Hi, Thanks Youness for explaining the idea. It was absolutely not true that Tcl/Tk would be dropped completely. Much more will be in C than currently in aMSN2, but still most of the logic will be in Tcl, only Tk will be dropped! And the logic will be implemented as a kind of 'core plugins'. (Phil, I must admit, the way I named those is not right, in the document is seems like those will be optional, but that is not what I meant. Needs to change in the next version.) Talking about languages, I will also clarify why I mentioned JavaScript. We will probably need it to implement some basic GUI behavior and to link the XUL-based GUI with the C code behind, but my intend is to keep the use of JavaScript as minimal as possible, and not use it al all if possible. In any case, no serious piece of logic will be implemented in JavaScript.
About glib and C++... I am more of a C++ hater than a glib lover. In my opinion C++ should never have been created. Many people seem to like it, but I don't see why. The way the OO principles are implemented in C++ is not usable for me (I learned about OO with VisualBasic and Java in the first place, then glib is close than C++, even though you need to code every thing yourself in glib, and C++ will do some for you, the principles followed by glib do match those of real OO as in Java better than C++ does). In fact Objective-C is still the best one around, but it is only supported well on Mac, support on Linux is limited, and Objective-C on Windows might not be possible at all. So for that reason it would not be a good choice for us. About my document missing much information... that is completely true. We'll need to develop that document. I put it in the Wiki too, and I will update it there regularly to reflect things discussed here, and once it clearly takes some direction, I will issue a new draft that contains the changes, and maybe new ideas emerging from that. About the Wiki page listing the modules and who will work on it, it is there indeed, but needs change. The final list of modules is different, and people will probably shuffle around a bit. For example I am at 'Telepathy to GUI Layer', but moved to XML2GUI in the meantime, finally dropping most of that in favor of an existing solution, but cannot go back, because there will probably not exactly be any 'Telepathy to GUI Layer'. What we'll probably have: - The GUI in XUL - D-BUS for Tcl - libamsn2_ui - libamsn2_tcl - libmsn - The 'core plugins' that will glue everything together. In the beginning anyone only able to code TCL can stay on aMSN1. They can join aMSN2 later, when some non-TCL parts they will need get near the alpha status. (Don't say this is unfair. aMSN 1.0 needs to get out too, and TCL coding for aMSN2 simply cannot be done in the first stage.) Harry Op maandag 17 juli 2006 01:00, schreef Youness Alaoui: > ok... I think we'd better discuss this on IRC, it will be a lot easier.. > unfortunately, as I said already many times I can't do much with my PC > right now... > I think Phil is wrong about glib.. I do understand that OO emulation is > not as good as real OO with C++ but problem with C++ is as you know, the > libstdc++ ABI changing every couple of weeks... so it's hard to package > something... > if it's not for that, I would say C++ all the way!!! > C is also prefered to C++ because of the linux commuity emotions where > everything everywhere is C and we don't want to switch to something else > (tell someone to compile a java version of the linux kernel, he'll > laugh/spit at you...) > about tcl/tk, we didn't say we'll drop it, on the contrary, it will be > our base... you need to see amsn2 as a 3 module project > the protocol - the logic - the graphics > the protocol will be farsight+telepathy+libmsn > the graphics will be XML files, XUL, etc... > the logic will be Tcl... > the logic will take care of saying "ohh, the protocol said I have a new > contact in my allow list, ok, then update the allow list data and do > whatever has to be done for that event"... > we'll also store in Tcl snit objects the different data types, the > contact list, the users, the groups, etc... > I don't know how XULRunner works but I think I saw a 'javascript' word > in your doc... what's that ? are you serious about using damn JS in > amsn2 ? I would say definitely NO! use Tcl if any scripting is required! > > Didn't we have a page on the wiki that described all modules and which > team member wanted to work on which module? it is very abstracted, > Harry's document aims to go more technical, but I think it lacks many > info... remember the mail I sent with details about the snit objects to > use ? the interactoin ? the directory structures, etc... ? read it and > use it... we need to join every mail on the subject. > Tcl should be the core of amsn2, this is what amsn2 is really about!!! > it's not because someone uses libmsn with telepathy to create his own > client that it's the same as amsn2.. the protocol is one thing and has > nothing to do with the client... the GUI is the same, it is our front > end, but a nice button is nothing if you click on it and nothing > happens... those events will be Tcl, so the aMSN spirit is not dead and > noone has to leave the team! > > KKRT > > On Sun, Jul 16, 2006 at 11:41:41PM +0100, [EMAIL PROTECTED] wrote: > > I agree as regards dropping tcl/tk - I cant program in anything else > > except PHP lol so i wouldnt be able to contribute if we switched langs > > (sorry Harry havent read the doc yet, just going on what Phil said, > > will read later) > > > > Tom > > > > On 7/16/06, Philippe Valembois - Phil > > > > <[EMAIL PROTECTED]> wrote: > > > Ouch ! > > > Hi, > > > between my so beautiful week end and my first day of work tomorrow, I > > > take the time to read mails and read yours too ;) > > > And the first point where I disagree is : > > > You want to drop Tcl/Tk ! > > > And it's a huge point ! Because if you drop it you will have many > > > devels who will leave the project because they don't know C or C++... > > > Tcl/Tk would be the only common point between aMSN and aMSN2 so if you > > > remove it why call it aMSN2 ? (not the same devels, not the same > > > languages, no incompatibility ) The only common point I can find is the > > > aim to have the better MSN client the Earth will ever have ! > > > KaKaRoTo you said I didn't created the amsn2 tree... But if I created > > > it I would have put in it our work with TkHtml... But where will be > > > TkHtml in this design ? It will be a GUI in TCL... But what said Harry > > > ? "Usage of Tk for plugin UI components will however be strongly > > > discouraged." so it will be a sort of "Do you I say but not what I do" > > > very strange... > > > In addition, you seem to be very tied with GLib and don't want to use > > > C++... Why ? Why use some emulation of C++ in C ? If you want to use > > > classes so we should use C++... Else I could advise you to code our > > > classes in assembly :p Well, in fact, I totally disagree with this > > > design but maybe I am the black sheep... Even if I know C and bases of > > > C++, I couldn't let all non C/C++ devels leave the team... > > > Please everybody say if I am wrong or not ? Maybe I have too much > > > feelings... Phil > > > P.S. Harry, it's not that your design is bad (maybe on Glib thing it is > > > bad :p) but I think it hasn't any relation with the aMSN that everybody > > > know... > > > > > > Le Sunday 16 July 2006 20:38, Harry Vennik a ?crit: > > > > Hi all, > > > > > > > > Last months I have done a lot of thinking about aMSN2. Especially > > > > about the GUI part of it. I don't know if everyone followed the > > > > XML2GUI discussion. Anyway, I changed my mind on that subject. I was > > > > convinced that, although XML2GUI solutions did already exist, those > > > > would not fit our needs well enough, so I started off designing a new > > > > solution, that would work for us, and probably a lot of others. But > > > > still I did not feel comfortable with that. It was just too much a > > > > matter of reinventing the wheel, and somehow I could not really > > > > believe that none of the existing solutions would fit. Especially XUL > > > > seemed so close, that I decided to go back one step, and > > > > reinvestigate the possibilities, and try to find out what I missed > > > > the first time. Eventually I found the answer, and chose Mozilla's > > > > XULRunner. > > > > > > > > After that, I started off writing the attached document, to specify: > > > > - What the aMSN2 project is meant to be > > > > - The aims of aMSN2 > > > > - The very basics of the technical structure of aMSN2 > > > > > > > > Please read it all, and send your comments on this thread! > > > > > > > > Harry > > > > > > ----------------------------------------------------------------------- > > >-- Using Tomcat but need to do more? Need to support web services, > > > security? Get stuff done quickly with pre-integrated technology to make > > > your job easier Download IBM WebSphere Application Server v.1.0.1 based > > > on Apache Geronimo > > > http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=12164 > > >2 _______________________________________________ > > > Amsn-devel mailing list > > > [email protected] > > > https://lists.sourceforge.net/lists/listinfo/amsn-devel > > > > ------------------------------------------------------------------------- > > Using Tomcat but need to do more? Need to support web services, security? > > Get stuff done quickly with pre-integrated technology to make your job > > easier Download IBM WebSphere Application Server v.1.0.1 based on Apache > > Geronimo > > http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642 > > _______________________________________________ > > Amsn-devel mailing list > > [email protected] > > https://lists.sourceforge.net/lists/listinfo/amsn-devel > > ------------------------------------------------------------------------- > Using Tomcat but need to do more? Need to support web services, security? > Get stuff done quickly with pre-integrated technology to make your job > easier Download IBM WebSphere Application Server v.1.0.1 based on Apache > Geronimo > http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642 > _______________________________________________ > Amsn-devel mailing list > [email protected] > https://lists.sourceforge.net/lists/listinfo/amsn-devel ------------------------------------------------------------------------- Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642 _______________________________________________ Amsn-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/amsn-devel
