> The database design in Axiom is old. It uses random access files which
> > follow VM/370 LispVM design. These databases should be redone using a
> > better overall design.
>
>
> What is this database ?
There are several databases, you can see them loaded when the system
starts. These include information that the interpreter uses to run
the system.
What kind of information ?
They are random access files (a concept that seems to have been lost
from computer science :-) ). The first number in the file is the
number of bytes into the file to find the index, which is an association
list of (item . index). You can look up an item in the association list
and then the returned number is the byte index into the file for the value.
We still learn that dont worry.
It's very efficient in terms of disk I/O which was a concern way back when
I was a lad but it would make more sense to keep these in memory as they
are relatively small. Such a change would probably speed up the system
quite a bit.
Yes the trade of between doing disk I/O and using memory space are reversed.