[linux-elitists] [doc@ssc.com: [Fwd: Plug 'n' Pray to Plug 'n' Play: What's it Going to Take?]]

[Doc Searls]

At 9:33 AM -0700 4/20/06, Greg KH wrote:

 > On Wed, Apr 19, 2006 at 02:47:48PM -0700, Don Marti wrote:
 >
 >>  What Needs to be Done for desktop (and especially laptop) Linux --
 >>  to work with whatever devices come along.
 >
 >
 > So you are looking for what to do for new devices?  Or for the 
devices
 > we currently have today?


Both.

 >
 >>  Or better yet, vice versa. As a user, I want any Nikon, Canon, 
Casio
 >>  or Olympus digital camera to work 10X better on any Linux laptop it
 >>  meets than it would on any Windows box (which today in too many
 >>  cases requires that the user run some install disk). Or (hey, why
 >>  not) better than any Mac running OS X as well.
 >>
 >>  Same with every printer, scanner, external HD, CF or SD card...
 >>  whatever.
 >>
 >>  What's it going to take? Also, what do people not know now that 
they
 >>  ought to know (such as, say, where it's best to run the driver).
 >
 >
 > Only 2 words, "vendor involvement".
 >
 > Yes, it's that simple.  Get the hardware vendors to care that their
 > devices work on Linux and they will work.  Look at how well 
things work
 > today for network and scsi devices on Linux.  It is that way 
because the
 > vendor companies care about Linux and work to make sure that their
 > drivers are in the main kernel tree and are stable and fast.
 >
 > Oh, and if a vendor doesn't know how to get involved in Linux
 > development, point them at the all-inclusive Linux Kernel HOWTO. 
  It's
 > in the main kernel source tree at Documentation/HOWTO, or on the 
web in
 > a variety of places:
 >     http://sosdg.org/~coywolf/lxr/source/Documentation/HOWTO


Good.

 >
 >>  My goal is to convince the audience that any device should relate
 >>  better with Linux than with any proprietary OS. I want to give them
 >>  the reasons why.
 >
 >
 > What is the audience?  Vendors or users?


Both. Plus developers. Variety of people interested in 
Desktop/Laptop Linux.

 >  > I want to convince them that next year's (or whenever's) 
generic Linux
 >
 >>  laptop will be a better device companion than anything from 
Microsoft
 >>  (via its captive HW OEMs) or Apple.  Because that is the Inevitable
 >>  Nature of Things. (That's what I believe, but I need facts and
 >>  informed insights to back that up.)
 >
 >
 > Take a look at how damm well the latest Linux distros work on laptops
 > today.  For example, I just tried out SuSE 10.1 (beta) on a 
laptop and
 > was amazed.  Plug in a printer, and a little dialog box pops up 
saying
 > it found the printer and loaded the driver and do I want to print 
a test
 > page?
 >
 > Plug in a camera, and f-spot pops up and asks if I want to 
download the
 > pictures from it onto the computer, and then do I want to export the
 > pictures to the web to flikr or any of a variety of different web 
sites.
 >
 > Plug in a usb CompactFlash reader and the same thing as above happens
 > (as the flash came from a camera.)
 >
 > Unplug the flash card, and it is instantly recognised and nothing 
"bad"
 > happens as I didn't unmount the device.
 >
 > Plug in a different flash card, and it's instantly mounted and 
offered
 > up in a file browser (no pictures on this one.)
 >
 > Plug in an ipod and Banshee pops up and I can drag and drop a few mp3
 > files into it and have them copied over to it properly.
 >
 > Plug in a network ExpressCard device and have it automatically 
connect
 > to the network properly (Windows requires you to do this with the 
power
 > off, and at least 2 reboots.)
 >
 > Scan for all availble wireless networks in the area and connect 
to the
 > one I want.
 >
 > And I could go on (in fact, Nat will, in the presentation after 
yours.)
 >
 > And this isn't just SuSE.  Ubuntu and Fedora are also this good 
today.
 > I really don't think that people realize how good things have 
gotten in
 > the past 6 months or so.  I sure hadn't, until someone pointed it 
out to
 > me (I only use X for a zillion terminal windows, and userspace 
only as a
 > test load on the kernel...)
 >
 > If you notice, at no time above did I mention having to load a driver
 > disk, find a driver, manually configure a device, etc.  It all, "just
 > works" today.
 >
 > Now sure, there are some types of devices that Linux does not support
 > very well at times (some odd wireless devices are one of the most
 > visible ones.)  But there are a whole range of devices that other
 > operating systems (Windows and OS X) don't support right now.  Are
 > people stating that they suck because they don't support their 
devices?


Since I'm not that technical, and tend to be a high-level guy (as 
with "Linux is a species"...

http://www.linuxjournal.com/article/8664 )

... I think I'd like to talk about changes in the market ecology 
caused, or invited, by Linux laptops and desktops that are ready to 
handle any device that comes along.

What happens when users, developers and vendor engineers collaborate 
without interference from marketing?

Can we relieve vendors of the need to make platform-specific 
software to run their devices?

Of particular interest to me right now is GPS. Garmin makes some 
great little hand-held GPSes that are useful and fun. I can go 
geocache-hunting with my kid, navigate a topo map, thread my way 
through a city, or follow the route of the airplane I'm riding 
through the sky (not sure if that's legal, but I do it anyway, from 
my window seat). But to get the maps I use into the GPS I need to 
use Windows, because Garmin only makes Windows software for 
installing maps. They also have a business in selling those maps. 
This is all a grrr for me, and I think There Must Be A Better Way.

Can we relieve Garmin of that development burden (or cost) of doing 
Windows-only software development? Can we describe to them a better 
way to relate to the GPS market? For example, can they use Web 
services to put maps in GPS devices, once the laptop or desktop 
detects the device hooked up by USB? I don't know the answers. I 
just know the Old Way is goofy -- and there may be a New Way that 
takes advantage of the Linux Way of driving devices.

 > Anyway, none of this probably will help you out in your talk, as 
you are
 > trying to convince people that we still need to do a lot of work.  I
 > guess my point is that the work is already done, just no one 
seems to be
 > aware of it :)


I'll change the title. That's no problem. Let's look at this as a 
"now what?" topic.

Thanks, everybody.

Doc