Is MySQL Disruptive?
from the read-the-quotes... dept
A couple weeks ago I was talking with someone from a small tech company that is using Oracle for their backend database. This is a company that needs a hefty database, and it’s doing quite a lot of work for them. It’s also costing them an awful lot, and they’ve been playing around with things like MySQL to see if it can allow them to do what they need to do at a much lower price point. While the company admits that MySQL simply can’t handle what they’re trying to do right now, they wonder how long that will be true. Apparently, they’re not the only ones. Wired is running an article looking at how MySQL is quietly sneaking up on the bigger players. Often, new companies are using MySQL because the engineers they’re hiring to build the product have used MySQL for personal projects. What struck me most about this situation, though, is the response from Microsoft in the article about the challenge from MySQL: “Typically, MySQL and other open-source database companies are used in small departments.” This is a similar response to the one I’ve heard from Oracle employees. It’s also absolutely true. However, someone should send these people a copy of Clayton Christensen’s work on disruptive technologies. That’s exactly how they start. The established players always ignore the upstart, pointing out that it’s targeting a lower segment of the market and doesn’t have all their features. What they forget is that the disruptive technology gets better – and becomes increasingly “good enough” for a larger and larger segment of their market.
Comments on “Is MySQL Disruptive?”
PostgreSQL
PostgreSQL at one time (and may still be) managing the SETI database. Back ‘in the day’ it was dealing with 1.5 Terabytes.
Back when MySQL was in the 2.X versions.
PostgreSQL is already a threat to Oracle. One day MySQL will catch up to where PostgreSQL is today.
Re: PostgreSQL
but, when MySQL passes PostgreSQL, will anyone notice? PostgreSQL has all these nifty features now, but no one knows. It’s like a well-kept secret.
Then again, I’m still confused by PG, but I moved from ORCL to MySQL without many problems outside of missing features. Could ‘ease of use for regular people’ be a feature that’s helping MySQL gain ground now while it works on other missing features?
While it’s better to be powerful than simple, I think Simple makes a great jumping-off point to launch the attack vs Powerful.
Re: Re: PostgreSQL
MySQL is building on the fact that every hosting company seems to deploy it rather than PostgreSQL. My guess as to why that is would be that MySQL is a lighter installation.
Having used both for years, ease of use for developers goes to PostgreSQL. I don’t see that regular people really inteface with their DB engine much, rather they just want to run software X, which 7 times out of 10, mentions MySQL first as a DB.
Re: Re: Re: MySQL has one advantage over PostgreSQL
It’s got marketing. MySQL is owned by a company. I compare MySQL to PostgreSQL with Red Hat to Debian. Red Hat is a company unlike Debian. It has been more popular. But Debian is starting to become more popular now due to the fact Red Hat has to do things to make money which aren’t always in the best interests of its users.
PostgreSQL is far superior to MySQL but doesn’t have the company backing for it. Its all volunteer like Debian. Eventually, I believe this will be its advantage over MySQL.
Firebird? - Relational Database for the New Millen
I quote: “Firebird is a relational database offering many ANSI SQL-92 features that runs on Linux, Windows, and a variety of Unix platforms. Firebird offers excellent concurrency, high performance, and powerful language support for stored procedures and triggers. It has been used in production systems, under a variety of names since 1981.”
While I have not used FireBird I have come close to using it when a MySQL database did not meet my needs (views). It is highly spoken of in some circles. Free, Open Source, easy to obtain, under active development, and cross-platform. As a corporate developer it seems like a much better solution.
— jb
MySQL
I think your analysis is not at all incorrect, and MySQL is definitely a good pick for a lot of people. I do not find it strange that Oracle would ignore the problem for two reasons:
1. There is nothing they can really do about it. They could give ORCL awya for free, and in some ways they do (there is not really license protection for it), but it is such a hog…
2. It is not the tables that make the difference to large companies, it is PL/SQL which has millions (billions?) of LOC written in it with all sort of fun business logic somewhere in it.
MSFT is in a similar position. Great many people who run a Windows environment have taken to using MS SQL Server, but for obvious reasons MSFT is not making it available for Linux. I may think it is a mistake not to provide a linux-based MSFT product, but Redmont does not care about my opinion. Thus, they really have no leverage to induce people into *not* using MySQL (or Postgres).
As to the main point of the article – whether MySQL is a disruptive technology I am ambivalent. To me, a disruptive technology is something that really changes the way you do something, dramatically lowering the cost of doing that something, and rendering all or most of the previous investment obscolete. Given how companies buy RDBMS products I do not think this works out for open soruce RDBMS products so far. As long as they are copying and re-implementing technology that already exists in the commercial products, they are at best an evolutionary technology that gives a small leg up to start-ups, but does not change the equation radically.
In the same sense I do not find Linux disruptive – Windows is cheap enough of a platform to run applications and is certainly fairly stable at this point. Moreover, other open source OS alternatives exist and have existed for a long time. It is the cumulative effect of open source — free to use and modify RDBMS, Web Server, App Server, and OS that are disruptive.
No Subject Given
I think that when MySQL has views and stored procedures, most people will have little reason to continue to pay for a DBMS.
No more CALs, no more per-processor licensing fees.
Brand names
I’ve also been using MySQL for a recent personal project (art4iTunes.com) and have been very impressed with its capabilities – coping admirally with caching 25Mb of data over a couple of weeks.
However in my day job I’m usually employed to work with the one of the usual big databases, and for enterprise use I believe it will stay that way.
Managers and salesman love the fact we have a big, brand-name as the database behind the project, so they can point to the “best-of-breed” solutions and wave the “unbreakable” flag.
SAP
you also can’t overlook the fact that mysql has the gorilla of enterprise business applications in their back pocket… SAP. Along with transfer of SAPdb, which was the old Adabas product, to mysql, SAP also moved a lot of developers to help them extend mysql for use in enterprise deployments. SAP would like nothing better than to see Oracle’s cash cow get squeezed, and of course Microsoft is breathing down their neck as well, so it’s an added plus to hit them where it hurts (M$oft generates cash from Windows, Office, Exchange, and SQLServer… uses that cash to develop their other businesses).
Re: Daffodil DB - http://www.daffodildb.com
Daffodil DB is a J2EE certified, SQL-99 and JDBC standards compliant 100% Java RDBMS. It helps you build comprehensive, efficient and powerful relational database applications in Java to handle complex needs.It comes in two editions- Server and Embedded edition. Visit http://www.daffodildb.com