OO Design Point

I'm simulating a "University Registration System" webapp based on the Java Servlet Technology. There're main entities like Course, Student, CourseSection, Registration.
CourseSection has a ( it's teacher ) Professor and a ( it's content ) Course reference.
Registration has a List of CourseSections available for current term, ...
My problem is here:
Student has a List of CourseSections that has registered in. that' ok! but somewhere, i need to have a List of course objects that any student has registered in.
One way is when i want this information, iterate through registered CourseSections List and collect a List of Courses. ( "anySectionObj.getCourse()" ) and in this case bad point is a "For Loop" ( lookup ) everytime you want this information. ( performance )
Another is to also have a List of registered Courses in Student Class, but it means that any update ( Add, Remove, ... ) to CourseSections List needs a update to Courses List, but i think there'd be somehow a "Coupling" between these two, however they are naturaly "Coupled." { registered Courses depend on registered CourseSections. } ( OO best practices )
as now i've choosen the second. and in my "addCourseSection" method i add the courseSection to CourseSections List then it's Course to CoursesList ( cuncurrent update ). but it dosen't sound me a good approach, also having two list's that totally mean the same.
How can i OO-ly ;) solve this kind of problems ? ( any patterns? )
I hope, i've clarified my point. any help of OO issues would be appriciated.

One word ... Managers
Ok to explain
Entities
Courses
Sections
Students
Schedules
Relationships
A Course knows what it is, but may or may not know what sections of it exist.
A Section knows what course it belongs to but not what other sections may exist.
Students know who they are, and know what their schedule is but they have to look at the schedule to get any specific information
A schedule knows what student they belong to and what sections of what courses they "own".
Problem, if you want to know (a) what students are in a section, (b) what courses does a student have ... there are others but you get the idea.
By adding managers the core system, or an external system doesn't need to know where the references to the students are, it asks the manager. Basically a call that semantically says, "Student Manager could I have a reference to the student with the id=123456?"
The student might or might not hold a reference to their schedule. For our purposes we'll say that they don't, so you might want to have a schedule manager sitting in the student center to direct the hapless freshmen who inevitably get lost. Student says essentially, "Schedule manager, I'm student (id=123456) and could I have a reference to my schedule please.
Do you see where I'm going with this ???

Similar Messages

Interaction design point of view

I took a look to almost all video and explanation i found on adobe.com about catalyst, and it seems adobe care about what designer will be able to do with, and how they'll be sharing their work.
My point is that i found in Catalyst a huge tool for interaction design (as ergonomist). I think adobe could push thoses customers to buy it (and i barely can't it gets out !!!). We used to develop our model on illustrator, then make a report inside indesign and then, show and talk about it with our customers.
Here's the first thing : instead of showing statics capture screen, and take a lot of time to explain how this component or this bouton will interract, we'll be able with FC to simply execute the swf file, and say : this is how it will "interract".
2nd point is : as ergonomist, we never can't be sure that our choise on the interface are fine or best. We can just say that's is not bad at all, so let's try it ! So we can spend a lot of time to test some elements of the interface with real user, and understand what's good and what's wrong. With FC, when we (our team of ergonmist) discuss about a behavior of a component, we can say : here's what i think about it, this can do that, and that will come here with a fade in. We'll be able to discuss directly on our component, visually, improving our efficiency and take better choices.
The last point is how we work with the designers and developpers. Usually, designers takes our document, and try to respect the most of it. Sometimes it's easy (when they're no ambiguity about the behavior of the component), and sometimes it's hard (cause the designer understand his own way the behavior, while we were thinking another. Or just because designer produce a lot of terrific design and behavior we didn't specify, and at the end, a lot of thing has not been respect even if it's beautiful or fun). FC allows us to show the designer what behavior will that bouton do, or how this menu can shows up or how he can't shows up.
And all of this changes a lot about our productivity, efficiency, and how we works with our workmate.
So maybe adobe can just speak to this people (the ergonomists) and even improve FC in that way. (we use axure to make some prototype for the users test, and we work on a layer on a jpeg, and spend a lot of time inserting component (often invisible) or hide texte to put just a hyperlink, etc...).
FC is really awesome, and i hope everything i have in my mind about this soft will become true
I just have to thank all the team for the developpement, they're making the software we all missed !

It's vey exciting to hear that Catalyst is so useful in your work as an ergonomist, and thank you for sharing the ways that you use Catalyst in your workflows.
Ty

OBIEE11g-Mobile App Designer Point Of View

Dear All,
Is it possible to keep dynamic Column selector in OBIEE Mobile App Designer as like Analytics ?
Thanks in advance.

Hi
I was facing the same error after installing the Patches for (Mobile app Designer) and so many restarts of all OBIEE components as well as the server,
nothing worked except the same Error 404 for the last couple of days.
then here goes the path for getting the issue fixed juts now.
1: Verify the Patches applied for the Mobile app Designer
2: followed the blog step by step for the process to get completed.
OBI Mobile App Designer Installation | OBIEE by Shiva Molabanti
3:after applying the patches and deploying theOBI Mobile App Designer (MAD) manually using WebLogic configuration the bounce all the services on the OBI (all components)
4:the Run the Security Configuration as described in the blog . it wil give error if the services of console is not started as it awaits for the port 7001 of weblogic
5: after the successful completion of the security configuration restart all bi components on the OBI side and wiat for all the services to come up
now i logged in and click on New Mobile APP it worked as expected and it resolved the issue.
Thanks
BSK

Trying to design Point of Sale page

hi gurus,
I am trying to design an EPOS page. I have a field which will hold barcode value.
Using this value, I will retrieve item name, selling price and other necessary info from table and will display these value somewhere on page. I want to have item name, quantity and amount fields to be inserted in some kind of editable multi row container but i am not sure which container will serve the purpose like in Forms Developer, it could be handled in tabular form but i do not know what is the right choice here and how can i add data to that container?
advice required
nadeem

Nadeem,
You don't really need a container, you are just displaying a report. I've written a couple of EPOS systems (PowerBuilder and C++) and you're basically going to do a submit when the UPC field is changed. That is the simple part, things like inventory control, returns, breakage, security, etc are the hard parts. You are also going to have to deal with UPC's that are not in the database and manual price entries. You might want to go out on www.sourceforge.net and search for EPOS there, it will allow you to see how others handled all these problems and save you from re-inventing the wheel.
Keep Smiling,
Bob R

Which design is best from a performance point of view?

Hello
I'm writing a small system that needs to track changes to certain columns on 4 different tables. I'm using triggers on those columns to write the changes to a single "change register" table, which has 12 columns. Beacuse the majority of tracked data is not shared between the tables, most of the columns will have null values. From a design point of view it is apparent that having 4 separate change register tables (one for each main table that is being tracked), would be better in terms of avoiding lots of null columns for each row, but I was trying to trade this off against having a single table to see all changes that have been made across the tracked tables.
From a performance point of view though, will there be any real difference whether there are 4 separate tables or 1 single register table? I'm only ever going to be inserting into the register table, and then reading back from it at a later date and there won't be any indexes on it. Someone I work with suggested that there would be more overhead on the redo logs if a single table was used rather than 4 separate tables.
Any help would be appreciated.
David

The volumes of data are going to be pretty small,
maybe a couple of thousand records each day, it's an
OLTP environment with 150 concurrent users max.Consider also the growing of data and if you'll put data constantly to an historical db or if the same tables will contain the increasing number of record.
The point that my colleague raised was that multiple
inserts into a single table across multiple
transactions could cause a lot of redo contention,
but I can't see how inserting into one table from
multiple triggers would result in more redo
contention that inserting into multiple tables. The
updates that will fire the triggers are only ever
going to be single row updates, and won't normally
cause more than one trigger to fire within a single
transaction. Is this a fair assumption to make?
David
I agree with you, the only thing I will consider, instead of redo problem, is the locking on the table that could be generated when logs of different tables will have to be put in a single table, i mean if after insert of a log record you could need to update it....
In this case if 2 or more users have to update the same log you could have problems.

What can be the Tech design for the following functional points

Hi All,
What can be the Technical Design preparation for the below Functional Design Points!
1. Search table Z100
Select all records where
Status (Z100-ZSTATUS) = G
Date (Z100-ZDATE ) = current month minus 1
Total price (Z100-ZTOTPR) => $5000.
Select Z100-MATNR
List Z100-MATNR and ZTOTPR
2. Then, for each MATNR, list all equipment records where
Equipment type (EQUI-EQTYP) = S
List Material (EQUI-MATNR)
WBS (EQBS-PS_PSP_PNR)
UID (EQUI-UIINR)
Serial Number (EQUI-SERNR)
3. Then compare the equipment record WBS (EQBS-PS_PSP_PNR) to the Contract Line Item WBS (VBAP-PS_PSP_PNR)
From VBAP, select the Sales Document (VBAP-VBELN) and Contract Line Item (VBAP-POSNR)
4. Then compare the Sales Document on VBAP (VBAP-VBELN) to the Sales Document on VBAK (VBAK-VBELN).
From VBAK, select the UID indicator (VBAK-UID) and Contract Number (VBAK-BSTNK).
5. If VBAK-UID is initial, go no further. No entry is required to the table.
If VBAK-UID is not initial;
Create 3 new rows on table Z200
6. Select the UID number EQUI-UIINR
If no record exists in table with UID number, create three new rows in table Z200 populate fields per the following table mapping. The table mapping shows default values, or sources in SAP, and parameters if needed. 
Can anybody tell me what is the technical design we can write for above finctional design.
Thanks in advance.
Thanks & Regards,
Prasad.

Hi,
The functional design is very detailed and has covered almost the technical part as well. I dont think you need to add anything more. If you need to add anything then it would probably be the Selects statements. But i dont think they are required.
But once you collect the data in internal table, what should be done. Should it be sent to smartform/script or a report output should be displayed. Or are any tables to be updated.If so, then specify the same.
Best regards,
Prashant

How to give more than 10 Reward Points

Hi Experts,
For your excellent performance for solving problems, if I would like to give more than 10 "Rewards Point", how can I proceed? Can any expert guide me how to give more than 10 Reward Points?
Thanks,
Hoque
Toronto, Canada

Hello
There is one more way to award more than 10 points
This is the table
Designation
Point Value
Awards of This Type Allowed
Solved My Problem
10
1 Time Only
Very Helpful
6
2 Times only
Helpful
2
Unlimited
So if you think you want to give more than 10 points, keep allocating 2 points as many times as you can and finally give the 10.
Reg
*assign points if useful

Query Designer - Variables - Replacement Path

I need a help.
In one query I need to select some data from the current Period until yesterday.
In another query I need to select the same data, but from the current Month until yesterday.
I created 2 restricted key figures:
SEL_PERIOD restricts 0FISCPER by the variable 0FPER, wich is an exit SAP for current Period.
SEL_MONTH restricts OCALMONTH by the variable 0CMONTH, wich is an exit SAP for current Month.
Now in both RKFs I want to restrict 0CALDAY from the beginning of the Period or Month until yesterday.
I found this variable 0DAT, wich is an exit SAP for current day.
I'm trying to create an interval variable, type replacement path, with From Value not filled and To Value filled with 0DAT - 1.
But it doesn't work, Query Designer points that 0DAT is not an allowed source for the target variable.
And the documents that I searched state that "the target variable must represent a single value" and "the source variable must be input-ready".
Does anyone have a suggestion on how I can restrict 0CALDAY, for this specification ???
Thanks in advance.
César Menezes
Edited by: César Menezes on Jun 25, 2011 11:51 PM
I forgot an important point:
Sometimes I need current Period or Month, in other cases I need the same Period or Month from last Fiscal Year or Calendar Year.
No problem with 0FISCPER and 0CALMONTH, I can off-set the variables by -12.
But I would like a help on how to do it with 0CALDAY.
Example: today is 25.06.2011. In some cases I need to restrict until 24.06.2011, in other cases I need to restrict until 24.06.2010.

Hi Ceaser,
This is a generic explanation to your scenario. Please make some changes as per your requirement. Just to give you some idea......................
We need to manipulate u201CKey Dateu201D values to achieve required analytics.
Step 1: As a first step, create a variable on u201CSales Dateu201D (or whichever date is used for Sales Date) field (ZPDATE) with user input. It was meant to capture u201CKey Dateu201D through user entry.
Step 2: Create a restricted key figure on Sales & restrict it on u201CKey Dateu201D value to get first analytics need i.e. u201CSales on Dayu201D.
Step 3:Next step was to manipulate the value of u201CKey Dateu201D on which I could restrict u201CSalesu201D (Key Figure) again and again to get other values.
Step 4: Create a new variable (e.g. ZPUTMNTH) and assigned it to be filled up by Customer Exit. (For simplicity purposes only u201CMonth to Dateu201D value is explained below. Same procedure can be repeated to fulfill other needs. Though in code section, ABAP code is available for almost all the variables)
Create Variable
Step 5: Create one variable (ZPUTMNTH) for u201CMonth to Dateu201D with processing by u201CCustomer Exitu201D. This variable was created on u201CDateu201D characteristics.
Sample ABAP Code
*& Include ZXRSRU01
DATA: L_S_RANGE TYPE RSR_S_RANGESID.
DATA: LOC_VAR_RANGE LIKE RRRANGEEXIT.
Data: v_loc_date type d.
Data: v_loc_date1 type d.
Data: v_loc(4) type c.
CASE I_VNAM.
to calculate "month to date" user input is "Calday" Key Date
WHEN 'ZPUTMNTH'.
IF I_STEP = 2. "after the popup
LOOP AT I_T_VAR_RANGE INTO LOC_VAR_RANGE
WHERE VNAM = 'ZPDATE'.
CLEAR L_S_RANGE.
L_S_RANGE-LOW = LOC_VAR_RANGE-LOW(6). "low value, e.g.YYYYMM (200606) part of key date (20060625)
L_S_RANGE-LOW+6(2) = '01'. u201C low value e..g. YYYYMM01 (20060601)
L_S_RANGE-HIGH = LOC_VAR_RANGE-LOW. "high value = input
L_S_RANGE-SIGN = 'I'.
L_S_RANGE-OPT = 'BT'.
APPEND L_S_RANGE TO E_T_RANGE.
EXIT.
ENDLOOP.
ENDIF.
to Calculate no of days up to "month to date" (user input Calday Key date)
WHEN 'ZDAYMTH'.
IF I_STEP = 2. "after the popup
LOOP AT I_T_VAR_RANGE INTO LOC_VAR_RANGE
WHERE VNAM = 'ZPDATE'.
CLEAR L_S_RANGE.
L_S_RANGE-LOW = LOC_VAR_RANGE-LOW+6(2). "YYYYMMDD - DD will indicate the no of days
L_S_RANGE-SIGN = 'I'.
L_S_RANGE-OPT = 'EQ'.
APPEND L_S_RANGE TO E_T_RANGE.
EXIT.
ENDLOOP.
ENDIF.
to calculate Previous year "month to date" (User input Calday u2013 Key date)
WHEN 'ZDAYPMTH'.
IF I_STEP = 2. "after the popup
LOOP AT I_T_VAR_RANGE INTO LOC_VAR_RANGE
WHERE VNAM = 'ZPDATE'.
CLEAR L_S_RANGE.
L_S_RANGE-LOW = LOC_VAR_RANGE-LOW.
L_S_RANGE-LOW(4) = L_S_RANGE-LOW(4) - 1. "low value previos year
v_loc_date = L_S_RANGE-LOW. "take input value in another variable
L_S_RANGE-LOW+6(2) = '01'.
L_S_RANGE-HIGH = v_loc_date. "high value = input, taken into another variable
L_S_RANGE-SIGN = 'I'.
L_S_RANGE-OPT = 'BT'.
APPEND L_S_RANGE TO E_T_RANGE.
EXIT.
ENDLOOP.
ENDIF.
to calculate "year to date" (User input Calday u2013 key date)
WHEN 'ZPUTYEAR'.
IF I_STEP = 2. "after the popup
LOOP AT I_T_VAR_RANGE INTO LOC_VAR_RANGE
WHERE VNAM = 'ZPDATE'.
CLEAR L_S_RANGE.
L_S_RANGE-LOW = LOC_VAR_RANGE-LOW(4)."low value, e.g.200001
L_S_RANGE-LOW+4(4) = '0101'.
L_S_RANGE-HIGH = LOC_VAR_RANGE-LOW. "high value = input
L_S_RANGE-SIGN = 'I'.
L_S_RANGE-OPT = 'BT'.
APPEND L_S_RANGE TO E_T_RANGE.
EXIT.
ENDLOOP.
ENDIF.
to calculate "Previous year to date" (User input Calday u2013 key date)
WHEN 'ZDAYPYR'.
IF I_STEP = 2. "after the popup
LOOP AT I_T_VAR_RANGE INTO LOC_VAR_RANGE
WHERE VNAM = 'ZPDATE'.
CLEAR L_S_RANGE.
L_S_RANGE-LOW = LOC_VAR_RANGE-LOW.
L_S_RANGE-LOW(4) = L_S_RANGE-LOW(4) - 1. "low value previos year
v_loc_date = L_S_RANGE-LOW. "take input value in another variable
L_S_RANGE-LOW+4(4) = '0101'.
L_S_RANGE-HIGH = v_loc_date. "high value = input, taken into another variable
L_S_RANGE-SIGN = 'I'.
L_S_RANGE-OPT = 'BT'.
APPEND L_S_RANGE TO E_T_RANGE.
EXIT.
ENDLOOP.
ENDIF.
to calculate "Previous year" User input Calday u2013 key date
WHEN 'ZPYEAR'.
IF I_STEP = 2. "after the popup
LOOP AT I_T_VAR_RANGE INTO LOC_VAR_RANGE
WHERE VNAM = 'ZPDATE'.
CLEAR L_S_RANGE.
v_loc = LOC_VAR_RANGE-LOW(4) - 1.
L_S_RANGE-LOW = v_loc. "low value previos year
L_S_RANGE-SIGN = 'I'.
L_S_RANGE-OPT = 'EQ'.
APPEND L_S_RANGE TO E_T_RANGE.
EXIT.
ENDLOOP.
ENDIF.
to calculate No of Days - "year to date"
WHEN 'ZYRDAY'.
IF I_STEP = 2. "after the popup
LOOP AT I_T_VAR_RANGE INTO LOC_VAR_RANGE
WHERE VNAM = 'ZPDATE'.
CLEAR L_S_RANGE.
v_loc_date1 = LOC_VAR_RANGE-LOW.
v_loc_date = LOC_VAR_RANGE-LOW(4). "low value, e.g.200001
v_loc_date+4(4) = '0101'.
L_S_RANGE-LOW = v_loc_date1 - v_loc_date. "get the difference between current date & first date of year
L_S_RANGE-SIGN = 'I'.
L_S_RANGE-OPT = 'EQ'.
APPEND L_S_RANGE TO E_T_RANGE.
EXIT.
ENDLOOP.
ENDIF.
Save and Activate the Include.
Go back to Exit and activate again.
Go back to Project and activate again.
Variable (ZPUTMNTH) is ready (its values filled up with the manipulation of u201CKey Dateu201D) to be used in query to restrict Key Figure u201CSalesu201D on it and get u201CMonth to Dateu201D (MTD) Sales Value.
Regards,
Suman

Javascript interrupting responsive design

I added javascript to my website which is great and the site is still response but I have to reload the page to see the website change to the proportions. Obviously this is no good from a design point of view as the website will look terrible if you were to change your iphone from portrait to landscape and the website doesn't react until it's reloaded. I think it might be a loading issue as I know that javascript increases the loading time for websites. Have a look at btccs.co.uk and change the screen size and you will see what I mean. Thanks

Massive distortion occurs when images are upscaled beyond their native size. This is even uglier on ultra wide TV or multi-panel displays. So it's not a good practice.
Let's say you have a banner that is 2000px wide. The following CSS rule lets images downscale in proportion to viewport width but never upscale beyond that native size of 2000px.
img {max-width: 100%}
The above rule is the default in Fluid Grid Layouts + other RWD frameworks. It works very well providing you don't override it in your HTML code with explicit img height and width attributes.
Nancy O.

Design view not displaying but code view is ok

Hi Everyone,
my design view in dreamweaver cs5.5 is not displaying anything but all the code is there and my page online is working fine http://www.exclusivewindowtint.com.au/car-window-tinting.html
How can i fix this as easy as possible? this is only happening to this one page on my website.
Thank You
George

I have to second Murray on this.
I will go a step further and say ignoring errors on your site because someone else's site has some is extremely poor advice, especially considering the volume of people visiting this forum alone with a random display problems that are solved by simply correcting the errors found in the validators.
Restoring from a backup may work this time, but why did it work? That's where validation, and understanding what the validator is telling you, pays off. You somehow inadvertantly mangled your table code, will that ever happen again? What happens if your backup also has the same mangled code next time?
Some validation errors are never going to affect display and can be completely ignored from a "how does this affect my design" point of view, like missing an alt attribute on an image for instance. Others, like missing or improperly nested html elements, can kill your design in browsers or rendering engines that can't handle major errors (like most versions of IE or DW's Design View) while everything looks fine in a browser better programmed to handle problems (like Chrome or Firefox).
Bottom line is, the validators are extremely useful tools, do not ignore the error report unless you understand which errors can be ignored.

Bad link "Using Oracle9i SCM with Designer"

The link on the page http://otn.oracle.com/products/repository/demos.html under the title "Using Oracle9i SCM with Oracle 9i Designer"
pointing to http://otn.oracle.com/products/repository/demos/designer/new_nov02/html/des_scm_demo.htm
does not work.
thanx for the correction,
trevi

Hello,
Thanks for pointing that out Trevi - I'll get onto it
and fix it as soon as possible.
Regards,
Dominic
Designer Product Management
Oracle Corp

Landscape design using J2SE

Hi,
I am hoping that you could assist in the following interface design matter.
Engen have Oil deports right over the country. These are connected via the WAN.
These depots runs a Business Application that needs integration with SAP. There are about +/-12 that operate 24/7.
Currently they are transferring files between the Deport and SAP via MQ Series.
These transactions files are time critical and are generated adhoc. They are shipment instructions to automated
truck loading equipment at the plant and the response back to SAP on the actual load taken.
The plants are sitting in places quite remote which geographically could be hundreds of kilometers away.
From a design point of view would it be best to just use a File Adapter on the Central Adapter Engine, via a share or would the use of
a J2SE Adapter Engine (Hosts does not support J2EE due to RAM requirements) with a local File Adapter be best. Keeping in mind that the support and maintenance will have to be done from a Central location (HQ).
My challenge so far has been that I have not been able to find SAP documentation to guide me.
I would greatly appreciate if you could refer me to some document that could assist or with any advice you might have.
Regards
Willie Hugo

Hi Willem!
To keep landscape and maintenance as simple as possible I would start to use the Central Adapter Engine and FTP on the local sites. Are all local sites connected via secure network connections like intranet or VPN? If not security aspects also have to be taken into account (FTPS and SFTP are not secure).
Please keep in mind just the simple and day-to-day situation that you have to apply patches to adapter engine software. In the worst case all remotely installed adapter engines must be patched by someone. Addintionally you have to ensure at least a 2-system landscape. That means centeral PI x 2 (or 3) plus decentral Adapter Engines x 2 (or 3).
But you can also have a look at service marketplace under Alias "instguides". There you can find e.g. a Technical Infrastructure Guide for PI 7.1 (and other XI/PI releases). There you will also find security guides if security will become a cirtical factor in your environment.
By the way: Is it planned to use the PI/XI instead of MQ Series or can it also be a possible solution to leave everything as it is in the locations and just connect MQ series with XI/PI?
Regards,
Volker

WiSM2/WLC design

Hi experts,
We are in middle of deploying WiSM2s on our network, from a design point of view i am confused on where to position the WiSM2s. We have 2 DCs and from best practice architecture view WLCs should not be placed on the Data Centre segment, but it seems that is the only option i have.
DCs host 6509s, i am planning to host 1 WiSM in each DC, all clients will be on seperate subnet. Do you foresee any issues with this deployment or any security issues?
authentications are followed as per cisco recommendation, clients authenticated against AD through ACS so it is fairly secure. Will appreciate any advice.
Cheers
AP

Amar,
Since you are going to install the WISM2 you will be limited to where your chassis is physically installed. If we are operating under the assumption that all, or at least most, of the traffic sent to the WISM2 will eventually hit the data center I do not se a reason to move them. On the other hand if you have a lot of peer to peer wireless traffic such as voice calls and pages you may want to consider some HREAP configuration. Placement of the WISM2 will not have any affect on the authentication/encryption that you will use. Ideally when planning a new WLC/WISM(2) deployment I like to use all local mode APs with a WLC/WISM(2) at each site as well as a WLC in the DMZ to use as a mobility anchor. If you provide more specific details such as client type, traffic path, and number of clients per location I can help you with something more specific to your design.
Thanks,

Simple game design,

so I have a basic understanding of java, I took a class on it this spring and did well. I am wanting to make a simple game, a single picture, a map, that is clicked.
based on the location clicked, the player gets an item. this is an online game.
I am going to use SQL for this game on the back end. it will have a db with two tables, users, and map.
users will have user name, user ID, items owned, tools, and searched today.
map Im not quite sure about yet.
so basically, Im trying to figure out how I should go about this. Im not quite ready to hack code yet, I need to get a good plan together and make sure I go about this correctly.
can I open this for discussion and get some help on designing this. Im trying to stay really very simple, so as not to undertake too much.
this is my outline right now:
login: a screen with name and pass fields, an ok button. when the button is clicked, the program checks the db for the name, and verifies the password, if everything is in order, goto the account screen, possibly a cookie or session of some sort is set.
account screen: show the username, items owned and number of each, any tools or special items owned. button to logout, button to goto map. radio buttons on the tools to determine which one, if any, is equipped. there should also be a special area near the top that shows what item was found on the previous day.
map screen: this is a small image of a map. the user clicks and shows a pin placed, as well as any other pins currently on the map for the day. a search button that when clicked, locks in the location untill the end of the day.
the user can search one location per 24 hour period, server time.once a location is picked, its locked in till server is rolled to new day. based on the location, and the tool selected, the player gains an item, these will be used for creating items and such in a later game that is tied to the IP.
pretty simple, login, pick a tool, pick a location, complain in the forum that you can't find good items, and that the main game isnt done, move on till tomorrow :)
I need to go over each part in detail and do this right. I should mostly be SQL with a graphic frontend, which as of now Im choosing Java for, so this can run in a web browser. Im also considering PHP, but I fear it wont be dynamic enough. Flash is my other choice. I would like to get a good idea of what I need to do for each aspect before I make it. The screens I will make in netBeans, probably using Swing, and all of my actual logic will be in click events.
Help me out guys, I need to talk about this step by step with someone.

lord_midnight wrote:
]morgair: I have done some small projects here n there, mostly just demos, I am almost finished with a retro styled platformer, just placing tiles and entities, finishing up the art, code is done except for a few custom routines for a boss. this project is actually pretty small, other than being online, its a good choice for a first game type of project. Im confident it is within my skillset.Thank you for the reply, I was unsure what level of help you needed, many on here come in and have never done any programming before, let alone, an actual game on-line or otherwise. It's nice to hear where you're at.
]mrw: yeah, its not so much that I have a problem, I just want to make sure I think everything through before I start. Im trying to get a good grasp of what Im doing, so I don't have very many problems, Im sure I'll have a few though, somehow no matter how much I plan, something goes awry.
]soultech2012: Im actually in college, I took java this spring, and Im taking advanced java as soon as its offered online. still, its not so much instruction I want, as discussion.
maybe I was not being so clear, Im not looking for anyone to write code, or tell me what to do, I want to talk with people, Im sure I will have direct questions at some point, but right now, I want to think it through. If anything, another set of eyes looking at the design, pointing out possible snags or logic problems. its pretty simple as far as games go, I did some prototyping in flash last night, I think this will be pretty easy. does anyone see any serious problems in the design ?
I think it will make a nice little game, its not meant to take much time out of the players day, just a 2-5 minute jaunt into my website. login, pick a spot on the map, select a tool to use to help gather items, and slowly over several weeks, have accumulated a collection of goods for a game Im doing at a later date.
getting the x,y will be easy, I have done a couple of very simple apps that interact with SQL,
the only real logic parts of this are some ifs or a switch, and adding to a field in SQL.
getting pixel color might be tricky, it was easy in flash, and Im pretty sure I saw a similar function last time I browsed the javadocs, so Im not too concerned.I like to use getRGB and setRGB from a BufferedImage.
honestly, I looked around and I can all but make this in javascript and PHP/SQL, java might be overkill.
anyone see any obvious ways to cheat the game ?If you store any logic on the client side they can easliy find and modify it, that way they can have a super character or all of the items. Runescape is perfect example of this, they run client server and there are groups that have made their own client to cheat with.
am I underestimating the complexity of the task ? I have plenty of time to work on it, no hurry, but Im hoping to have something on the site within the next month or so.That is doable, If you are careful in your implementation, you can build a basic game engine and then work on map and support making for facilitate different levels.
I need to make a list of terrains, the colors Im using for them, and items found, and I need to draw out a table that Im using to plot out the sql.
Im taking PHP/SQL this fall, Im starting to get the hang of it, and I have a pretty good grasp of basic programming concepts. this should be well within reach, honestly, its glorified tic-tac-toe, or match-findingThe big things that seem to pop up for people are:
Picking up items
putting down an item
weilding an item
using an item
combat
breaking off combat
automatic path plotting (least cost algo from map)
frequency of random events
regeneration of items
regeneration of health
real language interaction with characters (pattern matching for content)
Just remember: choose what you want and don't let the scope of your game creap or you'll have a bunch of code that has no implementation end in sight. Keep your compiles short and changes departmentalize so you can have running code that works from one feature to another. I see a lot of people--experirenced too--that literally throw hundreds of lines of code out and then try to debug it after they make changes thorughout the stable code base to allow integration of the new feature; this can be very daunting, keep it as simple as possible to integrate new features--small steps are good.
thanks for the input, and comments are welcome, best I can offer is thanks and bonus tools in the game once its done, so thank you.

Design Pattern: Is the Universe Object a Singleton or Static or either way.

Hi All,
1. I've read this thread: static versus singleton
http://forum.java.sun.com/thread.jsp?forum=425&thread=401035&tstart=105&trange=15
2. Now, specifically if you were to model the Universe Object, what would you choose? a Singleton
or a Static Class or either way depending on your design point of view?
(either way depending on your design point of view imply there is more than one solution to a problem.)
Basically, I'm looking for is the justification of one (singleton) or the other (static) or doesn't matter
in addition to the pure technical pros and cons (or avantage/disavantage) of singleton versus Static (see pt. 1)

<dubwai>
What's 'the Universe Object'?
</dubwai>
Sorry, for not being clear. My assumption is that every body would undertand the word 'Universe' immediately. So with this clarification. I hope you will have more input. Thanks.
public class Universe //Singleton
 private static Universe instance = new Universe();
 private Universe()
 public static Universe getInstance()
 return instance;
 public void do()
public class Universe //Static Class like Math class for example
 public static void do()
 ...all other methods are static
<os>
Personally, I'd make the universe a singleton.
The universe is an 'object', not a class, and if alternate universes are proved to exist you can create new instances,
and not treat it as a singleton any more, without much rewriting (a static implementation would need a total rewrite).
</0s>
1. Keywords: Personally and alternate universes are proved to exist.
Yes, this is the kind of reasoning I'm looking for. By that I mean when we design a class, our reasonning should not depends
on the 'pure' technical concepts of what Singleton class or Static class can do but rather depending on the reality
of the world. And then from that understanding we would choose a Singleton or Static class. This is what I meant by 'Either way, it doesn't matter' which depend on one's view about of existence of the universe whether it's unique or not. In your case (Os), you prefer Singleton because of the possibility of alternate universes.
2. Now, let's admit, there is only one Universe, would you still prefer Singleton class over Static class? for all the techincal reasons that you said
"As a singleton,..."
"It would also probably be useful to treat the universe as a generic Object..."
OR just because a Singleton would be 'safer' to cover the possibility of design extension in that can cover all cases (alternate as well)

OO Design Point

Similar Messages

Maybe you are looking for