Just have to Vent

[KernelPanic]

quote:Originally posted by Solo:
I certainly dont have a computer science degree (I am still in high-school) but I believe if I took some time to learn how the drivers worked, how the interfaces worked etc I could write them. Not to say that the first ones I wrote would be good    . I do understand what you are saying though, all with the not many people able to do drivers. But maybe more in general we could have a network of general coders. Let's say Microsoft/Apple comes up with some good feature or in M$'$ ca$e buy the $oftware that ha$ the cool feature. We could direct the community to help the important projects implement the features quickly (one of the big reasons features take awhile is because most of the developers that contribute are busy or not interested. If we had a large base of developers willing to code more generally on projects, we could get a killer feature implemented in KDE/GNOME or Redhat/SUSE/Mandrake/Slack etc very quickly and synchronously.

Ok let's pretend people are just getting the XP leak builds now (travel back in time people). There are reports of a cool multi-display system for multiple users. A proper group of people are recruited from the developer network to research and implement a similiar feature in XFree86. Each member in the group is assigned a position, such as Research leader, Engineering (implementation) leader, and coders. The positions would be based on the skill and amount of free time the developer has. The Engineering leader instructs the coders to wait for further instructions, and gives them the email address of the Research leader. Meanwhile the research leader is looking through the code and looking for the ways it could be done, designs an X extension, etc, and brings his/her ideas to the Engineering leader. The engineering leader creates a report on the way it will be implemented, and takes feedback from the coders. The coders begin implementing the changes (perhaps on a different CVS server for the dev network). Once the code is stable, the engineering leader performs the proper steps to submit the code to XFree86.

You do realise this system already exists?
Perhaps not word-to-word as you have described, but these open-source projects don't just fall together through some form a hacker black magic you know  
The different project have different command organisation, for example Xfree has an elected board, the kernel team has several hot-shots under Linus, Kde and Gnome have their own things too.

[SAJChurchey]

So,

What should we do in the light of all this?  We can't just stand to the side and let Linux be taken down by M$.  We'd be just as guilty as them for our inaction.

 

quote:
  When they came to take the jews, I said nothing.  When they came to take the catholics, I said nothing.  When they came to take the homosexuals, I said nothing.  When they came to take Linux, I said nothing.  When they came for me, nobody was there.

Or something like that, the point is they're Nazis or some other form of fascist dictatorship waiting to happen in some form or another.

[realist]

Open Source software is the future. There will be a lot of gnashing of teeth before then but who cares? Sit back and spectate. Oh and as we all know, Unix Labs tried this exact same lawsuit already against Berkely System Design and they lost. Microsoft's involvement will change nothing unless it is to pay off a corrupt judge and jury.

[KernelPanic]

It was Martin Niemoller - he was a German pastor or something.
I love that quote, although SCo can only harm the Corpaorate view of Linux at the moment.
They can't even do that well as their case is going to get shot out of water and they are a laughing stock. A little like Claire Short...
In the words of Atticus Finch 'It's not time to worry yet'

[SAJChurchey]

I hope you are right, the both of you.  I agree Open Source is the future, but until M$ is gone, there will always be a threat, and my Tux, aren't we the historian and book-full-of-quotes

[solo]

Ahh I feel another long post coming on  

I am a member of OpenOpen and have been for quite awhile thank you.

It is absolutely NOT the time to "sit back" and watch. We *must* *never* let Linux lose it's track. Yes, we will always have Linux, but it is so important that Linux makes it to desktop and the server market. Not only because it is a superior system, but to triumph over Microsoft's illegal and immoral business practices.

We need an open source development organization. One that can command MASSES of freelance-type developers to work on certain open source projects. I know all open source projects have a system like that but we need a system that would equal the 'management' part of SomeSoft that would look over Linux/KDE/GNOME and other parts of the typical Linux-related open source projects and offer large amounts of help to improve that OSS product to better Linux. It would take time, but the longer we wait, the more time Microsoft has to pay developers to work full time to get something half-assed, and the more time they have to rub it in our faces because we don't have it yet. If we can improve on the proprietary model and keep OSS development's current form (volunteer development) while putting out quality source code, we could beat Microsoft much quicker.

Perhaps this idea could materialize into something like a minutemen militia. Volunteer coders sign up, and they outline their skills, then when they are needed they are called upon to code.

Do I have any supporters? Even if this doesn't work, we have to try. We'll never know if it will work until we've tried.

[SAJChurchey]

I have been and will always support open source in any way I can.  Although I do not have technical skills or resources for such a massive endeavor at this time.  

Open source if full of volunteers who are also faced w/ the same dilema of time and other resources.  Are you talking about centralizing open source into one large body, where the leaders dictate the speed of development.  Such a thing like that would not work for the simple reason that these are volunteers doing this on their own time and of their own volition (if I spelled that correctly).

Centralizing all of the projects would be too chaotic b/c there are so many of them.  That is what GNU is here for anyway.  Right now, I advocate and convert/educate people when I get the chance.  I do back open source 110%, and my actions reflect such.

More management is really not necessary in most areas.  Kernel 2.6 is expected out soon, and KDE and GNOME are already gr8 desktop environments where the features even surpass that of Windows (Virtual Desktops, Tabbed Filemanager, stability, etc.).

The real work needs to be done in advocacy and on the legal front.  The developers are not the problem it's the uneducated public blinded by M$, and perhaps the lack of proprietary companies (i.e. Macromedia and gaming companies) not porting key software to Linux.  The weakness lies there.

Hardware is another significant issue.  People do not want to have to manually set up their internet connections and configure hardware.  Linux will need to be plug and play and have an easier driver installation system, and I believe such tools could be easy to make in Python or Java and shellscripting.

Maybe we should get a project of our own together aimed at making these hardware config tools and getting them to distros.  It may seem small and insignificant but it is the first significant step in achieving the goals that you are outlining.

[solo]

I definitely do not want to create a dictatorship system that forces all OSS projects to comply. My idea is to offer a place where volunteer developers can sign up, and when features are implemented in competing operating systems, we can mobilize troops to work on getting it in Linux. So when someone says "Does Linux have feature xxxx?" we can tell them honestly that there's a team working on it, and we can have the feature in quickly. A lot of times developers in open source projects work on features they would benefit from (or think many people could benefit from).

If the OSS project is already working on it and is fully-staffed then we wouldn't have to deal with it. If they are understaffed we could contribute a few willing developers.

When a developer signs up, he/she would create a profile of the skills they have, how much time they would be willing to devote, etc.

The users would log on to the organization website and see that their are two feature requests that they have been mobilized on. They can accept or decline any of them, but perhaps there would need to be a minimum on contribution. It would be like sponsored OSS developers from commercial companies except without the money   .
If the developer accepts the feature request, he/she will be added to the workgroup and will get a notification email about who else is in the group. A discussion-starting email (defined by the creator of the request) would be sent to all participating members and discussion would ensue. The request would be subject to review every two weeks or so by two or three independent developers, who would check if planning/discussion/development is happening smoothly and the request is being fulfilled (a strict system of deciding this should be posted to the website, any breaches of the system could be contended by any developers registered in the organization). OSS projects would delegate one or more feature request managers. Anyone could create feature requests, but they must be reviewed by the project's request managers. The project would decide how to elect/appoint request managers. If the projects decline to appoint someone a request manager when the project started, I'm sure we could find people from the project's community who would take care of it until the project decides to officially delegate one/two/three etc.

Collaboration could be done on Freenode/temporary feature request mailing lists.

If this were to catch on, and work correctly, maybe some commercial companies would tack on some moola for developers of feature workgroups.

The point isnt to make people work on projects, but to encourage them and inform them about what needs to be done. A lot of people don't keep up on all of the projects they could contribute to, having a centralized organization that posts feature requests and organizes developm

[solo]

Comment on hardware config stuff: better driver system yes, we need a versionless driver system so drivers can be closed-source (i would MUCH rather have open source ones but closed source ones become impossible because it needs to be recompiled for each kernel).

hardware detection is doing quite well, I've installed redhat 9 on two computers, one an HP and one a generic bargain PC and it worked great on both. Only thing I had to set up was the printer: by clicking Printers, Add->HP 712c lol.

I never touched the shell   .

I want to see an in-kernel device detection system. That would standardize device detection and perhaps developers from other device detection projects could contribute.

[SAJChurchey]

kudzu is an adequate hardware detection system for Linux, and that's the reason that RH is so good about hardware changes.  It would be better if the LSB made it a standard to package kudzu with LSB compliant distros.  

We have the various tools out there.  The Linux community just needs to pick a good one and standardize on it.

In the meantime, I've decided to try my hand at the Software Installation Wizard.  I'm probably going to write it in Java and will probably need to use some shell scripting.  Most GUIs like these are written in Python, but i see no problem in using Java.

[ June 07, 2003: Message edited by: SAJChurchey ]

[mushrooomprince]

Yea i've had this epiphany before and i hate microsoft to.


 

[billy_gates]

Sry.  Didn't read all of the posts.

You do have to hand it to MS and Bill Gates though.  They came from the (near bottom) and got to the top.  Now they can hold the top and make even more money.  They have smart business strategies.  They know what they are doing and they do it well.