THE SQL Server Blog Spot on the Web

Welcome to SQLblog.com - The SQL Server blog spot on the web Sign in | |
in Search

Kevin Kline

Continuing Industry Consolidation and Other Trends.

I'm not sure whether I should be suprised, disturbed, or complacent.  But this morning, I was greeted by a couple major acquisitions in our overall market segment of Database and Application systems.

First off, Oracle has made another major acquisition, this time of BEA Systems.  BEA is a major provider of JAVA-based middleware technologies and service-oriented architectures (SOA).  This move shores up Oracle's middleware offering and goes a long way towards giving the 2nd largest software maker a more comprehensive set of product offerings to compete with the largest software maker, Microsoft.  I'm not really surprised by this move and feel like it makes good sense for Oracle to gain ground in this space.  The product lines are complimentary, even synergistic, and the acquisition allows it to gain on competitors like Microsoft and IBM.

In other news, I was a bit more surprised to learn that Sun Microsystems has acquired MySQL in a deal valued at around $1B. MySQL had been a major threat to the low end of the Microsoft SQL Server stack for quite some time.  Even more threatening, imo, was MySQL's ability to rally a strongly devoted community to help drive the open-source code base.  However, I'm not too sure about this move.  Sun has a wide range of hardware and software products, and has had a rocky road at several points in the past.  Can a company with such a wide, even diffused focus enable a database product to shine?  Tim O'Reilly, one of my heros, seems to think so in his blog commentary.  He believes "The acquisition is also a great fit because Sun has staked its future on open source, releasing its formerly proprietary crown jewels, including Solaris, Java, and the Ultra-Sparc processor design".  I'll reserve judgement, personally, on the situation but feel like there's great opportunity and great risk in this move for MySQL.

If you're wondering why any of this should matter to a died in the wool SQL Server person, I encourage you to take a look at Linchi Shea's article on "Checking out the Competition".  Linchi makes the great point that database platform vendors often try to foist off their latest features as new ideas, when in fact they've been around in other products for year.  So it's always good for a SQL Server person to know what the competition is up to because it makes you a better SQL Server profession. 

And finally, since I'm talking about broad trends, I really enjoyed Shashank Tiwari's article about dropping backward compatibility.  Although his discussion is about Java, I think he makes a very interesting point - that by carrying very heavy loads of backward compatibility features, we limit or even halt the ability of a product to evolve in directions.  I don't believe we've reached this point with SQL Server yet.  But I feel like we're starting to see a number of new "dead-end routes" for SQL Server, particularly with CLR and LINQ, that may in the end wind up being evolutionary deadweight for the database platform.  Undoubtedly, CLR has very valuable use-cases and LINQ might too.  But the resistance to these feature sets is quite strong throughout many enterprises, possibly resulting in major areas of investment by the SQL Server development team for what turns out to be, in an evolutionary metaphor, vestigial organs.

Thoughts are welcome!  Cheers,

-Kevin

Published Wednesday, January 16, 2008 3:41 PM by KKline
Filed under:

Comment Notification

If you would like to receive an email when updates are made to this post, please register here

Subscribe to this post's comments using RSS

Comments

 

steve dassin said:

Hello Kevin,

Good article. Apparently readers have no opinions on all these subjects. Let me see what I (we) can do:)

I think the Sun acquisition of MySql was a checkmate move. It was Oracle that attempted to slow down MySql with their acquisition of the InnoDB transaction engine. Now they have been paid back in spades. Now it's Sun vs. MS in a new competition for the minds and hearts of developers. The broad view is LINQ/net and Sql Server vs. java and MySql.  This is not about databases per se but a new war based on the object/entity model inspired by LINQ. I don't see IBM and Oracle in this war. They will have to be content to battle amongst themselves in the old database wars. (I'll come back to LINQ:)

As for "Checking out the Competition" of course I too applaud Linchi. But honestly we're first 'now' recognizing the advantages of eyes wide open?:) This attitude should be a given for a professional.  Perhaps better late than never, perhaps an airing of cheerleading vs.reality checking:) For most ansi features that MS has pre-annouced all one has to do to study them is read about them in Oracle or DB2 documentation (ie. OVER).  And as Linchi pointed out it often goes the opposite direction for other types of features. This attitude contributes to the great divides that are common in the industry.

And now we come to the "dead-end routes" like LINQ. I take the opposite view you do. There's a compelling case to be made (I've commented on this matter on this site) that if there is a deadweight MS see sql as it. LINQ is not just a piece of new technology, it's not just a local variable, it's a global one. LINQ is both an affirmation of the runtime environment for 'all' application development using an object model and a rejection of the sql interface.  MS can live with the physical file structure (the idea of relational data, rows and columns) but they don't want to live with the sql interface for application development.  MS explains this global move in terms of solving the historic impedance mismatch between two completely different environments. And they have picked their winner and at the same time the loser.  The rows and columns abstraction now ends at physical storage. The object abstraction and LINQ will take up everything else. Sql server is now something quite different than it used to be. Almost all development changes in server will be based on servicing objects and quite probably at the expense of features associated with a furtherance of the relational model.  Look at all the work on partitioned views in S2008. This lies at the heart of how LINQ will translate entity updates.  LINQ is still in its enfancy. I would expect it to appear to many just like sql did when it was intially introduced in the early eighties.  It will take time to get the matured version. What is truely ironic is I see no real argument in the sql community that LINQ represents a great opportunity for sql developers. MS is inventing a declarative language in the spirit of sql. Don't people see an opportunity to jump in and at least influence the design of the language? Or get involved in the LINQ translation/optimizations to sql. Over time as MS integrates LINQ deeper into server (returning entities) I can assure you the current translations will change:) Sql was most certainly not an implementation of the relational model. So sql folks shouldn't get hung up over this. The relational model would require the same strong typed runtime as net but MS is certainly not going there. But they are going to a place that sql skills can be used. And now Sun is going to go along with them. It's actually a small world if your eyes are open:)

best,

http://www.beyondsql.blogspot.com

January 19, 2008 11:05 PM
 

KKline said:

Excellent response, Steve, especially on LINQ.  I hadn't thought about it before in the terms that you've laid out.  So now I feel compelled to go back and re-examine the technology in this new light.

I think Linchi has a really good point and one that applies in a much wider sense to the SQL Server and possibly developer world.  Why is it that activities, behaviors, and processes that are fundamental to professionalism are so seldom known in the wider workplace?  I'm exasperated that nothing really seems to be established as a baseline of knowledge for what a SQL Server professional should know to be truly valuable to their employer.  For example, surely you'd want a database programmer to write decent SQL code but there are many who can't.  Ugh!

-Kev

January 22, 2008 12:01 PM
 

steve dassin said:

Hello Kevin,

>For example, surely you'd want a database programmer to write decent SQL code >but there are many who can't.  Ugh!

MS is also aware of this and is attempting to offer a solution:)

'The ADO.NET Entity Framework Overview'

http://msdn2.microsoft.com/en-us/library/aa697427(vs.80).aspx

"While the relational model has been extremely effective in the last few decades, it's

a model that targets a level of abstraction that is often not appropriate for modeling most business applications created using modern development environments."

Is this marketing nonsense or is it really the current thinking of database professionals within MS backed up with 7/8 years of research and development?

Should server professionals really be concerned about:

'Microsoft SQL Server 2008 and Microsoft Data Platform Development'

http://www.microsoft.com/sql/techinfo/whitepapers/sql_2008_dp.mspx

"Developers often spend countless hours deciphering database schemas and writing  

complex queries to retrieve the data that they need in their applications. The ADO.NET Entity Framework simplifies these tasks and enables developers to focus on the business logic of their applications."

Or is this something of no consequence and can be safely ignored by server folks?

I continue to be surprised by the lack of interest and discussion in the server community in the entity framework and especially esql. I wonder if it's an attitude of   lets ignore it and it will go away:) Does discussing katami in terms of features here and there really get to the heart of what MS is shooting for? It seems to this outsider that most folks are heavy into the 'how' of things and not they 'why'. Why is MS so concerned with the entity framework? What does it mean to the database programmer? Where does the sql language stand in the MS idea of application development? Is the idea of katami as a  'platform' and not a 'database' really significant? Sometime the trees really do hide the forrest:)

January 27, 2008 11:54 PM

Leave a Comment

(required) 
(required) 
Submit

About KKline

Kevin Kline is a well-known database industry expert, author, and speaker. Kevin is a long-time Microsoft MVP and was one of the founders of PASS, www.sqlpass.org.

This Blog

Syndication

Powered by Community Server (Commercial Edition), by Telligent Systems
  Privacy Statement