in Operations Automation, Covenant College, Lookout Mountain, Georgia

2009, Instructional Design for New Designers, Langevin

Red Hat Post-Mortem

2012-02-25T20:16:48Z

Right now I'm looking back at the last 5 years, where I first learned I enjoyed technical teaching and where Red Hat gave me the opportunity to do so, even with little proof on their part that I could do so. The experience was priceless, as it taught me to be able to deal with many sorts of planned and un-planned situations.

On at least one occasion I was asked to teach a class I'd never seen before. I won't reveal the client, for they seemed not to be able to tell that I was learning along with them. I got good reviews on the class! Amazing. There are many stories to tell about being on the road and experiencing a new host of students each week. I'll spare the reader those stories, for now.

Suffice it to say that I thoroughly enjoyed teaching technology to corporate students, each one who seemed to take their time in class quite seriously, and who seemed to care a great deal about learning. Many of those students are still my friends today.

When I came in-house to work in courseware development, my experience turned somewhat sour, as the management team and I didn't read from the same songbook. I valued customer experience too much, they said, so therefore I was "mentored" by folks who should never stand in front of a class. This they didn't realize, and the company (I believe) has suffered from the current group running things. This is simply my opinion, and I share it only with a few close friends. This must surely include you now!

After leaving the courseware team, I joined up with the Amentra group of folks, who were at the time a subsidiary of Red Hat (as of January 2012 they've become Red Hat Consulting.) I've been working with that team for the last 18 months or so. I've had a great time.

The timing of my leaving is not in any way related to the new absorption, since the management team I knew before never changed. I loved working for Amentra, and my management team there was par excellence. I wish to especially mention Brad Davis, who stands out in my mind as both a great customer representative AND a very capable people manager. Brad will go a long way in this company, or anywhere he ends up. Many thanks go to Brad for his excellent skills and mentoring.

Now it's on to another adventure, possibly the last one in my IT career. I'm moving on to Hortonworks, and I'm very excited to be classroom teaching again. This is new technical territory for me, but I'm pretty confident I can swing it.

Advent of the Eyephone

2010-03-25T17:45:00Z

Well, the time has arrived. The wireless application is now accepted.Everybody's doin it, and we'll all have apps galore on our iPhones and Droids. See, I've been doing this stuff for years. It's been a long,hard slog to get people to see the value of real applications held in the hand, providing real value for users, both for recreation and business. What has caused this adoption of the technology, to the point that we all will be using this stuff in a few more months?

I see several factors at work:

Ubiquity of the platform - hardware has always been the issue. We went through the stages of having kludgy cell phones connected via unreliable lines to providers that only reluctantly gave us an internet connection. Now, my iPhone will use 4G, admittedly fast, and auto-switch to 802.11 whenever it's within range of an access point. This ubiquity had to happen, or we'd still be walking around outside, looking for a halfway-decent signal.

Real Applications - I attended a seminar in Rome (yes, Italy) several years ago where billing for time through a provider's system was the only way to envision getting revenue from an application. Now we've got multiple revenue streams, all available with simple 2-step clicks and our PayPals will be ticked for $0.99 or something like that. This is really where it gets exciting. If only I can create a decent, simple program that will allow me to get a quarter million users signed up, that's my retirement plan right there!

Motivated Users - there's a desparate public out there, who need all sorts of applications that will make their lives better. We just need to think out of the box a little, and imagine a world where the desktop, keyboard an all, is floating around inside someone's brain, ready in a moment to respond to a thought, idea, event, or disaster, and provide answers, directions, suggestions based on fact, that can make life easier. It's all ahead of us, and we can do it!

I'll mention one more need out there, and it's capable developers. We should be ready to create world-class applications for those who are looking for it, and need the ability to find information when and where it's needed. Developers need to stop looking at the screen, and walk around looking at other people for awhile. Watch them. They'll tell you what they're looking for. Keep your eyes open.

I'm finally Flex-able

2008-09-21T18:36:00Z

I have to say that I've been around these 20+ years, and I've developed rather large complex projects on just about every tool available. I'm a deep developer with J2EE, PHP, and .NET Compact Framework, and I've always been very unhappy with the tools we were given to produce user-interface-heavy applications. For most of my career, RIA meant a hip town in Brazil.

Now, we're being required to deliver RIA, on the desktop or not, and the dearth of good tools is still disappointing. Until Flex. Adobe's really intelligent RIA generator is one of the best things I've ever experienced in GUI tools. It's brilliant approach to achieving the end game is so easy to implement, and there's been so much activity out there, that you simply cannot go wrong with it, if your goal is to create really cool, hip UI's. If you're a Java developer, Flex will become second nature to you in a short period of time. If you're a .NET developer, Flex's use of components to create forms and it's use of namespaces for categorization will be very familiar.

There are too many good things to say about Flex that I can't and shouldn't mention them here, as you can get them best on the product site. I will add a couple experiences that may help you in your decision to use it.

I was working recently at a large financial services firm near NYC that is a Red Hat/JBoss shop with lots of users and over 1000 servers running all sorts of applications. These guys have a real need for immediate visualization of server and network status, and so on the side I developed a little application in Flex that would present JMX data from their JBoss servers in a concise Dashboard.

This dashboard was created with ILog's Elixir plugin, which gives you terrific looking dials, linear reg tools, and charts with no rival.

With very little learning curve, I was able to connect these little guys to a small .ear file I installed to each JBoss instance. As JBoss provides hundreds of metrics available via simple JMX queries, I am able to tell the Flex application to poll each server every X minutes, and set the value on the associated dial or level. With some additional Actionscript magic, we can tell each dial to glow redder if an error condition is neared.

A Brave Old World

2008-07-19T04:15:00Z

I gotta say a little bit about the gig I've been on for the last month. Located in New Jersey, a major financial house has purchased tons of Red Hat technology, including JBoss, and has a large group of sharp folks that are working hard to get all this new cool software integrated into the corporation. It takes lots of testosterone, however.

They've got a web site that rumor says generates millions of dollars per week, and this site needs 100% uptime and high-SLA level responsiveness. I must say that Red Hat has the right stuff. All the way from RHEL 5, their flagship Linux enterprise product, to JBoss' new line of world-class application servers, this brave new company is adopting all the latest that open source can offer.

I think much of their bravery is due to a small group of empowered folks. For example, they've got this Architect there who isn't afraid at all of proposing lots of new cool stuff. Some might consider it rather risky, and the company has no doubt suffered some from being a ground-breaker, but I'm pretty impressed with how they have dealt with that risk. The main thing is that they don't suffer from the ridiculous crippling fear of new technology I've seen in other clients.

A good example of poor adoption ability is a company I'll call The Brown Company. TBC was a gig I worked on shortly in Atlanta. They've got an entire open venue of San Jose-like cool cubes to work in, and they make googles of cash from their clients by leveraging purchased software modules. A pretty brilliant business model. By many appearances, cutting edge place. Behind the guard desk, though, far from true.

When TBC hired me, they brought me in as a J2EE consultant to write a solution that essentially moved incoming EDI files from one system to another. I found some rather well-heeled open source software that would perform about 80% of the process we (myself and other team members) were to implement. I showed this really cool software to them, and my supervisor tells me, "we don't use open source". Notwithstanding their endemic use of Apache and other components that were clearly open source, it was my supervisor who didn't have the cahunas to make a decision to incorporate pre-built software code that would reduce development costs by 50% or better.

At TBC, it was folks with no guts that were valued. Instead of hiring people who can make logical decisions for the company, save the stockholders tons of expense, and better the company's brand, this firm values people who stop the process.

So kudos to my latest client. May they ever be. I like a company that values people who can make solutions happen.

And kudos to Red Hat. I am really enjoying working with them as a consultant. Instead of just teaching their new technology, I get to actually use it. Pretty neat.

My Baby's Gone

2008-03-10T01:14:00Z

Well, it finally happened -- I got rid of my Baby. She was an adorable, exciting, and demanding partner for over 6 years. Now she's gone -- and I'm not too sad.

"She" is the wireless hand-held solution for USDA compliance known to the Food Safety industry as Mobilitee, and she was a very cool software product. In the space of a few short months, she became the darling of the industry, and some called her the "gold standard." And she came about quite by accident.

Mobilitee started life as a brainchild of mine when I owned a consultancy with 2 partners that had grown to over $10MM in revenue. Life before 2001 was great, and we all felt the same immortality that the remainder of the information industry did. Everyone knew that it was a bubble, but we were having too much fun building houses and becoming rich to care.

In late 2000, my partners and I started being at odds over the age-old "product vs. services" discussion, and I opted for the product. They opted for the services. It turns out that neither of us were right, but we both went through the same pains in 2001 after Mr. Bin Laden was finished with us. Evey business in the country was affected by 9/11. I was in a taxicab last week, and the driver lamented how bad times were during that period.

After 2001 was over, I had employees to support, and expenses to pay, and life took on an ominous turn, and we all knew that it was going to go badly. Business fell off, customers looked for other options, and we found ourselves looking for another way to make a living.

It was then that I was told about a Kraft Foods plant in South Carolina that needed help with USDA issues. We went down to the plant, listened to the Quality people and promised to return in 2 weeks with a solution. We did that, and presented a solution that has now become Mobilitee. After 7 or so demos over the space of the same number of months, we sold the solution to Kraft. They funded our initial development, and allowed us to keep the intellectual property.

Over the next several years my sales guy and I saw every protein company in the country (it seems) and did demonstrations to over 20 different groups of people. 19 of them loved it, but we were unable to get any funding for the solution. Some say we were before our time.

About that time, a reseller in Virginia heard about what we were doing, and let us know that they wanted to sell the product. They took it on, we sold a couple of small accounts, and our customers seemed to like it. We didn't sell very many, but we knew that there was a market out there.

As of yesterday, the Virginia reseller has purchased the product from me, and now they will take it to the next level. I will help them make the transition, and wish them Godspeed as they make it the #1 solution in the industry.

The next New Thang

2008-02-19T01:12:00Z

So what is the next new language? Why need a new language? Isn't where we are good enough? I believe that the answers to these 3 questions are Dunno, Yes, and No. Let me explain.

We are truly in the Software Dark Ages. It is crazy in 2008 that we have to jump through such hoops when attempting to create elegant solutions for enterprise users. Web applications, for example, are such that no current platform readily adopts to the web model. So we use workarounds like Struts, JSF, Ajax, and others because there are simply no credible alternatives. J2EE has been greatly reduced in the 3.0 EJB specification, so we no longer have to use a sledgehammer to kill a gnat. But for web applications, there is no real value added there. ASP and .NET technologies are possibly closer to a real model, but they suffer from lack of real re-usability, as well as still leaning on the shared back-and-front end technologies we see in other options. Being a Java guy has me thinking, "where's the next new thing?"

Speaking of Java, Winn in 2003 wrote in a blog that "Java is becoming increasingly hard to apply" in many situations. It's over 10 years old, and new adds to it aren't making it that much cooler any more.

So where to turn to now? Are there any breakthrough ideas out there? Let's look at several of them.

1. Scala - according to Wikipedia, a pure object-oriented language in the sense that every value is an object. Types and behavior of objects are described by classes and traits. Class abstractions are extended by subclassing and a flexible mixin-based composition mechanism as a clean replacement for multiple inheritance.

Scala is designed to interoperate well with popular programming environments like Java and the .NET Framework. In particular, the interaction with mainstream object-oriented languages like Java and C# is as smooth as possible. Scala has the same compilation model (separate compilation, dynamic class loading) like Java and C# and allows access to thousands of high-quality libraries.

Making use of these libraries addresses the common drawback of using advanced functional languages which is that the small community, much of it centered around academia, often does not get to implementing quality libraries for common real-world tasks such as relational database access, XML processing, regular expressions, and so on. Scala can accomplish those tasks in a manner very similarly to these core languages.

2. Jura is a 2nd-generation to Java. It has some neat features:

Jura is a data language: it can be used to describe any structured data, including source code.
Data/program equivalence: the Jura representation of some structured data and the source code required to programmatically create an instance of that data are one and the same thing.
Programmatic design patterns: these are design patterns which can be imported and implemented automatically.
Generics: allows generation of classes from generic templates. Generic templates in Jura are not limited to changing the type of methods, parameters etc. but can change any aspects of a class, including implementation.
Abilities: these are tiny interfaces which allow common functionality to be declared at a fine degree of granularity, so allowing the creation of highly reusable algorithms.
Transforms: allow transformation of object graphs in a similar manner to XSLT (except clean and object-orientated) implemented using only a small set of programmatic design patterns.
Restricted types: allows restrictions to be added to types, such as enforcing that an Integer is in a particular range or a String satisfies a particular regular expression.