Menu
▾
▴
mentalunix-developers
|
From: Mentalspice <clu...@ya...> - 2000-12-07 20:36:45
|
I've been really bogged down with school work, so I haven't been able to do much work(that and my hidden addiction...IRC). I, unfortuanetely, have become addicted to debian. I say, let us use dpkg and apt, as well as some of the other debian programs. I'm not saying base the distro on Debian, but just use the nice apps that make the user experience good. Then, package everything from the ground up. I envision mpkg as the universal packager. It could use the debian database as its internal format, but use librpm, the slackware stuff, and the stampede stuff to modify their databases. And, keep the package databases for the different formats synchronized. And, eventually, have a daemon that monitors the /usr partition for changes, and checks a file databse to see if they belong in a package. For the file system, the user should have the option of using ext2 or reiserFS. Ext3fs is a waste of time, since it is so slow. When Tux2fs is released, it should also be included(since it is so nice). Using the 2.4.0testX kernels would be a good idea, since they will be final come january. Including support for every known peice of hardware, by including everything as modules, should be a goal. X: well, Xfree86 4.0.1 is good, but include XFree86 3.2.6 as well, since not all cards are supported by xf4(like my S3 trio64). The xlibs should all be 4.0.1, because the 3.3.6 servers can use them. Installation program: Using the Mandrake / Debian Potato installers as references is good. Grab the source for them, and admire the beauty. The MentalUNIX installer has to be graphical, and text based. My solution is to use ncurses, and port the library to X(requiring a simple relink of a program for it to work in graphics mode), or to use QT(since framebuffers QT is pretty nice). Everything is best done in C++(I have seen the dark, evil light), since objects are nice to have when doing GUI work(and since I am becoming good at coding in it). Using a modular install should be a goal as well, just write a seperate program for each step, and integrate it into one fully functioning app that calls them(or just uses their function compiled as a library). Also, try reading the source to the caldera and redhat installers...it can never hurt. The goals for 0.0.1 / the 0.0.x series should be to get the base system running. I'm talking, the package managers, core programs(update-menus, etc), and the shells all running. Just as a huge, bootable image / a downloadable tarball. By 0.5, the text mode installer should be ready. 1.0 should feature an up to date, functioning system. By 2.0, Yozilla and mpkg should be finished, but I don't know about vk(imagine it as a kernel module...). Oh, and yozilla. Yozilla should be built on Gtk+, and integrate mozilla using GtkMozEmbed(). Basing it on skipstone is not a bad idea, and then adding all of the communicator features to it. Of course, rewriting the MPL portions of gecko to be GPL is neccesary. And, why not improve the engine too? Full CSS2 would be nice... --------------------------------------------- DO NOT REPLY TO THIS MESSAGE BY EMAIL! --------------------------------------------- You have chosen to receive messages from "mentalunix" by email. Reply to this message: http://clubs.yahoo.com/clubs/mentalunix/bbsfrp?action=r&tid=mentalunix&sid=1600006580&mid=337 Unsubscribe from the Club mailing list: http://clubs.yahoo.com/clubs/mentalunix/config/change_mb_list Return to "mentalunix": http://clubs.yahoo.com/clubs/mentalunix ---------------------------------------------- Not a member? Remove yourself from this list: http://edit.clubs.yahoo.com/config/unsubscribe_mb_list?.userID=munix_devr&.groupID=mentalunix&.groupType=&.code=IVyJGMuRxG |
|
From: CyDust <clu...@ya...> - 2000-12-07 21:31:26
|
same here, so no prob. Yesterday was the last day of the fall quarter, so i have about 3 weeks to work on all this. My ideas were similar, but i dont like the idea of using GTK+ for much of anything. The reason for that is that GTK is Gnome TK and requires the gnome libs. There isnt necessarily anything wrong with that, but it would be better to be able to create system wide stuff. So, my plan is learning and using python. I've used a few apps created in python, and their great. It may be a little more work, but I'm all for it. I agree with X stuff, and kernel. I'm running 2.4 test 10 and its great, except my network card still aint working under linux. Filesystem stuff, sounds good, I have heard much about ext3 lately, thats why I listed it. And Reiserfs is likely to replace ext2 soon (or so I've heard) and as for tux2fs, havent heard of it, so Im not sure there. I just started working on the docs and stuff, so I'll consider your ideas, and try to find some common ground. Since I'm out of school for a while now, I think its time we actually set up a irc chat or something. But, we'll set that up later. I really want to try debian again, I know its great, but im stuck with my current favorite SuSE!! Initially, during the development of mpkg, we should use either rpm or deb for the overall installation of stuff. i havent had much experience with dpkg cause i havent used debian much, but, I think we need to start fresh, and create a entirely new system. I'm not going to take the time to explain it, here, I'll type up my thoughts and post my thoughts of how it should work. As for install stuff, I'm thinking we need at least some simply text install by 0.0.x. that way it can be tested. All the install has to do is beable to run fdisk, and install the kernel and some basic apps like dpkg (or rpm), gcc, bash, and other basic apps. the 0.0.x series wont be much, just a base with a lot of unstable software. by 0.5 yozilla should be partly done, and I'll post my ideas on that later as well, and I'll write up a general work plan and idea paper that we can work off of to get a roadmap in place. Yozilla will be semi functional by the 1.0 release (think lynx functionality) and by 2.0 by a fully fledged web browser. Mozilla isnt exactly fast under linux, so I plan on taking the gecko engine, and rewrite it (both statically linked and dynamically) to speed it up. New name will be decided later. Also, I strongly suggest you read the docs I post later, because the will explain, in much detail what I think, and how I want to develop this stuff. We are not going to base the distro off of another to create a reliable interface and system. It will make everything easier in the long run. deb and apt aren't bad ideas, and neither is using debian specific apps, but their are problems with that, and those will be explained in my docs. So expect to need a lot of time reading by friday or saturday night! --------------------------------------------- DO NOT REPLY TO THIS MESSAGE BY EMAIL! --------------------------------------------- You have chosen to receive messages from "mentalunix" by email. Reply to this message: http://clubs.yahoo.com/clubs/mentalunix/bbsfrp?action=r&tid=mentalunix&sid=1600006580&mid=338 Unsubscribe from the Club mailing list: http://clubs.yahoo.com/clubs/mentalunix/config/change_mb_list Return to "mentalunix": http://clubs.yahoo.com/clubs/mentalunix ---------------------------------------------- Not a member? Remove yourself from this list: http://edit.clubs.yahoo.com/config/unsubscribe_mb_list?.userID=munix_devr&.groupID=mentalunix&.groupType=&.code=IVyJGMuRxG |
|
From: CoolProgrammer <clu...@ya...> - 2000-12-07 23:24:59
|
For the kernel version, I'd recommend whatever the most recent stable version is when we compile. For XFree86, 4.0 is nice for those that can run it--if we can spare the space, it might not be a bad idea to include that and the last 3.x version so that those who are running older video cards for which support has been dropped can still run X. As to filesystem, I believe there should be a choice--I'm personally going to try ReiserFS if/when I get S.u.S.E. However, some might still want ext2 or ext3. Oh, after experiencing the pain of patching the kernel for PCSP support (I wouldn't generally recommend this patch, actually, but the TBS Montego is, to the best of my knowledge, not supported in Linux), I think it might be a good idea to patch the kernel for DVD ioctl support if we choose a kernel that doesn't have the support built in. Of course, there are so many kernel options that I'm in favor of using Slackware-type tactics--let the user choose one to install based on hardware support needs. ~CoolProgrammer --------------------------------------------- DO NOT REPLY TO THIS MESSAGE BY EMAIL! --------------------------------------------- You have chosen to receive messages from "mentalunix" by email. Reply to this message: http://clubs.yahoo.com/clubs/mentalunix/bbsfrp?action=r&tid=mentalunix&sid=1600006580&mid=340 Unsubscribe from the Club mailing list: http://clubs.yahoo.com/clubs/mentalunix/config/change_mb_list Return to "mentalunix": http://clubs.yahoo.com/clubs/mentalunix ---------------------------------------------- Not a member? Remove yourself from this list: http://edit.clubs.yahoo.com/config/unsubscribe_mb_list?.userID=munix_devr&.groupID=mentalunix&.groupType=&.code=IVyJGMuRxG |
|
From: CoolProgrammer <clu...@ya...> - 2000-12-09 01:50:15
|
Or, could we just modify the Debian package manager to work with librpm and .tgz files? Then, we only have to worry about one database that contains data for all three types. Using C or C++, I could probably fairly quickly come up with a text-based install (I have done some ncurses coding before) if someone could tell me how to change the keyboard input mode so that it does not buffer but gives characters as they are pressed. Oh, yeah--I am convinced that every feature that can be built as a module that we choose not to build directly into the kernel should be built as a module (in other words, get option-happy). This way, it will work for many situations without the user having to immediately learn how to compile a kernel. CSS2 support would be sweet... ~CoolProgrammer --------------------------------------------- DO NOT REPLY TO THIS MESSAGE BY EMAIL! --------------------------------------------- You have chosen to receive messages from "mentalunix" by email. Reply to this message: http://clubs.yahoo.com/clubs/mentalunix/bbsfrp?action=r&tid=mentalunix&sid=1600006580&mid=345 Unsubscribe from the Club mailing list: http://clubs.yahoo.com/clubs/mentalunix/config/change_mb_list Return to "mentalunix": http://clubs.yahoo.com/clubs/mentalunix ---------------------------------------------- Not a member? Remove yourself from this list: http://edit.clubs.yahoo.com/config/unsubscribe_mb_list?.userID=munix_devr&.groupID=mentalunix&.groupType=&.code=IVyJGMuRxG |
|
From: Mentalspice <clu...@ya...> - 2000-12-09 02:51:16
|
CSS2: possible. If bender is based on gecko, then it will require a lot, but not a huge amount of hacking to get it fully implemented. Of course, cleaning up the code and killing all of the XUL stuff for Gtk-- and removing any MPL code should be the first priority. My idea for the installer is to get a text based version first, using just the standard c++ text mode stuff(cout <<'s and such). Then, start to port the modules to run on ncurses. My grand scheme is to port ncurses itself to run on gtk or svgalib. Easy GUI mode. But, of course, getting the installer just to work should be the first priority. And packaging everything. I'll try to apt-get source kernel-2.4.0test11 on monday, and compile it with everything set up(and of course tar up all of my compilation stuff). I'm still using 2.2.17pre8! It has worked great for me(300K kernel, 500K with modules). For the base system, dpkg and apt are musts. As well as w3m(since it is so much better than lynx). SysVinit and init are also needed, as well as some essential libraries. No rpm yet, I'll package it later and we can include it with a later version of the installer. No emacs, but I think including mc and mcedit are feasible. Of course, it needs a kernel, and the fileutils, so, my sketch of a base system would be: kernel (2.4test11) shell (bash) init fileutils mc / mcedit apt(apt-get, cache) dpkg glibc(libc6) libstdc++ any apps I'm forgetting? --------------------------------------------- DO NOT REPLY TO THIS MESSAGE BY EMAIL! --------------------------------------------- You have chosen to receive messages from "mentalunix" by email. Reply to this message: http://clubs.yahoo.com/clubs/mentalunix/bbsfrp?action=r&tid=mentalunix&sid=1600006580&mid=346 Unsubscribe from the Club mailing list: http://clubs.yahoo.com/clubs/mentalunix/config/change_mb_list Return to "mentalunix": http://clubs.yahoo.com/clubs/mentalunix ---------------------------------------------- Not a member? Remove yourself from this list: http://edit.clubs.yahoo.com/config/unsubscribe_mb_list?.userID=munix_devr&.groupID=mentalunix&.groupType=&.code=IVyJGMuRxG |
|
From: CyDust <clu...@ya...> - 2000-12-09 09:32:22
|
hmm, sounds good. but you did forget egcs (a c compiler would be very good, and should be included) I have a Linuxcare business card cd that has a base version of debian on it, but i've been having trouble getting it to work cause it doesnt have much of anything on it. Possibly something similar to that would be good, ill post details on it later after i get them! Also Im starting to wonder if maybe we should start off with dpkg as the main system until at least 1.0.0 release or so, and later implement mpkg. This will require a bit of work on changing the database over and whatnot, but that shouldnt be too difficult. Also, straight deb packages shouldnt be used cause they are debian specific. Of course i have very little experience under debian at the moment, i may be quite wrong. But I should know soon! RIght now im downloading counter strike and a half life patch, so my bandwith sucks right now (damn dial up) so I'll check the updated docs later. As for gecko and blender. My goal isnt to implement gecko in, but rewrite it (maybe that will be blender). The advantage im going for is cleaner, faster code, and something I can work with easier then implementing in static code. Also, I plan on going with the rest of YoZilla and making gecko modular. This stuff will be very, very difficult, and will take a long long time, but in the end it will be worth it. depending on how much work we start doing, and how far we get, im thinking maybe a generic, semi-usage distro (say 0.5.0) around June, if not earlier. The main thing is that we get to work. --------------------------------------------- DO NOT REPLY TO THIS MESSAGE BY EMAIL! --------------------------------------------- You have chosen to receive messages from "mentalunix" by email. Reply to this message: http://clubs.yahoo.com/clubs/mentalunix/bbsfrp?action=r&tid=mentalunix&sid=1600006580&mid=347 Unsubscribe from the Club mailing list: http://clubs.yahoo.com/clubs/mentalunix/config/change_mb_list Return to "mentalunix": http://clubs.yahoo.com/clubs/mentalunix ---------------------------------------------- Not a member? Remove yourself from this list: http://edit.clubs.yahoo.com/config/unsubscribe_mb_list?.userID=munix_devr&.groupID=mentalunix&.groupType=&.code=IVyJGMuRxG |
|
From: The_Micea <las...@ad...> - 2000-12-09 19:15:33
|
Ok, it sounds like we are leaning towards making a minimalistic OS. One thats good for developers, hackers, and such and runs smothly and effeciently. Debian sounds great for that. I think we should keep the install as small as possible. The other Linux Distributions like redhat, are getting to bloated, more like windows. A gui is nice, but its the apps, and speed that attracts a user. Right now even BSD is abandoning the developer and hacker in order to get more users. I think we should include Vi.editor. The newest version is very small but usefull, I do all my coding with it. Also Vi and emacs are big rivals, you can include one without the other or else risk angering some hardcore hackers. Heres my ideal list [Main] [2.4.x] kernal init fileutils apt(apt-get, cache) dpkg mc/mcedit Bash shell Zsh shell csh shell [editing] Vi Emacs [Graphics] xwindows with enlightenment maybe [networking] pump(for detecting dsl and cable ip numbers from a dhcp host) dhcp client ftp client http client. telnet whois ping tracert [mail] pine [internet] lynx [development] glibc6 libstdc++ gcc and related libraries python perl This would easly fit on one cd, and also be small enough for people to download. nice and simple and contains many useful programs. |
|
From: Mentalspice <clu...@ya...> - 2000-12-11 19:46:46
|
No debs. No compiler. Just a small, nice base system. The base system only needs the basics - a package manager, a simple text editor(no emacs or vi), and the file utilities. That is it. It should fit in under 30MB, and be stored as a large .bin file. The installer would mount it to a loop device, and copy its contents to the drive the user chooses(since it will be mounted, it will just copy / to /target, which is the target drive...just like in slackware). And, s/blender/bender/g ! Rewriting gecko is good, at least the non-GPL parts first. Then, comes code optiomization, then adding new features. Things like gcc(not egcs...its code was merged back into gcc!) and X would be debs. I like the idea of hacking dpkg to use rpms/tar.gz/sli packages(using things like librpm). If you've never noticed, rpm is just a nice little app that makes librpm do most of the work. This makes writing rpm apps easier than writing a new package manager. So, since I've never looked at the dpkg source, we could just add a switch and have it look for .rpm as the last 3 chars of the package name, and then run the InstallRPM(char[]) function we would write. --------------------------------------------- DO NOT REPLY TO THIS MESSAGE BY EMAIL! --------------------------------------------- You have chosen to receive messages from "mentalunix" by email. Reply to this message: http://clubs.yahoo.com/clubs/mentalunix/bbsfrp?action=r&tid=mentalunix&sid=1600006580&mid=349 Unsubscribe from the Club mailing list: http://clubs.yahoo.com/clubs/mentalunix/config/change_mb_list Return to "mentalunix": http://clubs.yahoo.com/clubs/mentalunix ---------------------------------------------- Not a member? Remove yourself from this list: http://edit.clubs.yahoo.com/config/unsubscribe_mb_list?.userID=munix_devr&.groupID=mentalunix&.groupType=&.code=IVyJGMuRxG |
|
From: The_Micea <las...@ad...> - 2000-12-13 02:32:36
|
Vi is the simplest and smallest text editor. I have a version of linux called LOAF, that runs off a floppy disk. A whole distrobution on a floppy. But it has vi. Vi is very small.100k maybe. its included with every distrobution ever made and we should have it.Unless we want to code our own text editor. I would be up for that, might be fun. ----- Original Message ----- From: "Mentalspice" <clu...@ya...> To: <men...@li...> Sent: Monday, December 11, 2000 2:46 PM Subject: [Mentalunix-developers] Re: MentalUnix, YoZilla, and other stuff [Yahoo! Clubs: mentalunix] > No debs. No compiler. Just a small, nice base system. The base system only needs the basics - a package manager, a simple text editor(no emacs or vi), and the file utilities. That is it. It should fit in under 30MB, and be stored as a large .bin file. The installer would mount it to a loop device, and copy its contents to the drive the user chooses(since it will be mounted, it will just copy / to /target, which is the target drive...just like in slackware). And, s/blender/bender/g ! Rewriting gecko is good, at least the non-GPL parts first. Then, comes code optiomization, then adding new features. > > Things like gcc(not egcs...its code was merged back into gcc!) and X would be debs. I like the idea of hacking dpkg to use rpms/tar.gz/sli packages(using things like librpm). If you've never noticed, rpm is just a nice little app that makes librpm do most of the work. This makes writing rpm apps easier than writing a new package manager. So, since I've never looked at the dpkg source, we could just add a switch and have it look for .rpm as the last 3 chars of the package name, and then run the InstallRPM(char[]) function we would write. > > --------------------------------------------- > DO NOT REPLY TO THIS MESSAGE BY EMAIL! > --------------------------------------------- > You have chosen to receive messages from "mentalunix" by email. > > Reply to this message: > http://clubs.yahoo.com/clubs/mentalunix/bbsfrp?action=r&tid=mentalunix&sid=1 600006580&mid=349 > > Unsubscribe from the Club mailing list: > http://clubs.yahoo.com/clubs/mentalunix/config/change_mb_list > > Return to "mentalunix": > http://clubs.yahoo.com/clubs/mentalunix > ---------------------------------------------- > > Not a member? Remove yourself from this list: > http://edit.clubs.yahoo.com/config/unsubscribe_mb_list?.userID=munix_devr&.g roupID=mentalunix&.groupType=&.code=IVyJGMuRxG > > _______________________________________________ > Mentalunix-developers mailing list > Men...@li... > http://lists.sourceforge.net/mailman/listinfo/mentalunix-developers |
|
From: CyDust <clu...@ya...> - 2000-12-12 06:16:07
|
ok, I tried installing debian from mounted hard drive, but i cant get it to find my modem, so I'm gonna wait for my potato cd's (should be here thursday or friday). I was using suse, but got really tired of those damn rpms. I've got a couple of revised ideas. First of all. I think initially, to get our this project moving, we might want to consider loosely basing the distribution on debian. The extreme robustness of debian is awesome, and makes it an excelent choice. The install is awesome, and we can use that as a start. The first installation program should be something based on something like loadlin, so it can be installed via windows, and just create a simple batch file for the loadlin starting a very simple install program, same thing can be done to create a bootable cd or floppy images. The install should be composed of a simple interface that will load cfdisk (or fdisk) to partition the hard drive, then the option to initalize the swap partition (or to skip that step) then initialize the linux partitions (or at least mount them), then finally install, configure, reboot. Since the first couple releases will be bare and highly useless for the most part (possibly highly unstable as well), we might want to consider single user install. (basically only root) if we do that, however, it will need to be changed before we stabalize the distro. So, to save space and whatnot, lets say single-user until 0.0.9 by 0.1 we want a little more usability on install. 0.2 or 0.3 release will be around 50mb and should include fileutils, and egcs (or gcc which ever it is) as well as gmake. 0.4 will be semi usable, and use apt (or something like that) for pkg installation. 0.5 will have a straight console gui install similar to debians. from 0.6 up to 0.9 the main goal should be packages and installation. the base should be frozen by 0.5 0.1 will have two versions, a cd version and a dl version. the cd will have everything needed include all the basic packages. the dl version will be the base, the drivers, kernel, a dialer, and a package installer, as well as a compiler and stuff. we should not consern ourselves with anything but the basics until 0.5, from 0.0.1 upto 0.4 we need to work on the filesystem pattern, the basic fs hierarchy, the kernel configuration and the overall stability and usability. 0.5 and up we start including user packages. Depending on what we do for package management, dpkg, apt, and apt-get would be perfect. or we modify dpkg into mkg, and use that as the base. If we go on my idea of being able to convert debs and rpms into mpkgs, then it will require some severe hacking, and require a lot of work, but if we are successful, we might be able to create a competeter to debian in terms of its overall excellence. --------------------------------------------- DO NOT REPLY TO THIS MESSAGE BY EMAIL! --------------------------------------------- You have chosen to receive messages from "mentalunix" by email. Reply to this message: http://clubs.yahoo.com/clubs/mentalunix/bbsfrp?action=r&tid=mentalunix&sid=1600006580&mid=351 Unsubscribe from the Club mailing list: http://clubs.yahoo.com/clubs/mentalunix/config/change_mb_list Return to "mentalunix": http://clubs.yahoo.com/clubs/mentalunix ---------------------------------------------- Not a member? Remove yourself from this list: http://edit.clubs.yahoo.com/config/unsubscribe_mb_list?.userID=munix_devr&.groupID=mentalunix&.groupType=&.code=IVyJGMuRxG |
|
From: CyDust <clu...@ya...> - 2000-12-12 06:17:03
|
previous post continued: Ok, gecko/bender/blender stuff. Name isnt important, although we do need a code name, any ideas? well, first off, ive heard netscape 6 is great under windows, but still slow under linux, and it also doesnt contain all the standards compliance that it needs. Therefore, gpl parts need to be optimized, closed stuff needs to be hacked apart, and converted into optimized, good clean linux code. But we should also make the code portable, so it can be used on winblows, mac os, etc, etc. for yozilla: 0.0.1 -- backend started, basic engine started 0.0.x upto 0.3 -- backend coded to be allpurpose html/fs viewer/browser 0.4 ^ 0.6 -- basic interface modules made, and gecko hack started by 1.0 -- basic, stable backend and interface. basic gecko by 1.5 -- standard gecko engine, jvm finished (stable) Basically, yozilla will be lynx until about 1.5 or 2.0, meaning, it aint gonna support much web wise, but my goal is to create a robust backend that can also be used to create a highly efficient file manager or to be used in other web browser products. all parts will be released via gpl. if you have any ideas (all of you) please post your ideas. Also if you want to help code, or know some one that does or anything, please let me know what you want to do or are at least capable of undertaking. Yozilla is a much more difficult then mentalunix, and help is much needed. --------------------------------------------- DO NOT REPLY TO THIS MESSAGE BY EMAIL! --------------------------------------------- You have chosen to receive messages from "mentalunix" by email. Reply to this message: http://clubs.yahoo.com/clubs/mentalunix/bbsfrp?action=r&tid=mentalunix&sid=1600006580&mid=352 Unsubscribe from the Club mailing list: http://clubs.yahoo.com/clubs/mentalunix/config/change_mb_list Return to "mentalunix": http://clubs.yahoo.com/clubs/mentalunix ---------------------------------------------- Not a member? Remove yourself from this list: http://edit.clubs.yahoo.com/config/unsubscribe_mb_list?.userID=munix_devr&.groupID=mentalunix&.groupType=&.code=IVyJGMuRxG |
|
From: The_Micea <las...@ad...> - 2000-12-13 02:39:53
|
Im unsure of wether my email is working, let me know if this posts, If I dont here from anyone Ill post on the website. ----- Original Message ----- From: "CyDust" <clu...@ya...> To: <men...@li...> Sent: Tuesday, December 12, 2000 1:17 AM Subject: [Mentalunix-developers] Re: MentalUnix, YoZilla, and other stuff [Yahoo! Clubs: mentalunix] > previous post continued: > Ok, gecko/bender/blender stuff. Name isnt important, although we do need a code name, any ideas? well, first off, ive heard netscape 6 is great under windows, but still slow under linux, and it also doesnt contain all the standards compliance that it needs. Therefore, gpl parts need to be optimized, closed stuff needs to be hacked apart, and converted into optimized, good clean linux code. But we should also make the code portable, so it can be used on winblows, mac os, etc, etc. for yozilla: > 0.0.1 -- backend started, basic engine started > 0.0.x upto 0.3 -- backend coded to be allpurpose html/fs viewer/browser > 0.4 ^ 0.6 -- basic interface modules made, and gecko hack started > by 1.0 -- basic, stable backend and interface. basic gecko > by 1.5 -- standard gecko engine, jvm finished (stable) > > Basically, yozilla will be lynx until about 1.5 or 2.0, meaning, it aint gonna support much web wise, but my goal is to create a robust backend that can also be used to create a highly efficient file manager or to be used in other web browser products. all parts will be released via gpl. if you have any ideas (all of you) please post your ideas. Also if you want to help code, or know some one that does or anything, please let me know what you want to do or are at least capable of undertaking. Yozilla is a much more difficult then mentalunix, and help is much needed. > > --------------------------------------------- > DO NOT REPLY TO THIS MESSAGE BY EMAIL! > --------------------------------------------- > You have chosen to receive messages from "mentalunix" by email. > > Reply to this message: > http://clubs.yahoo.com/clubs/mentalunix/bbsfrp?action=r&tid=mentalunix&sid=1 600006580&mid=352 > > Unsubscribe from the Club mailing list: > http://clubs.yahoo.com/clubs/mentalunix/config/change_mb_list > > Return to "mentalunix": > http://clubs.yahoo.com/clubs/mentalunix > ---------------------------------------------- > > Not a member? Remove yourself from this list: > http://edit.clubs.yahoo.com/config/unsubscribe_mb_list?.userID=munix_devr&.g roupID=mentalunix&.groupType=&.code=IVyJGMuRxG > > _______________________________________________ > Mentalunix-developers mailing list > Men...@li... > http://lists.sourceforge.net/mailman/listinfo/mentalunix-developers |
|
From: The_Micea <clu...@ya...> - 2000-12-13 02:49:41
|
I agree the first few releases should work mostly with struture and fs. we need something really radical to make us stand out. there are many linux OSs out there, and they are all drifting towards "userfreindliness" instead of inovation. even debian and slackware(the holy grail of hackers) are doing less inovating and more cosmetics. yes they have nifty new programs and installers, but they are basicly all the same. we need to do somethine really awesome that will make our OS first on the list of every geek, sysadmin, and power user who doest really care about the GNome over KDE controversy, or anything like that. most sysadmins do everything from the consol. --------------------------------------------- DO NOT REPLY TO THIS MESSAGE BY EMAIL! --------------------------------------------- You have chosen to receive messages from "mentalunix" by email. Reply to this message: http://clubs.yahoo.com/clubs/mentalunix/bbsfrp?action=r&tid=mentalunix&sid=1600006580&mid=354 Unsubscribe from the Club mailing list: http://clubs.yahoo.com/clubs/mentalunix/config/change_mb_list Return to "mentalunix": http://clubs.yahoo.com/clubs/mentalunix ---------------------------------------------- Not a member? Remove yourself from this list: http://edit.clubs.yahoo.com/config/unsubscribe_mb_list?.userID=munix_devr&.groupID=mentalunix&.groupType=&.code=IVyJGMuRxG |
|
From: Mentalspice <clu...@ya...> - 2000-12-13 15:04:52
|
I say, let's use XFce as the default desktop enviroment. We can bundle it with nautilus or gmc as the file manager, and aterm as the default term. None of the other distributions do this. Also, by making linuxconf actually work right, then people will also like it. Making the package manager able to use any kind of package is another step in the right direction...it allows linux to gain a universalness that other Operating Systems have. Also, stictly adhering to the LSB and FHS is a good idea. And, creating symlinks to all of a packages files in /opt/packages/pkg_name bin src doc etc.. If it modifies a file in /etc or /bin, then their may be problems...any ideas? I'm thinking they can be linked into main_bin and slash_etc or soemthing. --------------------------------------------- DO NOT REPLY TO THIS MESSAGE BY EMAIL! --------------------------------------------- You have chosen to receive messages from "mentalunix" by email. Reply to this message: http://clubs.yahoo.com/clubs/mentalunix/bbsfrp?action=r&tid=mentalunix&sid=1600006580&mid=355 Unsubscribe from the Club mailing list: http://clubs.yahoo.com/clubs/mentalunix/config/change_mb_list Return to "mentalunix": http://clubs.yahoo.com/clubs/mentalunix ---------------------------------------------- Not a member? Remove yourself from this list: http://edit.clubs.yahoo.com/config/unsubscribe_mb_list?.userID=munix_devr&.groupID=mentalunix&.groupType=&.code=IVyJGMuRxG |
|
From: CyDust <clu...@ya...> - 2000-12-14 06:26:54
|
sounds like a good idea. Been a long time since ive used linuxconf, so i need to check up on taht. asd for the package stuff, not a bad idea. maybe we should have at least a 4disk set (disc 1-base, 2-debs, 3-rpms, 4-mpkg) or (disc 1-base, 2-mpkg, 3-src, and then the user desides what kind of package system, if mpkg then disc two, if deb or rpm, then disc 3, and packages are compiled, installed as needed (or disc 3 will be binary tarballs) and then installed into the appropriate database. Now, if I read the last part correctly, i dont think we should have symlinks from /opt/packages/pkg_name to whatever, but instead /opt/packages/pkg_name will have a text file (lets say pkg_name-pkg_type-installed) which will list all of the files installed by that package, then maybe (pkg_name-pkg_type-modifies) that would be updated every time the package database is updated and that would list changed files and whatnot. --------------------------------------------- DO NOT REPLY TO THIS MESSAGE BY EMAIL! --------------------------------------------- You have chosen to receive messages from "mentalunix" by email. Reply to this message: http://clubs.yahoo.com/clubs/mentalunix/bbsfrp?action=r&tid=mentalunix&sid=1600006580&mid=356 Unsubscribe from the Club mailing list: http://clubs.yahoo.com/clubs/mentalunix/config/change_mb_list Return to "mentalunix": http://clubs.yahoo.com/clubs/mentalunix ---------------------------------------------- Not a member? Remove yourself from this list: http://edit.clubs.yahoo.com/config/unsubscribe_mb_list?.userID=munix_devr&.groupID=mentalunix&.groupType=&.code=IVyJGMuRxG |
×
Want the latest updates on software, tech news, and AI?
Get latest updates about software, tech news, and AI from SourceForge directly in your inbox once a month.
Thanks for helping keep SourceForge clean.
X