Linux-Wildo! is an initiative to create something similar to the windows program Roger Wilco and Teamspeak, allowing users from different platforms talk with each other in real time with minimal CPU and bandwidth usage. Currently the project is being developed using C++ and we are heading towards a totally object oriented scheme.
We wish to offer users from different platforms the possibility to talk to each other; this way they can have a soft migration between operating systems. Nowadays when even Apple, the "closed everything" philosophy creator, is moving to a much more confortable open philosophy the obstructionism offered by some other companies leaves me without words. It's with this spirit We hope to success in creation of a program that unites users from different platforms instead of separating them.
The project is licensed under GPL.
To check out latest development version:
svn co https://linux-wildo.svn.sourceforge.net/svnroot/linux-wildo/trunk linux-wildo
Time is allways scarce
Time is something you cannot save and put apart, you cannot store it in a purse and you cannot buy more. To be more exact what you can do is work all the time and save the money, then use your own money to buy your own time or pay someone so you can save time on something, that is the meaning of the statement time == money. Problem, for me it is different
For me it different is because I am still a student so I did not get paid for what I do, but I still need to cook, wash, clean and organize my life, this simply causes me to not have much free time and so development goes on slowly.
There have been some bug fixes and some real minor feature in the main core. I mainly dedicated myself to cleaning small bits all around and adding features to the gtk2 client that now supports displaying of room list. At least we have some new buttons for the gtk2 client, they are by far more readable than the older ones
Some bugs are still pending and I will try to fix them instead of going on with the client, but this is also sad, because since LW is working being not able to use it cause there is no decent client is terrible.
During the summer you can expect some progress in the project, even if my girlfriend will close me in a deserted island without Internet or anything (It is true!! Not a joke!!), so wait patiently and you will finally get something you can use
Best Regards
Gabriele Dini Ciacci
I could have filled this news entry by telling you about all the design sessions we had during the last few days; the many hours of effort we spent to get something nice and clean; the knowledge about the obscure autotools arcane we gathered during those sessions
But I will not.
Why?
Because for once, you'd never believe it; I know it - you are not *that* naive. And besides that, that would ruin my reputation of miracle worker... err, coder; that would definitely not be nice, wouldn't it?
So, instead, I'll just tell you how we suffered, how many tears we dropped, and how the difficulties nearly torn us apart before making us closer and stronger than before. You'll not believe it either, but at least, it is a better advertisement for our reputation,isn't it?
Anyway, I strongly encourage you to test Linux-Wildo as much as possible, so we can get the best return about bugs you find and get a better software (and, well, we suffered a lot to make it work, so there is no reason you do not suffer the same, after all :)).
Oh, and don't listen Gabriele about the testing framework when he says it isn't "exactly user-friendly"; it is in fact "definitely not user-friendly" :D.
Dwarvenly Yours,
Y.E.J. Chachkoff
Meeting is funny
I write a news now because we are in the middle of a meeting, yes not a but meeting, but one. It is a meeting of two persons only, but that happen to be the two main developers of the project or at least the oldest ones!
Yann Chachkoff was so nice to spend some of his holidays to come and meet me, so we could talk of the project, take some design decisions and look forward for future plans
The result is that we managed to assemble a new release that will be out in the next days after we extensively tested it. So keep the eyes open for it.
As a side not i have to warn everyone that the final client is not ready yet and so to test the program you will be forced to use our internal testing framework that is not exactly user friendly.
Best Regards
Gabriele Dini Ciacci
Cooperation is success
Lot of time has passed since last status report, in that report I was telling the story of how much cooperation is complex. Now I will save you the story about how much a successful cooperation is proficient.
There was a great advancement in the status of the project. First, even if only for testing, it is again working. Second the error management code has been completely rewritten to use exceptions. Third a lot of the code has been converted to use some more of C++ features.
We are actually anyway facing some performance problems and that is why I am not releasing a new version just now. I want first to try to fix those problems since they can be caused by various spots or even by a mix of some of them. My priority now is to identify those sharp edges and smooth them out, so the new version could be used to toy a bit with friends and hear their voices.
Personal matters are sucking my time heavily and I cannot dedicate to LW all the time I would like to dedicate to it. Very sad is that my university exams are a bit in the back too, I hope to stabilize myself in the shortest possible time to be able to dedicate more to LW again.
Here I take the occasion to thank Yann Chackhoff for his hard work on LW and for his moral support in any moment I had need of it. I wish to thank all the others from the Linux-wildo IRC channel too, where I often idle, for their suggestions and constant presence.
Happy Christmas to everyone
Best Regards
Gabriele Dini Ciacci
Cooperation is complex
To recapitulate from the last post we had implemented all missing features and I have already added some improvements over them. I had many internal testings of the code and I successfully talked to some of my supporter users.
Many of you awaited and trilled for the release of the alpha version for the first of November 2005, time passed, nothing happened, release postponed. Sadly there has been delay, reason of it are explained on sourceforge page, some them are unrelated to my will, most of them are dictated by my wisdom.
Better release something working but complex to set up, and that is going to likely change in the short period or something more stable? I opted for the second.
Finally maybe I have found some help. A guy from Belgium started to help me, doing touch ups in the parts of the code where I am less skilled, but starting a cooperation is complex.
After in the past months he branched the code enhancing it, we have then failed to merge back the changes, killed by some bugs. So we tried it again and I declared a CVS freeze to ease, or really zero, the merge back issues.
CVS is frozen by 5 weeks and I am dedicating myself only to writing a GUI using the GTK toolkit. We are all awaiting the moment where we can merge back the changes and see the new features, that even if are only at code level (plug-ins, better error handling, better thread support, etc...) will really boost the project quality.
In the next month after the merge back I will dedicate myself to finish the work on the GUI, to clean the client API and to make it more robust. This way building another GUI should become an easy task.
Best Regards
Gabriele Dini Ciacci
Time passes, code rests.
As an alone coder I slowly search for time to implement a new feature or add something that brings us a step closer to paradise.
So here I start with the nice news, I deleted the old server code from Vanya Sergeev, rewrote LW_ServerCore class, reorganized the server structure and finally ported lwtestserver to his now not anymore test domain.
This means 3 things, the First is that we have a working control server that is the definitive server, not just some testing code placed there to try the client.
Second is that we now have 2 classes LW_ControlServer and LW_DataServer that does all the job about the 2 server and actually you can have the bin start one, the other or both, like a real server should be able to do.
The Third is that I am really writing that data server class and it has more or less 300 lines just now, it has all the basic things working, maybe with just some feature missing.
This is more or less the list of missing things:
Best Regards
Gabriele Dini Ciacci
As you may have noticed the project has been a long time freeze, no commits, no status reports, no activity on the irc channel. This has been caused by my new renewed effort to finish my engineering studies, a sad news for you is that my time shortage will continue until I really finish them.
The nice news that convinced me to make this post is that I have worked a bit on he code anyway and that now lw is able to send the audio data it read from the audio card, server is not ready yet to receive them and client cannot manage a receive too, but some code got in and that is the important thing.
A nice fact is that if I find myself in a good mood after my next exam I will add receive ability to the server and then we will see what happens. Expect another report and wish me good luck. Salutes to everyone.
Best Regards
Gabriele Dini Ciacci
Life is Life, none know what will happen the next day. So a day some time ago our server programmer Vanya Sergeev reappeared. He motivated with quite surprising arguments his long absence, so I took him back in the team.
First thing he did was the cleanup of the server code, now he has mantainership of only the data part, while I do the control part. We moved from the old fast to code array for storing clients to a new nice linked list architecture that both the data and the control server share. The nice addition here is the lwtestserver program that can accept a connection from lwtestclient, it is very nice seeing that all the network code seem to work and the protocol look like rocking.
Documentation has been improved a bit, I preferred focusing on the code now that I had some spare time. Network code after some testing with the above programs has been cleaned and got some new functionality to improve his flexibility and ease of use.
The latest news is lwgsmtest that, after some problems, shown that the coding-decoding functionality is working perfectly at least with gsm. Now you can enjoy encoding and decoding your voice, storing it to a file or replaying it on the fly. Some design changes were necessary in the codec and soon I will change many of the internal naming to avoid conflicts with commonly used names.
The future for me show a new feature for our logger, so in the next review I can use it functionality in all the code. It will be nice and you will notice it if you will ever mess with lw code. After that I plan to add some features to lwtestclient and lwtestserver and then try to merge lwtestserver with data server code.
In the middle of that I released a package version for sourceforge, so everyone can do some joking with the code, do the usual configure steps but before doing the make cd in the test directory, so you will get no compile errors. That version sadly still miss lwgsmtest program, you have to get cvs version for that.
Best Regards
Gabriele Dini Ciacci
After really a lot of time I finally find time to post a report. We have had very big problems with coders, the fact is that none is active and as you can see my university is eating up all my time, so I personally have no time to code by myself. Anyway for all of you that have been on the Mailing list you may have noticed that I never lagged more than 1 day to reply to mails.
"Not is bad all that look bad" in all that time I had some time to layout the documentation and learn some TeX to write it. It is nice and is available in the cvs in the documentation directory, a dvi version is available too for those missing a TeX interpreter.
Now the bad news, the biggest is that we lost our main server developer, Vanya Sergeev aka Br33zy left with a discrete silence saying, some months ago, "I will commit my code on Wednesday", I never seen or heard him again. Minor bad news are not available, sorry, you have to keep stick with that one big.
Proposition for the future are that I will take over server coding too, rewrite it from scratch taking good ideas from V. Sergeev code, hoping there are good ideas in it. I am sure something good can come out, especially when my version will finally support our protocol and be able to do some negotiation and login.
Best Regards
Gabriele Dini Ciacci
I posted some screenshots of the Tcl/Tk GUI that i've been developing. (see the screenshots page.)
I'm working on the room dialog at the moment, if anyone has any ideas, submit them to the dev list and i'll get them.
- Tim Hentenaar (aka Matriark TerVel)
Many of use are engaged with their respective works and/or studies, but the project is slowly growing, we have nearly all Abstraction Layers necessary to code various pices of the project and the audio and codec parts are ready to go alpha testing phase. The network is still a thing on the work, after changing for the 3rd time it's impelemtation. We all really hope that this is the good one. We have choosen to use RTP as protocol for data transmission while leaving all other comunication between the client and the server to TCP.
Windows audio support is still missing, since our windows developer is currently busy with the network code. Help is glad, the audio part is really a small thing, about 2 files, totaly abstracted, so you will have to only look at the sound.h and make your class conforming to it's really basic and easy API. In later review maybe the work will be harder, but for now basic support is enougth
GUI has been recently abstracted (just yesterday) and one of our coder is willing to write a Tcl/Tk inteface, so probably even if it will be not the "offical" one, it will be the first ready. Planned GUI are many, we want users to be confortable as much as possible, so a small list comprend: QT, Ncurses, native mac, fltk and maybe PicoGUI if we go for the PDA port.
Comments are wellcome from anyone and a some opinion feedbeck and/or feature request is even more.
Regards to all
Gabriele Dini Ciacci
