Re: Scalable DB Solution?

new topic     » goto parent     » topic index » view thread      » older message » newer message

On 4 Feb 2001, at 18:52, CK Lester wrote:

> Kat, can you post a snippet of your database? I'm curious as to the
> format... 

</p><p>    
<h1>applicant </h1>  <fld01>ap-pli-cant</fld01> <br>     
<fld29></fld29> <fld18></fld18><br>   
1. <defblk> <fld20>01</fld20> <Syntax>noun,sing</Syntax> 
<Semantic>human,htitle</Semantic> <def>a person who applies, as for employment, 
help, etc.</def> </defblk><br>    
   
    
</p><p>    
<h1>applications </h1>  <fld01>ap-pli-ca-tions</fld01> <br>     
<fld29></fld29> <fld18></fld18><br>   
1. <defblk> <fld20>01</fld20> <Syntax>plur</Syntax> <Jmp>application</Jmp> 
</defblk><br>     
 
applicant
ap-pli-cant

1. 01 noun,sing human,htitle a person who applies, as for employment, help, etc.

applications
ap-pli-ca-tions

1. 01 plur application

The Eu code can be told to find the word within the h1 tags, then subseq each
defblk,
and search within each for syntax, semantics, or any tags i or it wants to add
in the
future. Any added tags won't break anything. Any removed tags won't break
anything.
So Tiggr can add pronunciation tags and specify to the DB manager to go find
them,
and it will, or examples of misuse as they are found, or anything i cannot think
of right
now. Tiggr can likewise spec a tag in a location be removed, and it doesn't
crash the
manager when the tag isn't there.

>Doesn't seem efficient to me to have a text file with delimiters
> as a database format... Show me the light? smile

I am striving for maximum versatility here, not max speed. If i wanted max
speed, i'd
have fixed length records, and spec the records at compile time, and be unable
to
change them,, and put up with them being the wrong length a vital 1% of the
time. And
not be able to add new fields or delete any. The XML form is in keeping with
Eu's
sequences. They are any length, any depth of subscripting,,, and in my code, i
can
slice them by column too. The tag/tag can contain any tag that's not the name of
the
immeadiate bounding tags. For instance, this won't cause an error: "<h1> <html> 
</h1>". It's deceptively simple, like chess.
 smile

Kat



 
> ----- Original Message -----
> From: "Kat" <gertie at PELL.NET>
> To: <EUforum at topica.com>
> Sent: Thursday, February 01, 2001 12:35 PM
> Subject: Re: Scalable DB Solution?
> 
> 
> > On 1 Feb 2001, at 10:18, CK Lester wrote:
> >
> > > Anybody have a good website for reviews of large scale databases? I'm
> > > thinking Oracle vs. DB2... Personal opinions welcome...
> >
> > Any XML/SGML database is scaleable, as far as the database management
> software
> > has the ability to read it. I did Tiggr's database in XML format from the
> start in 1991,
> > and i have not regretted it. My getxml() in Eu can read any tags i add, of
> any length or
> > subdivision. But if you are wanting to read existing databases that
> someone else
> > wrote, i can't help.
> >
> > Kat
> >
> 
> 
>

new topic     » goto parent     » topic index » view thread      » older message » newer message

Search



Quick Links

User menu

Not signed in.

Misc Menu