By Karen Lopez on Monday, March 12, 2007 10:11 AM

Garry Gramm has found more information and a new video of Perceptive Pixel's wonderful computer interface.  I found a working copy of the video on Google Videos.

Can you imagine, instead of whiteboards, having this functionality on your walls?  No more copying, or photo graphing - just do the work once and refine it later.   I wonder what the agile folks would do with this?  Perhaps erase it right away?

By Karen Lopez on Tuesday, March 06, 2007 12:13 PM

We arrived in Boston on Sunday to attend the Wilshire Conferences' Meta-Data Conference and DAMA Symposium 2007.  This is my 10th DAMA Conference, I believe.

Sunday there were half day workshops, but due to the glories of air travel, we arrived late, so we skipped these events.  There were two I wanted to attend, so I was disappointed.

Monday offered full day workshops, so I spent time in Graeme Simsion's Consulting Skills workshop, John Zachman's Framework workshop, and Steve Hoberman's Data Modeling Challenges.

Today I am presenting on "Implementing Industry Standard Data Models". I have already attended the ERwin SIG, which I have posted about on the ERwin User Discussion Group.  Check out that group to see my reports. 7:15AM events are sometimes a bit strained for everyone.

Earlier this morning I attended Gordon Everest's presentation on ORM and NIAM.  I started my data modeling career working with NIAM.   This modeling technique leaves you with every fact as an object on a diagram.  I think it is important that data architects be familiar with all kinds of modeling techniques, so search Object Role Modeling (ORM).

Right now I'm sitting in Michael Ley's presentation on Globalization, Localization. Code Pages, Unicode, etc.  This presentation is very detailed and very informative.  Who knew that UPS supported Egyptian Hieroglyphics?

The next presentation I will be attending is Len Silverston's case study on culture and human dynamics.  Len has been working a great deal in this area.  I wish that IT educational programs focused more on these topics.

More to come.

By Karen Lopez on Thursday, March 01, 2007 2:16 PM

Arthur Fuller, of Redgate, has written an article for the RedGate Simple Talk newsletter about database design for systems that have to preserve data - that don't delete data, but mark it as no longer current.

http://www.simple-talk.com/sql/database-administration/database-design-a-point-in-time-architecture/

Point in Time Architecture (PTA) is a database design that guarantees support for two related but different concepts – History and Audit Trail.

• History – all information, both current and historical, that as of this moment, we believe to be true.
• Audit Trail – all information believed to be true at some previous point in time.

The distinction is that the Audit Trail shows the history of corrections made to the database. Support for History and Audit Trail facilities are notably absent from typical OLTP databases. By "typical", we mean databases that support the traditional Select, Insert, Delete and Update operations. In many cases, typical OLTP databases are perfectly fine for their requirements, but some databases demand the ability to track History and Audit Trail as core requirements. Without these abilities, the database will fail.

Typical OLTP databases destroy data. This is most obvious with the Delete command, but a moment's thought reveals that the Update command is equally destructive. When you update a row in a table, you lose the values that were there a moment ago. The core concept in PTA is this: no information is ever physically deleted from or updated in the database.

 

 

By Karen Lopez on Thursday, March 01, 2007 2:11 PM

Michael Coles has a couple of good articles on NULLs, ANSI, SQL Server, the Four rules of NULLs, and unknowns:

http://www.sqlservercentral.com/columnists/mcoles/fourrulesfornulls.asp

http://www.sqlservercentral.com/columnists/mcoles/2829.asp

 

In one of the first articles I wrote for SQL Server Central, I talked about SQL NULLs and three-valued logic (Four Rules For NULL). In this article I take it all back...

No, not really, but stay tuned as we talk about the darker side of ANSI NULLs.

The Original Four Rules

The original four rules I proposed for NULL-handling are all reproduced here in Figure 1.

Figure 1. The original "Four Rules"

 

 

By Karen Lopez on Wednesday, February 14, 2007 10:13 AM

This information comes to us via Tony Shaw of Wilshire Conferences:

...the folks at Exeros have come up with a fun contest that ties in with the upcoming DAMA+Wilshire Conference in Boston.  You have to map two datasets within 2-hours.  The entries are judged on the number and accuracy of the mapped table columns.  They're offering cash prizes up to $2500 for the winners. 

 

First prize also includes a free ticket to the DAMA + Wilshire Conference in Boston, travel expenses included! 

 

The contest is this Saturday.  Full details are here:

http://www.exeros.com/html/mapoff.asp

 

 

 

By Karen Lopez on Wednesday, February 07, 2007 1:38 PM

If you work in the data management industry, we have e-mail, web, and newsgroup based discussion groups.  The groups, with about 1000 users, are intended for user-to-user support.  Yes, vendors can answer questions about using their tools, but sometimes you might want to know how others are using them.

You can click on the Groups tab at the top of the page to join.  If you are interested in seeing what sort of discussions are happening, you can check our our preview page  for recent posts to our groups.

 

By Karen Lopez on Monday, January 29, 2007 2:56 PM

Spam (unsolicited, unwanted e-mail) has become a huge problem.  Our mail server here at InfoAdvisors is bombarded with unwanted mail -- and most of it concerns scams.  We do use good spam fighting software which reports that 87% of the mail that hits our server is known spam.

Today, while checking my filtered mail, I came across on piece that shows that the spammers are getting smarter or luckier:

From:  Agelina K Cantrell

To: *******

Subject: or ifyour database normalizesdata differently from

Date: Thu 25 Jan 2007 02:27:59 AM MST

Status:  Message quarantined by Spam module (score 99.9%)

Of course, I'm deleting the spam.  I Googled the phrase above and found it in a Java-related blog. 

So either the spammer knows that I would open a mail with those words, or he just got lucky.  Which do you think it is?

By Karen Lopez on Wednesday, January 24, 2007 4:13 PM

On many occasions I've had the opportunity to discuss data design issues with team members who would prefer to have a quick workaround rather than a durable design.  The discussions almost always end up with a debate of whether the data structure is necessary.  And I almost aways am heard to exclaim:

"Yes, the system we are working on will probably be around for only a few years, but the data will be around for much, much longer - probably decades."

...And the person on the other side of the debate almost always disagrees.  So I go to the data and show them...That customer signed up for our utility service in 1954....and that bank account was opened in 1963.  The original systems are long gone, but the data is still there.

I thought about these discussions while reading an article on Microsoft's patent application on 'immortal computing'.:

E-mail from the grave? Microsoft seeks patent on 'immortal computing'

By TODD BISHOP
P-I REPORTER

In this culture of instant information, some Microsoft Corp. researchers are pursuing a radical notion -- the concept of saving messages for delivery in decades, centuries or more.

The project, dubbed "immortal computing," would let people store digital information in physical artifacts and other forms to be preserved and revealed to future generations, and maybe even to future civilizations.

After all, when looking that far in the future, you never know who the end users might be.

Hmmm... If Microsoft is thinking long term about data, will I have more ammunition for my discussions?  Then I read more, and it it sounded as if their proposed applications for this technology might be a bit...weird:.

And here's where the notion of immortality really kicks in: The researchers say the artifacts could be symbolic representations of people, reflecting elements of their personalities. The systems might be set up to take action -- e-mailing birthday greetings to people identified as grandchildren, for example.

What do you think?  Will you be sending birthday cards to your great-great-great-great grandkids? 

Do you believe this technology will help sustain information for decades?

By Karen Lopez on Friday, December 22, 2006 12:47 PM

It's probably no secret to most people who know me that I'm a real fan of the film It's a Wonderful Life.  Since most organizations in North America will be slowing down next week, I thought I'd share a quick 30 second rendition of the film, as protrayed by bunnies.  If you'd seen the movie, you'll reconize most of the scenes.

Happy Holidays to everyone, and have a wonderful New Year.

By Karen Lopez on Monday, December 04, 2006 11:26 AM

No, it's not about paychecks running late, or tax bills for $145 billion.  It's the fact that companies are still sending mail to the WTC offices and to people who are no longer with us.

NEW YORK—It's the kind of holiday mail that might have been tossed aside, discarded like any other piece of junk mail: a special offer for a facial at a local spa.

Only the address on the letter no longer exists. And the woman the letter is addressed to died more than five years ago in the terrorist attack on the World Trade Center.

http://www.thestar.com/NASApp/cs/ContentServer?pagename=thestar/Layout/Article_Type1&call_pageid=971358637177&c=Article&cid=1165187408545

(The link above is good only for the next 7 days)

If you still manage data that results in mailings to ZIP Code 10048, it might be a sign that those addresses are no longer current.   I realize that one can't always just delete accounts or change addresses, but it seems to me that after 5 years, come companies could take some time to find out what the new address is, or whether these addresses should be marked Do Not Mail.