Sergey,

My quarrel was simply that the "clarification" really didn't clarify much.

SNIP
>As result a record pointer has been moved to the inserted record.
>Now lets go back to the help: "Do not include any commands or functions in your validation rules that attempt to move the record pointer in the current work area".

Right. Now here's an extract from VFP7's INSERT - SQL documentation:
---- Begin extract of INSERT - SQL
INSERT INTO dbf_name
Specifies the name of the table to which the new record is appended. dbf_name can include a path and can be a name expression.
If the table you specify isn't open, it is opened in a new work area and the new record is appended to the table. The new work area isn't selected; the current work area remains selected.

If the table you specify is open, INSERT appends the new record to the table. If the table is open in a work area other than the current work area, it isn't selected after the record is appended; the current work area remains selected.
---- End extract of INSERT - SQL

Note that it makes NO MENTION of [ work-area | alias ] even being valid here!
It does say: "If the table you specify is open, INSERT appends the new record to the table" but again no mention of alias/work-area, and the staatement after that one could be interpreted as the work-area not actually being used anyway.

>Does INSERT-SQL fells under this warning? Yes it does if you do Insert-Sql into the current work area.

You've proved that it does, BUT the documentation for it hasn't. That's my point.

>The fact that "INSERT-SQL" isn't in the list of commands means that this list isn't complete.
Yes, that would be the easier "fix". But possibly the real problem is that INSERT - SQL is not performing as designed. It looks to me that, if INSERT - SQL executed as documented, then its omission from the list could be legitimate.

Cheers