However I'm not sure this is a bad idea. How often have we had issues reverberating through the Internet because of insecure computer systems? So many companies pay lip service to the idea of security, especially with their own corporate systems. And home users are even worse, many times not even realizing that they are behind on patches.

I've been a big fan of not building something you think people need, but rather working with them, getting feedback along the way, and tailoring your efforts to something that works well for the user. There are exceptions, but for the most part I think that if you are building a product you need to work with your audience, get feedback and thoughts from them, and adjust your product as you proceed.

First you figure out what has happened and what is happening. You don't take any actions, and I'd argue that you might not even want to stop what's occurring. Someone might have a great reason for starting that restore, so don't stop it for now. Besides, the database has already been whacked at this point. Stopping the restore doesn't help you.

When I ran across this article recently, it reminded me of the Borg. The article talks about tapping the disparate knowledge and wisdom of employees to solve problems and innovate inside companies. It's an interesting idea, and one that I think is good for data professionals.

It's plain, simple advice. And it's the advice that I, and many others, often give to people that ask about regular maintenance or settings for their SQL Server. In a recent discussion with someone, they noted that there wasn't any other way to say this.

Or is there?

To me, this is one of those areas that starts to separate out the beginners from the mid-level, from the senior people that write code. It's when you realize that not only can the computer do work for you, but also write the code that tells it what work to do. When you are getting started, you often don't realize that the investments in writing code to write code pay back many times over. It takes some time to learn that this is a way of becoming more efficient in the way you approach your craft.

We used to only allow people into the party that used our referral code. That was good for me since it was less work and the PASS staff managed the admission process. However so many people wanted to come that either didn't, or couldn't, use our code (hint, it's SSC2010 for the 2010 Summit), that we decided to sell tickets. So someone had to work the door and handle money.

"From a personal standpoint this worked out poorly: I was out with friends and after more than a few drinks. "

Bill wasn't on call, but he ended up needing to work. I can appreciate that, but I also think that this is a place where someone shouldn't be pressured into working unnecessarily. Being a little tipsy could result in more mistakes, or even longer delays in service restoration than necessary. It could also open you up to some level of liability if you made a mistake while impaired.

Is this a good fix? Does it make sense to spent the $$ and time for another instance for a few users?

I ran across an interesting article talking about usage based pricing for software. A way that you would pay for what you used as opposed to paying a flat rate for all users. Using SalesForce and Amazon EC2 as references, there's a report saying that enterprise customers would rather pay for what they use than pay a flat rate and possibly not get the value for their purchase.

About 30k worth of it.

The post has some good advice about where we might be overusing identities, though I'm not sure I agree with the idea that a OrderDetails table doesn't need one. I've had places where you order additional quantities, ship separately, re-order on the same order, etc. that make having a simple handle to that particular row nice. But in general I agree with Aaron.

Dealing with times, time zones, and the proper way to calculate data is complicated. We recently had a question of the day that asked if times were equivalent, given different numerical times across time zones, and there was quite a debate as to whether the times given were actually equivalent or equal. That was a frustrating debate for me to referee and I am very hesitant to allow other questions dealing with time zones for the time being.

Whether you think this is a good or bad idea, I do think that this idea of showing your customers what data you have about them, and then giving them a way to manage it, is an idea that might start to be required in other industries. Especially those places that have privacy or identity information of a personal nature. I know I've love to be able to understand, and even manage, the data that my doctor, my banker, and even the credit agencies have about me.

I wasn't sure, and didn't see it documented explicitly in a few places, so I asked the question and got a definitive answer from Paul Randal, who pointed me to this white paper outlining how Log Shipping and Mirroring work together. However should we evolve mirroring to work differently? Would the publisher model make sense here? Is there some value in having a separate mirror server setup on a lower priority that can act as a tertiary failover point?

"Write a program that prints the numbers from 1 to 100. But for multiples of three print “Fizz” instead of the number and for the multiples of five print “Buzz”. For numbers which are multiples of both three and five print “FizzBuzz”."

Sounds easy enough, and while some set based solution would take a little time, do you think you could solve this easily inside of 5 minutes during an interview? I think I could, since the algorithm that I'd use is fairly simple, though I'd be using a simple loop.

Once the next version is released, you often cannot buy the previous version.

So even though he might need to put up a new SQL Server 2008 server, he would essentially need to "buy" the new version of the software, SQL Server 2008 R2, pay for that (more expensive) license, and then downgrade to SQL Server 2008.

However you don't have to be a superstar to improve your chances of getting a better job, or keeping the one you have. You just need to show that you can do the job required, and that you bring value to the company. Not necessarily the same value as the next individual, but you bring good value for the salary that you are paid.

Backup and recovery is the core of what a DBA is responsible for and I thought this would make an interesting Friday poll.

"I consider this a strange obsolete limitation"

But what to get? The geek in me had to search around and examine different models. I'd bought an Acer for my son, and it seemed like a nice machine. I found some nice ones with long battery life on Amazon and was tempted to get one. However I've owned a couple Dells, and always been happy with them. My friend, Andy Warren, says to stick with what works, so Dell was under consideration.

Apparently not. There's a back door in that you can start SQL Server in single user mode, and if you are an administrator, you can connect and you'll be a sysadmin. You can then fix your mistake and go happily on your way. That's pretty cool, and it's a good thing, right?

The data from this whitepaper has been around for quite a few months, but as I talk with people and mention it at times, I often see surprised looks. It seems that there are many DBAs that don't understand it, or don't think it might apply to them.

If we require encryption for credit cards, medical records, and the like, shouldn't we also be thinking about encryption for the military? I would like to think that this issue should be fixed soon, even at the expense of replacing some other hardware, but I'd also like to see some responsibility taken by the people that arrogantly allowed this system to be developed and implemented.

I'm not certain that I like that idea, especially as storage systems become larger and more complex. I could argue that the network administrators might be the closest skillset to managing and handling large storage systems, but I'm not sure. If an IT department is indeed a series of armed camps, it's not a skills challenge we have, it's a cultural issue among different groups. Delivering a service or a set of applications to a user requires different talents and skills working together, not removing people from the loop. It requires a forging of new relationships, not the elimination of old ones.

On the other hand, I think we often end up with a group of people that lack depth, miss out on some of the fundamentals, and don't appreciate where they are because they haven't paid their dues. Too often I see people thrown into jobs that they have to teach themselves about as they go along, or they have success in one area, like desktop administration, and find themselves running another system, like a database or mail system.

Over the years I've had mixed feelings about the Summits. As a DBA I wanted it in Seattle so that I could meet and ask questions of dozens of MS developers during the Summit. As the editor here, it's still good to meet MS people, but I'm less concerned about finding the mid-level developer that worked on PBM to ask about an issue than I am about getting together with Dan Jones and talking higher level issues, and sharing beverage.

Read the rest of Moving PASS at SQLServerCentral.

Now SQL Server might be expensive, or might be cheap, depending (I love that word) on your point of view. Oracle lists for $47.5k for the Enterprise Edition of it's v11 database server (per CPU). DB2 runs about $20k/core I think. Hard to tell with them. MySQL, with support is about $5k/server/yr, and SQL Server Enterprise is $25,000/CPU. Now there are all sorts of variation for clients, sockets, add-ons, etc. that make comparisons difficult, but if you assume you are licensing 2 cores/sockets for your application, that's a pretty hefty sum.

There are always horror stories about laptops, from every manufacturer, and it doesn't matter if you have a cheap Acer, or a high end Apple, all manufacturers will have issues. When you make thousands of anything, there's bound to be a few lemons. That's OK, and I get that. I think I've just had a failure, and I understand that can happen.

Why is that?

I mean I understand why the backup structure changes, and why the database structure changes to take advantage of the new changes in the next version of SQL Server. You want your new version server to run more efficiently, and often there are core engine changes that ensure that happens, regardless of the compatibility level of the database. That is, almost all the time, a good thing.

If you read the goals of Andy Warren, Tim Mitchell, and others, you'll see that most of them talk about growing their professional reputations and brands quite a bit. They all talk about speaking at events or doing some other growth activity that puts them out in front of other professionals. That's great, and it's those attitudes that ensure that the rest of us get great opportunities to learn at a SQL Saturday or the PASS Summit.

Not that most of us want to write that low level code, but I'm sure it's of interest to a few people. I know I found myself checking the list to see how many I'd read and which ones I might want to tackle at some point. Being in technology, I think we're naturally curious, even to the point where we'd like the chance to write an Operating System (or maybe just the idea of trying that).

Read the rest of Do You Want to Learn a Thing or Two? at SQLServerCentral.

As the site as evolved, we have also limited advertising on the site. We have never really allowed advertising or promotions in the forums since vendors often seem to annoy users with their posts. They seem to feel that if they post once and get a good response that posting 100 times should get them a great response. It doesn't, and my inbox tends to fill with complaints. So I remove advertising posts.

I typically do something with regards to New Year's resolutions, and this year is no different. However I'm looking to get you to focus a bit more this year on something that I think is valuable for everyone: professional development. No matter where you are in your career, working on improving some aspect of your work will help you to continue to move forward. For this Friday's poll, I'm asking this question:

My first prediction was that SQLServerCentral would pass 1,000,000 members. We did, and that was a pretty easy one. We had a minor celebration, and continued growing throughout the year, though at a slower pace. We still get a lot of traffic, but we've removed the need to register to read most of the content, and so that has slowed down our growth. I think we're still doing well, and I thank everyone for all the support that keeps me employed.

He also had posted a query on Superuser about recovering his data. It's back up now, but it was a hard lesson for him to learn, and a little embarrassing for someone that's running a programming site teaching people how to better build systems.

At the same time I see many technical people frustrated at times by the lack of resources being devoted to issues. This might be people or hardware that could help make something run smoother, save time, save the company money, etc. I actually saw a post recently with someone venting about the expenditures on the Christmas party by their company when they were regularly spending time deleting files because of a shortage of disk space on a server.

There was an interesting interview from by Phil Schiller of Apple talking about their approval process for their App store. For those of you unfamiliar with this, Apple has a store that allows sales to iPhone and iTouch users of various applications. However all of the programs listed on the App store must be approved by Apple, and with Apple controlling the hardware, they have the final say on what's available.

If Microsoft is serious about helping clients secure data, this is one feature that should be deployed across all editions. I could even argue that it's more important to have this on Express than Enterprise. After all many of the data loss issues reported in recent years have come because of missing laptops, which often contain data from databases.

I realize that many people don't usually use Express to work with data, preferring another tool like Excel, but we could look to rectify that. We could store the data in an Express edition instance and set up queries against the local database from Excel. And in the places where Express is being used, this would instantly make the data on laptops more secure.

If you read some of the descriptions in the article, you realize that the term "database" is being taken loosely by the author. The number one listing, the World Data Centre for Climate, has 6PB of data in "magnetic tapes" for "easy access." While I'm sure they have a lot of data, I'm not sure that those tapes qualify as part of a database.

I know that the really good consultants and architects out there, the successful ones that constantly find work, often bring guidelines with them at a new job or engagements. The decision they make, which might not seem to be related from project to project, are often actually related by a set of rules which help to create that regular success.

I've worked in a lot of American companies in my career and it's been rare that I find a group of people that provide strong leadership, or sometimes even any leadership at all. I heard recently from one person that their entire management committee had resigned without a word to the employees. That's typical of what I'd expect, with management not having respect for the staff that does most of the work for a company. There's often a fundamental lack of respect from those in charge, somehow viewing everyone below them in the organizational chart as less important than those at their level or above them.

They asked customers what would be a good interface for working with sets of data. The customers wanted something that looked like a spreadsheet. They asked about how customers wanted to manipulate data and heard that something was needed like the formulas in Excel, which were easy to use. They asked about how data should be shared and presented, and customers said that ease of building reports and charts in Excel would be preferred.

The perception of all of our systems and the services we deliver is what really counts. I heard someone recently paraphrase a well known saying. I don't remember it exactly, but it was something like "People never remember the quality code we deliver, but they never forget the lack of quality in our code."

Whether you think you'd like this in your life or not, it's an interesting knowledge idea for companies or individuals. Maybe it’s a good idea for programmers or DBAs. What if you could have all the code you'd ever written? What about tests of that code and the applications where it had been used? Would a walkthrough of tuning efforts you'd made for some types of SQL or even queries that solved particular challenges be useful? What about a walkthrough of the changes to your schema across time?

It seems that sometimes we are confronted with too many choices, and it ends up confusing us. We freeze up, often unable to make a decision.  You could view this as a complicated piece of software like Word or Excel. Or maybe the shutdown menu from Windows, which is way too complicated for Joel Spolsky.

It's funny how much people seem incorporate their work into the rest of their lives somehow. I've seen all sorts of web sites, blogs, license plates, even Twitter handles that use some part of your work as the name. So for this Friday, I thought I'd put a fun poll out there.

When I started in my career, there were relatively few ways to network with others, and they all required a lot of time. You had to spend time out of work going to happy hour, some professional event, or playing golf. You spent time with people, but you could touch relatively few people for your effort. While you can build strong relationships this way, they can be hard to maintain.

Recently I saw a report that T-Mobile employees had sold customer data to brokers, who in turn sold it to rivals. Now the details weren't released as to how this occurred, but there are plenty of issues that I have seen in many companies that would make this easy. You can't always prevent a user that needs legitimate access to data from mis-using it, but you can take some steps to prevent it from being easy or undetectable. Things like auditing, obfuscating data in test or development environments, and using encryption when data is at rest.

Recently I saw a report that T-Mobile employees had sold customer data to brokers, who in turn sold it to rivals. Now the details weren't released as to how this occurred, but there are plenty of issues that I have seen in many companies that would make this easy. You can't always prevent a user that needs legitimate access to data from mis-using it, but you can take some steps to prevent it from being easy or undetectable. Things like auditing, obfuscating data in test or development environments, and using encryption when data is at rest.

Recently I found a post where someone had a problem with their transaction log. They had installed snapshot replication, but the log would not shrink despite repeated log backups. Eventually through some advice of others, they discovered that someone else had set up transactional replication on that instance and they were not aware of it.

I noticed a post from an ISV awhile back asking if there was a way to remove Windows authentication. Most of us know that we can't remove it, and it concerned me a bit that an ISV didn't know this. The poster wanted to know if we could make the database more secure, potentially limiting the ability of a client to muck around with the database.

That was my big argument for going to a conference when I was employed as a DBA. I'd tell my boss that a conference is a chance for me to learn something new from all the amazing speakers that attend. Many of the well known MVPs, like Kimberly Tripp, Brian Knight, and Andrew Kelley, teach me something every time I hear them speak. Many of the other speakers that do interesting work, like Michelle Ufford of Go Daddy, help me learn how to better build my environment based on the experiences they present from their jobs.

That got me thinking about something that would be handy for examining hardware. What if a tool could take the workload of a server, looking at the volume of queries, and then determining the rate, ratio, and types of I/Os that the application performs. Rather than look at the queries, look at the work that the queries do.

But is that viewpoint valid? Are there things that you can't do in VB.NET that you can do in C#? Especially with all the OOP changes in VB.NET that have occurred in the last few years. So for a Friday poll, for the developers and hybrid DBAs out here:

Virtualization is really a one-time cost savings. Once you've virtualized a server, you can't do I again, so there are no repeat savings from this event. Is consolidation the same? After all, the main driving force for that process is also cost, and once you've consolidated a server you can't do it again.

I had to explain that the CUs are cumulative by definition, and you only need to apply the last one, along with any service packs (SP) on which they are based. You do know that CUs are based on SPs, don't you?

When I read this quote recently, I was surprised: "If the next programmer you interview can't identify at least one of the programmers interviewed in Coders at Work and tell you roughly what they're famous for … I'd say that's an immediate no-hire. "

I think most of you that are at the Summit in Seattle this week have felt the same way, but I wanted to get your opinions. This has been a tough econonmic year for many companies, and it's a large expense to send someone to the Summit for a week. It's Friday, and for this week's poll:

Why is that so? Why can't the "restore" script be stored as part of the backup file and let the user open it, and get notified if there are issues. Or walked through the process? If you open a diff, can't it include the "name" of the last full? Or search for that file in the folder? What about logs, can't SSMS just "figure out" what LSNs are in which files and load them up automatically? Shouldn't a log have the name of the last full backup in it? Or the name of the last and next log backup files?

There is a blog at ZDNet that talked about how Sharepoint is six products for Office bundled together, it has thousands of people working on it's development, and no one is quite sure where it fits in. Is it a content management system? Is it an enterprise social networking platform? I've never been quite sure what the big deal is.

Read the rest of Low Hanging Fruit at SQLServerCentral

I thought this might make a good Friday poll, so I'll ask you this week:

This election was a change from previous years in that the voting was done before the annual Summit. Typically I've scrambled to find some time during the Summit to vote, and haven't always remembered to do so. This year the candidates were a little more accessible electronically, with bios posted on the PASS site, and a great series of Interviews from Brent Ozar (Thomas LaRock, Brian Moran, Matt Morollo, and Jeremiah Peschka). The one with Matt Morollo was particularly engaging with lots of comments and debate in the discussion. It was a little controversial, and I think with good reason. It's worth a read, and it got me thinking about what we want from PASS, and what we want our board members to accomplish.

Why was it worth it? That's an easy one, and it's the same reason I went to the Business of Software conference last year, and why I'd recommend it for other people. It's why I go to the MVP Summit, and why I usually try to attend keynote speeches.

As the world changes, it seems that so many old paradigms and methods of doing business either evolve or die off. I think this is just the way the world changes, but I think that the last decade, and possibly the next one, will change the world quite dramatically.

Most of the time I get the sense that a particular person is just unaware of how to solve a small problem and some hints, sample code, or suggestions on what to try will help them solve a problem. However there are times that it seems that someone is way out of their league. I don't want to specifically cite a thread, but I have seen a few posts that were asking for help with a large problem, such as how to set up a high availability solution for their servers.

I thought about it over the years, wondering what it would be like to work at Microsoft. I'd known a lot of employees over the years, thought they were interesting people, and it seemed like an exciting place to work. There was a time I even wondered if I could take over Robert Scoble's job and be the "voice" of Microsoft.

For those of you hiring people, do you want to hire people like this? I'd like to think that you do and you look for people that present themselves as self-motivated. A person's history, the types of problems they've solved, the way others talk about them, show the way that they handle themselves at work.

For this Friday, I wanted to take the temperature of the readers out there with regard to their career. The poll I have in mind is:

I tried to read through the quickstart on Apache's site for the common core installation and walk through a few examples, but it's a little hard to tell what exactly the buzz is about. Wikipedia was more help, pointing me to the MapReduce papers that Google published. I'll see if I can work through them at some point. Hadoop is available under a free license and the list of companies using it for large data set processing is impressive: Yahoo!, Amazon, Facebook, and more.

That seems to be a story played out in many companies all around the world. Lots of creative, talented people, all stuck in boring jobs cranking out code that supports the business. Over time they become jaded, un-motivated, or burned out at their jobs, perhaps even questioning their choice of career.

I've mostly abandoned the goal of becoming a CIO, but I still read about the position, trying to better understand the management point of view for many issues. Recently I ran across an article on how CIOs can better support their CEOs and thought it had some good advice, and not just advice for current CIOs.

Read the rest of Support the Boss at SQLServerCentral.

When I came across this blog talking about information workers and how they use technology, I was pleasantly surprised. It seems that Gen X people are more likely to participate in the Web 2.0 technologies out there. That seems to match my experience where I find many younger, Gen Y-type people having very short attention spans and not being willing to give back as much to the world. They don't post as often, and tend to just want to consume things quickly and without effort. At least that's been my general experience. There are always people that don't fit the mold, but I see a lot of those trends in people that are younger than 30 years old.

As SQL Server professionals should we care? I think we ought to be aware of these issues because you never know when some of this software might be in use on a server. What if you have old versions of Quicktime or Adobe PDFs that you generate in one of your servers. An attack on that software could be aimed to move against the database supplying the information.

We have a very talented group of professionals that donate their time to help others, something that I do as well.

Having been in a number of technology departments, from small to large, I can say that I see both sides. On one hand technology departments spend a lot of time and money cleaning up mistakes and problems from users. On the other hand, new applications and enhancements can often increase the efficiency or effectiveness of workers that find a new way to do their jobs.

In most of my past jobs, I've implemented security like this:

But did you clean house? In other words, while getting ready for this upgrade, planning ,testing, and preparing, did you re-examine your system to see if there were objects, users, jobs, or something else in your database server that is old, obsolete, and ready for retirement? Did you look for things to delete while you were getting ready to make everything else better?

Since I'll be there for 3 days, I don't have a party to plan or run (this year), I decided to check out the list of sessions at the event. I've been watching the banners rotate by, and seeing a fantastic list of speakers. Most of them have a lot to teach me and you, so I'll give you my top list of sessions and why. Note that I'm not going by the schedule for now. I'll figure out what to do with conflicts later.