Ellison faces pressure from both sides of the database market — IBM on top, and Microsoft below. A little more than a year ago, Gartner Group principal analyst Colleen Graham issued a study that showed IBM taking the lead over Oracle for the first time in the cutthroat corporate-database market.
Is Oracle about to Get Netscaped?
2003-05-28 Databases 26 Comments
I think it’s pretty sad to say, but yah, I’m quite sure they’ll be netscaped… in 1 year? 5 years? most likely the latter, but things can go pretty fast.
The big question though… will MS get Netscaped by MySQL or PG?
Even better question really is if databases will be in 5 years from now what they are today… I think it’ll be a completely different technology differing so much that the work placed in it today will be worthless.
Unfortunately the cost of the database isn’t the only thing that needs to be looked at when making a purchase decision for a major database. When you figure in support costs suddely IBM and ORacle are very similar and Oracle is a more flexible database. DB2 still feels like a mainframe database. And comparing the decision on which database to buy to internet browsers is so ridiculous it can’t even be seriously commented on.
Just something funny to point out, the title of the article on money.cnn.com is “Is Oracle About to Get Nescaped?” and it is on CNN.com–which is owned by AOL/Time-Warner and actually has the Netscape bar above it. Odd that they would use that choice of words.
Netscaped? You mean release the source code to the community, providing servers to host it on, and become a geeks favorite software? Try not to make up vocabulary.
“Even better question really is if databases will be in 5 years from now what they are today… I think it’ll be a completely different technology differing so much that the work placed in it today will be worthless. ”
I seriously doubt this considering RDBMS technology has been around quite a while and has matured, and is virutally everywhere. I just don’t think that new technology is going to even begin to replace this stuff any time soon.
There are still companies running COBOL and IMS after all.
Yes, I think somebody is about to eat Oracle’s launch. And rightly so.
Hey, what else does Oracle do? Just databases and related stuff? Even MS, with its successful product lines, is still venturing into new markets. Apple too.
Ellison and Scot McNeally, on the other hand, are just there, marking time, basically putting finishing touches to existing product lines, and waiting for their competitors to catch up.
I’ve personally used PostgreSQL, MySQL, and MS SQL, and quite frankly I like PostgreSQL best. Not necessarily the best featured, but a good solid product that I didn’t have any serious problems with. Just my opinion though.
I’ve used all the major commercial databases including ORACLE, DB2, SQL Server by Microsoft and Sybase. In addition to a couple of open source databases like MySQL and Firebird. In my opinion, ORACLE is the gold standard of databases. It is by far the most flexible and capable of the different offerings. However, the author is correct to point out that in the low end market this only goes so far. By and large, the vast majority of low-end users, particularly small businesses don’t need the number of high-end feautures that a full fledged database such as ORACLE have to offer. This is why databases like MS SQL Server and to some extent MySQL are growing. However, as customers become more sophisticated and grow their needs may change and they may one day need a high-end database and move up the chain. Although, they may never need to move up that far and you can’t make them.
I think the battle is more akin to Windows and Solaris then IE and Netscape. Although, Solaris has more enterprise data center features then windows in most cases it really doesn’t matter. Buyers will gravitate towards software that provides the lowest perceived costs and seems the most accessible. Sun, Oracle, etc. have a bad habit of moving upmarket when faced with competition. The problem with this is that although the high-end market may be profitable it is also finite. There will always only be five hundred Fortune 500 companies.
In a recent interview Larry Ellison remarked that Microsoft growth has been largely attributed to the fact that they have embraced the low-cost per unit commodity market. If Oracle wants to grow it will have to embrace that market to and put a package of products that provides a
The funny thing is they have the tools to provide such a package at their disposal. With the availability of linux(and in particular their partnership with Redhat) and partners such as Dell and even Apple. The can sell a compelling package that would be very attractive to small businesses. However, they need to get their ego’s out of the way and realize that the database business is changing. Microsoft and IBM, as well as open source alternative like MySQL and Firebird are good enough for most users. As those products improve they will continue to be good enough for most users.
From experience, I can tell you that changing your database engine in an application can be fairly costly(why do you think people still use the soon to be defunct Sybase and the already defunct Informix). Oracle would be better served by lowering their prices on the low-end, expanding their market, and ensuring that their is a clear upgrade path up the ORACLE database software chain. The battle is not likely to be won with features but by costs and the presentation of a cohesive package that doesn’t require a full time dba to perform the most menial tasks.
In the short term this may hurt the bottom line but is likely to be better then trying to hide at the top when Microsoft, IBM, MySQL, Firebird, etc own 95% of the market because they are good enough.
I think the author is remiss in not mentioning MySQL or postgreSQL as additional factors.
“Hey, what else does Oracle do? Just databases and related stuff?”
They also create App Servers (Oracle9iAS) and Development tools (Oracle9i JDeveloper). Although BEA’s WebLogic and IBM’s WebSphere are kings of the App Server market and Borland and IDEA or pretty much kinds of the Java IDE markets.
Brett, do you really think that when someone considers buying a database they debate between Oracle and MySQL? The only edge MySQL and postgress have over Oracle, DB2, and MS SQL is thier price point, and lack, in a great way, in features, comparetively.
CPUGuy, you are right to say that the open source alternatives and in particular MySQL don’t match up against ORACLE feature wise. However, it is my contention that a large number of users don’t really need the majority of those features. What I am advocating is a retail friendly version of ORACLE that small business users can build their application platform on. When a user requires those additional features then they can upgrade to a more feature rich version of ORACLE. Although, SQL is SQL. Those of us who have worked on the different platforms know that this is not always the case. Most users of MySQL and other databases are only in need of a glorified file system with standardized query abitlities. That is why these products are successfull in web applications, etc.(besides their price point).
The whole features issue is kind of like putting a Porshe engine in a Pinto. Most people just don’t need it and won’t use it, so why pay for it.
Market share is the key to keeping a company on top. Don’t ignore the small market. They shouldn’t fear the small business market, but see it as a new avenue for selling ORACLE and packaging their product accordingly.
They are sitting on a GOLD MINE with their Collaboration Suite, but are too naive to realize it. Anyone looking to find a full replacement for Exchange on Linux need look no further. It even might be Exchange done BETTER as it uses the Oracle Database for message storage, not some lame proprietary crap MS makes up jsut for the reason of locking other out.
BUT Oracle has their heads up their ass when they put this thing out and continue to do so. Look at the requirements for it! It is only geared for Huge Enterprise rollouts. Meant to replace Exchange in the Corporation when what the world needs is to replace Exchange at the real small level with something on Linux by a big company such as Oracle. Small Business is the one who would chomp at the opportunity to try an alternative that is cheaper, and built better. Small/medium business are already the largest integrators of Linux. I have been looking for Exchange replacement on Linux for over a year now, nothing comes close to what Oracle has.
All other attempts are piecemeal components. And Openmail is RPM ONLY (More FOOLS).
So would someone from Oracle read this and slap the face of the one responsible. Find a better TARGET!
Oracle serves several tiers of the DB market already with Enterprise,Standard,Personal and Lite (with different licensing options for each depending on how much you want to spend). They have had these editions for years but people don’t seem to know about them. People also seem to ignore Adaptive Server Anywhere (formerly SQL Anywhere) which is a kick-ass database product that is a perfect fit for small business. Up until recently at work we used SQL Anywhere with 10GB+ databases and they were up 24/7. Very stable database.
#1 Informix is not defunct. It was purchased by IBM 2 years ago. A new version, IDS 9.4, was just released last month. More releases are already planned for the next 5-7 years. 9.4 has a boatload of new feature enhancements. And Informix is the only database with advanced object-relational technology like the Time Series and Spatial DataBlades. Granted, Informix is under the DB2 brand of data management products at IBM. You don’t see the name plastered everywhere, but it is still a viable product with steady continued renewal licensing and new licenses. Financial companies, especially, are getting into the Time Series DataBlade and Real Time Loader, which no one else can do. And Informix still rock as the easiest to administer and most reliable at staying up. Don’t have to pay several Oracle “consultants” just to tune queries.
#2 As for DB2, IBM is making strong strides with this. It is already combining the best of DB2 (best query optimizer, data federation, good admin and developer tools), with Informix (ease of admin, object-relational, XPS data warehouse features), Red Brick’s star schema, and more. It will be a technical force to reckon with. Oracle will still probably build better *brochures* but their product is technically inferior from the ground up.
#3 Oracle’s main problem is arrogance. They think they are the best and they make you pay like it is. They also compete with their partners by trying to build applications. IBM has settled on middleware, development tools, open standards, and some hardware and services. They don’t compete w/ the likes of PeopleSoft, SAP, etc. Oracle won’t disappear, but their days of being #1 are over.
From a satisfied Informix/IBM customer
The thing is Oracle is one of the few firms out there that has all the pieces together to really put together a cohesive package that would be very very attractive to small business. However, they probably won’t do it because they seem to be more interested in selling to the large enterprise market then expanding their distrubution channels downstream. Don’t get me wrong, it is not easy to change from an enterprise vendor with expensive salespeople to a small business merchandiser where volume is the key to profitabiltiy. To tell you the truth I don’t think they should “change” in this regard, but if they want to have an enterprise market in the future they need get with the program and understand that there are a greater number of customers out there looking for a minimal set of features then those willing to overpay for software that they won’t use. Between NetLedger, Collaboration Suite, their application server, Redhat linux, etc. they could put together a really strong comprehensive package. Oracle should try to grow with their customers as opposed to only dealing with them when they are all grown up and comfortable with solutions from other vendors.
Your right about Oracle being arrogant. Instead of trying to get a handle on the needs of small businesses and putting together a cohesive package to serve that need their probably going to spend considerable effort to search for “studies” that indicate Oracle is #1, etc. The bottom line is who cares. Whether or not they are #1 in the market is irrelavent, because they should be acting like they are #3.
BTW, without getting into a flame war: I realize that IBM is still supporting Informix and putting some work into it, but the fact remains that DB2 is their focus. They are only trying to keep those customers that are on Informix happy until they can create a “seemless” transition to DB2. It is not in their best interest to push Informix over DB2 and they never will.
Well, it’s true that perceived arrogance may kill Oracle.
Just like apple except that they can’t seem to publicize their products as well. The proper thing for them to do is simply said, find new markets. For this I think that one thing they can do is buy Mysql, or create an embedded version of their database, then form a subsidiary to create applications that will sit ontop of this database and actually create new markets for their other products.
This means not just the back end, but full applications with clients targetting small to midrange companies…
Examples are telecom, telecom billing, health sector, data warehousing, transportation, payrol…
Then, they will stand to make money not only on their databases, but on the desktop clients, addons and perhaps on hosting the applications online.
Oracle *does* have an “embedded” database – Oracle Lite. Works on Win CE, Palm, Symbian EPOC. $100 per named user. Kind of funny to see the people talking about Oracle and why they’re doomed here and these same people obviously have never looked at the Oracle product line-up. Same thing with the guy chiming in about Informix and Time-series and Spatial plugins. Oracle has the same thing and has for a long time. I’m happy that person is a happy Informix/DB2 user but sheesh…at least do some research. Hell…Oracle has had ORDBMS technology for years as well. XML is now handled natively in the DB as well.
Oracle may have products called “time series” or “spatial,” but I doubt they measure up to the technology in Informix DataBlades (rtree indexes, NAG functions built-in, etc). Heck – even DB2 doesn’t have this level of technology yet. One of the main reasons they acquired Informix.
It is true that IBM is stressing DB2. It is the code base for a common vision that Informix and IBM already had independently. DB2 was a bit broader in some areas so it is the codebase for the “next generation” database that can handle all the different content-types and DBMS uses (OLTP, DSS, etc.). OTOH – there are more developers working on IDS now than when Informix was acquired. It isn’t just in maintenance mode. The “next generation” vision is going to take awhile. And there are too many companies depending on Informix for IBM to cut it off. Besides, it is profitable and gave IBM the database market share crown. Finally, Informix was good at a market that IBM hasn’t done well in, the small-medium businesses and the vertical ISVs that serve them.
Informix won’t take over the world, but it is an important piece of the IBM Data Mgmt strategy and a viable data base choice in it’s own right. Plus – no one can offer a technology stack like IBM (Websphere, DB2/Informix, Tivoli, Lotus, MQ, etc.). Microsoft is trying, but theirs is limited by their shoddy “single” OS platform.
“Heck – even DB2 doesn’t have this level of technology yet.”
Does DB2 still not have an equivalent to PL/SQL or TransactSQL for stored procedures?
DB2 has stored procedures. I think you can do them in Java or a stored procedure language. Not sure what it is called. Informix allows you to write stored procedures in Java (or C) and has a JVM built-in to the server engine. Very nice feature!
DB2 isn’t multi-threaded like Informix, but again I think that is a unique feature of Informix allowing parallel index builds and table scans and such. We don’t need/use that kind of power in our small business. For folks that really get into database architecture, there is much to love about Informix. They just never got the marketing right. As MS and Oracle have proven, you don’t have to have a good product to sell, just good selling of a product (or its “future”).
However, has SUN and Oracle ever thought of merging into a mini-IBM? Integrate Oracle DB into the iPlanet line up, start bundling this software with their hardware with a “nice price”(tm) and I am sure they could easily provide a bit of competition to Microsoft and IBM.
I’m sure someone has thought of it. However, Oracle runs on many platform besides Solaris. A SunOracle would have to still support them, but keep up the Sun mantra that Sparc and Solaris are superior. The recent “compact” between the two for marketing a “complete” solution is probably best for both.
Besides, it seems that Oracle knows what it is trying to do (all software except the OS). Sun doesn’t seem to know.
Also, unless you have a decent marketshare/economies of scale, a proprietary operating system can be an expensive undertaking. Just look at BeOS as one example, mind you, they made their own hole, however, that is an entirely different issue.
Oracle database is an incredible product. Please take your blinders off. I think before making value judgements about Oracle you’d probably want to actually look at and use the product first. I’d suggest otn.oracle.com and looking through the documentation to see what the product is capable of. They charge big money for the enterprise version for a good reason: the product is worth it.