I have been thinking about the issue with "massive amounts of data" and this
time I am going to approach this from a different angle. Its starts with a
few simple questions:
1. Why are we not using the locality principle more: keep the processing
close to the data.
2. Is this really useful information to the user. I think it maybe be
already answered by the use of the word data instead of information. Why do
we not concentrate on providing better ways of
filtering/soring/formating/collating data into information instead of
relying on the poor user having to extract what he needs. We should think
thin at all layers in our architecture. I find the closer we are to
modelling what the user requires not want they user thinks the closer we
come to building a fast and usuable system. The problem above stems from the
fact that this industry just is dominated by systems programmers how are put
in charge of writing up some use cases because they know OO. What is really
needed is people who can think outisde this box and come up with novel ideas
to help the user do there work more efficient.....you just have to do proper
task analysis and hire HCI consultants. My experience as both a HCI
consultant and EJB/CORBA specialist has enabled me to achieve success on
most projects because I live in both worlds.
3. We have multithreaded operating systems so why not use backrgound thread
evaluation to give the user a preceived faster response time instead of
blocking until all data is displayed. (note this is dependent on the type of
client application). So we could initially return 50 beans details while in
the background loading the rest. I hate the way that my email client hangs
when I ask do a send and recieve request. Why can't I continue doing other
work while this folder is loading.
more later... have to do some work
William
> -----Original Message-----
> From: Laird Nelson [SMTP:[EMAIL PROTECTED]]
> Sent: Saturday, March 04, 2000 3:14 PM
> To:
> Subject: Re: Patterns for massive amounts of data?
>
> Assaf Arkin wrote:
> > IF you use a session bean, it must load all the records, serialize them
> > and return a serialized copy.
>
> {grok grok grok grok}
>
> Why can't a stateless session bean return, say, an Enumeration or an
> Iterator backed by a JDBC ResultSet? Is it because neither Enumeration
> nor Iterator nor ResultSet are Serializable instances?
>
> Puzzled.
>
> Cheers,
> Laird
>
> ==========================================================================
> =
> To unsubscribe, send email to [EMAIL PROTECTED] and include in the
> body
> of the message "signoff EJB-INTEREST". For general help, send email to
> [EMAIL PROTECTED] and include in the body of the message "help".
***********************************************************************
Bear Stearns is not responsible for any recommendation, solicitation,
offer or agreement or any information about any transaction, customer
account or account activity contained in this communication.
***********************************************************************
===========================================================================
To unsubscribe, send email to [EMAIL PROTECTED] and include in the body
of the message "signoff EJB-INTEREST". For general help, send email to
[EMAIL PROTECTED] and include in the body of the message "help".
