Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Technology Stocks : IFMX - Investment Discussion -- Ignore unavailable to you. Want to Upgrade?

To: treetopflier who wrote (11699)	8/19/1998 1:19:00 PM
From: Charles Hughes	Read Replies (1) \| Respond to of 14631

>>> they are still talking about the emerging SQL3 standards. There have been 'emerging' since 1992. It they ain't emerged yet, they ain't a comin'. What is their current scope? That was the primary problem in the early 90's. Can't imagine it has gotten narrower. Trying to incorporate 'procedural' standards which are essential language based, not mathematics oriented was the stumbling block. <<< You know, Codd always complained that his predicate logic from his original designs was never implemented in RDBMS. But the problem is much bigger than that. SQL3 and mathematical relational logic is just more features for pros to use on the same old regular data. I like OO and the multimedia extensions, and they are a useful increment which give you many more kinds of applications you can do, but again, this is for data that a professional with lots of tech skills can set up in advance to be accessed in a very particular way. Predicate logic, or maybe any binary logic, would end up the same way, professionals operating on well-characterized data in rows and columns or some other predictable format. The big advance in databases has been the access to data that is not organized in advance in any but the loosest manner. That an amateur can search without DB skills. Most of the world's accessible data is now presented in this fashion, and yet the traditional database industry has completely ignored it. What am I talking about? The data on the Internet. Free text, FTP and hierarchical directories, keywords in html pages. That is the major advance in hypertext, multimedia retrieval, searching that the DB companies completely ignored as a technical challenge. That is where most of the data is, though. Sure, they sell RDBMS as a way to organize the data that can be organized that way, which is the minority. So we have sales records in RDBMS, and customer lists. No big deal, really. Now the search engine interfaces out there are weak and inconsistent. You enter boolean operators in one, wildcards in the next, only keywords in the third. Each search engine tries to address most or all of the web but ends only addressing a third of the sites, mostly in your language, and in your country. It is extremely difficult to come up with a search strategy that pinpoints the data you want. Very. There isn't even an interface that allows you to enter your query in a way that would let you get exactly, and completely, what you want in one operation. Further, the search engine indexing schemes are primitive compared to what needs to be done. This is a technical database problem, an indexing problem, a retrieval problem, an interface problem. This is a giant database, hundreds of thousands of times bigger than the world's biggest relational or OO databases. What are the DB companies doing about it? Nothing. They can't think outside their little box. The box of 'We sell individual packages to individual companies that have well characterized data on particular machines that we can stuff into our own system before dealing with it.' Instead, they leave the solution of the world's most important database problem up to 22 year old Java developers at what are basically advertising carriers. Now this is a known problem, and they are doing nothing effective about it. I wonder what would happen to a developer or other person who had their hands on a truly new data storage/access idea, working at a big DB vendor? Dump them, probably. To a gorilla, a human is just an ugly, hairless gorilla. Defective. To a traditional DB vendor, a guy with a new idea is just a programmer who has lost appropriate focus. Folks on the Internet who understand these things have been talking about this problem now for a decade or more. A lot of initial approaches have been suggested: Taking some of the associative logic ideas from expert systems, making use of formal fuzzy logic systems, (which work, by the way.) There are actually ideas from all over the map that come out in conversations like that about data and searching between those who get excited about these kinds of things. I can think of a half dozen approaches that could be added together in productive ways to do a much better job. But I doubt very much that any of us interested in this kind of thing could wake the brain of a DB company executive from it's serotonin-induced slumber. Too much golf, too much press attention, too much butt-kissing by too many people, from doing things the old way. So will they ever get it? I sure don't know. But the opportunities to do something truly new are right out there on the scope. You just have to look. Oops, lost a sixteenth so far on my IFMX since yesterday. Darn. Cheers, Chaz