Open Source Challenge

  • Open Source Challenge

    Open source software is a challenge to the way traditional software vendors operate. I'm not sure how great a challenge it is, or if it matters that much. OpenOffice is in many ways compatible with Microsoft Office, at least for most people. However I'm not sure OpenOffice has made significant inroads into this market.

    Or if it matters if it does. As long as the data in the text document, spreadsheet, whatever can be read, does it really matter which product is used? People will prefer different ones for different reasons and I don't think either can kill the other.

    What about databases? Apparently there is an open source competitor for Oracle that's abailable. EnterpriseDB Advanced Server is designed to be Oracle-compatible and work with applications that are supposed to run on Oracle database servers without changes.

    That's big news.

    Oracle database servers are expensive, both the server and client licenses, and if you can eliminate that cost it can impact bottom lines. It's one thing to port applications to MySQL or PostgreSQL, but another to just point the application at a new server. I can see companies seriously considering the switch in the latter case.

    I don't know of any SQL Server compliant competitors in the open source world and I'm not sure how vulnerable SQL Server is. Without some of the subsystems, it might not be as cost effective to change databases. Especially if you are using some of the subsystems available in SQL Server, SSIS/DTS, Service Broker, etc.

    Keep a close eye on this new product and be ready to sell some Oracle stock. They may take a big hit on this.

    Steve Jones

  • That's the most ridiculous statement you've made, and believe me, you've made quite a few.  As if companies are just going to 'switch' over to an unsupported product at the drop of a hat.  And the fact that you talk about Oracle as merely a db server while you blow smoke up SQL Server with all it's 'subsystems' proves that you know nothing about Oracle.  If you must make such an absurd statement it's best you back it up with some valid research, rather than just say 'apparently'.

  • Steve is not making a ridiculous statement. If a project is initiated to make an Oracle-compatible database, the guys who designed it must have in mind the idea that it could be used in some circumstances where Oracle is now used. Oracle will kill it stone dead by producing a free 'Lite' or 'Express' version, in much the same way as Microsoft did to potential competitors with 'Express'.

    The threat is more apparent than real. The real value of a system is in its performance in a production environment. Try using PostgreSQL in a production environment and you will grow to love SQL Server once more. The open-source databases are fine in the academic environment but come nowhere close in the features that really matter.

    I see databases such as SQLite as being a much more serious threat, as they address an area of the market that the big players have neglected. Oracle and SQL Server add more and more features with each release, thereby making them unlikely choices for small websites or PC applications. A database that has a tiny footprint and good performance is doing something the big players can't, or won't do, because there seems to be no immediate commercial reason for it. By the time one of the new generation achieves significant market share, it will be too late for the industry behomoths to crush it.

     

  • I'm going slightly adjacent to the main topic but I don't get why people equate "Open Source" with "unsupported". How many of you get much of your support from Microsoft? I have called Microsoft on maybe four or five issues, ever. Any time they were actually able to solve the problem it was because they had information that was not made available to the public. I run into lots of issues with software of all sorts, Microsoft and other providers, and I can almost always find a solution to my problem (or at least an explanation of what's going on) via the internet. With the proliferation of bloggers out there now it's only getting easier to correct problems for free. Open Source communities are often just as big as communities based around commercial products so do the math (or the boolean logic).

    [Open Source] != [unsupported]

    [font="Tahoma"]Bryant E. Byrd, BSSE MCDBA MCAD[/font]
    Business Intelligence Administrator
    MSBI Administration Blog

  • I agree with Tatsu: open source != unsupported, it is a different support model (provided by local experts rather than a central organization). Nor does published source necessarily mean free, one does not give up copyright by publishing source for the benefit of one's customers.

    As was pointed out, most major organizations are unlikely to toss SQL server or Oracle simply to adopt these alternatives (one possible exception would be an organization interested in substantially customizing its DB engine for special uses), but there is a significant market in small operations that don't truly need the massive capabilities (and massive costs) of the big guns.

    On the other hand, the availability of these alternatives helps keep the pricing/licensing structure of the big guys more friendly. If they had total control of the market, one can imagine the 'take it or leave it' mentality that would develop.

     

     

    ...

    -- FORTRAN manual for Xerox Computers --

  • I think this is fantastic news.  I've been a PostgreSQL supporter for years, even as so many applications and people have switched to mySQL.  It's great to see someone leveraging the power of Postgres in this fashion.   For the SMB marketplace it makes sense to have a reliable enterprise level database that's less expensive that the big 3 (Oracle,SQL,DB2).  You can purchase support from companys that sell it and it's still much less expensive than Oracle and SQL server. 

    As Tatsu pointed out Open Source != Unsupported.  If you went to the EnterpriseDB Website you would have found pricing information for supported deployments. 

    Also there's this quote from Tom Goguen, vice president for system software, Sun Microsystems. “We chose EnterpriseDB to work with us on PostgreSQL support, because we’ve been continually impressed by their PostgreSQL expertise, customer focus and support capabilities"  from this press release stating that EnterpriseDB will be helping out Sun's suport staff when it comes to PostgreSQL.

    To help us help you read this[/url]For better help with performance problems please read this[/url]

  • I'm sure I've made some ridiculous statements, with probably more to come, but I'm not sure this is one of them.

    Porting costs are high and they subsantially impede the move for companies to switch products. It's hard enough getting software written without adding in costs of moving between platforms.

    I never said companies would switch at the drop of a hat but rather they'd consider the move. Licensing costs are small compared to people costs, but they do add up to significant numbers in places. Epsecially for smaller companies. If you're an Oracle/J2EE guy and working at a small company, this may be a great alternative. A supported alternative.

    PostgreSQL, like MySQL, has been around a long time and there are quite a few places you can get support from. It will cost something, but I'm sure Oracle suport costs as well.

    As for the free Oracle version, maybe it will help, maybe not. Depends on how crippled it is.

    And for Microsoft, in my decade+ or working with their products, I've called a few dozen times. I think they have some of the best support, along with Cisco. I have used them and appreciated their level 3/4 support where I can work with someone who can dig deeply into the products and solve an issue.

  • I also agree with opensouce != unsupported.  Sure I can get the source for MySQL and do my own builds.  Why bother.  They give me the Windows builds and actually perfer that I use those.

    Look at Apache vs IIS.  It takes at least a full day to properly secure IIS.  Apache can be at least water tight in an hour. I've integrated Apache/PHP with SQL Server just fine.

    I move both sides of the line.  My employer has so many Microsoft plaques that we are looking for more wall space.  I love SQl Sever and VB .Net.  I tolerate most other things from Microsoft.  I also love MySQL, Apache, and PHP.

    When working with an Oracle customer I went in, hat in had, confessed my ignorance, and worked hand in glove with their DBA's to get the project done.  A very successful coooperaion.  I still don't know much about Oracle except that I have to code my SQL statements differently when deling with dates.

    Stay flexible.

     

    ATBCharles Kincaid

  • Interesting article, interesting question.

    However there is an ENORMOUSLY HUGE gap between "compatible" and "supposed to be compatible".

    Even if a product is an exact clone of another, bug fixes in the first product will not be made in the second. The team working on the "clone" product may not even be aware of the bug fixes.

    Are you going to trust your corporate payroll data to the "clone" ?

    I love open source, believe me. But lets be serious. Oracle, SQL Server, DB2.....they are mature products with incredible power and scalability. Nobody is going to be able to clone them.

  • "Are you going to trust your corporate payroll data to the 'clone'?" is a good question.

    If the clone has support options (a la MySQL) then why not.  With something like Apache or PHP there is enough comunity support that many major organizations rely on them every day.

    I have not looked at EnterpriseDB specifically so I'm just preaching on general priciples.

    ATBCharles Kincaid

  • Even more: they (EnterpriseDB) are targeting not only Oracle.

    Read this blog article

    They're now offering Oracle-compatibility layer named "Redwood" mode.

    And : "At the same time, they are looking at a “Redmond” mode for SQL Server compatibility. "

    I'm waiting for this mode.

    Maybe, X years later, we'll see EDB topics on SQLServer sites. =)

  • In there he says, "PostgreSQL, on the other hand, requires you to do a little tuning here

    and there and know a little bit about databases."  Ther is one reason that we have not used it.  In SQL server we have been doing "zero maintenance" deployments long before the term came into fasion.

    In some cases (all too often) we are the customers IT department.  It (whatever we are talking about at the moment) just has to run all the time without anyone there to baby sit.  Sure we go pay a visit and watch the thing run and do tuning.  Mostly we start a job like the tuning wizard remotely and then go get the results in the morning.

    OK, so I went to http://www.enterprisedb.com/ just to check.  What the gives here?  Who said EDB was open source?  They don't!  It may be based on an Open Source database at the design level.  Just look at the download page to see what kind of source code is available.

    Sure there is the free version and you could run lots of different bussinesses on it alone, including payroll.   How is this different from SQLExpress?  It's a loss leader to get you to buy the real stuff when you outgrow the free one.

    At least with SQL Server I can have a database on the server and the handheld.  By the way:  Hey Steve Jones! When are we going to start seeing SQL Mobile articles around here?

     

    ATBCharles Kincaid

  • "...Oracle, SQL Server, DB2.....they are mature products..."

    SQL Server 2000 is a mature product.  SQL Server 2005 and it's management tool are not.  MS's penchant for adding unneeded bells and whistles to sell more products is an excellent reason to seek open source alternatives.  Of course, each new release requires service packs.

    "Porting costs are high and they subsantially impede the move for companies to switch products. It's hard enough getting software written without adding in costs of moving between platforms."

    This comment also applies to porting from SQL Server 2000 to SQL Server 2005 and from DTS to SSIS.  Another reason that open source alternatives are appealing.

  • I have used Oracle and SQL Server in production, I have not had reason to call Microsoft their documentation is the best in the business; I had called Oracle Metalink support they are good if you know the in and out of your problem.  Oracle have data type problems while Microsoft have limited none standard TimeInterval implementation, so if you need time in 200 countries format any developer can do it for you in Oracle but it is not that easy because Unicode in SQL Server is very easy to use.  So businesses use both as needed.  And the subsystems SQL Server Agent with Integration services can replace a lot of system jobs in the right hands.

     

    Kind regards,

    Gift Peddie

    Kind regards,
    Gift Peddie

  • >>EBD is open source

    EDB is not open source. Now they're just "cheap Oracle"

    >>How is this different from SQLExpress?

    Differences:

    1. EDB challenging not free versions, they're trying to challenge enterprise versions.

    2. What do you think about 99.99%-SQLServer-compatible DBMS on Solaris or Linux?

    Many our customers (possible customers) want to utilize their Unix-boxes, but we are limited to x86-NT.

    If EDB will produce "Redmond-mode" - we'll look at possibility of EDB-port of our system.

Viewing 15 posts - 1 through 14 (of 14 total)

You must be logged in to reply to this topic. Login to reply