This post is Day 8 of the 12 Days of SQL, a series that was the idea of Brent Ozar (blog | Twitter) to have members of an elite group of foodie friends bloggers and writers who get together in person and virtually to improve our writing, branding, blogging and business skills. This series has us picking our favourite SQL Community blog entry from outside the group and sharing with our readers.
On Day 7, Yanni Robel (blog|Twitter), DBA extraordinaire and superwoman at allrecipes.com, wrote about Jonathan Kehayias’ post on No such thing as Small Change to Production database and the Change Management Process. I first met Yanni during SQLCruise. She also took us to a Disney Character Breakfast at Disneyland last week. This is significant because being kid-free, we don’t get out much when it comes to kid stuff. We had a blast – a future blog post to follow.
For Day 8 I suppose that I should write a post that has to do with “Eight Maids a Milking”…but that’s just a bit too weird for a part-time vegan hippie woman in IT to pull off. So I’m going to go with a football theme. That’s my George Bailey in a football uniform ornament on the left. Let’s call this day’s present “Eight Tight Ends”. Or “Eight Ends a Tightening?” Either one works.
I have chosen Louis Davidson’s Can you over-normalize? post. Louis is a frequent speaker, book author, and blogger who focuses primarily on database design issues. I met him in person recently at SQLPASS and volunteer with him on the PASS Data Architecture Virtual Chapater (DArcVC).
In his post, Louis asks an important question, “Can you over normalize?” His answer is that no, you really can’t. But before you get your tinsel in a tangle, read what he said:
My rule of thumb is that:
1. The requirements dictate the database design
2. The requirements and the relational engine dictate the implementation
So unless you understand the requirements, you can’t design the optimal database, and if you don’t understand SQL Server, you are not going to end up with an optimal implementation. Over-normalize? No. Over-engineer? Definitely.
You know that I love contentious issues and levels of normalization is one of the major ones. What I find most people miss in discussions about normalization is that it’s relevant to inserts, updates and deletes. We normalize to avoid data anomalies that would decrease data integrity. But when people complain about “over normalization”, they are usually looking only at retrieving data, not maintaining it. All normalization looks excessive when you aren’t concerned with data integrity. This leads to people wanting to trade off data integrity for performance gains, which is something we might do based on costs, benefits, and risks for the entire lifecycle of data. That’s why we denormalize data. Sometimes the trade off is acceptable and sometimes it isn’t.
If you read them comments on Louis’ post, you’ll see just how many people still don’t understand the basics of normalization – why it is done, why it might be undone, why there isn’t a single number for the “right” level of normalization…which in a way was his point.
As you may have heard me say in a presentation:
There is no one right answer for all projects, all designs, all organizations, all environments.
This is why sports books have an Over/Under for a specific game. They can’t have a single Over/Under for all teams, all locations, all days, all weather conditions. There is no one right number. Your normalization number should be based a on a whole sleigh ride full of inputs, including performance, data quality levels, trust, data architecture, etc. One project’s normalization design choices might be over done and a similar approach for another project might be undernormalized.
In fact, even George Bailey was calculating his Over / Under for Mary Hatch as she hid naked in the hydrangea bush. That football uniform was serving him well in his calculations.
Take a couple of minutes and Buffalo Gals your way over to Louis’s post and let me know what you think.
Next up is Kendra Little (blog|Twitter) who is going to write about Nine Ladies Dancing for Day 9, I hope. Kendra is an amazing presenter and DBA who works in the Pacific Northwest. I first met her at SQL Saturday Miami. If you ever get an opportunity to attend one of her presentations or webinars, Charleston right to it. She has fabulous slides and she shares a ton of practical information about SQL Server.
Brent O’s 12 Days of SQL post
Day1: Jeremiah Peschka
Day 2: Grant Fritchey
Day 3: Dave Stein
Day 4: Andy Leonard
Day 5: Erin Stellato
Day 6: Tim Ford
Day 7: Yanni Robel
Day 8: [This post] Karen Lopez
Day 9: Kendra Little
Day 10: Crys Manson
This post is a rerun from 2009, but the truth is still there.
It’s a Wonderful Life is a part of many a holiday tradition. While most of you probably know this as a Christmas film, the links to the Christmas holiday are fairly weak. There’s an angel, some snow, and a background Christmas tree with a bell. Other than that the story itself could take place at any time of the year. So if you’d put off seeing this film because you think it is all happy, shiny, Christmas cheer-ish, you need to think again.
George Bailey is a protagonist in a small town, Bedford Falls, located in upstate New York. He finds himself on the brink of financial ruin due to no fault of his own. He goes to visit the town big wig and resident bully, Henry Potter, to get a loan. Potter tells him that with no equity in his life insurance policy, he’s “worth more dead than alive.” In a fit of despair, George decides to sacrifice his life so that his family and company (The Bailey Building and Loan) can survive. However, his plans are side-tracked by Clarence, his guardian angel. George rambles that he wished he’d never been born. Sounds pretty dark, doesn’t it? Not that Christmas-y at all. Clarence sees this as a great opportunity to show George just what a great impact he has had on not just his community, but on the world.
So George is shown what his family, friends, town and world would be like if he’d never been born. The new town is called “Pottersville”, after the antagonist Mr. Potter. It’s not a good thing. Clarence tells him:
“Each man’s life touches so many other lives. When he isn’t around he leaves an awful hole, doesn’t he?”
“No man is a failure who has friends.”
The “feel good” part of the film is when George gets to return to the world he wished he’d never been born into. I’ll leave the assignment up to you to see how this all happens and how George and his family do once he returns.
I’ve seen this film at least 50 times in my life, perhaps even more. In fact, most people would probably think that I’m a bit obsessed with this whole story. To which I say: everyone needs a hobby. This just happens to be one of mine.
So what does this have to do with architecture and data management? I think plenty.
- Enterprise-class projects can’t be done by one person. No one is a failure who has good team mates who collaborate well. You don’t have to be friends with them, or even like them that much. But you do have to find a way to collaborate with them.
- Architecture done well can have all kinds of impacts elsewhere. Each architect’s work touches so many other parts of the architecture. When it isn’t there, it leaves an awful hole…to be filled by a non-architect to do. One small great architecture component can have huge impacts on solution quality for a long time in the future. When George is a kid, he saves his brother Harry’s life. But in the Pottersville world, Harry dies. Clarence tells George: Every man on that transport died! Harry wasn’t there to save them, because you weren’t there to save Harry.
If you aren’t there to create the architecture, or if the architecture you create isn’t used, then the good stuff it could deliver won’t be there when it is needed.
- Fastest isn’t always “bestest”. In running George and Clarence out of his Pottersville bar, Nick the Bartender explains what works in the bad world: Hey look, mister – we code fast here for people who want to get lots of stuff done fast, and we don’t need any characters around to give the joint "atmosphere". Is that clear, or do I have to slip you my left for a convincer? Well, I paraphrased that quote. But you get the point. Perhaps that was Nick’s collaboration method – using bouncers to run people out. Some people feel that data and other architectures are just there for some sort of religious checklist nirvana. It’s our jobs, as architects, to show them why architecture plays a key role project success. A left convincer might work in the short run, but the best way to get support for an architecture is have one that works. Good architectures need to be designed. Hacking away on a pseudo architecture is more Pottersville than Bedford Falls.
- Architecture is more than drawing boxes and lines. George was fabulous at motivating people to do the right thing. The first way he did this was by living by his own principles. When put in a tight spot, he did the right thing. We architects need to do the same thing. We can’t tell development teams that they must treat data with respect and then treat our own meta data as if it weren’t important. This means ensuring that our architectures are managed with real tools, backed up, disaster-proofed, and generally treated as production data – which they are.
- Sharing a vision is important. George was also great with expressing vision. He could get people to rally ‘round a cause by getting others to see what was in it for them. His monologue to Potter about why the Bailey Building and Loan should carry on after his father’s death is a classic. But his best was saved for Mary, in the moonlight:
George: What is it you want, Mary? What do you want? You want the moon? Just say the word and I’ll throw a lasso around it and pull it down. Hey. That’s a pretty good idea. I’ll give you the moon, Mary.
Mary: I’ll take it. Then what?
George: Well, then you can swallow it, and it’ll all dissolve, see… and the moonbeams would shoot out of your fingers and your toes and the ends of your hair… am I talking too much?
I’m still not sure how I’m going to work that line in with one my vision/architectural reviews, but I’m still thinking about it. Look for it on call soon.
For those of you who have seen IAWL and might appreciate some derivative works, I leave you with:
Carolyn’s Sill’s video and song “George Bailey”. This is one of my favourite holiday running songs for the tempo and overall good feelings it leaves me with. Her songs are available on iTunes. While you are watching this video, head over to iTunes and buy it. As an independent artist, she deserves the 99 cents for putting this together.
Next up is Angry Alien’s It’s A Wonderful Life in 30 Seconds with Bunnies. A 30 second overview of the film. Be sure to click on the bunny outlines at the end to see some clips that couldn’t be part of the 30 second summary.
Finally, dear reader, I want you to know that you personally touch many lives by being part of our communities, both here and on Twitter.
Happy Wonderful Life, everyone.
I knew it would happen. As I blogged previously, SQLPASS is hosting a 24 Hours of PASS event in March and is using this event to honour Women in IT (WIT) by having 24 sessions given solely by female data professionals.
Having worked on WIT committees, programs and events for more than two decades (I was a national spokesperson for WIT here in Canada for two years), I knew that someone (and there will be more) would eventually anonymously complain that this one event should not be run with only female speakers. Trolls must post anonymously because they don’t want to contribute to the discussion; they just want to make a good thing look bad. This is my letter to Anonymous (an infrastructure DBA).
Hi Anonymous –
You forgot to mention the creepier one that most anonymous trolls give:
"What about the lack of overweight, old, ugly, grumpy white guys in the Supermodel profession?"
It’s a classic. It is the most common response I get in letters to the editor, live events and articles from anonymous posters. If you are going to go for it, please go all the way.
This witty questions does not contribute to the discussion of diversity in any profession because it makes a huge leap of logic: that there is some physical trait in females that should keep them out of the IT profession. That is flat out wrong.
By the way, there are people who are concerned about the lack of diversity in the nursing profession. I support all kinds of programs that want to address real issues of diversity in all professions.
The reason society should be concerned about diversity isn’t about making the numbers somehow magically match demographics of the full population, it is that we should investigate the reasons why certain professions aren’t diverse as the full populations and make corrective action to ensure that silly obstacles aren’t there. The most successful WIT programs focus on ensuring that young women understand the opportunities available to them and remove roadblocks they might have to considering a career in IT.
Much research has shown that young women don’t consider IT (and other STEM) careers because they:
- Don’t even know what the career is about and therefore think it is all about grumpy evil-doing nerds working alone in a dark basement drinking Jolt Cola and typing all day. Think of the Wayne Knight character in Jurassic Park.
- Hear from grumpy people that women aren’t smart enough to work in IT.
- Don’t realize soon enough that they should have taken more math and science during their schooling and therefore can’t get in to certain programs of study, even though they have the aptitude to work in IT
- Think that Computer Science programs are the only career path into IT
- Read computer science program “marketing” materials, which most programs fail miserably at creating, and think “wow, what a boring technical wasteland”.
- Think that IT is only about programming…alone, in a dark basement, typing all day. GOTO point 1.
But let’s focus on one of the main reasons (I presume) why SQLPASS wants to hold an event featuring WIT. The reasons that most women give for not submitting abstracts for speaking:
- Much more often than men, they don’t think that they are enough of an expert to give a presentation.
- Much more often than men, they think there are so many "celebrities" in the field that the shouldn’t even bother submitting.
- Much more likely than men they are more likely to feel that they are an "imposter" in the field and therefore shouldn’t even try to speak at an event.
- They have so many more outside-of-work responsibilities that traveling a ways to speak and attend a conference is a significant roadblock to participating.
- More often than men, they believe that they should be specifically invited to speak rather than just nominate themselves.
- They are more likely to worry about the catch-22 of doing anything new: you shouldn’t do it until you have more experience doing it.
- They think that no one will attend a session they give because they haven’t written a book (see point 4), they don’t travel the world giving presentations, or that someone else has already given a presentation on that topic.
I talk to many women who have wonderful thoughts, observations, scripts, data models, ideas, opinions, and other knowledge to share but won’t even consider submitting an abstract. Most of the time they give one or all of the reasons above. Please ensure that you understand all those “more often” words in the above list.
The idea of featuring only females during 24 Hours of PASS isn’t going to solve all these problems, but it can go a long way to getting more women to present because it takes away some of the obstacles that many female IT professionals give as reasons to not even try. With more women presenting at this one event, we will most likely have more women presenting at other events during the year. You may not want that, but I want that if the reason women aren’t submitting is because they’ve never been encouraged enough to submit an abstract or to gain speaking experience.
Should SQLPASS bend to address those issues? I think they should not have to do so, but often all it takes is a slight change in how women are recruited to make a real difference. Personally, I’d like to figuratively whack all these women on the side of the head like Cher did on Nicolas Cage in Moonstruck: “Snap out of it”. Sometimes, though, helping people realize their potential is one of the most wonderful thing we can do for them.
Yes, I realize for the one event, some presenters will be excluded. For this one event. I would love to have a professional, insightful conversation about whether or not the one shift in a variable is acceptable, desirable, laughable, or even hurtful. But we can’t have those conversations when one posts anonymously in short bursts of accusations. It really doesn’t help the conversation at all.
I’d love to hear opinions on this, but in a way that advances the conversation.
The call for speakers has gone out for 24 Hours of PASS, a virtual conference of the Professional Association for SQL Server. But this isn’t an ordinary call for speakers. PASS is going to do something extraordinary: in honour of Women’s History Month, only female speakers will be be presenting.
Normally I’m not a fan of any special accommodations or “help” for female workers: quotas, waiving of requirements, etc. I sometimes think that those sorts of programs send the wrong message, too. But in this case, I’m a huge fan of what PASS wants to do here. There are plenty of qualified women to speak on these topics, but usually the problem is that women tend not to submit to speak, for a variety of reasons.
24 Hours of PASS To Celebrate Women’s History Month
By Thomas LaRock
Mark your calendars! The next 24 Hours of PASS event is taking place March 15 and 16. We are sticking to the two day format with 12 sessions presented each day.
Since March is also Women’s History month we’ll be carrying that theme through to the online event. As a result we plan to feature 24 prominent female speakers during the course of the event with session content as always, focused on SQL Server.
If you have an abstract in mind or have suggestions for specific speakers or topics, send us an email at firstname.lastname@example.org. Deadline for abstract submission (max 250 words with a 125 word bio) is January 14.
You can read Tom’s blog post about how and why he came up with the idea of a female-only event.
I welcome this special event for a few reasons:
- PASS is doing something to recognize the vast amount of knowledge women in technology
- Choosing to promote the wonderful female speakers out there during Women’s History Month (which also includes International Women’s Day and Ada Lovelace Day) is a great way to honour and recognize all the good work that millions of women do in IT around the world.
- Encouraging female speakers is the best way to build a pool of qualified speakers for other events like SQL Saturdays, SQLRally, PASS Summit, Enterprise Data World, DAMA Chapters, etc.
- I believe that this one event will do more to encourage more women to speak at events that all the encouraging e-mails and blog posts could ever accomplish.
While this is a PASS event, not every presentation will be just about SQL Server code. I typically give my Database Design Contentious Issues presentation at in-person events and one of my database design-related presentations for virtual events. They might include SQL Server content, but they aren’t just about the DBMSs. If you have a presentation that you’ve given at a DAMA event, there’s a good chance you can present it at a PASS event.
So I need you to help:
- If you are a female and work with data, please put together an abstract and submit it. Now. I’d love it if they had 300 abstracts to choose from.
- If you aren’t female, please personally ask one of your female co-workers to submit an abstract. Do it now…it will only take a couple of minutes.
- Please retweet this post, post to Facebook and LinkedIn about this amazing opportunity to highlight female IT professionals. Let’s show the world what #WIT has to offer.
I am excited about this event – I can’t wait to see it unfold. Please help us by getting the word out. Let’s make something happen.
As I blogged last week, I participated in a webcast on social networking for data management professionals. That webcast was recorded and is now available for viewing.
Handouts of the slides I presented on the cost, benefits and risks of social networking are also available.
If you are reading my blog and on any of these social networks, I’d love to friend/follow/link to you. My contact information for those services are in the handouts. If you do send me a request, please mention that you are a blog reader, attended an event I presented at, or where we met.
I’ve uploaded handouts for my Database Design Contentious Issues presentations for SQLSat59 (New York City) and SQLSat61 (Washington, DC).
Both audiences were Contentious, which is just perfect for this presentation.
As a reminder, if you attended these sessions, please take a moment and leave a testimonial/ration at www.speakerrate.com/karenlopez. It helps me and event planners. Please help us participate in more of these events.
I was excited to see that I’ve been recognized as a Rockstar Blogger by Thomas LaRock (blog | @SQLRockstar) at the tempdb level (entry level). Tempdb is a SQL Server system database and all of Tom’s rockstar bloggers are awarded a level based on these system databases.
As a TempDB level blogger, I fit this profile:
The tempdb group has the bloggers that I want to recognize for doing good work. However, they are also the group of bloggers that could most easily fall off and never been seen again, just like a temp table. There is no line separating names in this group, because they are all equally eligible for promotion or relegation back into my general RSS feeds.
Read more: http://thomaslarock.com/rankings-faq/
It is a great honour to be listed with all the bloggers on Tom’s list. You should check out Tom’s blog, plus the others he has listed. I recommend all of them as well.
Subscribe via E-mail
- September 2016
- August 2016
- June 2016
- May 2016
- April 2016
- March 2016
- February 2016
- January 2016
- December 2015
- November 2015
- September 2015
- July 2015
- June 2015
- May 2015
- April 2015
- March 2015
- February 2015
- January 2015
- December 2014
- November 2014
- October 2014
- August 2014
- July 2014
- June 2014
- May 2014
- April 2014
- March 2014
- February 2014
- January 2014
- December 2013
- November 2013
- October 2013
- September 2013
- August 2013
- July 2013
- June 2013
- May 2013
- April 2013
- March 2013
- February 2013
- January 2013
- December 2012
- November 2012
- October 2012
- September 2012
- August 2012
- July 2012
- June 2012
- May 2012
- April 2012
- March 2012
- February 2012
- January 2012
- December 2011
- November 2011
- October 2011
- September 2011
- August 2011
- July 2011
- June 2011
- May 2011
- April 2011
- March 2011
- February 2011
- January 2011
- December 2010
- November 2010
- September 2010
- August 2010
- July 2010
- February 2009