The Vision

[golgoj4]

Hello again kids, its time for a fun discussion!

Lets get this out of the way from the start: There is no right answer. So there is no reason to bash other peoples views. Since ive become part of the LinuxMCE project, both as a user and a sort of jr developer, Ive noticed marked difference from what used to be my day job as an construction operations specialist (read: guy who cracks whip to make sure schedules are met). And Its been quite the learning process, both in learning linux, and the associated culture in the F / OSS world. So im a bit of a newbie around here.

Thom asked an interesting question the other day in IRC:
'If you knew it would take you 50 years to do what a commercial entity could do in 10, would you still do it?'

Everyone has their own answer to this, as for the most part, we are all driven by some selfish aspect that drove us to try and work on something like tagging in the 1st place.

But, it led to an interesting discussion on what is our Vision. It was pointed out that uniting under a singular vision generally has positive results, and my experience in life tends to agree. But, but the nature of open source, design is by committee. Which isn't a bad thing, and shouldn't be taken as a negative. But what im getting at here is that we as developers should come to an agreement on what our vision is, and attack it with laser-like focus.  

I feel like that will lead us to a clearer path of where we all want the project to go, what we want it to be, and how we can get there without wanting to murder each other

So, in conclusion, NO im not trying to manage stuff. But i do think that having something we can point to and say 'THAT. Thats what we want to be' and go at it until we cant. So, let the discussion begin! I only ask that we keep it civil and not dump on others views. Im guilty of that myself sometimes, admittedly, but im really trying hard not to be. For the good of the project, I feel that its a good idea. But, also a little selfish as part of what i do want to contribute is not only code, but something in the way of a banner that allows us to draw more developers and help us fix/ evolve faster.

Btw: My vision of linuxmce is the 0704 video i saw combined with a little hal-9000 level ai. Plus some kinda anti-monolith patch. Jus sayin.

Also, i plan to make some dumb shiny graphics that can hopefully be included on the home page or somewhere that make it easier for people with a specific skillset to contribute.

Good DaY!
-golgoj4

[Marie.O]

I've said it before, I say it again.

LinuxMCE promised set of features needs to be fulfilled. We do not need new features, we need the existing feature set bulletproof in place, wrapped up by a simple install process.

That is my foremost goal for LinuxMCE for the foreseeable future

[bongowongo]

I've said it before, I say it again.

LinuxMCE promised set of features needs to be fulfilled. We do not need new features, we need the existing feature set bulletproof in place, wrapped up by a simple install process.

That is my foremost goal for LinuxMCE for the foreseeable future

I agree.
 
If you built it, they will come.

[seth]

 
Ok. Here is my take. And while I have not been contributing much lately, I have been training at work on using eclipse. And starting in january, Android.

When I came into this project, from knoppmyth, which for a media client was rock solid. Ncurses installer and all, I saw the video. I attempted several times to install 7.04 with the 2 cd method, both x86 and x86_64. It was painful, but after it was over, it ran extremely well.

After that there was 7.10. Which by far was my favorite. No long download installer scripts. You popped the DVD in, booted up, chose your disk, and it untarred a complete system, while you went for coffee/beer. This was magnificent. Now with 8.10 in final, and 10.04 being used more in it's stead, with the promise of new hardware support, and a newer "supported" version of Kubuntu, I find myself still longing for that, choose your drive, walk away, come back and reboot.

I use satellite internet because I live way up in the mountians, and there are no cable/dsl/fios options for me. I have to wait to install LinuxMCE because of the HUGE amounts of downloads after a DVD install. A DVD install. I have not yet attempted 10.04 yet. My data usage has dropped below 60% of allowed, so I will try this week at some point. But 8.10 is final. It requires a DVD to install it, you have to do Kubuntu first, and then spend a lot of time downloading the rest of it. On a DVD install.

I guess my question is. Is there anyone who has a good working install, that can create this tarball method of new installation, with the choice of which drive to install it on, like the video. To save the need for all of this additional time spent downloading?

I wish I could say my install could be used as such. But my RC1 upgraded has issues. I am going to do a 10.04 install now, as that installing 810 final, from DVD, would push my download limit too high, and rather than take that time, I should use my bandwidth for 10.04.

My situation is unique. I have limited broadband downloading, limited speed.

So I guess my vision for LinuxMCE is a combination of golgoj4's and posde's.

A final product as rock solid as the original video that kidnapped me, and the ease of installation of the 7.10 DVD. i.e. choose drive, wait for tar to complete, boot into working system.

But I am no one.    I have contributed little other than wiki info, and testing. I hope to amplify my voice in the coming months, after my training is complete.

Best Regards to All,

Seth

[Murdock]

I agree with posde in terms of vision with the exception of the comment on new features (specifically Thom's game player which is sweet  ), though I personally feel that new features keep the project alive and keep contributions coming. While I agree that older components need to be maintained and kept in a working state it doesn't seem to hold developer interest.

Hi Seth - posde is working on the new dvd installer for 1004 and his goal is to make it like the one used in 710. I'm working on hardening the internet install and MD installation process. All issues I've encountered and am aware of have a trac ticket, if you have any issues with the 1004 installation process please let me know or better would be to log a ticket!

[davegravy]

It all comes down to being accessible to SWMBO. Though I can't contribute much code, I do have some insight into what makes "them" tick. I think it comes in this order:

1) Bug-free, consistent operation
2) Snappy (responsive), intuitive, efficiently organized, and attractive UI
3) Ease of adding new media and finding/organizing media

 

[Marie.O]

davegravy,

I took the liberty and split your stuff a bit:

1) Bug free
2) consistent operation
3) intuitive
4) organized
5) snappy

ease of adding media and finding media is, imho, part of organized.

And, yeah, well, the part about the attractive UI was never brought up by my SWMBO. She is still unhappy about 1 and 2

[mkbrown69]

Just some thoughts to share with respect to golgoj4's post...

What is LinuxMCE now, and what is it meant to become?  What doesn't it do that it should, what does it do that it shouldn't, and what will it never do?  That's where Vision comes into play... Vision provides scope and direction, which are really important when developers (and developer's time) are limited.  It also affects decisions about architecture and infrastructure, frameworks and integration points...

For example, the requests forum is a real smorgasbord of people's wants and desires for LinuxMCE.  What makes it in, and what doesn't?  The usual answer is, those that write the code can put it in; but does it always make sense to put some something in, just because you can?  Many of the requests are for portal and app server functionality, like the stuff Amahi does.  Is that a space LinuxMCE wants to or should be playing in?  If so, what framework(s) get used/re-used to provide that functionality to maximize on the features delivered vs time invested by developers.  Another example would be Dianemo using MythTV and saying that it provides everything that VDR could do.  Does that mean VDR support is a duplication of effort?  Should it be deprecated/removed for 10.04?  (Note: I'm not saying that, but with a common vision and direction, these kinds of things have to be examined and decided upon).  Vision provides scope and direction, which influence quality, time and effort.

I bring this up only because one of my roles in my day job is to encourage the DBA's and app admins I support to consider the full life-cycle of the services they support, both for the here and now and for the end state, because they do effect one another.  The end state (the vision) effects the here and now, and the attitudes towards the here and now (scope and direction, and consequently time, effort, and quality) effects the end state.  I do this as someone who inherits the "lost turds"; the services that were rushed out to meet the immediate needs of the here and now, with no consideration for the long-term life-cycle needs of that vital service.  Specifically, how to keep it alive and well, and evolve it as needed.

LinuxMCE is different from most projects or applications, and not just because of it's complexity.  When fully integrated into a home environment, it becomes a daily use tool, one that people can literally grow up with; in some cases, children may become teenagers or adults, not remembering a time before LinuxMCE was a part of their lives.  That's why Vision is so important... As a collection of OpenSource components, LinuxMCE's foundations have a life of their own, which will drive the evolution of LinuxMCE to a large extent.  As the LinuxMCE project itself grows and evolves, more people will come about to try to help out in some way.  With a shared collective vision and direction, it becomes easier to rally the troops and make strategic choices; without, it may be like herding cats...

I'm going to duck back out now and continue to examine the plumbing in 10.04... still got a lot to figure out before I can hope to make a contribution, but I hope to make a dent in posde's points 1 & 2 eventually...

/Mike

[Marie.O]

A note on VDR and MythTV, and why we have both: There some things, where VDR excels, and there are other things where MythTV excels. And there is no easy way to have the stuff where MythTV excels in VDR, or the other way round.

[bongowongo]

And, yeah, well, the part about the attractive UI was never brought up by my SWMBO.

That explains a lot why she chose you.

[davegravy]

That explains a lot why she chose you.

OH SNAP!

[Marie.O]

That explains a lot why she chose you.

Yep, one attractive thing in the house hold is enough for her to look at. No need to have it on the computer display as well.

[golgoj4]

I believe i said be nice bongo!

[davegravy]

After talking with Posde (http://forum.linuxmce.org/index.php/topic,12065.0.html) I need to add one other bullet point to my list:

Affordable

Dunno where this fits in terms of priority, but probably somewhere near the top. SWMBO wants to spend more money on taking vacations (spending quality time together) and less money on electricity bills due to idling MDs and such