Able to read geometry files created with most CAD packages, COMSOL Multiphysics(TM) v3.2 delivers finite-element based scientific-modeling package with moving-mesh feature and CAD or mesh file import capabilities. COMSOL Script(TM) has command-line interface, scripting capabilities, and 500 commands for numeric computations and visualization. Software allows user to perform time-domain simulation of electromagnetic waves, and GUI promotes use of consistent system of engineering units.

BURLINGTON, MA (September 6, 2005)-COMSOL, Inc., is releasing version 3.2 of COMSOL Multiphysics(TM), a scientific-modeling package whose new features boost productivity throughout the entire modeling and simulation process. The software now reads geometry files created with all major CAD packages. It introduces COMSOL Script(TM), a standalone product featuring command-line modeling. The graphical user interface encourages the use of a consistent system of engineering units, and a moving-mesh feature allows a model to simulate moving parts and parametric geometries. Improved solvers handle models with millions of degrees of freedom and calculate the answers faster than ever before.

Perhaps most obvious to existing customers is the change of the product name to better reflect the company's offerings, which now address many areas of scientific computing. Company president Svante Littmarck remarks, "We are renaming our leading product from FEMLAB to COMSOL Multiphysics. That's already the software's name in Japan, and it's one we find better suits our growing product line. Although we started with finite-element method (FEM) software, COMSOL products today and those planned for the future cover considerably more in terms of functionality and appeal."

CAD import addresses all major formats

To make it easy for users to import CAD drawings for modeling in COMSOL Multiphysics, a suite of optional CAD-import modules read a wide range of industry-standard CAD and mesh file formats starting with those for SolidWorks[R], Solid Edge[R], NX(TM), and NASTRAN[R]. Importing an existing CAD or mesh file enables users to bypass the geometry-creation step, which makes the first step in the modeling process fast and convenient.

The CAD Import Module is based on Parasolid[R] geometry kernel and includes ACIS[R] to support the SAT[R] format. In addition to the native Parasolid and SAT formats, the CAD Import Module also supports the STEP and IGES file formats. Live synchronization with the SolidWorks CAD package enables a truly productive design and modeling environment. Separate CAD-import modules accept the CATIA[R] V4, CATIA[R] V5, Autodesk Inventor[R], Pro/ENGINEER[R], and VDA-FS file formats.

COMSOL Script(TM)-a technical-computing language for modeling

The scope of modeling takes on entirely new proportions with the release of COMSOL Script, which integrates seamlessly with COMSOL Multiphysics but as also runs as a standalone package. On its own, this interpreted language handles most computation tasks through its command-line interface, scripting capabilities, and 500 commands for numeric computations and visualization. For users of COMSOL Multiphysics, COMSOL Script also offer an alternative to MATLAB[R], which up to now was a prerequisite for command-line modeling.

When run within COMSOL Multiphysics this new language enables command-line modeling whereby users can access all functions available in that modeling package, or they can call COMSOL Script functions from within the COMSOL Multiphysics GUI to define any property of a model. They can also save work performed in the graphical user interface to a Model M-file and run that text-based file in COMSOL Script. By working with such scripts, users can conduct iterative parametric studies and optimizations as well as perform any model explorations and simulations. Further, COMSOL Script's graphing and visualization capabilities set new standards for packages in this category, and a set of GUI tools allow users to quickly construct graphical user interfaces.

Consistent units throughout

Other features make it far easier to set up a model. For instance, COMSOL Multiphysics' graphical interface presents each parameter with an appropriate unit; users select from nine common engineering unit systems so that consistent unit labels appear in all dialog boxes, next to data-entry fields, and in postprocessing plots. Examples of supported unit systems are SI, MPa, CGS, and British Imperial Units. This feature avoids the confusion that can arise especially when trying to determine the proper value to enter for a parameter that has compound units, and it eliminates many unnecessary user errors.

Moving meshes for fluid-structure interactions and parametric geometries

Extending modeling into new areas is a Moving Mesh mode that allows COMSOL Multiphysics to easily simulate geometries with moving parts such as those in MEMS (microelectromechanical systems), piezoelectrics, and biology applications as well as free-surface flow and natural wave effects. Users define the desired type of motion such as for the deflection of a flexible barrier in a strong flow of liquid or gas, or even fluid sloshing in a tank. Coupled with the moving-mesh engine is geometric parameterization, where it is possible to describe a how a geometry changes without the need to set up a loop in a script file.

Sustainability and human settlements; fundamental issues, modeling and simulations.

Monto, M. et al.

Sage Publications

2005

211 pages

$49.95

Hardcover

HC79

M. Monto (sustainable technologies, Indian Institute of Science, Bangalore), L.S. Ganesh (management), and Koshy Varghese (civil engineering, both IIT Madras) explore the complex relationships between humans and water in an urban context, identifying fundamental issues within a framework of integrated models and simulations for sustainability assessment and forecasting. They write for students and researchers in environmental and urban studies, and for policy makers from the local to the international.

Every year, bit designers, like car designers, use newly developed technology to make drill bits run faster, drill deeper and last longer in an effort to give operators more value for their bit dollar. In some cases, the added value is real, while in other cases, it is simply perceived. In the end, it is the operators who must determine which bits actually do perform better. Here are some of the latest technologies being applied to drill bits.

TECHNOLOGY OVERVIEW

Generally, the increased use of computer modeling, for roller cone and PDC bit design and manufacturing, is producing a new generation of bits that delivers breakthroughs in rates of penetration (ROP), increased durability and longer life. These computer models utilize proprietary algorithms to model forces and bit behavior to assure maximum bit performance.

Additionally, computer modeling of the dynamics of interactions between PDC bit cutters and rock allows bits to be custom-designed for specific applications. Computer-aided design tools are being employed to model, with 3D and 4D graphics, the drag, axial and radial forces acting on the bits' cutting surfaces.

One manufacturer's cutter/rock interaction model divides the cutting edge into three surfaces: cutting face, chamfer and cylinder surfaces. The computer calculates the cutter's engagement area by meshing each surface into grids, so the cutter orientation's effects on the engagement area can be considered.

Data on advanced cutter wear is one of the results of this modeling. Cutter wear depends on cutting force, relative speed, temperature, cutter material properties and rock properties. Previously, computer models estimated only the wear flat without considering its orientation, as well as the actual diamond thickness, the interface geometry of the diamond layer, and carbide and abrasive resistance. With the newer computer models, cutter wear can be considered three-dimensionally, and all factors neglected by previous models are now easily considered. Bit designers then use this information to devise a bit specifically for a particular job.

Bit balancing. Another development that is becoming more important is bit balancing, Fig. 1. This concept considers the forces acting on the bit to create designs in which no single cone or cutting system is overstressed. This increases cutting efficiency and extends bit service life. Two types of balancing methods are used--force balancing and load balancing.

[FIGURE 1 OMITTED]

Force balancing. Of the three forces acting on a bit--axial force, lateral force and bending moment--it has long been recognized that balancing the lateral force is very important for preventing whirl. In fact, previous concepts of PDC bit force balancing referred only to lateral force balance, due to the belief that once lateral force was balanced, the bit bending moment was balanced also.

However, further study revealed that bit bending moment contributes not only to bit lateral motion or whirl, but also to tilt motion, which significantly affects directional control. Even a perfectly force-balanced bit may exhibit tilt motion, if the axial forces are not balanced. Therefore, balancing the axial forces is equally as important as balancing lateral force.

A PDC bit that is balanced, both in terms of lateral force and bending moment, is a "global force-balanced" bit. Designing such a bit involves adjusting the cutting structure to reduce the imbalance numbers. For example, newer series bits are force-balanced according to a specific set of design criteria that consider the summation of cutter forces to a global, lateral and axial bit imbalance, resulting in a global force-balanced design.

Load balancing. A bit in which the drilling forces acting on each individual cutter are balanced and are evenly distributed across the entire cutting is said to be "load balanced." This technique is meant to prevent cutter wear and excessive point loading that can break or damage cutters.

Roller cone bits are load balanced in two ways--by volume and by force. Volume balancing almost equalizes rock removal among all the cones, while force balancing ensures that all cones are subjected to nearly the same loads, including weight-on-cone, bending moment and force-on-bearing.

For PDC bits, load balancing, which was employed originally on roller cone bits only, is now being used to improve the PDC bit performance. The concept of load balancing is based on the fact that the amount of formation removed by each individual cutter differs and, as a result, the force acting on each cutter also differs. Furthermore, the number of cutters differs from blade to blade. Therefore, the forces acting on each blade differ. To avoid overloading individual cutters and blades, it is necessary to control these load distributions.

Equally distributing the forces minimizes the change in work, or force, among zones of the cutting structure. Thus, designing a "torque- and drag-balanced" PDC bit involves analyzing the distribution of work and forces acting on a cutting structure, with the goal of controlling force distribution over both the blades and cutters. By controlling the force distribution, these bits are able to reduce impact damage and uneven wear while promoting improved ROP.

Steel Catenary Risers (SCRs) offer several advantages in deepwater production. In particular, they are compatible with all host types and can often be the cheapest riser option, depending on specific environmental, field and load constraints. In design terms, however, the issues to consider are few, but complex. As water depth increases, so does the SCR diameter, especially in more hostile environments and with hosts that have limited station-keeping. This creates a design challenge, especially from a fatigue perspective, to the point where costs become prohibitive, or the risks too great, in deep water. SCRs, renowned as a cost-effective option at 5,000 ft, are therefore widely viewed as unviable at depths below 6,500 ft. Now, with the latest advanced modeling, this could change.

Traditionally, because of a lack of precise understanding of the cyclic stresses to which the SCR is subjected, fatigue assessment of SCRs has been highly conservative, erring on the side of extreme caution--given the massive cost of failure. Now, however, where initial analysis indicates insufficient fatigue life in an SCR, advanced riser modeling capabilities independently developed by deepwater consultancy DeepSea Engineering & Management enables more accurate fatigue prediction. This, in turn, avoids excessive conservatism in system design, which optimizes maximum cost-efficiency, while full technical assurance and reliability is retained. SCRs could prove, after all, to be the optimum riser system for some deepwater projects.

For more accurate prediction of riser system behavior, the model must account for all the major physical influences. For example, advanced Touch-Down Zone (TDZ) modeling of the riser/seabed interaction accounts for factors such as trench formation and development, time-dependent and loading-dependent effects of seabed behavior, and varying seabed properties along the riser--none of which are considered in conventional modeling. Equally, accurate pipe-in-pipe SCRs can be modeled, looking at individual representations of the flowline and carrier pipes, and the spaces and gaps between the spacer and inner surface of the carrier pipe. Additionally, the ability to undertake time-domain Vortex Induced Vibration (VIV) analysis using Computational Fluid Dynamics (CFD), coupled with Finite Element (FE) structural dynamic analysis, provides the most accurate modeling and, therefore, the greatest understanding of this cause of riser fatigue available.

Touch-down zone fatigue analysis. Fatigue in the TDZ of an SCR on the seabed is the governing factor in the SCR's durability, and is heavily influenced by the riser-seabed interaction. This interaction has the greatest uncertainty when modeling SCRs. It is known that this fatigue is proportional to the soil reaction force, which increases with soil rigidity, but complex interactions of the seabed soil composition, its nonlinearity, and the random and cyclic nature of loading, combine to make predicting SCR behavior difficult in the TDZ.

The high degree of conservatism in design to address TDZ fatigue is because traditional modeling used in current design practice deploys an elastic (or more crudely, a rigid) seabed, and does not allow for soil softening under repeated loading, despite the fact that this phenomenon is well recognized. while not a problem in more moderate water depths, for deep and ultradeep developments, these conventional, linear seabed models may not allow SCRs to pass minimum fatigue-life requirements.

This advanced riser-seabed interaction modeling draws on published works, including two JIP studies, Carisima and Stride, as well as the company's own expertise. It allows the influence of physical phenomena on SCR performance to be identified and quantified, including axial friction, lateral resistance, soil suction forces, vertical seabed stiffness, and trench formation and development.

It is believed that most of the fatigue in an SCR's TDZ comes from the lifting and setting down of this section, where it undergoes repeated bending between zero and the maximum curvature, which changes with seabed stiffness. A small change in seabed stiffness can result in a small change in bending stress, but this causes a significant change in fatigue life,

Accurate modeling is not easily achieved. The soil lying on the seabed-to-water interface undergoes cyclic loading and generally does not obey well-established laws of soil mechanics used in most geotechnical engineering. Moreover, the dynamic motion of SCRs causes repetitive loading and unloading of the soil under the pipe, causing soil degradation and creation of a trench, reaching, in some reported cases, over 100-m long and five diameters in depth, which changes the seabed profile in the TDZ and results in reaction load redistribution on the SCR, significantly influencing its fatigue life,

The model, valid for cohesive soils, must therefore take into account all aspects of soil behavior characteristics, such as compressive, unloading, peak-to-peak stiffness, hysterisis and so on, as well as the seabed profile (capable of gradual trench development) and axial variation of soil response through the TDZ.

Physics SDK developer Havok and 3D graphics designer Nvidia have teamed up to develop a software package allowing physics calculations to be run on Nvidia graphics chips.

The Havok FX package will be provided to developers later this summer, Nvidia said on Monday, and will be designed for the GeForce 6 and GeForce 7 family of graphics cards.

According to Nvidia, the SDK will support the inteaction of "thousands" of objects calculating friction, collisions, gravity, mass, and velocity; using the SDK, debris, fluids, and smoke will be able to be modeled. The cards must support Shader Model 3.0, Nviida said, although it wasn't clear whether shader calculations would be needed to actually compute the physics model.

"Moving physics processing to the GPU is a natural progression enabled by the high programmability in today's GPUs," said David Kirk, chief scientist at Nvidia, in a statement. "By combining expertise with Havok, we have produced a fantastic solution for game developers that will lead to more compelling game-play and more realistic gaming experiences."

Havok FX will be demonstrated at the Game Developers' Conference this week in San Jose.

Denali Software Inc., has announced that Cray Inc. (Nasdaq NM: CRAY) has selected its Databahn memory controller intellectual property (IP) cores and MMAV verification IP software for next-generation supercomputer product development.

Cray plans to use Denali's Databahn IP in the design of the DDR-SDRAM memory system for its next-generation chips. Cray will also use Denali's MMAV product for modeling and simulating the interactions between its chips and external memory devices for design verification and performance analysis.

Says Dave Kiefer, vice president of engineering at Cray: "Designing supercomputers demands leading-edge EDA tools and IP. We selected Denali's MMAV and Databahn controller cores based on the product capabilities as well as Denali's reputation for quality and reliability."

"Cray stands apart as a leader in the supercomputer race," adds David Lin, Denali's vice president of applications engineering. "By providing Cray with memory controller cores and verification software, we are helping them continue to meet their rigorous standards for quality and performance."

Licensed for use in more than 100 designs by leading semiconductor and system companies, and with more than 35 chips in production, Databahn is the industry-leading memory controller IP solution. Databahn cores are configurable for a wide range of performance and power requirements, as well as ASIC interfaces.

To ensure compatibility with all the latest high-speed memory technologies, the configuration process is tightly integrated with Denali's database of memory component specifications, including all the latest SDRAM, DDR1-SDRAM, DDR2-SDRAM, and Mobile DDR-SDRAM devices from all major memory vendors.

Deliverables include: register transfer level (RTL) and synthesis scripts, silicon-independent DDR PHY, verification testbench, static timing analysis (STA) scripts, programmable register settings and documentation.

The silicon-proven Databahn IP is library independent and covers solutions from .18-micron to .08-micron technologies, and DRAM device frequencies from 100-400MHz (200-800MHz data rate).

In this paper we combined keyword search and content analysis to investigate the employment demands of ERD vs. UML in the context of data modeling. We found that data modeling is one of the popular skills required by many of the general business information systems professionals without specific methodology attached to the employment requirements. ERD and UML are required in narrower and often separate fields of building data models for information systems, in which ERD is mentioned in the domain of system analysis and design and database administration, while UML is required mostly in software engineering and application development. The division between ERD and UML in professional data modeling remains, although there is a small number of employment requirements that demand both methodologies and require a mapping capability between the two approaches.

Information and communication technology is one of the fastest changing fields in employment skills. This has resulted in constant revising of the academic curricula and textbooks to best match the education objectives and demands in professional skills (Gorgone, et al. 2002). In IT education, the various correlations of the recent IT employment opportunities and college students entering this field further stress the needs for examining the employment skills. The available employment opportunities published on the Internet have provided not only a job search tool for many job seekers but also a systematic monitoring of changes in employment skill demands in real-time (SkillPROOF Inc. 2004).

In this study, we analyzed IT employment data published by employers and focused on the ERD(Entity Relationship Diagram) vs. UML(Unified Modeling Language) requirements in the field of data modeling. We hope that our findings can supplement education decisions on what we should include in our teaching scopes and identify trends in the required professional skills

2. EMPLOYMENT DEMANDS

The employment data used in this study is published by individual companies on the Internet and collected daily by SkillPROOF Inc. since the beginning of 2004. The data is collected from up to 137 IT-focused companies. Each data sample contains attributes of company industry, posting date, job title, job responsibility, skill and education or training requirements. The general and background information of the data collection and categorization can be found on the website ofSkiIlProof.com (SkillPROOF Inc. 2004).

From the archived data, there are total of 35,932 jobs. The job counts from the top 12 industries (among a total of 46 industries) are plotted in Figure 1 to illustrate the overall distribution of the employment demands. The distribution reflects the post dot.com and post Sept/11 IT employment demands.

3. DATA ANALYSIS

We first applied keyword search to the job description to categorize the relevant ERD vs. UML skill requirements according to industry and then according to job types or job functions. We further sampled the contents of job description to investigate the implications for the job requirements in ERD vs. UML.

3.1 Keyword Categorization According to 'Industry'

We used the keywords of 'data model' or 'data modeling', 'ERD' and a few commonly referenced design tools like ErWin(2006), Visio(2006) and a database tool 'TOAD'(2006) to search for data modeling and database analysis, design and management related employment requirements. Similarly, we used the keyword 'UML' to search through the same data sets. We classified the search results according to the top 12 industries identified in Figure 1 and plotted the job counts in Figure 2(a), (b) and (c) below for comparison. One extra industry 'pharmaceutical' is added because the job counts in that industry are within the range of interest in the new aggregation.

Both ERD and UML keyword searched job counts are notably reduced to about half compared with data modeling searched job counts. Overall, the distribution for data modeling and ERD are similar in that both job spectra are broadly distributed across the industry line. They both share three common demanding industries: defense, high-tech and IT consulting. One exception is that the retailing industry appears to be very significant in requiring data modeling skills whereas the telecommunication is an industry requiring significant employment of ERD skills.

For the skill demand in UML, the outcome is significantly different. The defense industry is very noticeable in its relatively large job number in requiring UML skill. In terms of the job counts, the demands for UML are similar to the demands in ERD in both high-tech and IT consulting. However, they are both less than 25% of those counted in the defense industry.

3.2 Keyword Categorization According to 'Job Type' Another classification use keyword search is to sort all the job requirements according to the job function or job title. Use a standard job classification, we were able to use keywords of 'data modeling', 'ERD' or 'UML' to plot the searched skills vs. the types of jobs defined. The search results are summarized in Figure 3(a), (b) and (c) respectively for 'data modeling', 'ERD' and 'UML'. In this classification, we find that, again, UML has a lone popular job type as 'software development' whereas 'data modeling' or 'ERD' skill demands are more evenly distributed among the various job types, such as business analyst, software developer, IT consultant, technical writer and database administrator. One exception is that the job type of 'project manager' shows a significant number in ERD searched results but is not as popular in the 'data modeling' group.

In this paper we combined keyword search and content analysis to investigate the employment demands of ERD vs. UML in the context of data modeling. We found that data modeling is one of the popular skills required by many of the general business information systems professionals without specific methodology attached to the employment requirements. ERD and UML are required in narrower and often separate fields of building data models for information systems, in which ERD is mentioned in the domain of system analysis and design and database administration, while UML is required mostly in software engineering and application development. The division between ERD and UML in professional data modeling remains, although there is a small number of employment requirements that demand both methodologies and require a mapping capability between the two approaches.

Information and communication technology is one of the fastest changing fields in employment skills. This has resulted in constant revising of the academic curricula and textbooks to best match the education objectives and demands in professional skills (Gorgone, et al. 2002). In IT education, the various correlations of the recent IT employment opportunities and college students entering this field further stress the needs for examining the employment skills. The available employment opportunities published on the Internet have provided not only a job search tool for many job seekers but also a systematic monitoring of changes in employment skill demands in real-time (SkillPROOF Inc. 2004).

In this study, we analyzed IT employment data published by employers and focused on the ERD(Entity Relationship Diagram) vs. UML(Unified Modeling Language) requirements in the field of data modeling. We hope that our findings can supplement education decisions on what we should include in our teaching scopes and identify trends in the required professional skills

2. EMPLOYMENT DEMANDS

The employment data used in this study is published by individual companies on the Internet and collected daily by SkillPROOF Inc. since the beginning of 2004. The data is collected from up to 137 IT-focused companies. Each data sample contains attributes of company industry, posting date, job title, job responsibility, skill and education or training requirements. The general and background information of the data collection and categorization can be found on the website ofSkiIlProof.com (SkillPROOF Inc. 2004).

From the archived data, there are total of 35,932 jobs. The job counts from the top 12 industries (among a total of 46 industries) are plotted in Figure 1 to illustrate the overall distribution of the employment demands. The distribution reflects the post dot.com and post Sept/11 IT employment demands.

3. DATA ANALYSIS

We first applied keyword search to the job description to categorize the relevant ERD vs. UML skill requirements according to industry and then according to job types or job functions. We further sampled the contents of job description to investigate the implications for the job requirements in ERD vs. UML.

3.1 Keyword Categorization According to 'Industry'

We used the keywords of 'data model' or 'data modeling', 'ERD' and a few commonly referenced design tools like ErWin(2006), Visio(2006) and a database tool 'TOAD'(2006) to search for data modeling and database analysis, design and management related employment requirements. Similarly, we used the keyword 'UML' to search through the same data sets. We classified the search results according to the top 12 industries identified in Figure 1 and plotted the job counts in Figure 2(a), (b) and (c) below for comparison. One extra industry 'pharmaceutical' is added because the job counts in that industry are within the range of interest in the new aggregation.

Both ERD and UML keyword searched job counts are notably reduced to about half compared with data modeling searched job counts. Overall, the distribution for data modeling and ERD are similar in that both job spectra are broadly distributed across the industry line. They both share three common demanding industries: defense, high-tech and IT consulting. One exception is that the retailing industry appears to be very significant in requiring data modeling skills whereas the telecommunication is an industry requiring significant employment of ERD skills.

For the skill demand in UML, the outcome is significantly different. The defense industry is very noticeable in its relatively large job number in requiring UML skill. In terms of the job counts, the demands for UML are similar to the demands in ERD in both high-tech and IT consulting. However, they are both less than 25% of those counted in the defense industry.

3.2 Keyword Categorization According to 'Job Type' Another classification use keyword search is to sort all the job requirements according to the job function or job title. Use a standard job classification, we were able to use keywords of 'data modeling', 'ERD' or 'UML' to plot the searched skills vs. the types of jobs defined. The search results are summarized in Figure 3(a), (b) and (c) respectively for 'data modeling', 'ERD' and 'UML'. In this classification, we find that, again, UML has a lone popular job type as 'software development' whereas 'data modeling' or 'ERD' skill demands are more evenly distributed among the various job types, such as business analyst, software developer, IT consultant, technical writer and database administrator. One exception is that the job type of 'project manager' shows a significant number in ERD searched results but is not as popular in the 'data modeling' group.

Data modeling is a difficult topic for students to learn. Worse yet is the fact that practitioners, who look to academia for methods and techniques to perform such model building have found little on which to standardize, although many techniques exist. Entity relationship (ER) modeling was developed in order to help database developers visualize their (relational) database design with its data stores and internal relationships. This technique was certainly an important step forward, yet data collected over the past 11 years would indicate database developers are still having difficulty learning, assimilating, and using design techniques (cf. Blaha, 2004). Confounding the issue is the arrival of the object-oriented paradigm. The Unified Modeling Language (UML) was introduced in order to speed, simplify, and clarify design of systems. Portions of the UML are derived from ER modeling and are useful in merging the front end portion of the system with the back end data storage so a picture of the entire system can be viewed by the designer. While providing functionality that ER modeling lacks, the UML approach to data modeling also leaves some developers indecisive and confused as to which technique to use in practice. The same indecision appears to haunt the academic world. So how should data modeling be taught? In order to shed light on this question, we asked contributors to focus on whether this new system of modeling (the UML) yields a better understanding of the database design to the extent that better database designs result. We detected a buzz in the literature and in the IT world that a dichotomy of opinion over this question exists, and so this special issue was born. Educators need to air their opinions, facts, and results and discuss this controversial topic to encourage refinement in this important area. We hope that research ideas can be generated and practitioners informed that this topic is being addressed in academia. As expected, the contributors to this issue provided a dichotomy of opinion but surprisingly, their experiences and opinions moved the issue in a direction far different than what we could have predicted. We now provide you with insight into this poignant topic by presenting this special issue.

Chen (1976) developed the entity relationship diagram (ERD) as a set of tools to assist the database designer in visualizing the internal workings of a complex design. Chen laid out the basic data model with symbols and later other extensions were added. The extended ER model was developed by Teorey, Yang and Fry (1986) to include the concept of generalization (inheritance) in the modeling technique. No clear standard has emerged and there are several in use, including the so-called Crow's foot model, or Information Engineering and the Integrated Definition for Information Modeling (IDEF1X) adopted by the Federal Information Processing Standards agency in 1993.

Object-oriented (OO) techniques began to infiltrate the database world because it seemed that this natural way of processing software "objects" in OO programming languages could be extended to the database. In fact, it seemed like a perfect match since each record in a database table fits the definition of an object as defined in the OO paradigm. Database designers rushed to market their new product, the OODBMS, because at the time, the writing was (or at least seemed to be) clearly on the wall-all software will eventually be subsumed under the OO umbrella. Growth rates of these products were predicted to be so high (~50%) that one wonders how manufacturers could have kept up. The harsh reality was this growth rate was never achieved, and the OODBMS approach has never completely caught on. Pockets of use may be in existence, but only because the technology has been suited to specialized applications-CAD/CAM, multimedia systems and network management systems as examples. The OODBMS has not effectively competed with the advantages provided by relational database's particularly strong roots in mathematical theory. Added to this is the extent of infrastructure held by the relational products making it is extremely doubtful it will be supplanted by the (pure) OODBMS in the foreseeable future.

In spite of many misgivings, the OO movement did force relational database designers to add object-oriented extensions to their products. Primarily due to the need to process complex objects, the object-relational database is now able to store and perform searches within audio, geographic data, telecommunications data1 and other complex data types. The OO movement also brought new data modeling tools. The Unified Modeling Language (UML) became the standard for OO systems and included tools for database design. The UML class diagram is probably the strongest contender in this realm since it maps directly into a relational (logical) design and is able to convey even more system information than just entities and their relationships. The use of UML diagrams by relational database designers is somewhat controversial and not entirely accepted. We detected this controversy among database instructors and systems analysis instructors and decided looking further into this issue would help to place the arguments for both sides on the table and foster a healthy academic discussion. Our goal for this issue was to advance the body of knowledge on the use of modeling techniques by airing this controversy and promoting cogent discussion of the topic. We hope this goal has been accomplished.

Traditionally, the typical undergraduate database course uses a form of Entity-Relationship (ER) notation when teaching conceptual modeling. While we have seen an increase in the academic coverage of UML in the database course, it is very rare to see UML as the primary modeling notation when teaching conceptual data modeling. However, outside of academe, there has been advocacy for the use of UML as an effective modeling tool for database design and for it to provide a unifying modeling framework. This paper examines the level of support for using UML vs. established ER notations for teaching conceptual data modeling in the introductory undergraduate database course. An analysis of textbook and tool support as well as a survey of what IS undergraduate programs are using in their introductory undergraduate database courses is included.

As data modeling has evolved in the last 50 years we have seen a shift from hierarchical and network models to relational and object-oriented models. While the term "data modeling" may imply a variety of different meanings (Topi, et al., 2002), in information systems (IS) education, data modeling is consistently used to describe entities and relationships within a real world domain (Hoffer, et al., 2005). In the past 20 years relational data models have dominated the market but today the Unified Modeling Language (UML) has emerged as the software industry's dominant modeling technique for application development (Siau, et al., 2001). In the past few years there has been an increase in interest in the applicability of UML class diagrams in data modeling.

While there are a wide range of issues one must consider when selecting an appropriate data modeling language, the aim of this paper is not to pass judgment on or comment on Avhich modeling technique is correct. It is to gain insight into the support for the different modeling techniques and the current state of data modeling in undergraduate database courses. As the paradigm governing modeling techniques evolves there comes a time when the academic environment may consider if the tipping point has been reached where the academic teachings in introductory database courses have the support to shift to UML. While we recognize that there are many theoretical and practical issues to consider when selecting an appropriate data modeling technique used in the classroom we have chosen to report on the current level of support for the use of ER modeling and UML class diagrams in undergraduate database courses.

This paper examines the viability of UML as a conceptual modeling notation for an introductory undergraduate database course by investigating the supporting issues, including: curricular fit; support materials (i.e., books and tools); and the use of UML in IS undergraduate programs. We then discuss the strengths and shortcomings of UML for teaching conceptual data modeling in light of these supporting issues. Finally, we highlight potential directions for future research and discuss conclusions and limitations of this study.

2. SUPPORTING ISSUES

In order to gain insight into the viability of using UML as a notation for teaching conceptual data modeling we examine some of the infrastructural supports for teaching UML in an introductory database course. First we discuss influence of the overall IS curriculum, specifically whether or not an object-oriented methodology is reinforced throughout the curriculum, on the readiness for teaching UML in database. We then examine the support of UML in eleven introductory database texts marketed to the academic community along with five popular software applications that support the diagramming of conceptual data models. Finally, we analyze the level of coverage of UML in current introductory database courses at nineteen undergraduate IS business schools in the United States.

2.1 Curricular Fit

When selecting the modeling technique for an introductory database course one must keep in mind the unique characteristics their academic environment provides. The database course is often part of an IS majors' curriculum and finding synergies between the courses in the curriculum may be a priority. While there may be multiple courses one needs to consider fit with, the most prominent course topics to consider are programming courses and systems analysis and design courses. If your program is using an object-oriented (OO) methodology and has embraced UML in the systems analysis, systems design and programming courses your students may have already been exposed to UML, thus potentially making the use of UML class diagrams a more natural fit and one that is more synergistic within your overall undergraduate IS curriculum. By giving the student multiple exposures to UML in different contexts the student will be able to get a more holistic view of systems design and integration.

Furthermore, the sequencing of the introductory database course needs to be considered. The earlier the database course falls in the sequence of required courses for the majors the easier it is for the database instructor to select the diagramming technique they are most comfortable with. However, if students have already been exposed to UML class diagrams, the database instructor may need to keep that in mind to reduce the confusion of the use of the various modeling techniques. The database course does not have to use UML if the systems analysis and design courses do but the database teacher may find that they need to address the differences in the modeling techniques in order to improve student comprehension of the integration between the various courses.