By Karen Lopez on Friday, December 21, 2007 12:46 PM

A while back I posted a blog entry about Microsoft's feature addition to Visual Studio - Entities (or ADO Entity Data Model).  In that post, I complained that the "redefining" of the term Entity Data Model was going to lead to even more confusion on development projects.

Today I watched an online video of using Visual Studio to work with Entity Data Model functions .  While it didn't help assuage my misgivings about calling this feature Entity Data Modeling, it did help me understand what Microsoft means by data modeling.  For now, it appears to be database modeling via reverse engineering of databases.

One of the things I did note, though, was that I preferred some of the graphical presentation features of the actual entities.  The gradient shading and the drop shadows gave the models more visual appeal than most tools I work with.  I know that these sorts of features come with a performance price, but I still think they were more visually pleasing.

I'm thinking of giving Visual Studio a try to see how these graphical features come out in printing and other presentations.

What do you think of them?

 

By Karen Lopez on Wednesday, December 19, 2007 3:33 PM

The CA Modeling Product Line Community (PLC) is looking for volunteers to serve on the PLC Modeling Board.  It is my understanding that you must be a current user of CA ERwin products and be employed by a non-partner organization.

Now through Janurary 31, 2008, we are seeking nominations from CA ERwin users in this community, who are interested in serving as a member of the PLC Board. The primary role of the board is to act as the voice of the global user community by providing leadership in supporting members interests and liasing with CA, and to support and encourage growth of local user groups.

These PLC board positions are being offered to you, as CA ERwin users. All members of the PLC Modeling community will vote for the candidates of their choice.

PLC governance is owned by officers of the board & it's members. For your PLC, the board will be representing the entire virtual product line community member base, including local user groups and users who do not have local groups in their area. This governing body will be responsible for these types of functions.

Develops and monitors budget
Has fiduciary responsibility of the PLC
Maintain communication channel with CA's Customer Program office, local user groups, PLC members, and the modeling tool development lab.

Facilitate the prioritization and voting of certain product enhancement requests upport local user group activities as seen appropriate by the PLC Board Serving a 2 year term, the PLC Board will need to have a minimum of 3 officers; maximum of 6 officers, as deemed necessary to govern your PLC.

In addition to being able to give back to the global CA ERwin user community, I'd think that being a member of the board would have great networking benefits -- with other ERwin user organizations and with CA staff and product management.

For more information or to nominate yourself, visit http://causergroups.ca.com/usergroups/News.aspx?ID=397 .

 

By Karen Lopez on Tuesday, December 18, 2007 4:40 PM

End of service has been announced for version 7.0 of CA ERwin Data Modeler, CA ERwin Model Manager, CA ERwin Model Navigator, and CA ERwin Model Validator in the next 12 months.

The announcement mentions a future general release of version 7.3 of these products.

...and if you are still working with ERwin 7.0 and not the current release of 7.2, I highly recommend you upgrade.

By Karen Lopez on Tuesday, December 11, 2007 7:05 PM

Today I am presenting Collaborating with Techs at Toronto IRMAC.  This presentation, which I've also given at DAMA Portland and the New York Enterprise Modeling User Group (NYEMUG), focuses mainly on collaborating with different generations of IT workers. 

Our work environment today involves collaborating with four generations of workers.  The gap between work- and lifestyles of our multi-generation teams can become a real issue with collaboration.  One of the main issues is how the different generations prefer to communicate, including formats and locations.

A very popular YouTube video, orginally developed to influence teaching styles in  a Colorado school system, touches on some of the same issues.  Specifically, it highlights differences between today's youngest generation of students and workers.  The video is 8 minutes long.  While it includes music, you can watch it with your speakers muted as the sound is only background (for those of you who fill older generations and are concerned with the appearance of work).

  

I wish I had time in my presentation to include this video. Perhaps I'll find time to include it in my related Tutorial at the DAMA conference.

By Karen Lopez on Friday, December 07, 2007 4:03 PM

An anonymous blogger has started writinga series of posts of his experiments with database design and performance.  His profile describes him as "A Seattle database guy who works at start ups."

In the last article the performance impact of joins was shown.  This one will demonstrate cases where denormalized joins are a bit faster, as will the third article with larger data volumes.  The fourth article, the most interesting one, will show where a denormalized data model can be 50 times faster than a normalized data model. 

Here are the tables that will be involved in the sql.  The normalized ProductSmall table has a 100 million rows and is about 0.67 gig. 

What I appreciate about his posts is the fact that he is supporting his positions with actual tests.  So far his two blog posts have focused on very large tables (more than a million rows) and the impact of memory usage.

I'd also like to see him post about working with smaller data volumes.  For instance, I work at times with new developers who tell me that our database or table is "very large" at 4,000 rows and needs a great deal of denormalization for performance reasons.  I usually ask them to run tests similar to what the DBScience guy is doing to show me all the great benefits of combining 6 tables into one table with a total of 10,000 rows.

Check out his blog as he adds articles. http://dbscience.blogspot.com/

By Karen Lopez on Wednesday, December 05, 2007 10:49 AM

On December 4, 2007, Danny Sandwell, one of the CA’s Enterprise Modeling Product Manager, conducted a session called Complete Compare 101.   For those of you that were not able to attend the live session, it was recorded and can be accessed on the PLC-Modeling website (in the News Section).  You must be logged in to view the replay information and presentation.

 

 

By Karen Lopez on Monday, November 26, 2007 5:29 PM

CIO magazine has interesting article on how to better recruit females for enterprise IT jobs.

 

Business technology needs broad-thinking candidates from a broad range of undergraduate and graduate curricula who want to learn how companies—not computers—work; who can work with a global project team, rather than with programming languages; and who can see business process linkages, rather than map out electronic connections.

Meanwhile, the collection of jobs that saddled business technology with its geeky image—network and data center administration, code maintenance, programming and help desk—may soon be centralized, automated or offloaded to outsourcers. The stereotypically inarticulate men with pocket protectors who hold these jobs—and who defined the image of the profession way back in the '70s—will soon retire en masse (taking with them their pocket protectors).

Now you need business analysts, program managers, vendor managers, relationship managers, information architects or process analysts. These jobs (any of which can lead to CIO) demand employees with excellent communication skills that many of the women you know have: the ability to speak, negotiate, influence others, write, analyze, manage projects or programs, and lead cultural change. These jobs are not about writing operating systems or learning programming languages. They are about helping companies change the way they work. "Driving changes that help the business generate more revenue, lower cost or improve customer service—cracking these business problems—that's fun!" says June Drewry, CIO of the Chubb Group of Insurance Companies.

 

Some good points in that article.  Having spent a few years as the Canadian Information Processing Society's national spokes person on this gender issue, I agree with a few points and disagree with others.  Let's start with the agreement part:

IT does a terrible job recruiting for IT careers. 

I'm not sure whether this is due to the fact that the marketers are mostly men and therefore us marketing messages that they'd want to hear or that all of IT is terrible at all marketing.  These recruiting errors start early, in the marketing literature of computing-related educational programs at colleges and universities.  If you visit some of the most prestigious computing programs you'll see bland descriptions of computing science and information systems programs -- all focused on things like "learn C++", "algorithms", and "good careers as programmers".   If the website is maintained by the computer science department you might even be wowed with some nifty flashing HTML or scrolling techniques from The Information Super Highway(tm). 

Visit a website of other professional-preparation programs like law or engineering and you'll see photographs of people actually talking to other people, smiles on their faces as the solve some intriguing societal problem.  At worst, you may find a link describing finite element analysis or even algorithms, but it won't be the focus of their recruiting pages.

You'll find the same scenario on the careers page of your local newspaper.  IT jobs will display some long list of foreign sounding terms  about computers (PeopleSoft, Microsoft, etc.) or even worse, a set of acronyms that sound as if the successful candidate will also suffer from food poisoning (SAP, ERP, CRUD, C#, CRM, etc.) .  Postings for IT jobs rarely focus to any degree on solving people problems.  In fact, they tend to describe jobs that require one to sit alone in a basement, staring at a monitor everyday.

We need to stop trying to drag techies up the Zachman Framework, against their will

I cringe every time I read a recruitment add for an enterprise architect or other strategic position that demands several years of hands on technical skills in very specific areas.  Sure, enterprise architects, data architects, and process architects need to have a thorough understanding of the technologies that help run the business, but they don't have to have done that job in order to understand the architectural role.  We don't demand that lawyers be police officers for 5 years before they can be legal experts.  We don't require that professional engineers be construction trades people or draftspeople before they can be engineers.  Why must all IT jobs start with 5-10 years of programming?  The reason why that may have been true10-20 years ago was that there used to be only a couple of routes to get into IT at all.  This was due to the fact that people didn't have 6 mainframes in their basement to get hands on experience with technology.  That left formal education and training as the primary route to a career in computers.  But those days are gone. 

Most DBAs I know are excellent technical professionals.  I think they'd do a wonderful job of data modeling or data architecture.  For many though, I think they'd hate it.   These guys (and most of them are men) love understanding, working with, and managing detailed technical nuances of their jobs.  They think I'm crazy for wanting to spend days and days with business users, working on process or data models.  As for me, I'd go crazy spending much time on indexing strategies or tuning someone else's SQL code day in and day out.   Whether it is differing Meyers Brigg profiles, DNA, or brain make up, people tend to be drawn to certain jobs.   When I see those job postings that are looking for the ideal candidate who will spend an equal time working in Row One and Row Five of the Data Column, I know that most likely the organization will spend a great deal of time talking to people who are very experienced at one end of column over the other end.  And with all the inherent risks associated with gender stereotyping, I'd say that women tend to be drawn to jobs that have more interaction with the business.  Not all women, just most.

If you are recruiting candidates for multiple roles (and who isn't), I'd recommend you recruit along the same rows but perhaps multiple columns.  Think Business and Data Analyst, not DBA Modeler.  Look for Programmer DBAs, not Programmer BPM architects.   Let people who are "good at what they do" do what they are good at.

 

 

By Karen Lopez on Thursday, November 22, 2007 7:06 PM

From a recent Embarcadero Press Release:

Embarcadero Announces New Versions of DBArtisan and Rapid SQL

Rapid SQL 7.5 and DBArtisan 8.5 Provide Enhanced Productivity and Support for the Latest RDBMS Features

SAN FRANCISCO--(BUSINESS WIRE)--Embarcadero Technologies announced today the general availability of DBArtisan 8.5 and Rapid SQL 7.5, the newest releases of its market-leading, cross-platform database tools. Both tools feature a redesigned user interface, a new query engine, and deeper support for Oracle, SQL Server, DB2 and Sybase.

A key feature of DBArtisan 8.5, a cross-platform database administration tool, is a new highly scalable engine that helps DBAs better manage large, complex datasets and produce faster queries. DBArtisan 8.5 also enables DBAs to exploit more features available in Oracle 11g, Sybase 15, SQL Server 2005 and DB2 v9. Also new to DBArtisan 8.5 is a set of user interfaces designed to increase user productivity by simplifying the standardized workflow across platforms. The interfaces enable the product to dramatically decrease the time to market for updates required to support features of existing or future RDBMSs.

If you'd like help evaluating DBArtisan or Rapid SQL For your projects, please contact us.

 

By Karen Lopez on Thursday, November 22, 2007 7:04 PM

We've all dealt with the risks associated with the proliferation of user-developed spreadsheets and desktop databases.  I've even heard that more business functions are run from data copied out of corporate databases than in the databases themselves...and I don't doubt that one bit.

I know why users want to copy and control locally their own copy of data (and I don't blame them for wanting to do this), but the problems I come across when gleaning requirements from users is often related to the whole "I'll just do it in a spreadsheet or MS Access" attitude.  Sure, it's a get-er-done attitude, but it often suffers from quality assurance and peer review steps that we typically see in enterprise-class application development projects.  At least we give the appearance of testing, right?

The staff at ITBusiness.ca have pulled together a very scary list of examples of spreadsheet errors, gaffs, and flaws that cost companies billions of dollars:

• Kodak spokesman Gerard Meuchner said the hefty $11 million severance error was traced to a faulty spreadsheet.
  
• Details of the $2.818 billion record profit result for the 12 months to September 30...were embedded in a template of last year's results and were accessible with minor manipulation of the spreadsheet. (Some news reports indicated an employee had thought that a black cell background fill would hide black text.)
  
• Fannie Mae, which finances home mortgages, stated in a news release of third-quarter financials that it had discovered a $1.136 billion error in total shareholder equity
  
• Shares of RedEnvelope Inc. lost more than a quarter of their value Tuesday after the company warned of a fourth-quarter loss due to weak Valentine's Day sales and a budgeting error that resulted in an overestimation of gross margins.
  
• A simple spreadsheet error cost a firm a whopping $24 million. The mistake led to TransAlta, a big Canadian power generator, buying more US power transmission hedging contracts in May at higher prices than it should have.

 

Think there are enough scary stories in the article to help you on your next quest to get enterprise data under enterprise quality control?  There's a website by the European Spreadsheet Risks Interest Group (EuSpRIG) that collects media stories of spreadsheet errors that have put companies at risk.  They currently have almost 100 stories.

I wonder what we'd find with a similar effort to track MS Access errors?

 

By Karen Lopez on Thursday, November 22, 2007 1:52 PM

As a reminder, we won't be approving messages today or tomorrow as this is a major US holiday weekend.

We do this to stem the flow of "out of office" messages the moderators have to deal with.

Here in Toronto it's not a holiday (our Thanksgiving was in October), but it is snowing and it sure feels like a holiday.

Speaking of Moderators and Thanksgiving, I'd like to thank each of our community moderators. These individuals donate time to keep our community active and the messages on topic. Every message posted to our groups is cleared prior to releasing it to your inbox.

Gracious thanks go to:

• Rick Davis
• Scot Fearnside
• Garry Gramm
• Jeremey Janzen
• Carol Lehn
• Ray McGlew
• Brett Medalen
• Frank Palmeri
• Karel Vetrovsky

I'd also like to thank each of you who has taken the time to respond to questions, gripes, and tips posted to our boards. The DM profession is a "small world" type community of like-minded individuals each with different needs but often common goals. These communities allow us to get our jobs done faster and easier with your help.

Happy Tofurky Day to all of you in the US.

Karen