Re: Slashdiot article about P6 -- MIDRANGE-L

On 1/7/07, PaulMmn <PaulMmn@xxxxxxxxxxxxx> wrote:

Steve--

I don't see the point of some of your wishes:

>long object and field names

Maybe 10 characters is a little short, but you can't tell me that
making a name longer than 10 characters will improve programming!
You can be obscure in any language.   COBOL has had long names for
years.  Names like "Input-File-Customer-Number" may sound appealing,
but once you get into a calculation, it quickly becomes a problem
because you get a massive run-on sentence:
"(input-file-quantity-ordered x part-master-file-selling-price x
customer-contract-file-special-discount-percentage =
line-item-discount-price)."


yes, within code super long names would be a hassle, but that is when
the name is used in a lot of statements.

In CL, something like:
  CALL  PGM( Payroll_MainMenu )

is prefered over:
  CALL  PAY0010C


>expand pointer size so the 16meg limit on spaces and strings can be
>eliminated. or scrap/redo the single level store architecture.
>Possibly, limitations and security problems of the SLS is what is
>preventing IBM from investing in our system.  You dont need the SLS to
>get the features of i5/OS.

The single-level-store is one of the advantages of the OS!  The main
advantage is that I can take a backup tape from -any- i/5, AS/400, or
System/38 and restore that tape on any of today's hardware (as long
as there's enough disk space).  I don't have to worry about drive
names, numbers, buss position, controller connections, or anything
else.  It loads, and I can run the programs!


what you describe here is terrific. But I dont think it has anything
to do with SLS. Object encapsulation and MI were invented back in the
day, I believe, by G Glenn Henry. Dr Frank is the one who developed
SLS.

Perhaps it's because the integration and SLS are so easy to
administer that IBM doesn't push the system-- they don't want to cut
into their consulting business!

>a garbage collection memory model that holds objects. this means you
>need a facility for defining and storing object types. ( you need a
>reflection type facility for calling the dispose method of the object.
>)   One use of something like this would be system APIs would return
>objects that are typed, that are self describing. In the case of a
>list objects API, the system would return an object that is holds an
>indexed array with the capacity to list all the objects on the system.

I'm not as far into OS modeling as you are-- but doesn't QTEMP handle
a LOT of the garbage collection on the i5?


activation groups is IBM's partial answer to the issue of garbage
collection. One problem with  AGs is the added work for the programmer
having to be aware of what AG their code runs in.  I am no expert, the
best I can say is .NET is very good and could not exist without its GC
model.

-Steve