nanodesigner-developers Mailing List for Nanodesigner (Page 2)
Status: Alpha
Brought to you by:
swinnen
Menu
▾
▴
nanodesigner-developers — Discussions by and for developers
You can subscribe to this list here.
| 2004 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
(7) |
Sep
(7) |
Oct
(9) |
Nov
(30) |
Dec
(3) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2005 |
Jan
(1) |
Feb
|
Mar
|
Apr
|
May
(2) |
Jun
|
Jul
(1) |
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
| 2014 |
Jan
|
Feb
|
Mar
|
Apr
(1) |
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
|
From: Elie De B. <el...@de...> - 2004-11-05 05:50:48
|
On Thu, 04 Nov 2004 23:32:40 +0100 Val=E8re Swinnen <sw...@us...> wrote: > Hi Javier, >=20 > > I am not quite sure why you want to keep to yourself > > > >the privilege of writing to the CVS. I would suggest > >you to let everyone to commit files freely to the CVS. > >Indeed, I believe it will speed up thing and encourage > >people to contribute anytime and anyway they want to > >the project. > > =20 > > > Because I want to keep things like directory structure, release data,=20 > labels and such well-organized. > Because of the size of this project and the number of participants I can= =20 > not allow 'everybody' to write to cvs. > It would undoubtedly tangle up the entire cvs tree within days. > No, sorry, I *can not* oblige you on this one. I hope this can change=20 > in the future so that I can give access to some selected people but=20 > certainly not to 'everyone'. > You can check other sourceforge projects but I doubt you will find many=20 > (if any) where *every* particpant has write access to the cvs tree (and=20 > I am talking about projects with more than 2 or 3 developers). >=20 I also think there should be a common place which version control where=20 we can put some code, call it a playground. For example, Fujy had to upload his sample code to some ISP, we had to download it, extract it, load it=20 in our ide to view it. While every other IDE has cvs access support. In the next fase I wrote a small testapp to illustrate the splitpanes, to do this I had two choices: a) Modifiy Fuji's code, package it, upload it to another piece of website so Fuji could (if he recieved the mail before that and didn't do the change=20 or another change before that) download it, replace his tree by my modified version, hope I didn't do anything wrong with it. Or open the archive look = for changes at first site and copy paste them. b) Just dump a little bit of proof of concept code in an e-mail hoping he w= ould be able to apply it to his code. Imo none of these solutions is a good one, there's still something missing.= =20 This doesn't necesserily mean we want access to _the_ entire cvs, isn't is possible to create a subset with different access rights ? Isn't there an alternative ? Somebody got a cvs server in their backyard ? greetings Elie |
|
From: <sw...@us...> - 2004-11-04 22:29:15
|
Hi Javier, > I am not quite sure why you want to keep to yourself > >the privilege of writing to the CVS. I would suggest >you to let everyone to commit files freely to the CVS. >Indeed, I believe it will speed up thing and encourage >people to contribute anytime and anyway they want to >the project. > =20 > Because I want to keep things like directory structure, release data,=20 labels and such well-organized. Because of the size of this project and the number of participants I can=20 not allow 'everybody' to write to cvs. It would undoubtedly tangle up the entire cvs tree within days. No, sorry, I *can not* oblige you on this one. I hope this can change=20 in the future so that I can give access to some selected people but=20 certainly not to 'everyone'. You can check other sourceforge projects but I doubt you will find many=20 (if any) where *every* particpant has write access to the cvs tree (and=20 I am talking about projects with more than 2 or 3 developers). Val=E8re |
|
From: Javier F. <jv...@ya...> - 2004-11-04 22:01:06
|
> >I think this is a great idea and it will speed up > >things a little bit. Well, it is up to everyone to > >decide. My vote is a "yes". What does everyone > think? > As I said before I am going to keep cvs write access > to myself for the > time being. So if you have some code you want to put > in cvs send it to > me please. I am not quite sure why you want to keep to yourself the privilege of writing to the CVS. I would suggest you to let everyone to commit files freely to the CVS. Indeed, I believe it will speed up thing and encourage people to contribute anytime and anyway they want to the project. Have a nice day. __________________________________ Do you Yahoo!? Check out the new Yahoo! Front Page. www.yahoo.com |
|
From: <sw...@us...> - 2004-11-04 21:39:12
|
Hi Javier, I am going to put some pages on basic chemistry in the Wiki as a=20 reference for people unfamiliar with the =91finer points=92 of basic chem= istry. Maybe you could give a hand with that? I think we should use existing web resources as much as possible for this. A site I found recently is=20 http://www.wwnorton.com/chemistry/index/site_map.htm Quite good actually... At the moment I think we need something about: 1. Bond types 2. Atom radii 3. Bond geometry 4. Proteins, Residues, amino acids =96 only very basic stuff here. >I think this is a great idea and it will speed up >things a little bit. Well, it is up to everyone to >decide. My vote is a "yes". What does everyone think? > > =20 > About creating a test cvs tree (or whatever). There is no need to do so=20 I think. We just put it in the appropriate place in cvs. We can always delete=20 files (i.e. move them to the attic directory) . Removing completely is=20 not possible on the sourceforge cvs. The directory structure will be something like this: org.nanodesigner.core /doc /nbproject /test /org/nanodesigner/gui /=85 /src /org/nanodesigner/gui /=85 I would also recommend using the JUnit test facilties for testing=20 critical classes (at least). As I said before I am going to keep cvs write access to myself for the=20 time being. So if you have some code you want to put in cvs send it to=20 me please. I hope to put the first classes I am writing at the moment in cvs one of=20 the next days. Hopefully this will inspire others... :-) Val=E8re |
|
From: Fuji H. <fo...@gm...> - 2004-11-04 17:39:22
|
I am also in favour of setting up a code repository. I'm interested to see how Subversion works, but CVS works too. -Fuji On Wed, 3 Nov 2004 11:51:48 -0800 (PST), Javier Fuentes <jv...@ya...> wrote: > Hi everyone. > > It has been a while since the last time I contributed > to the project with something useful. I have to admit > that I am guilty of doing nothing. :-) I will try to > catch up on lost time. > > > >Btw, should it be better to create a small cvs > > testzone or a separate CVS tree > > >or whtvr. Just for making the sharing of sources > > easier ? > > > > > > > > Hmm. Maybe. > > I think this is a great idea and it will speed up > things a little bit. Well, it is up to everyone to > decide. My vote is a "yes". What does everyone think? > > Have a nice day. > > Javier Fuentes > > > __________________________________ > Do you Yahoo!? > Check out the new Yahoo! Front Page. > www.yahoo.com > > > > > ------------------------------------------------------- > This SF.Net email is sponsored by: > Sybase ASE Linux Express Edition - download now for FREE > LinuxWorld Reader's Choice Award Winner for best database on Linux. > http://ads.osdn.com/?ad_id=5588&alloc_id=12065&op=click > _______________________________________________ > Nanodesigner-developers mailing list > Nan...@li... > https://lists.sourceforge.net/lists/listinfo/nanodesigner-developers > |
|
From: Javier F. <jv...@ya...> - 2004-11-03 19:51:54
|
Hi everyone. It has been a while since the last time I contributed to the project with something useful. I have to admit that I am guilty of doing nothing. :-) I will try to catch up on lost time. > >Btw, should it be better to create a small cvs > testzone or a separate CVS tree > >or whtvr. Just for making the sharing of sources > easier ? > > > > > Hmm. Maybe. I think this is a great idea and it will speed up things a little bit. Well, it is up to everyone to decide. My vote is a "yes". What does everyone think? Have a nice day. Javier Fuentes __________________________________ Do you Yahoo!? Check out the new Yahoo! Front Page. www.yahoo.com |
|
From: Elie De B. <el...@de...> - 2004-11-03 17:17:19
|
> >That's what I meant, an action subclass with an class extending Abstract= > Action=20 > >for each action. > > =20 > > > Yes, also important is the ability to add action classes from plug-ins=20 > later. > This will be the real tricky thing I think. That shouldn't be a problem either, we could load them with Class.forName(); so we should only get the name of the classes (and their location in the menu structure) from a configfile. Or from a factory class which contains a list/hashmap/... of all possible plugin classes. This is more a matter of convention than of technical difficulty. The code snippet posted was only proof of concept so Fuji could easily copy paste it to his current sourcetree. greetings Elie -- <============================> Homepage: www.de-brauwer.be <============================> |
|
From: <sw...@us...> - 2004-11-03 16:58:43
|
Elie De Brauwer wrote:
>>>Another design question now, like you might have noticed,=3D20
>>>adding some simple menu options in MainView without actually doing=3D2=
0
>>>something makes your class huge, now already 300 lines. So why not
>>>make a subpackage 'action' to contain all the menuactions, which can
>>>easily access the main application thru the mvcapplication singleton.
>>>=3D20
>>>
>>> =20
>>>
>>I think you should look at the command design pattern. You can put the=
=3D20
>>actions into a separate class actually.
>> =20
>>
>
>That's what I meant, an action subclass with an class extending Abstract=
Action=20
>for each action.
> =20
>
Yes, also important is the ability to add action classes from plug-ins=20
later.
This will be the real tricky thing I think.
>>I think we should first try to figure out how we solve the three-in-one=
=3D20
>>window and the other view, three separate windows. We essentially have=
=3D20
>>six items:
>>1. a menu bar
>>2. a message toolbar
>>3. a 3D pane
>>4. a top tool pane
>>5. a middle tool pane
>>6. a bottom tool pane
>>Each of them can be represeted by a class, each of them should have a=3D=
20
>>separate controller class I think
>>On top of that we have a class (or more than one) that creates one or=3D=
20
>>three frames depending on a preference variable.
>>With the use of three splitpanes we can create the panes mentioned abov=
e.
>>We have to work out the details a bit but I think it should be doable=3D=
20
>>like that.
>>Let me know if it is not clear... or, if you have another idea about it=
...
>>
>> =20
>>
>
>It certainly is doable with three splitpanes, consider the small proof o=
f=20
>concept code:=20
>
><code (Split.java) >
>import javax.swing.JButton;=20
>import javax.swing.JFrame;
>import javax.swing.JSplitPane;
>
>
>public class Split extends JFrame{
> private Split() {}
> public Split(String s) {
> super(s);
> JSplitPane main=3Dnew JSplitPane(JSplitPane.HORIZONTAL_SPLIT);
> JSplitPane upperMiddle =3D new JSplitPane(JSplitPane.VERTICAL_SP=
LIT);
> JSplitPane left =3D new JSplitPane(JSplitPane.VERTICAL_SPLIT);
> upperMiddle.setLeftComponent(new JButton("LEFT"));
> upperMiddle.setRightComponent(new JButton("MIDDLE"));
> left.setLeftComponent(upperMiddle);
> left.setRightComponent(new JButton("LOWER"));
> main.setLeftComponent(left);
> main.setRightComponent(new JButton("RIGHT"));
> setContentPane(main);
> }
> =20
> public static void main (String [] args) {
> JFrame frm =3D new Split("Test");
> frm.setSize(400,400);
> frm.setVisible(true);
> }
>}
>
></code>
> =20
>
I did not doubt it would be possible.
Initially we have to create three frames and add the different panes=20
depending on which view is chosen (one or three windows) depending on a=20
default or user preference. A preference listener keeps track of this=20
preference. If it is changed the appropriate method is called and panes=20
are added and removed from the appropriate frames. I don't think it is=20
more difficult than that.
>Btw, should it be better to create a small cvs testzone or a separate CV=
S tree
>or whtvr. Just for making the sharing of sources easier ?=20
> =20
>
Hmm. Maybe.
Regards,
Val=E8re
|
|
From: Elie De B. <el...@de...> - 2004-11-02 21:04:15
|
> >Another design question now, like you might have noticed,=20
> >adding some simple menu options in MainView without actually doing=20
> >something makes your class huge, now already 300 lines. So why not
> >make a subpackage 'action' to contain all the menuactions, which can
> >easily access the main application thru the mvcapplication singleton.
> > =20
> >
> I think you should look at the command design pattern. You can put the=20
> actions into a separate class actually.
That's what I meant, an action subclass with an class extending AbstractAction
for each action.
> I think we should first try to figure out how we solve the three-in-one=20
> window and the other view, three separate windows. We essentially have=20
> six items:
> 1. a menu bar
> 2. a message toolbar
> 3. a 3D pane
> 4. a top tool pane
> 5. a middle tool pane
> 6. a bottom tool pane
> Each of them can be represeted by a class, each of them should have a=20
> separate controller class I think
> On top of that we have a class (or more than one) that creates one or=20
> three frames depending on a preference variable.
> With the use of three splitpanes we can create the panes mentioned above.
> We have to work out the details a bit but I think it should be doable=20
> like that.
> Let me know if it is not clear... or, if you have another idea about it...
>
It certainly is doable with three splitpanes, consider the small proof of
concept code:
<code (Split.java) >
import javax.swing.JButton;
import javax.swing.JFrame;
import javax.swing.JSplitPane;
public class Split extends JFrame{
private Split() {}
public Split(String s) {
super(s);
JSplitPane main=new JSplitPane(JSplitPane.HORIZONTAL_SPLIT);
JSplitPane upperMiddle = new JSplitPane(JSplitPane.VERTICAL_SPLIT);
JSplitPane left = new JSplitPane(JSplitPane.VERTICAL_SPLIT);
upperMiddle.setLeftComponent(new JButton("LEFT"));
upperMiddle.setRightComponent(new JButton("MIDDLE"));
left.setLeftComponent(upperMiddle);
left.setRightComponent(new JButton("LOWER"));
main.setLeftComponent(left);
main.setRightComponent(new JButton("RIGHT"));
setContentPane(main);
}
public static void main (String [] args) {
JFrame frm = new Split("Test");
frm.setSize(400,400);
frm.setVisible(true);
}
}
</code>
greetings
Elie
Btw, should it be better to create a small cvs testzone or a separate CVS tree
or whtvr. Just for making the sharing of sources easier ?
--
<============================>
Homepage:
www.de-brauwer.be
<============================>
|
|
From: <sw...@us...> - 2004-11-02 20:41:33
|
Elie De Brauwer wrote: >As suspected we didn't see much of you first mail, but the link and the >app works fine here. It's basicly what valere did in paintshop written i= n java >;-). >Anyhow some comments: >* The framework looks nice, do you have any sources on that ?=20 >* Perhaps put the framework classes in a separate package.=20 >* Perhaps using the sun java graphics repository for a start:=20 >http://java.sun.com/developer/techDocs/hi/repository/ should make >the images look more java'ish. > =20 > They look fine although the toolbar should not be used as it is used in=20 this example. The toolbar I envisioned is not used for toolbar buttons but only as a=20 message bar to help the user in using the tool panel buttons on the left=20 of the 3D panel and to notify the user if the program is=20 calculating/working on some problem. >Another design question now, like you might have noticed,=20 >adding some simple menu options in MainView without actually doing=20 >something makes your class huge, now already 300 lines. So why not >make a subpackage 'action' to contain all the menuactions, which can >easily access the main application thru the mvcapplication singleton. > =20 > I think you should look at the command design pattern. You can put the=20 actions into a separate class actually. I think we should first try to figure out how we solve the three-in-one=20 window and the other view, three separate windows. We essentially have=20 six items: 1. a menu bar 2. a message toolbar 3. a 3D pane 4. a top tool pane 5. a middle tool pane 6. a bottom tool pane Each of them can be represeted by a class, each of them should have a=20 separate controller class I think On top of that we have a class (or more than one) that creates one or=20 three frames depending on a preference variable. With the use of three splitpanes we can create the panes mentioned above. We have to work out the details a bit but I think it should be doable=20 like that. Let me know if it is not clear... or, if you have another idea about it... Regards, Val=E8re |
|
From: <sw...@us...> - 2004-11-02 20:21:58
|
Hi Fuji, >I'm not sure if my last message got through or not, I tried to send an >email attachment. Instead I will just give you guys a link to the >file here: http://www.ualberta.ca/~jliske/SampleGUIDesign.zip > =20 > I didn't get your last mail either, Fuji, but the link is okay. I=20 downloaded and compiled it. Works fine. I think it is a good starting point to have this code and look how MVC=20 works in a practical, not too large, application. However Nanodesigner=20 is a lot more complex. I will write a bit more about it tomorrow. >Hi All, > >I have attached some code that I threw together based on an example >framework. I am not sure how you guys feel about using existing code, >nor am I familiar with the rules by GNU/GPL for open source project >code. However, I feel this might be good start to approaching the GUI >design. So take a look at the java classes. The classes that start >with "Wmvc" I took straight from the Wampler book (and are clearly >documented as being Wampler's), and the classes that start with "Nano" >I wrote based on his examples. It is mostly incomplete skeleton code >that will use the Wampler MVC framework. Let me know what you guys >think. We don't have to use this code in our project, but at least it >may give us some ideas as to how we should be thinking about the >design of the GUI using MVC. > >The Wmvc code is so short and elegant that I didn't feel the need to >rewrite it, although you guys may want to for the project, or even >take a different approach. The classes *should* compile and run on >Java 1.4.2. Let me know if the attachment goes through OK. > =20 > The rules of GNU/GPL are not difficult. You can use GPL code but you=20 have to keep the original copyright in place and you have to keep the=20 name of the author in place also. You can change the code and add your=20 name and point out the changes you made, no problem there. You also=20 have to keep it in the original package I think. Obviously if you use somebody elses GPL code in your code, your code has=20 to be distributed under the GPL copyright also. This is, by the way,=20 the copyright we will use. Regards, Val=E8re |
|
From: Elie De B. <el...@de...> - 2004-11-02 19:26:16
|
Hello, As suspected we didn't see much of you first mail, but the link and the app works fine here. It's basicly what valere did in paintshop written in java ;-). Anyhow some comments: * The framework looks nice, do you have any sources on that ? * Perhaps put the framework classes in a separate package. * Perhaps using the sun java graphics repository for a start: http://java.sun.com/developer/techDocs/hi/repository/ should make the images look more java'ish. Another design question now, like you might have noticed, adding some simple menu options in MainView without actually doing something makes your class huge, now already 300 lines. So why not make a subpackage 'action' to contain all the menuactions, which can easily access the main application thru the mvcapplication singleton. greetings Elie Quoting Fuji Hakayito <fo...@gm...>: > Hi All, > > I'm not sure if my last message got through or not, I tried to send an > email attachment. Instead I will just give you guys a link to the > file here: http://www.ualberta.ca/~jliske/SampleGUIDesign.zip > > *Original Message: > > Hi All, > > I have attached some code that I threw together based on an example > framework. I am not sure how you guys feel about using existing code, > nor am I familiar with the rules by GNU/GPL for open source project > code. However, I feel this might be good start to approaching the GUI > design. So take a look at the java classes. The classes that start > with "Wmvc" I took straight from the Wampler book (and are clearly > documented as being Wampler's), and the classes that start with "Nano" > I wrote based on his examples. It is mostly incomplete skeleton code > that will use the Wampler MVC framework. Let me know what you guys > think. We don't have to use this code in our project, but at least it > may give us some ideas as to how we should be thinking about the > design of the GUI using MVC. > > The Wmvc code is so short and elegant that I didn't feel the need to > rewrite it, although you guys may want to for the project, or even > take a different approach. The classes *should* compile and run on > Java 1.4.2. Let me know if the attachment goes through OK. > > -Fuji > > > ------------------------------------------------------- > This SF.Net email is sponsored by: > Sybase ASE Linux Express Edition - download now for FREE > LinuxWorld Reader's Choice Award Winner for best database on Linux. > http://ads.osdn.com/?ad_id=5588&alloc_id=12065&op=click > _______________________________________________ > Nanodesigner-developers mailing list > Nan...@li... > https://lists.sourceforge.net/lists/listinfo/nanodesigner-developers > -- <============================> Homepage: www.de-brauwer.be <============================> |
|
From: Fuji H. <fo...@gm...> - 2004-11-02 19:01:53
|
Hi All, I'm not sure if my last message got through or not, I tried to send an email attachment. Instead I will just give you guys a link to the file here: http://www.ualberta.ca/~jliske/SampleGUIDesign.zip *Original Message: Hi All, I have attached some code that I threw together based on an example framework. I am not sure how you guys feel about using existing code, nor am I familiar with the rules by GNU/GPL for open source project code. However, I feel this might be good start to approaching the GUI design. So take a look at the java classes. The classes that start with "Wmvc" I took straight from the Wampler book (and are clearly documented as being Wampler's), and the classes that start with "Nano" I wrote based on his examples. It is mostly incomplete skeleton code that will use the Wampler MVC framework. Let me know what you guys think. We don't have to use this code in our project, but at least it may give us some ideas as to how we should be thinking about the design of the GUI using MVC. The Wmvc code is so short and elegant that I didn't feel the need to rewrite it, although you guys may want to for the project, or even take a different approach. The classes *should* compile and run on Java 1.4.2. Let me know if the attachment goes through OK. -Fuji |
|
From: Elie De B. <el...@de...> - 2004-10-30 15:46:59
|
> I have a question for you (and all others)=85 > I am currently reading a book about the differences between Java 1.4 and= =20 > 1.5. It seems there is quite a difference and I am wondering if it might= =20 > not be a good idea to start with coding in Java 1.5 instead of in 1.4.x.= =20 > We haven=92t really started coding yet so instead of having to move to=20 > Java 1.5 next year somewhere, if v1.5 is supported by Eclipse and other=20 > IDE=92s, and recode everything we will have by then, it seems starting=20 > with 1.5 is a logical choice, isn=92t it? What do you think? Is there an= =20 > IDE already supporting 1.5? >=20 > I will put the book on the developers page. >=20 Hello, My 2 cents, it's not 1.5 anymore it's java 5.0 or tiger. I believe sun is a= lready offering JDK/netbeans bundles but I don't know for sure if netbeans supports the templates already. But the templates would be nice especially when=20 thinking of creating a framework and using native datatypes and the=20 collections framework. I also think that by the time some real coding will be happening the support for tiger will be much better. greetings Elie |
|
From: <sw...@us...> - 2004-10-30 15:35:33
|
Hi Fuji, > I haven't heard from you guys in quite a while over the mailing list. > What's happening? Anything new? Unfortunately I didn=92t have much time to work on the project the last=20 weeks. This will change however. I intend to step up =91the development=92= =20 considerably through a number of ways. > And where did the example JOGL file go from the developers only page? I removed it because I thought it didn=92t have much relevance anymore=20 especially after I decided to use a different layout (without using=20 Toolboxes for the Buttons) and I also thought everybody interested would=20 have downloaded it by now. > I'd like to take another look at it if someone can repost it please. Unfortunately (again) I didn=92t retain a copy of it. Maybe somebody else still has a copy? > Also, I've started coding up the MVC > framework for the GUI in Swing, it's going kinda slow but I will have > something together soon hopefully for you guys to look at. I was planning to have some chat sessions on November 1 and 2. I would=20 suggest to have a chat with you (others are invited also) about the UI=20 to make sure there is no misunderstanding. What do you think Fuji? Do=20 you have time on one of those days? I will be online (via Yahoo) during=20 most of the two days so if you can make some time to chat=85 I have a question for you (and all others)=85 I am currently reading a book about the differences between Java 1.4 and=20 1.5. It seems there is quite a difference and I am wondering if it might=20 not be a good idea to start with coding in Java 1.5 instead of in 1.4.x.=20 We haven=92t really started coding yet so instead of having to move to=20 Java 1.5 next year somewhere, if v1.5 is supported by Eclipse and other=20 IDE=92s, and recode everything we will have by then, it seems starting=20 with 1.5 is a logical choice, isn=92t it? What do you think? Is there an=20 IDE already supporting 1.5? I will put the book on the developers page. Val=E8re |
|
From: Fuji H. <fo...@gm...> - 2004-10-29 22:37:56
|
Hi All, I haven't heard from you guys in quite a while over the mailing list. What's happening? Anything new? And where did the example JOGL file go from the developers only page? I'd like to take another look at it if someone can repost it please. Also, I've started coding up the MVC framework for the GUI in Swing, it's going kinda slow but I will have something together soon hopefully for you guys to look at. Thanks, Fuji |
|
From: <sw...@us...> - 2004-10-11 19:25:09
|
Hi Javier, >This is in principle not a bad idea. However, I do not >know how well it will work at the end. The person in >charge of the class may lose interest and slow down >the development process. Additionally, a developer >will not feel completely free to colaborate because >someone else is already in charge of what this person >wants to help with. Who knows. It may perfectly work >for this proyect. I really hope so. > =20 > I do not think there is a perfect solution to these matters, not in an=20 open source project with volunteers anyway. In case there are disputes=20 or delays it is the project manager (yours truly) who will resolve the=20 issue. I also expect volunteers to act in a responsible way. =20 Val=E8re |
|
From: Javier F. <jv...@ya...> - 2004-10-11 12:36:21
|
Hi everyone. > I think it would be easier if the person who created > a certain class > stays responsible of that class at all times. Other > people who want to > change something in that class should send/discuss > their modifications > with the person in charge of the class. As it is the > original writer > he/she, I think, is the most knowledgeable person to > assess if the > change 'fits' in the overall design of the part he > is working on. > If for some reason the original writer does not want > the responsibility > of that class anymore it can be transferred to > someone else. If *that* > person changes something in that class he can add > his name as a > co-author unless it would be a complete rewrite of > course. This is in principle not a bad idea. However, I do not know how well it will work at the end. The person in charge of the class may lose interest and slow down the development process. Additionally, a developer will not feel completely free to colaborate because someone else is already in charge of what this person wants to help with. Who knows. It may perfectly work for this proyect. I really hope so. > > The MVC design > pattern might be a good > > choice when implementing the GUI. It is always a > good idea to plan > > out the OO design before jumping into code. I completely agree with you. I think we need some initial plan of how we will structure the project. MVC is the best option. __________________________________ Do you Yahoo!? Read only the mail you want - Yahoo! Mail SpamGuard. http://promotions.yahoo.com/new_mail |
|
From: Elie De B. <el...@de...> - 2004-10-08 04:34:02
|
> > NanoWiki for the UI Component. The MVC design pattern might be a good > > choice when implementing the GUI. It is always a good idea to plan > > out the OO design before jumping into code. I will try to get some > > brainstorming ideas posted into NanoWiki sometime this week. If > > anyone else has design ideas about the UI Component let us know, maybe > > some quick UML diagrams might help too. >=20 > I think the MVC design pattern is the best choice, yes. > Just in case there are people who are not familiar with Java design=20 > patterns I will see if I can find an ebook and add it to the main=20 > =91developers only=92 page to download asap. >=20 > Val=E8re An ebook covering design patterns in java is freely available at: http://www.patterndepot.com/put/8/JavaPatterns.htm greetings Elie |
|
From: <sw...@us...> - 2004-10-07 19:29:04
|
Fuji Hakayito wrote: >I noticed in the NanoWiki posting for "Software Development" we are >using a version tag for our code that includes author/version/date >(I'm guessing this is for CVS/Subversion). I am wondering if it is a >good idea to keep track of the original author AND the person who last >modified and committed the code into the repository. It might be nice >to see who the original designer was and also nice to see who has made >any recent changes. I think debugging will benefit if we can see who >touched the code last. > =20 > My first idea, as you know, was not to add *any* authors because I=20 thought it would become too complex to acknowledge every single=20 contribution to a certain class. After some thought I changed my mind=20 because it would be too elaborate to track down the person who=20 introduced a bug (as you mentioned). I think it would be easier if the person who created a certain class=20 stays responsible of that class at all times. Other people who want to=20 change something in that class should send/discuss their modifications=20 with the person in charge of the class. As it is the original writer=20 he/she, I think, is the most knowledgeable person to assess if the=20 change 'fits' in the overall design of the part he is working on. If for some reason the original writer does not want the responsibility=20 of that class anymore it can be transferred to someone else. If *that*=20 person changes something in that class he can add his name as a=20 co-author unless it would be a complete rewrite of course. >Secondly, I have been looking at the specifications that are posted in > =20 > > NanoWiki for the UI Component. The MVC design pattern might be a good > choice when implementing the GUI. It is always a good idea to plan > out the OO design before jumping into code. I will try to get some > brainstorming ideas posted into NanoWiki sometime this week. If > anyone else has design ideas about the UI Component let us know, maybe > some quick UML diagrams might help too. I think the MVC design pattern is the best choice, yes. Just in case there are people who are not familiar with Java design=20 patterns I will see if I can find an ebook and add it to the main=20 =91developers only=92 page to download asap. Val=E8re |
|
From: Fuji H. <fo...@gm...> - 2004-10-06 19:31:46
|
Hi Team, I noticed in the NanoWiki posting for "Software Development" we are using a version tag for our code that includes author/version/date (I'm guessing this is for CVS/Subversion). I am wondering if it is a good idea to keep track of the original author AND the person who last modified and committed the code into the repository. It might be nice to see who the original designer was and also nice to see who has made any recent changes. I think debugging will benefit if we can see who touched the code last. Secondly, I have been looking at the specifications that are posted in NanoWiki for the UI Component. The MVC design pattern might be a good choice when implementing the GUI. It is always a good idea to plan out the OO design before jumping into code. I will try to get some brainstorming ideas posted into NanoWiki sometime this week. If anyone else has design ideas about the UI Component let us know, maybe some quick UML diagrams might help too. Thanks, -Fuji |
|
From: <sw...@us...> - 2004-10-02 15:04:15
|
Hi all, I have added a preliminary document with an alternative view on the data=20 structure. http://nanodesigner.sourceforge.net/developers/wiki/index.php/NanoWiki_ta= lk:Javier_Fuentes Have a look at it and if you have any comments or ideas about it we=20 would like to hear them. I have only mentioned classes and their relation to each other for the=20 time being. I will add attributes and methods one of the next days. Have a nice weekend! Val=E8re |
|
From: <sw...@us...> - 2004-09-17 16:28:40
|
Fuji Hakayito wrote: > <>I've been reading up on JavaSpace technology and it's really > interesting, I look forward to seeing how it works with the project. > Anyways, I have a question about the Nanodesigner architecture: Why do > we need two separate clients for Designing (DC) and Computing (CC)? I > understand from the design description that the CC will only be > responsible for processing job requests (workunits). Is the intention > here to utilize machines dedicated for processing and processing only? > Or could we have one machine running both the CC and the DC clients > concurrently? > Let me know if you understand my question and what you think. Using two or more systems is only necessary in case there is a need for=20 large amounts of raw processing power not uncommon in molecular=20 modelling and design. Obviously to take advantage of parallel processing=20 you also need algorithms that can be parallelized which is a problem on=20 its own. So in case parallelizing is not possible or not desirable the=20 processing needs to be done on the Design Client itself. I think it should be possible to use both on the same machine although I=20 am not sure what the benefit would be in this case, except maybe for=20 debugging purposes. You are correct, it is the intention to use machines for processing only. Mind you that over time the newly designed molecules will become larger=20 and larger so obviously the processing power needed will become more=20 extensive. To give an idea about it: an ab initio (read: exact)=20 molecular modelling calculation of a molecule of 10 atoms or so takes=20 all the processing power of the biggest supercomputers. There is no need=20 to be alarmed however because I doubt it will ever be needed for the=20 goals we have set ourselves. There are a lot of other algorithms, less=20 accurate of course, but who will do quite nicely for the purposes of=20 molecular nanotechnology and other fields. Those kind of algorithms only=20 give approximations, true, but they do it a lot faster than ab initio=20 algorithms. More details about such an environment can be found on=20 http://math.nist.gov/mcsd/savg/parallel/screen/index.html Javier was kind enough to point me to it earlier on. I have contacted=20 the person in charge, William George, and he replied that their software=20 will be open source and available around the end of this year. I see no=20 reason why we couldn=92t adapt it to our goal. Also it would be=20 interesting to adapt it I think. No need to reinvent stuff. That holds=20 true for other parts of the software also. We only need to be careful=20 about license issues. Does this answer your questions Fuji? >Also, let me know if you guys have any GUI chat sessions. I >unfortunately don't have the Yahoo messenger, but I do have the >following: >ICQ: 18059890 >MSN: atm...@ho... >(I like IRC as well) > =20 > There has been no chat session for the UI group yet. You are in the=20 UTC-7 time zone if I am not mistaken, Javier is in the UTC time zone and=20 Chandresh is in the UTC+4 time zone I believe. It might be difficult to=20 arrange a chat session and I must confess I didn=92t take into account th= e=20 time difference while creating the groups. I only took into account your=20 apparent interests. Val=E8re |
|
From: Elie De B. <el...@de...> - 2004-09-17 08:42:31
|
Hello, Afaik your assumptions are correct. When visualising molecules the client will use a (big) part of its processing power (or call it wasting just to show something on the screen). So it's easy to imagine that distributing the workload over multiple clients crunchers will be much faster than rendering molecules and calculating them on the same cpu. Nevertheless a machine rendering a molecule will also have some idle time in which it can also dedicate to numbercrunching. So your question comes to this: "Can the setiathome webserver run a seti client ?". For small molecules/testing purposes I think all three servers can easily be one. But once the molecules and calculations increase you'd either want to see the workload spread acros machine or see smoke coming out your own machine ;-). Hope this helps (and i'm not telling any lies here). Elie De Brauwer > Hi Team, > > I've been reading up on JavaSpace technology and it's realyly > interesting, I look forward to seeing how it works with the project. > Anyways, I have a question about the Nanodesigner architecture: Why do > we need two separate clients for Designing (DC) and Computing (CC)? I > understand from the design description that the CC will only be > responsible for processing job requests (workunits). Is the intention > here to utilize machines dedicated for processing and processing only? > Or could we have one machine running both the CC and the DC clients > concurrently? > > Let me know if you understand my question and what you think. > > Also, let me know if you guys have any GUI chat sessions. I > unfortunately don't have the Yahoo messenger, but I do have the > following: > ICQ: 18059890 > MSN: atm...@ho... > (I like IRC as well) > > Thanks, > > Fuji > > > ------------------------------------------------------- > This SF.Net email is sponsored by: YOU BE THE JUDGE. Be one of 170 > Project Admins to receive an Apple iPod Mini FREE for your judgement on > who ports your project to Linux PPC the best. Sponsored by IBM. > Deadline: Sept. 24. Go here: http://sf.net/ppc_contest.php > _______________________________________________ > Nanodesigner-developers mailing list > Nan...@li... > https://lists.sourceforge.net/lists/listinfo/nanodesigner-developers |
|
From: Fuji H. <fo...@gm...> - 2004-09-16 23:45:16
|
Hi Team, I've been reading up on JavaSpace technology and it's really interesting, I look forward to seeing how it works with the project. Anyways, I have a question about the Nanodesigner architecture: Why do we need two separate clients for Designing (DC) and Computing (CC)? I understand from the design description that the CC will only be responsible for processing job requests (workunits). Is the intention here to utilize machines dedicated for processing and processing only? Or could we have one machine running both the CC and the DC clients concurrently? Let me know if you understand my question and what you think. Also, let me know if you guys have any GUI chat sessions. I unfortunately don't have the Yahoo messenger, but I do have the following: ICQ: 18059890 MSN: atm...@ho... (I like IRC as well) Thanks, Fuji |
