Older blog entries for baretta (starting at number 15)

Wow! Finally I got some free time to work on my projects. It's been a long time since my last entry here, but there isn't many "O.S. news" since then, unfortunately. All I can say is that I've resumed working on Seis again, and I started a new project called via.
Regarding Seis, I've changed some core issues in its design. All data is now handled by "Reactive Cross-Linked Directed Graphs" (tm) (a.k.a. directed graphs with triggers) :-). Also, player code don't need to call or maintain a frontend object now. It means that you can focus on writing good IA for you bot-player and you get it working on your preferred interface for free! :-). Of course, if you want it to be interactive, you will have to call frontend methods (it is obvious). I can say that everything but the game itself is already written and tested :-) The game code is still "Under Construction"(tm), but I think I'll have a beta out in 5 to 8 weeks. It will be an excelent game and a nice tool for some Genetic Programming research. :-)
Via is another (small) project that's almost finished (I'll probably release a beta today). It's purpose is to allow you to have multiple configuration profiles (both user-based and system-level) and let you select one (manually or test-driven) at boot time. It's something very useful for notebooks, and I've been using it for more than 3 months by now. As an example of why you'd like such thing: I use my own notebook to work on Conectiva and to do my stuff at home. At Conectiva I use DHCP to get an IP address, need to fetch my emails from its internal email server (it can't be seen outside Conectiva's network), I must use its email server to send every email, my browsers must use its proxy server, and so on. At home, I have static IP address, my own DNS server, my own email server, my personal email accounts, I want to fetch my personal emails only when I'm home (and work-related emails only when I'm working), I'd like to have a different set of applications starting up automatically, and so on. Changing my notebook configuration every day would be a PITA; making my own network resembles Conectiva's would be a PITA squared (and that would not work if I had a third place to plug my notebook at). So, that's what via came for. Via solves all these problems and many other ones! In future releases, it will also solve the hungriness in the world!!! :-) (Mmmmkay, I noticed you're interested in it and you don't know where to get it. By now, send me an email. Soon I'll put it avaliable on the net, probably at SourceForge).

Today I released UCLA 0.1. It's far away from achieving the objectives it's been designed for, but it is somewhat usable by now, and I'd like to get some feedback from the comunity about it.

Last night I did some work coding SEIS. Now it's a great one-channel IRC client/server. :-)
I also coded some stuff to handle the client-side data file, which contains player profiles. It seems to be working fine. The next things I should do is to code the server-side data handler, to store player information, such as password, level, experience, score, and so on, and then work on a simple authentication scheme.

Yesterday I finished coding the graph generation stuff for the CVS log analyzer. I also coded some procedures to draw the graph in ASCII-art. UCLA (the analyzer) is generating a separate graph for each file by now. The next step is put them all together and implement a nice drawing routine (maybe generate a .ps file). I'm also planning not to discard this ASCII-art stuff. It's proving to be useful! :-)
Ah, I created a project at SourceForge for it. I'll put it there ASAP.

Almost a week without posting a diary entry :-)
Today I finished "autoconfusing" the CVS log analyzer. I think I'll host it at sourceforge too.
Regarding SEIS, the Truco game, it evolved a lot in the last days! :-) Now it's almost a complete IRC client/server with only one channel (Oooohhh)! You can say things, whisper things to an user and emote things. Also, you can see who is logged on. All network communication stuff is done but handling some connection errors efficiently (disconnecting dead players, handling broken connections)

Today I finished the information gathering code, to retrieve useful information from the parsed CVS log. Now I just have to process the data (building a graph seems to be the best way out) and then generate a visual representation of the information graph...

Yesterday I started coding a CVS log analyzer. It is (will be) a tool that processes the CVS log of a project and generates a graphical representation of its branches. We can add other features later. :-)
By the way, I was told about a (uugh) perl/Tk script which does that, but I couldn't find it anywhere. If you know of it, please let me know.

And so it is!
I couldn't finish releasing G.H.O.S.T. yesterday, but now it's already there! From now on 2001/01/18 will be a internetional holiday to celebrate its death! While it decays into dust, I'll be working on the second version, which will add at least the following features:
  • Python-based ectoGHOST
  • SNMPv3 support
  • IPv6 support
  • Generic diagnosis algorithm support
  • Hi-ADSD, Hi-ADSD with Timestamps, Hi-ADSD with Detours, sync Hi-ADSD, and other Hierarchical Adaptive Distributed algorithm plugins
  • A yet-to-be-mathly-proven log(N) adaptive distributed algorithm plugin
  • Dynamic handling of diagnosed hosts
  • SNMP information gathering from the ectoGHOST
  • Management functions avaliable at ectoGHOST
  • Fully functional diagnostic performance analyser: epiGHOST
  • Lots of other amazing features
17 Jan 2001 (updated 18 Jan 2001 at 17:02 UTC) »

I decided to release G.H.O.S.T. version 1.0 today! It was a dificult decision, because I'd like to release a really good package, no known bugs, everything working correctly, blah blah blah, but , in fluent en_BR, I got tired of trying to "get milk from a stone". So I decided to release it the way it is and start working in a new version, rewritten from scratch. Thus the first version has born dead. But don't be sorry about that! As a G.H.O.S.T., it must die to exist, and so be it! As soon as I finish uploading it, you can check it out at SourceForge: http://SourceForge.net/projects/netghost/

Java sux!

6 older entries...

New Advogato Features

New HTML Parser: The long-awaited libxml2 based HTML parser code is live. It needs further work but already handles most markup better than the original parser.

Keep up with the latest Advogato features by reading the Advogato status blog.

If you're a C programmer with some spare time, take a look at the mod_virgule project page and help us with one of the tasks on the ToDo list!