February 3, 2009 at 11:01 pm
Comments posted to this topic are about the item Guest editorial: Reasons to be Cheerful
February 4, 2009 at 6:48 am
I hadn't been following the MySQL developments, but this is good to know in case I am presented with a situation in which I have to deal with it. I guess Sun bit off more than they could chew at the moment. Hopefully they haven't alienated the MySQL people and they get it sorted out. If they don't, at least we still have SQL Server.
February 4, 2009 at 7:14 am
No doubt that Microsoft does a good job with SQL Server as a mature platform, but as an old geezer, I recall the year following the release of SQL Server after Microsoft purchased the code for Sybase to use as a basis for their enterprise database. They took the same type of criticism from admins and journalists that we are hearing about Sun and MySQL. Sun is a good organization and will eventually get it right but it takes time to tackle a beast of this magnitude.
February 4, 2009 at 9:12 am
SharePoint.
Sharepoint bloat because the Sharepoint team have decided not to talk to object relational experts that will improve architecture because most important components still use custom collections instead of SQL Server tables. And when SQL Server is used the code is crappy.
PerformancePoint.
One word greed because Microsoft bought what Ralph Kimball called MDX scripting tool instead of keeping the original and using parts of it to provide Winform and Webform for SSAS, Microsoft created a tool that most people are not willing to pay for. If Microsoft have created configurable Webform and Wniform doing MDX SAS and other tools vendors would have taken notice instead of ignoring Performance point.
LINQ-to-SQL.
The existing problems are pedestrian algebraic problems that can be fixed in the next version of Visual Studio by adding a LINQ to SQL designer that allow direct Store proc and .sql files execution. There are changes provided by both Microsoft employees and third parties it is not over because the main issue is to drop the sqlmetal from a 30th floor window. This product and Performance Point show brains solve problems aggression just cost development funds.
entity framework.
Drop it from the 30th floor window because most tasks can be better performed with automated SSIS package and the old DataSet.
On a side note everyday Microsoft leaves billions of dollar because departments don't work together.
MySQL
Sun will make it if it leaves Peter Gulutzan in Canada as the MySQL team did before Sun bought MySQL it is the same arrangement Microsoft had with Jim Grey. Peter Gulutzan created a relational engine before MySQL became popular and Oracle's Jim Melton and Peter Gulutzan are the only people to wrote books about ANSI SQL persistent stored modules. So Sun needs to keep the bureaucrats from the MySQL team and things will only improve because many web developers both Microsoft and PHP uses MySQL. On a side note there is MySQLTransactions in .NET easy move the JDBC Java data access code to .NET to improve MySQL ADO.NET implementation.
Kind regards,
Gift Peddie
February 4, 2009 at 9:58 am
I have written here before on the development history of MySQL, and indeed, my current employ uses it, with much pain. Don't get me wrong: I love that there are other ideas and competition in the DBMS space. However, for anyone to think MySQL is at this time a comparable platform to MS SQL Server or Oracle, has not done their homework, or is just flat out biased and/or without real knowlege of the other systems.
One must understand that MySQL is a "plugin" architecture, where the SQL service engine is a layer over potentially many different types of engines (15+ types). This is a feature and a bug. Only one of these engines was advertised as ACID compliant (INNODB), and MySQL did not even own that engine. This is what chokes progress on that platform. In my opinion, Sun is using it more to sell more servers, not databases... and give the market another big reason to buy their hardware.
They do have a market... primarily the anti-MS crowd.
February 4, 2009 at 10:06 am
That was an oversite because I know enough to say MySQL is not even up to SQL Server 2000 more like SQL Server 7.0 with service packs. And yes Oracle owns INNODB storage but Sun with time will only improve on the existing product.
I will never say it can be compared to Oracle or SQL Server but money and brains improve products overtime Sun have both.
Oracle paid more than 200 million dollar for INNODB storage so Sun did not pay too much for MySQL. Both Oracle and Microsoft improved their products by buying other companys with features needed to improve each products. If Sun did not buy MySQL things would only get worse IBM runs on all operating systems including Windows same thing with Oracle and SQL Server runs only on Windows.
On a side note Microsoft owns most none ANSI SQL Transaction through Jim Gray.
Kind regards,
Gift Peddie
February 4, 2009 at 12:22 pm
Gift Peddie (2/4/2009)
... but money and brains improve products overtime Sun have both.
OK, let's consider the "money and brains" situation:
Considering MySQL and MS SQL Server: both products have had basically the same amount of time, many people do not know that both products started very nearly at the same time.
One product had paid development thrown at it (translation: large amounts of money and brains). The other sat for a time with a far smaller pool of few, then went opensource (translation: trickle amounts of money, and functionally not more brains as everything still flowed through a single (or few) gate keeper(s)). One somehow grew 10 years ahead of the other (witnessed even by your own words above, in your comparison to SQL Server 7/2K).
I sincerely hope MySQL does improve. But it won't be on the "free" nickel. In TCO, sadly many only consider the blaring initial cost... but the cost in time and money in any required work-arounds, make-dos, and caveats far exceeds the initial cost of any product... many times over. Bad or inappropriate choices translate to lost opportunity costs. One must consider how much that choice will affect the enterprise, compared to other products, for at least a decade into the future. If one product is a decade behind another, what does that mean to the enterprise? How much is at the fingertips, and how much will cause endless sleepless nights?
Too many times the choice is made by personal familiarity and predjudice (free vs. cost of initial investment) instead of real world experience and reality.
Sun will make up their billion dollar investment in boosted server sales alone. Look at the main page of the MySQL site these days: what is advertised more? Servers or DBMS? Good luck on the MySQL platform going forward in big ways. It will only do so if it becomes a major paying stream and owns their ACID-compliant DB engine.
MS SQL Server, on the other hand, is and must be a paying stream for MS, and has far more "money and brains" that are actually fully dedicated to the improvement of the product thereof. This is not biased chatter... just looking at the situation as it has rolled on, over time.
I see MS making strides and substantial in-roads with light platforms of the flagship DB product, negating the intial startup cost issue. I see Sun pushing hardware more than ever.
Your mileage may vary.
February 4, 2009 at 12:46 pm
This will not be the first time Sun have failed pushing hardware I remember back in 1996 I could not get a price quote on Sun's Spark. The problem is Sun, IBM, HP and Red Hat and many other companys sells some version of Unix. I don't think it is practical for companies to buy MySQL instead of any other RDBMS Sybase included because it comes with limited features.
But MySQL getting ACID compliant storage is not complicated because ANSI SQL Transactions are a Unit of Work nesting is added by vendors and all uses Save Points to enable nesting but Save Point implementations differ because you can only roll back Save Point if your SQL Server is 2000 Sp3a and above. This MySQL can get with Peter in MySQL team because ANSI SQL have done the complex basic engineering.
The way I see it Sun could make money keeping web developers happy something MySQL was doing before Sun. And again I cannot compare MySQL to Microsoft because Microsoft had the multi talented Jim Grey from the start MySQL just found the money to hire Peter Gulutzan. Jim Grey can only be compared to Oracle's Jim Melton and I will take Jim Grey over Jim Melton any day because I can understand what Jim Grey writes and creates for software development.
Kind regards,
Gift Peddie
February 4, 2009 at 1:33 pm
Gift Peddie (2/4/2009)
But MySQL getting ACID compliant storage is not complicated...
Unfortunately, the evidence shows that it was/is no small undertaking for MySQL.
February 4, 2009 at 1:51 pm
Transaction is very complex math some of the reasons I prefer Jim Grey I interpret his work which Microsoft implemented in more than five places to developers like me.
So MySQL had a lot of work waiting for Peter Gulutzan like stored procedures and DRI and he was hired not very long ago so I still think it is easier to implement quantifiable mathematical Unit of work than the application layer version which require resource manager or provider to resolve the atomicity. That is not needed in the ANSI SQL version. And I am not saying there is DRI in MySQL either but both are complex tasks.
Kind regards,
Gift Peddie
February 5, 2009 at 8:13 am
dphillips (2/4/2009)
One must understand that MySQL is a "plugin" architecture, where the SQL service engine is a layer over potentially many different types of engines (15+ types). This is a feature and a bug. Only one of these engines was advertised as ACID compliant (INNODB), and MySQL did not even own that engine.
To counter Oracle's purchase of INNODB, MySQL hired Jim Starke to help them develop the Falcon engine for MySQL 6. Alpha versions of this have had some performance issues though. I wonder how much effort was taken away from the 5.1 product line to help build 6.0? I can't imagine that's a 1 man job to create a new database engine plugin that's expected to be enterprise class.
While there may be some people who move from MySQL to SQL Server, I imagine people in that area would be more likely to move towards PostgreSQL, Firebird, or one of the other open source databases out there. Both of those are full featured databases that have ACID compliance, enterprise capabilities, and SQL Standards for years that MySQL has only recently gained. They also both run on many platforms, so if you're running MySQL on Linux, BSD, or Solaris, you're more likely to move to a different database that can run on the same hardware/OS you already have.
February 5, 2009 at 12:11 pm
Chris Harshman (2/5/2009)
dphillips (2/4/2009)
One must understand that MySQL is a "plugin" architecture, where the SQL service engine is a layer over potentially many different types of engines (15+ types). This is a feature and a bug. Only one of these engines was advertised as ACID compliant (INNODB), and MySQL did not even own that engine.To counter Oracle's purchase of INNODB, MySQL hired Jim Starke to help them develop the Falcon engine for MySQL 6. Alpha versions of this have had some performance issues though. I wonder how much effort was taken away from the 5.1 product line to help build 6.0? I can't imagine that's a 1 man job to create a new database engine plugin that's expected to be enterprise class.
While there may be some people who move from MySQL to SQL Server, I imagine people in that area would be more likely to move towards PostgreSQL, Firebird, or one of the other open source databases out there. Both of those are full featured databases that have ACID compliance, enterprise capabilities, and SQL Standards for years that MySQL has only recently gained. They also both run on many platforms, so if you're running MySQL on Linux, BSD, or Solaris, you're more likely to move to a different database that can run on the same hardware/OS you already have.
I completely agree with the first paragraph. I wonder about Sun's efforts to continue this?
On the second paragraph, it does put in perspective where MySQL really is in the pecking order. However, having used most of those other systems, I wouldn't call them enterprise replacements. There is huge gulf between the class, features, and bugs of products, even amongst the ones listed here.
February 5, 2009 at 4:47 pm
Microsoft is looking for a few good BI developers to help with the next version of BI toolset.
This study starts on February 6 and runs till July 31. It will be a remote study and is open to all U.S.-based participants. We are looking for developers who are familiar with ETL, data warehousing, OLAP cubes, and MDX. We will be offering all participants a gratuity option in appreciation of your time. If you are interested in participating in this study please e-mail us at itusable@microsoft.com with "BI Developer" in the subject line.
Kind regards,
Gift Peddie
February 7, 2009 at 3:18 pm
I just read some news that Sun is back to it old Solaris tricks again so MySQL founder quit that is not good. Sun needs the Linux and Windows running web hosting and developing users.
http://tech.yahoo.com/news/pcworld/20090205/tc_pcworld/mysqlcofounderquitssun
:Whistling:
Kind regards,
Gift Peddie
Viewing 14 posts - 1 through 13 (of 13 total)
You must be logged in to reply to this topic. Login to reply