If you haven't already heard, Microsoft is upping the ante in the
certification arena by introducing a new line of certs called the Microsoft Certified Master Program.
The gist of it is that these are certifications (currently only for SQL
Server 2008, Exchange Server 2007, and Windows Server 2008) that
represent the highest skill level in building solutions based on the
specific product. A much more detailed blog post from the program owner
HERE.It's a lengthy post, be sure you have a minute to digest all of that information. Here's
the rub: many people already feel that certifications, particularly MS
certs, aren't all that valuable in the real world, because the certs
can be obtained by anyone patient enough to read the study materials
and take the test, without having any actual, useable, real world
experience. And on top of that, with the changes in recent
certifications from MS, many managers and HR reps don't even know what
the certs represent. So, how can adding a higher level cert be helpful
in any way? Well, personally speaking, I think they've got a good
deal with this one, with one caveat I'll speak to in a minute. This
particular certification, much like the Microsoft Certified Architect,
requires (as in MANDATORY) classroom training. For the SQL Server
Masters Cert, they are going to require that you already hold the
MCTS:2005 cert, as well as both the MCITP (Microsoft Certified IT Pro)
SQL Server certs for DBA and Developer. And, they will need an
application, as well as a resume to see if they will accept you into
the program in the first place. All of which seems to, in my opinion,
make the cert a little more "weighty" than it's predecessors. Primarily
because this will make it hard for the cert to become watered-down by
literally hundreds of thousands of people holding it. Plus,once
accepted to the program, you have to pass 3 written exams AND a lab
based exam (similar to the Cisco CCIE).Again, I think that's a huge
plus. The idea behind all of this is that if you encounter someone that
holds the MCM for a given product, you should be able to reasonably
assume that they know, very well, what they are talking about. This
increases their value to a company, as well as value in their career. However. The
blog post says (and it may be an early report, but I don't imagine it's
too far off), that the program fee, IF you are accepted, will be in the
neighborhood of $18,500USD. This includes the mandatory 3 WEEK course
and ONE round of exams. If you fail any exam, retakes are extra: $250
for written tests and $1500 for the lab exam. In other words, this is
not likely to be a cert you can afford on your own. In fact, just to
meet the pre-requisites, you or your company has already forked over at
least $225-$250 in exam fees for the earlier exams, not to mention
training. Not only would an individual find this expensive, even a
small or medium business might find this to be too much to fork over.
The argument, of course, is that a business that employs someone that
holds a cert of this level has not only a strong technical resource,
but a good marketing tool (particularly those companies that are MS
Partners). So the question is not only "Am I good enough?", but "Who's
going to pay for this?". I would love to acheive this cert. I
currently have the MCITP:DBA, but never pushed on for the Dev MCITP (or
any others) because I felt like it was pointless. I really felt that if
you've spent more than 6 months with SQL Server, you could probably
pass the test(s). And if you've already built up a healthy resume of
experience, the MCITP cert doesn't really impress anyone as much as
your background and references will. However, I think the MCM would be
a much more difficult certification to earn, not only requiring you to
demonstrate expertise but probably deepen it during the training
courses. If nothing else, think of it this way: Many experts in
our field (think Kalen Delaney, Kimberly Tripp, Itzik Ben-Gan) have
made careers out of sharing their deep knowledge, and don't generally
advertise or glorify certifications (I don't know if they do or don't
hold any MS certs, but they certainly don't advertise it if they do).
Their body of work shows what they know. But, for those of use
somewhere in between "Senior DBA at XYZ Corp." and "SQL Server
Community Guru", a cert like this might help us get that slightly
higher paying job or even help us branch out on our own as consultants,
authors, and trainers. I'm curious to hear if anyone else out there has any thoughts. Please feel free to respond, thanks for stopping by!
--Josh
 Digg It
After tons of feedback from the community, MS has decided to release another Service Pack for SQL Server 2005. More information on the Data Platform Insider Blog. Of course, there's not much detail about what will be IN the SP, but I think this is a huge step in maintaining good standing with customers. While many people jumped at the chance to go from SQL Server 2000 to 2005, the jump to 2008 will be a little more hesitant, mostly because of the difficulty in moving database platforms in production environments. Hopefully, this SP will give people a "longer life" in SQL 2005, while giving them the opportunity to more carefully evaluate SQL Server 2008.
--Josh
Digg It
Just a quick note today. Our buddy, SQL Server MVP and author Paul Nielson has decided to throw some code out there (two stored procedures) in the wild that will help DBAs understand and utilize the new data compression feature in SQL Server 2008. Head over to his site, www.sqlserverbible.com, to get the details (I don't want to steal his thunder here)! Thanks Paul!
--Josh
Digg It
Microsoft announced today (via the DataPlatformInsider blog) that Katmai, AKA SQL Server 2008, will have a feature complete CTP released in Q2, with the RTM release planned for Q3. This doesn't affect their plans to include SQL Server 2008 in their launch of the new "platform" (Windows Server 2008, Visual Studio 2008, and SQL) in Feburary. Check out the blog for more info.
There are some questions bouncing around the internet about what impact this may have on customers. And the fact is, it will have very little. Many organizations are still migrating from SQL Server 2000 (or even 7.0) and will simply have to change their timelines a little bit, and decide whether or not they should move to SQL Server 2005 versus 2008. Couple this with the naturally slow migration process for migrating a company's data to a new platform, a delay of a few months will have little impact. Though it's always nice to hear what everyone out there thinks; please feel free to leave comments and let us know what your plans are!
Digg It
Hi all,
We noticed last night that episode 2 of our podcast, "CS TechCast", got a few hits over at digg.com. So we wanted to say thanks, and let everyone know that we appreciate any and all feedback. Something like this is not only makes us feel good (they like us, they really like us!), but helps our business out tremendously. Thanks! And remember to tune in each week!
Digg It
Anyone who has spent time analyzing someone else’s SQL code has had the horrific experience of trying to make sense out of poorly formatted (or completely unformatted) code; indiscriminate case usage, lack of tabs or indentation, and wrapping text with no carriage returns are just a few of the readability problems you can encounter. This is enough of a challenge when trying to determine the purpose of a given piece of code. Then you have to factor in that the SQL you are reading was more than likely written to take function, syntactically, with a specific RDBMS, such as Microsoft SQL Server, DB2, or Oracle. Additionally, each of these implementations has its own set of “current” syntax conventions, as well as older conventions supported purely for “backwards compatibility”. Given this, it can take hours to evaluate what the code is even trying to do, much less figure out how to fix it, much less performance tune it.
Obviously, one of the major purported solutions to this problem is to always use ANSI standardized SQL.
/*
For the uninitiated, ANSI is an acronym for the American National Standards Institute, an organization in the US that sets technical standards. They have published a series of standards relating to SQL in the last few decades, notably SQL-92, SQL:1999, SQL:2003, and SQL:2006. Each version generally includes updates to the standard that reflects new functionality, i.e. SQL:2006 introduced standardization around concurrently accessing traditional SQL data as well as XML data inside a relational database. ISO (the International Standards Organization) generally ratifies these standards as well, since ANSI is a US based organization.
*/
Since many of the major RDBMSs meet at least ANSI-92, if not later, using pure ANSI standard syntax will work for queries that do literally nothing but SELECT, UPDATE, INSERT, or DELETE data. That is, ANSI is used to standardize ONLY the data access and manipulation aspect of what DBAs and developers tend to think of as “SQL”. However, that standard does NOT include anything procedural, such as control-of-flow syntax. This is where the usage of ANSI SQL starts to break down.
Try to remember some of the recent SQL you’ve written. Was there a CONVERT statement? How about a FOR or WHILE loop?
None of these are ANSI compliant; they are extensions supported by specific vendors.
Now, can you imagine a world where none of your SQL had any procedural extensions? You’d be returning raw recordsets back to applications for sorting, traversing, etc. Doesn’t sound like much fun, does it?
Additionally, learning and using vendor specific SQL extensions can have a huge impact on performance. There are always nuances in query design that you can take advantage of when customizing your SQL for a specific RDBMS. But remember, the trade-off is portability and maintainability. If you use ANSI SQL, your code will likely run on any RDBMS with little or no changes. Also, *most* other developers will be able to understand your code, at least at a basic level, without having to dive into the vendor documentation.
So, like any other debate, there are multiple sides, and the answer is always “it depends”. But this is a choice that all database developers should consider when writing new SQL code. Be sure to evaluate each project and environment accordingly to choose what’s best for that situation.
I'm interested to hear what people think on this topic; we've all heard/read about various opinions on this, but what are the people in the trenches actually doing? Post a comment and let me know what you think!
Digg It
Hey-O!
Eric Johnson and I (Josh) are doing a webcast in conjunction with CA and Redmond Developer News on Data Modeling. To register for the event, click here. Below is the description. Come on by; we'll be talking about why data modeling is important, covering data model basics, and how CA's ERWin tool can help you build data models. Plus, we're trying to work out a deal to give away some sample chapters from our upcoming book, "Architecting Database Models for SQL Server".
--Josh
|
Understanding and Applying Data Modeling in Application Development
|
 |
|
Date: Tuesday, November 13, 2007
Time: 2:00 PM Eastern; 11:00 AM Pacific
|
|
|
|
About the Webcast:
Data modeling is the process of mapping real world information requirements to logical representations of the supporting data. Once your logical model is complete, you can start thinking about transforming it into a physical database model. Keeping the logical and physical model separate, but "in sync" will help ensure that you build a solid database that reflects the business needs.
This technical Webcast gives you a thorough understanding of data modeling and shows you how to apply data modeling to application development. This presentation:
Takes a close look at entities, attributes, and relationships, and how you can use them to build a logical model.
Demonstrates how to take entities, attributes, and relationships and turn them into tables, columns, and keys.
Looks at how you can use modeling tools like CA ERwin® DM to help in the process of creating a logical data model and mapping it to a physical database structure.
Don’t miss the audience Q&A session following the presentations – always a hit with developers!
REGISTER NOW!
About the Presenters:
Joshua Jones; Co-founder & Operating Systems Consultant; Consortio Services
Eric Johnson; Co-founder & Database Technologies Consultant; Consortio Services
Danny Sandwell; Product Manager, CA ERwin® Modeling; CA
This Webcast is sponsored by CA.
|
|
 |
Digg It
|