UH2@PSUVM.BITNET (Lee Sailer) (03/28/88)
This is a little long.
Some time ago, someone posted Codd's 12 rules for a *real* relational
database. Below you will find my estimation of how well RBase V 1.0, 1.1,
and the new "for DOS" and "for OS/2" (half an operationg system?)
stack up.
These are short comments, not completely qualified by ifs, ands, and buts,
so certainly some folk will disagree. I like RBase, even though they
do lie about the products true capabilities.
===============================================================================
E. F. Codd's relational rules :
---------------------------------------------------------------------{ RULE 1 }
* The information rule
"All information in a relational data base is represented explicitly at the
logical level and in exactly one way - by values in tables."
ASSERTIONS and VIEWS are stored in tables, as are screen and report forms,
at least partially. However, the column and table definitions themselves
are not. They are available as the output of certain LIST commands.
---------------------------------------------------------------------{ RULE 2 }
* Guaranteed access rule
"Each and every datum (atomic value) in a relational data base is guaranteed to
be logically accessible by resorting to a combination of a table name, primary
key value and column name."
Yes.
---------------------------------------------------------------------{ RULE 3 }
* Systematic treatment of null values
"Null values (distinct from the empty character string or a string of blank
characters and distinct from zero or any other number) are supported in fully
relational DBMS for representing missing information and inapplicable
information in a systematic way, independent of data type."
RBase seems to be up to speed on this.
---------------------------------------------------------------------{ RULE 4 }
* Dynamic on-line catalog based on the relational model
"The data base description is represented at the logical level in the same
way as ordinary data, so that authorized users can apply the same relational
language to its interrogation as they apply to the regular data."
No.
---------------------------------------------------------------------{ RULE 5 }
* Comprehensive data sublanguage rule
"A relational system may support several languages and various modes of terminal
use (for example, the fill-in-the-blanks mode). However, there must be at least
one language whose statements are expressible, per some well-defined syntax, as
character strings and that is comprehensive in supporting all of the following
items:
- Data definition
- View definition
- Data manipulation (interactive and by program)
- Integrity constraints
- Authorization
Yes.
- Transaction boundaries (begin, commit and rollback)
No. RBase's biggest failing is that it has no transaction logging,
commit, or rollback type capabilities.
---------------------------------------------------------------------{ RULE 6 }
* View updating rule
"All views that are theoretically updatable are also updatable by the system."
No. VIEWs of only one table are updatable, but VIEWS of joins are not.
---------------------------------------------------------------------{ RULE 7 }
* High-level insert, update and delete
"The capability of handling a base relation or a derived relation as a single
operand applies not only to the retrieval of data but also to the insertion,
update, and deletion of data."
Yes, but the CHANGE command does not allow all the expressions one might
hope for, so sometimes you have to resort to a cursor and walk through
the table.
---------------------------------------------------------------------{ RULE 8 }
* Physical data independence
"Application programs and terminal activities remain logically unimpared
whenever any changes are made in either storage representations or access
methods."
This is pretty easy for RBase, since about the only thing about the
physical database that can change is whether or not you build fast
access paths, i.e., B-trees.
---------------------------------------------------------------------{ RULE 9 }
* Logical data independence
"Application programs and terminal activities remain logically unimpared
when information-preserving changes of any kind that theoretically permit
unimpairment are made to the base tables.
Yes, but the applications programmer has to be careful not to use certain
constructs that defeat this.
---------------------------------------------------------------------{ RULE 10 }
* Integrity independence
"Integrity constraints specific to a particular relational data base must be
definable in the relational data sublanguage and storable in the catalog, not
in the application programs."
Yes. RBase has RULES which make enforcing the PK and FK assertions
pretty easy, though a "trick" is required for tables with compound
keys.
However, cardinality and domain constraints often have to be enforced
in the applications code.
---------------------------------------------------------------------{ RULE 11 }
* Distribution independence
"A relational DBMS has distribution independence."
Yes, trivially, because RBase databases cannot be distributed.
RBase has limited "networking" capabilities.
---------------------------------------------------------------------{ RULE 12 }
* Nonsubversion rule
"If a relational system has a low-level (single-record-at-a-time) language,
that low level cannot be used to subvert or bypass the integrity rules and
constraints expressed in the higher level relational language (multiple-
records-at-a-time)."
Yes, with willpower. You can SET RULES OFF, but you won't, right?
So, actually, the answer is NO.
==============================================================================
Additional Issues :
-> Are data base requests compiled or interpreted ?
Either interpreted, or intermediate code.
-> Are domains supported ? (a domain is a pool of values from which one or more
columns draw their actual values from - semantic data type.)
No.
-> Concurrency mechanisms acceptable ?
With care, RBase can support low volume concurrency.
-> Optimizer quality ? (how SMART is the optimizer)
What optimizer?
-> Composite field support ?
None.
-> Hierarchic clustering available ?
???
-> What application building tools are available ? Do you like them ?
The application tools that come with RBase are pretty good.
-> How complete is the data sublanguage (SQL ?) implementation ?
Despite claims to the contrary, RBase is NOT SQL. It has been
patched up to look like SQL, but critical features are missing.
The data language is complete enough, but is is highly irregular.
Sometimes variable names need '.' prefix, sometimes not, sometimes
expressions must be enclosed in parens, sometimes not.
It is typical to find users who write record-at-a-time while loops
to process a table, when a simple table level command is provided.
I have seen applications where the user sorts a table after evry
insertion.
-> Does the data reside on a mounted file system (one big file or multiple
files ?) or on a raw device (choice ?) ?
Data, definitions, indexes, in three files. Applications live
in their own files.