when should you use Cluster Records? I have never wondered about this since i encountered DataEase 100 yrs ago

100 ? we had used DataEase clay tablets in Babylon already.  

"UNSW Sydney scientists have discovered the purpose of a famous 3700-year old Babylonian clay tablet, revealing it is the world's oldest and most accurate trigonometric table, possibly used by ancient mathematical scribes to calculate how to construct palaces and temples and build canals." Mathematical mystery of ancient Babylonian clay tablet

Clustered records is in many ways a very "old fashioned" way of looking at tables and should in many way "never" be used.

Instead of using an index to find and sort data, the data is re-sorted in "index" order so when listed out sequentially they will appear to be in "order" without being sorted based on an indexed field etc.

The only good reason for using it is in lists (dropdowns) etc that "never" change as the table will need to be re-clustered if you add or change data.

So in the "Olden" days Multibox (combo/dropbox) you would use it to get the choices in chronological order.

In LE9 You will be able to sort/exclude and search in the Multibox without the need to cluster data and it will be based on the live dataset where you can even add new entries from the Combo and get it sorted on next use.

Thank you for that reply, nothing is better than to know the complete truth to one's curiosity

in fact, all end users prefer non abstract databases, but tuned datasets related to a certain data management task. so did guys from Software Development Laboratories, in 1985, under their  new name Oracle ;) really in our life we are not solving general problems so  clusters records of data are our usual things... missile attacks over our territory are related to or caused by solar activity and its influence on humans. this is not general opinion but clustered info ;)  we use ancient DataEase SQL 4.53 for  prediction of awful  attacks on our civil persons now. Solar storm surveys by ancient Assyrian astronomers

"Humans have been looking to the skies for as long as we have been around. Some of the observations made by ancient Assyrian and Babylonian astrologers more than two millennia ago survive in the form of cuneiform records."

Users don't care how data is stored but how it is presented.

A clustered database store the data in the order its sorted which in reality is very impractical if the data isn't static. 

So back in the day when you for instance put the phone book on a CD you would produce a clustered database so people could look it up chronologically like you did in the old paper version, but in modern databases it is an outdated idea the same way warehouses no store things according to free space rather than according to chronology as you have an index system that will tell you exactly where the stuff is.

yup! but as abstract algebraist from math department of my uni I consider other abstraction levels as those ancient clusterisation ;) and if this my level of understanding can be never achieved by finite automata like computer progs then it is better for me to consume ancient babylon DataEase SQL 4.53 for new DE9/LE9 are still in design ... Glory to DataEase Team!

it was in 199* we used DE 4.53 SQL with option to choose between Native or Oracle type of db. that was sales system for a huge cardboard paper mill near Kyiv based on Novell netware, Oracle and DataEase. this true story became a legend now ... but clustered records were on Oracle db 

DQL from 1981 is real AI now

so DataEase AI =DFD