Hi Frederick.<br><br><div class="gmail_quote">On Mon, Apr 26, 2010 at 6:37 AM, Friedrich W. H. Kossebau <span dir="ltr"><<a href="mailto:kossebau@kde.org">kossebau@kde.org</a>></span> wrote:<br><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
Hi Leandro,<br>
<br>
thanks for your introduction.<br>
<br>
Vendredi, le 16 avril 2010, à 06:49, Leandro Melo de Sales a écrit:<br>
<div><div></div><div class="h5">> Hello guys, I'm very glad to get to all of you in this e-mail, finally!<br>
> I'm sending this e-mail because I'd like to discuss more about the use of<br>
> the UPnP stack in the KDE projects. As some of you already know, me, as the<br>
> BRisa Project Manager and a group of some developers are implementing a<br>
> UPnP stack in Qt called BRisa UPnP Framework. I have recently found the<br>
> project HUPnP at the same time you guys were talking about this in this<br>
> list but unfortunatelly I was very busy with other stuff. We are<br>
> developing the BRisa project in partnership with Nokia Institute of<br>
> Technology [1] and the Federal University of Campina Grande at the<br>
> Embedded Systems and Pervasive Laboratory [2], both in Brazil, with a goal<br>
> of make a powerful<br>
> implementation of a UPnP stack in Qt.<br>
> We started the implementation in the middle of the last year after some<br>
> discussion with Kate Alhola (Nokia/Maemo project manager) in the ELC<br>
> (Embedded Linux Conferece) 2009. Recently we have released the first, but a<br>
> complete version, of a UPnP stack, which is available at [3] and it was<br>
> based on our mature python version of a UPnP stack. In fact, we have being<br>
> working with UPnP stuff since 2005 and many issues with the UPnP specs has<br>
> being studies etc. Now, after a great efforts from many developers, we have<br>
> the complete stack available in Qt with a strong goal of providing a native<br>
> UPnP stack implementation in three main programming languages: Qt, Python<br>
> and Java, but with a unique programming language API.<br>
> I'm glad to say that we are currently supporting not only windows, linux<br>
> and mac, but also maemo platform and with the targets for MeeGo Platform (a<br>
> new mobile platform built as the result of a merging of maemo platform,<br>
> from nokia, and moblin, from Intel).<br>
<br>
</div></div>Impressive. So your complete stack is running on all of these platforms, or<br>
does it reuse things already available?<br>
<br></blockquote><div><br>It basically uses LibQxt project and QtSOAP. The rest is from BRisa project.<br> </div><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
What is the relation to GUPnP/Rygel/..., which seems to be part of the<br>
current/next maemo platform? Is there cooperation or rivaling?<br>
<div class="im"><br></div></blockquote><div><br>Neither one. They have different strategy and we are an alternative for GUPnP since it is written in Qt. GLib is available for maemo, which enables GUPnP to be run on it. Qt is available also, which will also enables BRisa to run on it, Qt is becoming a standard for maemo platform, we are looking for this and that is it.<br>
</div><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;"><div class="im">
> This is why I'm writting to all of you, unfortunatelly only now. I'm<br>
> currently at the Embedded Linux Conference 2010, where I gave a talk about<br>
> our project, but most importantly, also in the Linux Collaboration Summit<br>
> (the conference merged in one they), I had opportunity to talk to some guys<br>
> from Nokia and Qt Framework to work in collaboration with them in this<br>
> initiative to make a unique and powerful UPnP stack in Qt.<br>
> It is in this moment where I'd like a little from the attention of all<br>
> of you guys. I'd like to discuss the possibility of integrate Qt-BRisa<br>
> and HUPnP in just one project where we can collaborate and make a decent<br>
> UPnP implementation.<br>
<br>
</div>If you could join forces that might be great.<br>
<div class="im"><br></div></blockquote><div><br>Yes, it is, I think. But, recently the HUPnP project manager answered me in a private conversation and he think it will be difficult to do, although he clean said that it would be great, but he didn't see any merging possibilitity. The only option he sees is to stop one of both project and all of us focus in just one, but he said that it will not do this with HUPnP. Since will cannot do this also, I also can't see a easy possibility for this.<br>
<br>Well, BRisa is part of my PhD researching in partnership with some companies interesting on supporting the project. The project is available, there is a set of developers working on it and it is also supported by the university where I'm taking my PhD. It is there and just want to let you know about it, but I really don't like to stress any deadlock situation on this. My PhD is going well, we develop many researching on UPnP stack for user authentication and service authorization (we gained a best paper award on this) and many others technical/researching projects on top of UPnP and this is our thoughts and interesting.<br>
<br>UPnP is a very good standard, although it is limited in many aspects, where we are always targeting on these issues, while maintaining a UPnP stack compliant with the standard. Although it has some limitations, the way how it was thought allows a object-oriented design that allow it to be implemented in different programming languages. This object-oriented design we believe that it is really possible and is it real in BRisa UPnP stack. Since we are developing based on a standard, I can see this is possible. It is **similar** to implement a TCP stack in different languages but provide the same BSD-Like Socket API.<br>
</div><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;"><div class="im">
> I have being also talking to Bart about the<br>
> possibility of integrating Qt-BRisa in the KIO and also, as a<br>
> representative of the research field (as the university) too, we have<br>
> being working on some improvements of the UPnP specifications, mainly in<br>
> the area of security issues, such as authentication and authorization<br>
> mechanism. In this field, we have recently publish an article that talked<br>
> about a possible solution on this and as a result we have gained the best<br>
> paper award of the conference.<br>
<br>
</div>Would be nice to have you as such an experienced player in the construction of<br>
a good UPnP support in the KDE platform.<br>
<br></blockquote><div><br>Well, this will depends on how the things will be decided. I'm open for discussion.<br><br>BR,<br>Leandro.<br><br>-- <br>Leandro Melo de Sales<br>Professor in Computer Science at Federal University of Alagoas, Brazil<br>
Pervasive and Embedded Computing Laboratory, UFCG<br>PhD candidate in Computer Science at UFCG<br><br></div><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
Cheers<br>
Friedrich<br>
<font color="#888888">--<br>
KDE Okteta - a simple hex editor - <a href="http://utils.kde.org/projects/okteta" target="_blank">http://utils.kde.org/projects/okteta</a><br>
_______________________________________________<br>
Kde-hardware-devel mailing list<br>
<a href="mailto:Kde-hardware-devel@kde.org">Kde-hardware-devel@kde.org</a><br>
<a href="https://mail.kde.org/mailman/listinfo/kde-hardware-devel" target="_blank">https://mail.kde.org/mailman/listinfo/kde-hardware-devel</a><br>
</font></blockquote></div><br><br>