Re: Multi row blocks

comments in line...

;-{ mcs

"Julesm" <julesm_at_ip242yh.plus.com> wrote in message news:2DMfc.31064$h44.4445336_at_stones.force9.net...
| Hi Mark,
|
| I guessed you would respond thus, and myself and the rest of the team are
| with you. 'dave' does have some issues :)
|
| And I believe you might have hit the nail on the head about him not
wanting
| to find a solution.
| Howevere, to be fair, we did employee the services of an Oracle
consultant,
| who also couldnt provide any answers; all we basically got from him was,
| 'yeah, that shouldnt be a problem', but no idea as to how to do it.

unfortunately, that happens too often -- and that would certainly give grounds for a feeling of 'those people' can't be trusted

in all fairness to dave, his comments were obviously not intended to be published to this group, but likely were for the internal audience (i.e., i'm not sure if he would have posed his concerns the same way if he thought they were going to forwarded to 'those people') -- how about if you invite him to join the thread or to provide comments for publication here?

|
| The performance issue was also raised with Oracle, in front of our £9mil
| paying customer, to help add some weight to the problem, and Oracle looked
| at it and said, 'yep, thats a bug, we'll get on and fix it', only to be
| jumped on from Oracle USA who said, 'no we wont!'
|

ooo! bad form! but kind of sounds like sales -vs- customer service -vs- cost center budgets

can you describe the performance problem? and the bug? for that matter, the project? is this custom development or oracle applications? is oracle providing custom development services? what phase is the project in?

| I guess our problem is simply that we dont have the expertise, and when we
| tried to hire some, it left us with a bad taste.
|

looks like your project is in somewhat in a war zone, and you're not getting reliable reinforcements. or is it a matter of bad project planning, not identifying the risks up front?

| Any links or pointeres to some functionaly rich 9i stuff might be a help.
| (or screen shots at least). We're just all struggling a bit to get the 9i
| environment to match up to our Functional Specification, and are thinking
we
| should do it in something else, but what?
|

could you give an idea of the functional specs that you're trying to match 9i to? if you want to continue offline, perhaps i could provide you with some less general advise.

| I hate thin client stuff, seems we take 10 steps forward in thick client
| development, only to be thrown back 10 leaps trying to run it all in a
| browser.
|

you gotta code for the target environment... it can be frustrating, and forms is not the friendliest environtment to code in -- especially if you've got another pardigm in mind. again, i invite you to continue the conversation offline with more specifics (or online if you don't mind publishing the specifics). i'd be happy to provide any insite that i can.

| Many thanks
| Jules
|
| "Mark C. Stock" <mcstockX_at_Xenquery .com> wrote in message
| news:V_qdnWk36PwsXePdRVn-sA_at_comcast.com...
| >
| > "Julesm" <julesm_at_ip242yh.plus.com> wrote in message
| > news:AVSec.29700$h44.4286475_at_stones.force9.net...
| > | Hi again,
| > |
| > | I ran that by our forms man, and got the following?
| > | Is he right do we think?
| > |
| > | -------------------------snip-------------------------------
| > | Jules,
| > |
| > | Thanks for this, but as ever, these people are saying yes it can be
done
| > | using Oracle Forms or a Java grid. This seams to be a favourite
| response,
| > | but noone is ever prepared or able to point us at something that
proves
| > it!
| > |
| > | The first problem is where to get a java grid that will do what we
want
| > and
| > | of course the problems that will inevitably follow implementing a
grid,
| > such
| > | as bugs in the grid, Oracle and the Grid supplier blaming each other
for
| > any
| > | problems, etc. etc.
| > |
| > | The other problem (and this will apply to both the grid and the
non-base
| > | table MRB) is going to be performance. It's all well to develop these
| > things
| > | on a small database on what is a reasonable sized application server
| with
| > | just a couple of developers connected over a 100MB network, but what
| > happens
| > | when the application is supporting 100's of users some over not much
| more
| > | than a 64K link? Very often, the developer solution is just to bung
more
| > | hardware at the problem (Microsoft!) but may not be possible here, and
| can
| > | never be the correct solution.
| > |
| > | I would take anything that you're told here with a large pinch of
salt.
| On
| > | your other question (re: performance) the reply suggested that we
remove
| > | triggers from the form but we (STL) know that it is a bug in the
| > Application
| > | server, Oracle know it's a bug in the application server, but it
appears
| > | that the chap that replied does not!
| > |
| > | I hope to look into whether good, simple code can be written to
provide
| > the
| > | functionality required (or palm the search off onto someone else!).
| While
| > my
| > | gut reaction is no, I'm happy to be proven wrong!
| > |
| > | Regards
| > | Dave.
| > |
| > |
| >
| > well, jules, i don't remember ever discussing this with 'dave' so i
don't
| > know about the 'as ever' part ;-)
| >
| > if 'dave' understands forms, my previous response about the non-base
table
| > block should be sufficient information about that technique -- if it's
not
| > sufficient for him, then you'll need to hire a consultant to mentor your
| > forms guru <insert another wry smiley here>
| >
| > there is sufficient information about extending forms with java at the
otn
| > site -- 'these people' (or 'this people') suggested that approach as an
| > avenue for investigation, but it looks like 'dave' wants the testimony
of
| an
| > expert witness. whatever the case, it kinda appears that maybe 'dave' is
| not
| > real interested in finding a solution
| >
| > i'm not sure what the background is for the performance comments --
seems
| > like a general complaint against not only oracle but also microsoft.
| perhaps
| > 'dave' had some bad coffee before he responded? whatever the case there,
a
| > slow client configuration will be slow against a big or a small database
| > (it's the bad SQL, not a bad client software tool, that that scales
poorly
| > as the database gets larger). is 'dave' concerned about forms as a fat
| > client in a client-server scenario? that would seem unlikely if he's
| talking
| > about dial up connections. if you're considering doing forms 9i via the
| web
| > server, then the user's network connection really has little additional
| > impact on application performance (compared to other interface options),
| but
| > the communications link between the application server and the database
| > would be a larger factor.
| >
| > i also don't know quite what the issue is with triggers and the
| application
| > server -- we deployed a nice robust forms 9i application on 9iAS release
2
| a
| > year ago for the state of virginia and never had any problems with
| triggers
| > on forms or the database.
| >
| > so -- hopefully 'dave' can find someone else to do his work and everyone
| > will be happy.
| >
| > let me know if i can be of any further help (and keep a close eye on
| 'dave')
| >
| > ps: i tried to attach a large pinch of salt, but that functionality is
not
| > supported in a text-based news reader
| >
| > ;-{ mcs
| >
| >
| >
| >
|
|
Received on Fri Apr 16 2004 - 06:54:28 CDT