Weekly Edition
Return to the Press page
| |||||||||||
Predictions for 2004 (IT-Director)
IT-Director looks forward to 2004. The first prediction is that desktop Linux will succeed, but that's not all. "Finally, I believe that 2004 will be the year of the MySQL database. Unlike Linux and Apache, MySQL has not been a publicity magnet, but its use is growing and it stands on the verge of being taken seriously as a database to compete with Oracle, DB2 and SQLServer. It is already eating into their market share through the word-of-mouth marketing that turned Linux and Apache into formidable forces in their own right."
(Log in to post comments)
Predictions for 2004 (IT-Director) Posted Jan 5, 2004 17:08 UTC (Mon) by mmarq (guest, #2332) [Link] "The paradox of the PC market is that even though over 100 million PCs are sold each year, the PC has not become a commodity ?!!!. As such, it is defying normal economic logic. As it is currently sold, it is a far more complex-to-use product than, say, a car, but as yet no-one has managed to do a good job of packaging it simply, say, with browsing and home office capability and little else.?!!!"...(the industry havent sold anything else)Hmmm... this guy defenitly its not connected to the PC business... The problem with the "PC" type of appliance, being it a complex general purpouse machine, is that it is sold like a radio a toaster or a a TV set,..., though being more complex, by far, than a car. IMHO new appliances will be decoupled from the general PC, but the general machine will still rule by far,... and Linux has to catch it up to rule with it.
Predictions for 2004 (IT-Director) Posted Jan 6, 2004 7:38 UTC (Tue) by haraldt (guest, #961) [Link] I'm not so sure. With finer-grained program packages as we see it in free software, businesses will also have more control on what they put on their desktops. With this, we'll see more ready-made solutions sold as "plug it in to your infrastructure and run". (As opposed to a general-purpose package you'll have to configure on each PC). There's nothing new in putting an office package and a web browser on a number of PCs and spread them out on office desks (as opposed to f.ex. a pc used for graphics production), but the point is, while it'll more or less autoconfigure to fit with your network (think easy-made special-purpose distribution) it won't come with a presentation package added in if it's not asked for. I agree he was less than lucky with his example. Both web browsers and text editors needs mucho memory, decent speed and a good screen, and there's no need to not make it a general-porpose PC while you're at it. This idea disappeared after the millenium rush I think (web browsers don't look good on TV screens), but you'll find plenty of other nifty combinations. Perhaps "The borders between special-purpose devices and the general-purpose PC will continue to blur" is a better way to put it. (But that's not much to write about, is it?).
Predictions for 2004 (IT-Director) Posted Jan 6, 2004 16:44 UTC (Tue) by mmarq (guest, #2332) [Link] Well it wont be in 2004,... perhaps in 2005 the "paladium PC" will light everybodys minds for the need of a general purpose "Linux (PC) workstation" defenition.The gross of the industry is still in the domestic and the SOHO market.This generaly called consumer market, where the PVR, like TIVO, will shine and a "hundreds" of other brands including home made ones out of general PC hardware in "Flex" and "Micro" sizes,..., i've build one just to myself. Defenitly there is a new trend here, but the general purpose PC will still rule to the point of M$ wanting to transform XBOX in a PC. The entreprise market always were more visible and profitable. It includes a lot more money for single unit, so the efforts for "autoconfiguration", installation ease, reliability, security,... but it is far from the gross of the market.
Predictions for 2004 (IT-Director) Posted Jan 6, 2004 10:52 UTC (Tue) by leandro (guest, #1460) [Link] One more falling for MySQL siren's song... or perhaps, he's just thinking that MySQL can really do for Oracle or IBM DB2 or PostgreSQL.Perhaps too it's not his fault, but the developers' who have been creating such lousy SQL apps that they feel nothing missing when running under MySQL. Less bad that MySQL is actually if unofficially deprecated for MaxDB, aka SAPdb. But SAPdb is really not quite on a par with PostgreSQL, being more a very old AdabasSQL brought to parity with Oracle 7.2. So one is baffled, why use substandard MySQL as a stepping stone to almost-standard SAPdb when one can just go for the real thing with PostgreSQL? One could think of a reason that developers used to simple file-and-records management with things like Sleepycat probably get less intimidated by a SQL DBMS if they start with something middle-of-the-way such as MySQL is. But then, MySQL isn't really SQL and isn't really a DBMS, and they learn incredibly bad practice which will be very hard to unlearn if favor of really best practice with PostgreSQL, IBM DB2 or Alphora Dataphor.
Predictions for 2004 (IT-Director) Posted Jan 6, 2004 12:33 UTC (Tue) by dps (subscriber, #5725) [Link] I am developing applications with a database behind them right now and they are real business applications. While "real SQL" might allow me, in theory, to implement almost the entire application as a stored procedure it is much simpler to use a general purpose programming language.None of these applications needs anything that MySQL does not support, in fact MySQL supports more of SQL than these applications require. A solid permissions system and support for multiple table queries is sufficient, and I suspect most applications of databases think the same. IF there was a compelling reson to use triggers, complex stored procedures, etc then I would suggest MySQL is not the tool for the job and head for postgres, oracle, DB2 or whatever.
Predictions for 2004 (IT-Director) Posted Jan 13, 2004 13:54 UTC (Tue) by leandro (guest, #1460) [Link] > they are real business applications. That doesn't mean they are elegant, meaning among other things maintainable and scalable. > While "real SQL" might allow me, in theory, to implement almost the entire application as a stored procedure This isn't a theoretical issue, and no one eve proposed to do so. In fact, quite to the contrary: theoretically speaking the relational model enables one to implement most of a system declaratively. > it is much simpler to use a general purpose programming language. Stored procedures can be coded in general purpose programming languages. PL/SQL and SQL/PSM are just alternatives. There are stored procedures in C, Perl, Java and what-not.
|
|||||||||||