IAS 9i Rel. 2 9.0.2 General

I made an ias course by Oracle (GNC Partner) in Vienna, Austria and i hearded about BUGS and BUGs and so an.
as good as Oracles DB 9i is as bad ist this release !!
Our (me and the trainers too :((( ) opinion is:
This Release is worse and cant be really be used in production enviroments for cutomers. Why does Oracle Release so bad versions ?
Microsoft is not better in realeasing badly versions very soon for marketing reasons but customer get very very angry about this .
when will a new a better ias released ?
9.0.4 announced as new release ?
Further Question: i have ias 9i Release 1.0.2.2.1 with
Portal and reports services in productiv use.
Can we update to ias 9i Rel. 2 or must we install it completly new ? Is export and import portal apps in news release ? ............ ??????????????
when

Hois,
There are couple of issues here.
1. Bugs: What are the bugs that you are referring here? Are they not fixed in 9021 or 9022 patchsets?
2. 904 is scheduled for release Q1FY04.
3. What exactly are you referring to when you "can we update to ias 9i Rel2 or must we install it completly new"? Are you asking whether you can upgrade? If yes, then please note that except for Portal, the upgrade path for all of Oracle9iAS exists.
Pavna Jain
Oracle9iAS Product Management

Similar Messages

IAS 10g rel 2 BI and Forms services

I have the IAS 10g rel 2 infrastructure on a RH v3 server and, as we need Discoverer, plan on installing BI on the same server. We also need Forms and Reports services as well.
As I don't see Forms services in BI (although the documentation says it is) the following questions arise:
- If I install BI without Reports services, is it possible to install Forms and Reports services from the stand-a-lone version into the BI $OH? Or, would it require another separate $OH for the Forms and Reports services?
- If I install BI with Reports services, is it possible to install just the Forms services into the BI $OH? Or, again, is a separate $OH required for just the Forms services?
Your responses are much appreciated!!

There are three installation types that are pertinent to this question:
From Oracle Application Server 10g CDs, you can install BI and Forms Services and configure Discoverer, Reports, Forms (and even Portal). This has dependency on Infrastructure.
From Forms and Reports Services CDs, you can install Forms and Reports (in a seperate Oracle Home, never into BI and Forms install O_H). This has no dependency on Infrastructure.
From BI Cds, you can install Discoverer and Reports with no Infra dependency. It has to be in its own Oracle Home only.
So if your need is to run Discoverer, Reports, Forms in one Oracle Home but have no infrastructure dependency, it is not possible.
Regards
Pavna

Connecting a Computer to an HDTV

Hey everyone,
One of the most frequently asked home theater questions we see here on the forums is “How do I connect my computer to my HDTV?” Well, depending upon the model(s) you own you may have a number of options.
For starters, the most common option to connect a computer to an HDTV is to use a VGA cable. Often referred to as RGB, D-sub 15 or even generic “PC” inputs on many TVs, these connections are present on a healthy majority of desktop and laptop computers. Some laptops will even feature miniature versions of these ports that can be easily converted to a full-size VGA connection through the use of a special adapter.
The second option, connecting the two using component cables, is primarily used by consumers that own an older HDTV. Most often found in the home theater section of a store, these cables are not only relatively inexpensive, but can also generally be identified by their red, blue, and green color coding as explained by Rodney-GSHQ in this other thread. There is, however, one slight problem: most video cards do not feature component outputs.
The next available option is to use a Digital Video Interface (DVI) cable. These connections are quite common on newer desktop computers, and are often considered to be the digital successor to VGA. That’s right – everything up to this point has been analog only.
Last, but not least, you may have the option of connecting your computer to your HDTV using an HDMI cable. HDMI outputs are typically only found on newer (and higher-end) computers, but are considered by many to be capable of providing the highest-quality digital signals available.
Regardless of which method you choose, remember that you will need a separate cable to run audio from your computer to your TV! Additionally, it’s strongly recommended that you double-check your TV and computer’s owner’s manuals for step-by-step instructions as there are a number of precautions that you should take before attempting to connect the two.
Still not sure how to proceed? Give us a call at 1-800-GEEK-SQUAD (1-800-433-5778) and we’d be glad to help you out!
Aaron|Social Media Specialist | Best Buy® Corporate
Private Message

I know for a FACT that HDMI and DVI are intercompatible. I was talking about the sets themselves. NOT the specs. Many HDTVs have EDID tables that are not compliant with PC signal specs or in some cases, they give off EDID information that isn't understood by various electronic devices that rely on such information. This is a well-known fact and has been a particular problem with older sets. Yes, you can tweak the signal from your PC using PowerStrip, the Nvidia/ATI driver hacks, or whatever applies to your particular setup, but those don't always work. As far as I'm concerned, it doesn't work until it works out of the box and there are only a few sets that will work with PC signals over anything other than VGA by design. I know this from my own testing done at Best Buy on nearly two dozen of the 1080P LCDs with my own HDMI/VGA equipped notebooks using the latest (non-Beta) drivers from Nvidia, ATI/AMD, and Intel. I tested both HDMI and VGA from the notebooks for nearly three hours. VGA was the clear winner for out of the box connectivity on most of the sets and HDMI lost big time because the signal was either not displayed at all or displayed improperly with no clear way to resolve the issue. The Sony Bravia LCDs were the only ones that passed both the HDMI and VGA connectivity tests that I did. You simply got lucky with the two LCDs you have used because they were likely intended for multi-media use. The Westinghouse certainly was at any rate since they have always catered to that market.

Modify INITIAL in partitioned table

Dear all,
Facts:
oracle 10.1
aix 5.3
i have a table with the INITIAL value 1M:
partition PARTITION_83 values less than (201103)
tablespace D_DML
pctfree 0
pctused 40
initrans 1
maxtrans 255
storage
initial 1M
minextents 1
maxextents unlimited
and i want to change that value to 512 k.
i review the next http://ss64.com/ora/table_a_part.html but how can i do that ???
thanks all.

Why? What is the problem you're trying to solve?
I'm assuming you're using locally managed tablespaces since you're using a recent version of Oracle. I'm assuming your partitions are relatively large since partitioning is generally used for very large tables. Given that, the INITIAL parameter no longer controls the size of the initial extent-- instead, in a LMT, it sets a lower bound on the total size of all the extents that are initially allocated to the partition. It seems unlikely that you would have less than 1 MB of extents allocated to your partition-- there's a good chance that it's in a tablespace with uniform extents that are at least 1 MB. If there is more than 1 MB of data in the partition, what would changing the INITIAL parameter accomplish?
Justin

Recommendations for Analog to Digital conversion

I am seeking the help of all the Pros out there. I am moving my extensive analog music collection (LP/Cassette) to the digital world. Are there any recommendations for creating a CD-quality sound/volume level. I am currently using the standards (click/pop/hiss reduction, compressor and/or hard limiting). I am recording at 44.1/16 Bit and saving files in mp3 PRO-320 kbps for sake of storage. I welcome any assistance in regard to Plug-ins, equalization or other suggestions.
Thank you in advance
Paddy 41
Pentium 4/3.2 GHz, 2.0 MB Ram, Soundblaster X-Fi Elite Pro, Windows XP PRO SP3, Adobe Audition 3.0, Sony PS-LX250H Turntable, Denon DRM-600 Cassette.

>I am recording at 44.1/16 Bit and saving files in mp3 PRO-320 kbps for sake of storage. I welcome any assistance in regard to Plug-ins, equalization or other suggestions.
You shouldn't save files as MP3s until all the processing you want to do has been done - this is a lossy format, and every time you open a file saved like this it gets re-decoded, and then re-encoded when you do a save. And the quality degeneration is progressive. And unfortunately, recording your files as 16-bit is also not going to be quite the thing to do if you are going to do any sort of amplitude processing at all, either, especially if you are recording at a lower-than-optimum amplitude, which is usually the case. 44.1k is fine, though!
So, to get over the potential problems, the thing to do is to digitise your files as 44.1k 32-bit floating point, and until all processing is done, store them as Windows PCM wav files in this format, because that is uncompressed, which is what you need.
If you want to reduce noise on cassette recordings, it's worth using a fairly high FFT setting, even though it takes longer to do the processing. Since most of the noise is not LF, this tends to work better, although there is also a case for doing the NR twice, once with a lower FFT setting and once with a higher one, but not trying to take out too much in one go.
As for click reduction - well, you have to experiment. But letting the software determine all of the levels tends to give you a pretty good starting point. Another trick that's sometimes worth it is to transform your files temporarily to M-S stereo instead of L-R, and treat each channel separately - you generally get different levels of clicks in each, usually more in the S signal, whilst the M cancels out quite a few. And in general, the less clicks you can get away with processing, the better.
Hard limiting? Never used it on a conversion like that at all, and wouldn't even consider it - transients are already likely to be distorted, and hard limiting them further seems like a
very strange thing to do...
Other tools that are useful, especially on cassettes are iZotope's Ozone (some of which is already in Audition) and HarBal. You can restore usually distorted frequency responses with the latter, and get a good average value automatically and easily. And judicious use of Ozone's enhancer can make them seem a lot less like the results came from a cassette...
One of the useful things you can do with Ozone is to selectively widen parts of the spectrum. Since most records of music tend (for reasons of tracking) to have had the bass forced to be virtually mono, with subsequent similar consequences for other instruments with substantial LF content, it's worth expanding just about everything under about 250Hz, possibly by more than you might think.
Generally, 'normal' EQ doesn't have a lot of use when processing vinyl, if it's relatively recent. And that generally means anything from the 60's onwards. If you think things are seriously wrong, the chances are that it's your monitors and environment that are misleading you. Just bear in mind that when this stuff was produced, it was all monitored in a professional environment - more so than it might be today, even.
No doubt there will be a few other responses - I don't think that the foregoing is anything like complete, but is at least a start.

Use of scope resolution operator in PHP development example

I am relatively new to OOP in general so this question may sound dumb but
In this documentation: Query Performance and Prefetching
To quote that text, "
The $db->setPrefetch() call is used to set the prefetch value. The microtime() calls are used to show how long the report took to generate.
A new Db::fetchRow() method is used to get one row at a time. It is called in a loop after the query has been run"
Why the does it use the scope resolution operator? Isn't it used for just static methods and overridden methods.
I do not see any static declaration when this method is entered into class.
To quote the method declaration:
/** * Fetch a row of data. Call this in a loop after calling Db::execute() * * @return array An array of data for one row of the query */
public function fetchRow() {
$row = oci_fetch_array($this->stid, OCI_ASSOC + OCI_RETURN_NULLS);
return($row);
Am I missing something really trivial thing about this concept?
Thank you.

It's shorthand documentation for "A new fetchRow() method in the DB class is used to get..."

Getting Administrative Priviledge error when trying to install Pro on 2008 Server

We are trying to install Adobe Pro 8 and recieve the error message that you have to be an administrator, i have tried using the local admin and the domain admin account...Please advise

We have no idea about your server's configuration or any of that nor should we. That aside you will have to work out your issues yourself. You are a) trying to install an Acrobat version from the stoneage on a relatively modern OS and b) generally Adobe desktop apps are not supported on any server operating system.
Mylenium

Create a "draggable" image in Flash CS6 Pro

Hi folks,
Firstly, whilst the answer would be great, i'm also hoping for people to point me in the right direction for finding these answers myself...
I'm new to CS6, and relatively innexperienced with flash in general (certainly new to AS3).
I'm working my way through a project, and right now, i'm trying to create a background image or layer, that that is bigger than my canvas, the idea being that you can click, and drag the mouse to look around it, but NOT pull the edges of the image over the canvas (i'm sure theres a more appropriate word than canvas too).
My main issue is that i'm struggling with terminology, so whilst I know what i want to acheive, putting it into words that experts understand, or gives relevant google hits, is proving to be an issue.
Hope someone can help.
Tim.

This is pretty easy but will require some code. Here's a sample file you can look at. (we call the canvas the "Stage").
In this example I made a 200x200 document size with an object that is 400x400. What the code will do is figure out how much larger the image is than the stage. It will then calculate a rectangle size and position where the object is allowed to be dragged.
Example FLA (saved down to CS4)
The code is pretty simple:
// what code to run when the mouse is pressed
mygfx.addEventListener(MouseEvent.MOUSE_DOWN, handleMouseDownF);
// what code to run when the mouse is released or leaves the
stage.addEventListener(MouseEvent.MOUSE_UP, handleMouseUpF);
function handleMouseDownF(e:MouseEvent):void
          // get the extra area we need which it's allowed
          // to move in (object size minus the stage size)
          var extraWidth:int = int(mygfx.width - stage.stageWidth);
          var extraHeight:int = int(mygfx.height - stage.stageHeight);
          // use that extra space to make a negative value rectangle
          mygfx.startDrag(false,new Rectangle(
                    (-1 * extraWidth), // x
                    (-1 * extraHeight), // y
                    extraWidth, // width
                    extraHeight)); // height
function handleMouseUpF(e:MouseEvent):void
          mygfx.stopDrag();
To start I converted the big ugly object into a symbol (F8), a MovieClip. I then clicked on that MovieClip and in the properties panel I gave it the instance name "mygfx". This allows me to access that object in code by just using it's name, like you see in the code above. I'm trying to keep it simple.
I attached a couple listeners for the events MouseEvent.MOUSE_DOWN (the user is holding down the left mouse button on the object) and MouseEvent.MOUSE_UP (the user let the left button go). I'm running a different function for those 2 different events as you can see.
The math is easy but the concept of what's going on here is a little harder to explain.
When I created the symbol, I used the default upper left registration point. That means any time you move the object on stage, the x/y position reported in the Info Panel will be the location ot the "upper left" of the object. Check the help section if you don't know what registration points are.
With upper left registration and the object placed at x:0, y:0 on the stage, you can only see 1/4 of the upper left of the image. In this circumstance you can use a special function called "startDrag()" on any display object to allow the user to drag it. To keep it on screen you can specify a rectangular area it must be confined to. That's where the simple math came in.
I subtracted the stages size from the larger image size which told me how many pixels to the left and up the drag was allowed to go. You want to subtract the stage size so you reduce the amount it can move so it won't go off the stage. We know the stage is 200x200 and the object is double at 400x400, so 400-200=200. So I want to make the limiting bounding Rectangle(-200,-200,200,200). That means I can move the object up negative 200 pixels and left negative 200 pixels. If you change the size of the object the math will calculate the correct amount (as long as the image is larger than the stage height/width). Just keeping it easy for now.
Lastly when the user lets the mouse up the dragging is stopped with the display objects "stopDrag()" function.

Packages & classpath on os x

Hi,
I had some questions about where i should be putting class files. i've got files in a package com.one2one3.io, so i've got a directory structure /com/one2one3/io/ and i'm wondering where is the generally accepted place to put that, its working for now in ~/Code/com/one2one3/..., but then i have to specify the classpath everytime i compile...
that brings me to the question of setting the classpath in my .tcshrc file so i don't need to do that everytime... google says this should work:
setenv CLASSPATH .:/Users/stephen/Code/:${CLASSPATH}
but it isn't i still get class not found errors when compiling.
can i put the packages i write in /Library/Java/Extensions/com/one2one3/...? from what i understand the jvm on mac searches there by default so i wouldn't need to specify the classpath everytime i compile. thats working for me, but i want to check and make sure it won't cause me any problems down the line...i guess mainly i'm wondering what you all do to deal with these things.
and is it usual to include the source .java files in the /com/one2one3/... directory?
Stephen

I'm relatively new to Java in general, but I've been developing on Linux and Mac OS X a bit and I have some thoughts I can share. Someone will jump in here and correct me if I'm wrong. But I think I've got it. In any case here is a web page you may find helpful on the javac command.
http://java.sun.com/j2se/1.5.0/docs/tooldocs/solaris/javac.html
And here is a link to a relevant thread in the forum discussing the classpath on OS X.
http://forum.java.sun.com/thread.jspa?forumID=54&threadID=570279
I had some questions about where i should be putting class files. I generally let the compiler take care of where to put the class files. Of course if no direction is specified using the "-d" option with the javac command, the files will be placed inside the current working directory. That's true on all OS systems including Mac OS X.
i've got files in a package com.one2one3.io, so i've got a directory structure
/com/one2one3/io/ and i'm wondering where is the generally accepted place to
put that, its working for now in ~/Code/com/one2one3/..., but then i have to
specify the classpath everytime i compile...Note that when the source file that is being compiled includes a package declaration at the top like
package com.one2one3.io;
public class HelloWorld {
}if the directive option "-d" is specified with the javac command and it points to another directory location other than the current working directory, the compiler will create a directory structure for you (assuming one does not already exist) in that destination directory which mirrors the package structure of the java file. That's quite convenient. So for example you can have all your source files in a top level working directory, like "code", and then direct the output of the javac command, the class files themselves, into another directory, maybe one called "classes".
javac -d classes HelloWorld.javaAfter executing this you will see that the first directory inside the classes directory will be another directory called "com", consistent with your package com.
that brings me to the question of setting the classpath in my .tcshrc file so
i don't need to do that everytime... google says this should work:
setenv CLASSPATH .:/Users/stephen/Code/:${CLASSPATH}
but it isn't i still get class not found errors when compiling. Once you can isolate all of your class files into the classes directory, then you define the classpath as the classpath using the "-classpath" option of the javac command. You may want to include the current working directory too by including "./" with the option.
javac -d classes -classpath ./:classes HelloWorld.javaNote though that when you use the -classpath option with the javac or java command, it does override whatever you currently have set in your system settings. Nonetheless, I understand that this is the standard method of defining the classpath.
Suppose though that you have a number of different external library jar files that you need to include on the class path. Or suppose you have large numbers of source files you wish to compile all at once. In this case you may want to consider creating command line argument files, as explained here...
http://java.sun.com/j2se/1.5.0/docs/tooldocs/solaris/javac.html#commandlineargfile
Then your javac command would look something like this.
javac @options HelloWorld.javawhere your classpath and directive would be defined in a file named "options" in the current working directory.
can i put the packages i write in /Library/Java/Extensions/com/one2one3/...?
from what i understand the jvm on mac searches there by default so i wouldn't
need to specify the classpath everytime i compile. thats working for me, but i
want to check and make sure it won't cause me any problems down the line...i
guess mainly i'm wondering what you all do to deal with these things.
and is it usual to include the source .java files in the /com/one2one3/...
directory? Actually I understand that it is considered poor practice to place files directly into the extensions directory of the Java directory, though you are correct about it's location on Mac OS X. It does work, but it's a bit of a cheat.

Integration events - Retrieve (get) and delete

Hello forum. First, I would like to send my new years best wishes to everybody, hope it will be good. Second, I am relatively n00b to siebel in general, however I got some EXP integrating applications, so even the technology is new, the concepts (overall) are familiar. Anyways, I am working as you may imagine by now, with the OnDemand CRM, which exposes some web services to deal with the integration. My first attempt was to create a consumer (this part is working fine) and retrieve N amount of integration events. Now I am on the part where I got to "delete" those events from the message queue and I am facing the problem that it will only accept two parameters to proceed with the deletion, which are based on an arbitraty datetime or based on a given last event.
On the documentation, it reads that you might delete the entire set of events or any subset, but with those only parameters expected (in the wsdl), how could someone achieve it ? I read and research before posting, of course, I found a post by some other guy that was doing the same and at least in the "getting" int events part, he was stuck until customer care said this kind some sort of typo and this cannot be done, I think this may apply here too, but I got to be sure before I proceed.
So, bottom line, it is possible ? What is the best way to do it ? One by one, taking a bunch of them ? Any recommendation is appreciated.
Thanks in advance.

I'm not really sure why your example is not working since we
are missing a good amount of the code that goes along with it
however whenever I need to check for either of those keys I
normally do it the following way as attached.

Char to bit string???

Hi everyone,
I have a small problem and I was wondering if anyone can help. I want to convert a char value to a string of bits. I have had no problem doing it with an integer but I "keep reaching for the Aspirin" when it comes to the char. Any ideas? I am sure I am doing something stupid - it should be relatively simple.

Here's the general idea:private static String toBitString(char c) {
return Integer.toBinaryString(c);
private static String toFixedLengthBitString(char c) {
return Integer.toBinaryString(c | 0x10000).substring(1);
}

Ichat not connecting to jabber account

Hi - I'm relatively new to Macs in general and especially new to ichat. I'm using my gmail account to connect to ichat, but it won't connect. I've checked the user name and password, I've checked the server and the port. I also have connectivity to the internet and I'm the admin on the computer so I know it's not a parental controls issue. I'm also not logged into gmail and ichat at the same time
Why is ichat telling me "iChat can't communicate with the Jabber account [email protected] Could not connect to host talk.google.com."
This is so frustrating! Any help is appreciated.
Thanks!
Rayna

Hi,
||iChat Ports||TCP||UDP||
|AIM|5190|5190|
|A/V|none|5678, 16393-16402|
|Bonjour|5298|5297, 5298, 5353|
|Jabber|5220,5222, 5223|none|
Note this is the iChat 4 and 5 Range
Add port 5060 UDP for iChat 2 and 3 and extend the group of 10 to 16384-16403 (20 ports)
iChat 4 and 5 also uses an undeclared Random port to do the Screen Sharing part and works if you can and do use UPnP (Or Port Mapping Protocol) to open ports in your routing device.
This Info is from http://support.apple.com/kb/HT1507?viewlocale=en_US for iChat 3 ports which lists the Bonjour and Jabber ports as well.
This one http://support.apple.com/kb/HT2282?viewlocale=en_US lists the changes (reduction in the A/V Group mainly) but does not refer to the first for Bonjour and Jabber.
Then there is this
http://docs.info.apple.com/article.html?path=ServerAdmin/10.6/en/im5945b424.html
This does also list port 7777 for Federation (things to do with linking with other Jabber servers)
It lists 1080 for File Transfers using a Proxy (Not sure if this is Jabber or not)
It also lists "5269 TCP This port is used for encrypted TLS server-to-server connections, as well as nonencrypted connections". I have not come across this on the Client side (but it does say Server to Sever)
As the Clients may still be using iChat 3 (or in fact earlier) port 5060 is still listed and the A/V group is wider as a consequence as they will need them.
The biggest issue will be the Random Port for Screen Sharing.
As far as I can tell this an Audio Chat alongside a VNC part for the Screen Pics.
Both need to work for iChat Screen Sharing to work.
I hope this helps.
10:09 PM Monday; April 11, 2011
Please, if posting Logs, do not post any Log info after the line "Binary Images for iChat"

Some questions before I install Awesome WM

Hi,
I'm new to Arch and relatively new to Linux in general but I do enjoy tinkering with it.
Recently I have tried many different window managers, Gnome, KDE, Openbox and Fluxbox, each of which have their own little perks. However whilst browsing through the monthly screen shot threads I stumbled upon something new, something I haven't heard of; tiling window managers. I find the idea behind them to be innovative and would like to give them a try, particularly Awesome WM, before I do so I have a few questions outlined below.
1). I understand that a mouse is not needed, which can be a good thing but an obstacle as well depending on what I want to do. What I would like to ask is that can you keep the mouse functionality as well such as the cursor, left click, right click and possibly the scroll wheel? (Is this possible in both Xmonad and Awesome WM)
2). Can you have a panel so that you can either easily launch your applications or when you minimize say pidgin you can see its status icon?
3). Most of the screen shots I have seen show all these terminal windows running, I assume thats just for show. You can run gui based aps too such as something for burning cds, openoffice, firefox and other gtk aps?
4). If neither apply to Awesome WM, could you point me in the direction of a tiling window manager more tailored for my needs.
Please only leave helpful comments, I don't want to hear comments such as why do you want to use Awesome WM if your planning on running gui aps or don't want to run their terminal alternatives....Linux is about choice and I am entitled to my own as are you.
If you can help that would be "awesome".
PS: Please excuse the rather verbose post; its a problem of mine, hopefully one you come to find endearing.
Last edited by blackbird (2008-03-11 11:50:19)

blackbird wrote:
chimeric wrote:
blackbird wrote:Your response leads me to believe that I was asking if Awesome WM lets you assign shortcuts to the mouse, such as your very own example where a combination of mod4 and middle click would launch the terminal. What I intended to ask was whether you could use the mouse like you use it in other environments such as Gnome...where moving the mouse would move a cursor on your screen, left click would be used to select a window, move it around...etc.
Well I said "Yes." . The mouse behaves exactly like in other WMs and additionally you can add other functionality if you like. "Moving windows" is a little different in tiling window managers because windows aren't supposed to be moved around with the mouse, because the WM arranges them for you. However, in most tiling WMs, windows can have a special state named "floating", which means the WM isn't responsible how the window is placed on the screen and you can move it around freely (with the mouse). This is for example useful for applications like mplayer, vlc, gimp or gajim.
Brilliant; thanks for the clarification, I just hope nothing extra is required to obtain this functionality. Well if anything else is required that wont be a problem considering how good the Linux community is.
Thanks again
PS: There is a slight drawback to using Awesome WM though; whenever you google something about awesome wm you get results about people stating how awesome their window managers are, not specifically awesome wm lol. Everything comes at a price.
Nothing extra is required....just already preconfigured keystrokes do the job.
Last edited by ph0tios (2008-03-11 16:20:22)

Migration vs re-coding

I have a ND3 application of about 40 screens. No experience of ND3.
Experience of J2EE.
By your experience, given 100 to rebuilding the site from scratch,
how much you reckon would take me if I use the Migration Toolkit.
Just a rough guide, of course.
Thanks
Giulio

Migration using the iMT is partly automated and partly manual. The actual
amount of automated conversion varies from project to project, but is
usually higher than 50%, sometimes much higher. We generally say on average
that the iMT migrates 75% of your application for you. We base this
percentage on a number of factors.
First, moving the structure and declarative data of the original application
to J2EE/JATO alone is about 50% of the work you'd otherwise do manually.
Most people don't understand the HUGE amount of information that is present
in the original app that must be carried over. Just look at the application
description file after translation to see what I mean. Doing this work
manually would be extremely time consuming.
Ultimately, the base conversion ratio is the amount of declarative vs.
programmatic features you used in your project. If you used 100%
declarative features (ND wizards, field binding, etc.), the iMT can migrate
your project 100%, meaning it will be runnable right after conversion. The
percentage drops from there based on what your project code looks like.
The iMT comes with a Regular Expression (RE) Tool that can do code
conversion. We supply a base set of rules with the iMT to convert
Spider-API-based code to J2EE/JATO-based code. For example, things like
usage of the ND session or CSpValues is automatically converted for you.
You can also write your own rules for use with the RE Tool. Since a given
project usually uses common application-specific idioms, your own RE rules
can migrate these idioms very easily.
Usually, the largest single cost in migration is porting of
application-specific frameworks away from Spider to J2EE/JATO. The
disadvantage is that this is an up-front cost; the advantage is that
migration of code that relies on that framework is generally quick and easy.
The nice thing is that when the application involves a framework, the RE
Tool is great at porting code that uses the framework. A typical ND app
(declarative + code) that had no significant framework is easily migrated
something like 80-90%.
The most important thing to understand about migration using the iMT is that
it generally doesn't require significant redesign of the application.
Unlike moving to J2EE on your own, the iMT is able to preserve the structure
of the application, and it places application code in analogous locations in
the migrated app. The migration is largely a matter of API replacement
rather than a rethinking of the application design.
For an ND3 application, I would generally expect the migration percentage to
be pretty high. This is because most ND3 apps rely heavily on declarative
features, and don't use the much larger API that was present in ND4-5.
Furthermore, ND3 apps are less likely to use low-level container features
that sometimes make porting of ND4 & 5 apps more difficult.
Todd
----- Original Message -----
From: "Jesudas Sundararajan" <das_mathy@y...>
Sent: Thursday, November 08, 2001 11:38 AM
Subject: Re: [iPlanet-JATO] Migration vs re-coding
I heard the iMT tool would migrate 40 to 50% of the
code. Rest has to be migrated manually.
--das
--- giulio.vuolo@c... wrote:
I have a ND3 application of about 40 screens. No
experience of ND3.
Experience of J2EE.
By your experience, given 100 to rebuilding the site
from scratch,
how much you reckon would take me if I use the
Migration Toolkit.
Just a rough guide, of course.
Thanks
Giulio
For more information about JATO, please visit:
http://developer.iplanet.com/tech/appserver/framework/index.jsp

Impact of network latency

We have developed an application based on j2EE architecture and are running it on iWS and iAS. I would like to know the impact network latency would have on the application.
What should be the threshold of the network latency for our application's good response ?
Regards,
Vibha

Network latency between where and where?
Between the user and iWS? Between the iWS and the iAS? Between the various iAS servers?
Latency between the user and iWS is pretty unavoidable. Except for technologies like Akamai, there isn't much you can do about it.
Latency between the iWS and the iAS is somewhat negative. The impact on the user is pretty much linear. i.e. an increase of 100ms latency between iWS and iAS is likely to cause an increase of 200ms in the perceived response time. (100 ms inbound and 100 ms outbound) The problem is that the application also has to keep the processing thread around another 200 ms because of the network latency. This means that are going to see some degradation in the performance of the server.
Large latencies between the iWS and iAS are relatively easy to resolve, however. Just increase the size of the network pipe and/or increase the capacity of the firewall between iWS and iAS.
Latency between various iAS instances is VERY bad. Large latencies between iAS server can interfere with clustering both degrading performance and data consistency. You should definately resolve any network latency between iAS servers.
David
http://www.amazon.com/exec/obidos/ASIN/076454909X/

IAS 9i Rel. 2 9.0.2 General

Similar Messages

Maybe you are looking for