August 19, 2009 at 9:49 pm
Comments posted to this topic are about the item New Databases
August 20, 2009 at 7:14 am
Sounds like the old issue of central command being consistent but distributed command being fast. Nothing new, just new ways to implement answers to an old question.
- Gus "GSquared", RSVP, OODA, MAP, NMVP, FAQ, SAT, SQL, DNA, RNA, UOI, IOU, AM, PM, AD, BC, BCE, USA, UN, CF, ROFL, LOL, ETC
Property of The Thread
"Nobody knows the age of the human race, but everyone agrees it's old enough to know better." - Anon
August 20, 2009 at 8:49 am
If there is one major thing I have learned in a 30+ career in technology, its that when you hear terms like "new", or "revolutionary" your best bet is to be as skeptical as is ever possible. So the concept of "new databases" is something I take with a very large grain of salt - about the size of a small Hummer!
Lets remember a few of the more recent "revolutionary" breakthroughs in technology were Windows Vista and Office 2007. Both have been major flops with such "revolutionary new technology" as asking "Are you sure?" whenever you even breath on Vista, and the Ribbon in Office 2007 which has driven users almost totally nuts. But these are just two of the very long line of rehashed old ideas, or mind-numbingly silly features presented as "new" and "revolutionary". Of course "Cloud Computing" is another concept now being presented as something "new and revolutionary" even though it was envisioned decades ago and has yet to show whether its hype or something truly revolutionary.
I would be happy to look at any "new" databases, but I fear it would simply be a repackaging of old, existing ideas - at least that is what happens to 85% of these kinds of concepts. I say 85% because that is the estimated number of lawsuits that get throw out when companies complain that other companies have 'stolen' their "revolutionary" or "proprietary" technologies. That is, in most cases, the "new" or "revolutionary" technology is found to be not new, not revolutionary, and usually something the company either directly 'stole' themselves, or stumbled upon not realizing they were infringing on others work.
Even the article you link to in your post doesn't really say much of anything - that is, it shows one company doing something different, and that is hardly revolutionary. The author then just talks in very nebulous terms about what might be coming.
Frankly, if it were my choice, I would prefer to see companies improve and enhance existing technologies and then agree more efficiently on standards. Enough with the new and improved, and revolutionary - lets get something correct and standardized across the database world - and then worry about inventing the next "big thing".
August 20, 2009 at 9:47 am
I'd have to agree with being skeptical about new revolutionary ideas. Besides being rehashed old ideas, or just passing fads that won't necessarily catch on, from what I've seen over the years a number of those revolutionary ideas are either solutions looking for problems, or solutions to actual problems that actually create bigger problems somewhere else.
In this case it looks like people from the "relational databases are bad and slow" camp had these built, becuase they didn't understand or want to deal with technologies like replication, or understand how to properly tune a database application to use the system resources most effectively.
I suppose these technologies have their place, but the danger is that too many people will build applications that depend too much on the non-relational system as opposed to just supplementing the RDBMS, that's when they will create other problems.
August 20, 2009 at 10:04 am
One thing I've learned over my long career is to try to keep up with "the latest thing" but don't put all your money on it. I remember when everyone was going to be coding in Pascal or Ada, because they ensured better up-front definitions, and now most companies demand C++ and Java derivatives.
August 20, 2009 at 10:27 am
I do agree that often things aren't "new" in terms of a revolutionary idea. But I do see things repackaged in new ways, which can make them much easier to use.
SimpleDB isn't anything new in terms of databases, but by pricing differently, and making this available on a large scale, it might be a better way do distribute some information than something like replication in an Express database (which I've seen done).
August 21, 2009 at 9:43 am
True.
But to take a classic old example, C++ did nothing that was not doable in C (early C++ simply translated to C), but properly used it enabled enormous improvements in black boxing, code management, data and code abstraction, library construction etc.
Eventually C++ (and deriviatives like C sharp) provided the management environment such that few would want to go back to doing an entire project in C.
...
-- FORTRAN manual for Xerox Computers --
Viewing 7 posts - 1 through 6 (of 6 total)
You must be logged in to reply to this topic. Login to reply