approaching vendors with a consolidation of effort

[Scott Serr]

Since the SOHO hacked AP for $30-50 is just too good of a deal... what
if [I know I'm a dreamer]

There was a petition. Not a dumb petition to try to convince one of the
companies to yield to our desire, but a petition that quantifies the
interest and lets these companies see the numbers.

Something like this:

1. Setup a web sign up sheet for 2 groups: people that would hack and
people that would use
2. Try to get as many of the hackAP projects members on there as possible.
3. Try to get hackedAP users to sign up, easy if you could get on
slashdot... right? ;)
4. Approach these companies like Dlink, Netgear, Linksys, USR, SMC, etc
5. Basically say, HERE is the demand. "We've identified you out of 9
potential corporate sponsors." Sponsorship costs no money. It only
requires technical information so our Linux engineers can utilize your
hardware. And potentially a small amount of email correspondence with
one of your hardware engineers.

The more I think of this... the more I like it. A subtle factor is
they are 1 of 9, if they decline.. they are withdrawing from a market
that their competitors are just being invited into.

I can setup a website on a big pipe, if you guys think this might fly.
[I don't think I could survive a slashdoting though]

-Scott

[Roy]

Idea is brave, but I dont think it will work
There are already sime ap's with linux
also the big problem is wlan card chipset makers
who ususly keep their driver code top secret,
as I know only admtek is more or less open.
(maibe someone at least know the reason why driver code is top secret?
when normal ethernet drivers are open)
I suppose they are afraid that we are going to use illegal channels or maybe
they stole them from competitors?
So this will be big problem

also some manufacturers are so dumb that they even dont know what are they
producing
Acorp is example (I faced that personaly)
I suppose uniden CAN be similar but they produce wery cheap.

others may give some information but dumb bosses of these companies will
newer agree
since then they will need to show everyone how bad works their staff,
as I see there is no product without bugs what is normal if company is
fixing them
but why bother fixing when we can put new sticker on the same box and
advetise new product with more features that are not worth mentioning in
comparison to linux posibilities.
I think it is quite good reason since only realy good hardware have official
linux support.

I think the good way should be continue this project and offer for all
companies free software design for any ap
I suppose this can interest few, but again directors of most companies wont
like this idea since they NEED someone to blame in case of bug (In the one
of my jobs director was asking who is responsible even in case of
earthquake), who you are going to blame and fire fom job in linux comunity?
Better hire few programes who will spend more time licking their boss feet
than developing software.
But basicaly this can work if we offer to create alternative firmware for
the most cheap hardware they have which will offer more features than any
other for free. but we need to find the original creator since most cheapest
hardware is clones made by license. thus manufacturer either dont know about
their product anyhing either signed nondiclosure agreement.
Or we can offer for someone who makes CPUs to create good reference project,
nobody will be smart enough to change something anyway.

In any way this would work 10 times better if someone can take
responsibility for sucess of the project

----- Original Message -----
From: "Scott Serr" <***@theserrs.net>
To: "Linux Access Point Development" <linuxap-***@linuxAP.ksmith.com>
Sent: Monday, February 02, 2004 2:28 AM
Subject: [LinuxAP-dev] approaching vendors with a consolidation of effort

Post by Scott Serr
Since the SOHO hacked AP for $30-50 is just too good of a
deal... what
if [I know I'm a dreamer]
There was a petition. Not a dumb petition to try to convince one of the
companies to yield to our desire, but a petition that quantifies the
interest and lets these companies see the numbers.
1. Setup a web sign up sheet for 2 groups: people that would hack and
people that would use
2. Try to get as many of the hackAP projects members on there as possible.
3. Try to get hackedAP users to sign up, easy if you could get on
slashdot... right? ;)
4. Approach these companies like Dlink, Netgear, Linksys, USR, SMC, etc
5. Basically say, HERE is the demand. '"'We've identified you out of 9
potential corporate sponsors.'"' Sponsorship costs no money. It only
requires technical information so our Linux engineers can utilize your
hardware. And potentially a small amount of email correspondence with
one of your hardware engineers.
The more I think of this... the more I like it. A subtle factor is
they are 1 of 9, if they decline.. they are withdrawing from a market
that their competitors are just being invited into.
I can setup a website on a big pipe, if you guys think this might fly.
[I don't think I could survive a slashdoting though]
-Scott
_______________________________________________
LinuxAP-dev mailing list
http://linuxAP.ksmith.com/mailman/listinfo/linuxap-dev

[John T]

Post by Roy
Idea is brave, but I dont think it will work
There are already sime ap's with linux
also the big problem is wlan card chipset makers
who ususly keep their driver code top secret,

Sure they keep their drivers top secret, but with the gaining popularity of
Linux, more and more wlan card manufacturers are going to make linux drivers
available... and if we have an open source linux distro on an x86 based AP,
changing the wlan radio in the unit is simply a matter of installing the
manufacturer supplied driver I would think. Yeah, you may not be able to
tweak the settings on the card, but surely you'd be able to still have linux
handle the router functions, no matter what wlan card is in the unit, so
long as it has linux drivers from the manufacturer.

[John T]

Post by Scott Serr
Since the SOHO hacked AP for $30-50 is just too good of a deal... what
if [I know I'm a dreamer]

That is truly the biggest problem... one cannot produce an x86 based AP in
that price range... they come in around $100 for the unit at best. a RISC
based AP is much cheaper to produce, and puts it into that $50 sweet spot on
price.

[Roy]

Post by Scott Serr

Post by Scott Serr
Since the SOHO hacked AP for $30-50 is just too good of a

deal... what

Post by Scott Serr
if [I know I'm a dreamer]

That is truly the biggest problem... one cannot produce an x86 based AP in
that price range... they come in around $100 for the unit at best. a RISC
based AP is much cheaper to produce, and puts it into that $50 sweet spot on
price.

We dont need x86 based AP linux supports almost all major cpu types.
The most widdely used ones even support linux officialy and since nothing
more than wlan card is on the board all we need is to get info from cpu
manufacturer to hack ap.
they are all done by the same referense shematic, but how to trick some cpu
makers to give that info?

Maibe someone have friend working somewhere who can leak it? like Nvidia
beta drivers leaks after 2 days when are released.

[Brian Wilson]

Post by Roy
We dont need x86 based AP linux supports almost all major cpu types.
The most widdely used ones even support linux officialy and since nothing
more than wlan card is on the board all we need is to get info from cpu
manufacturer to hack ap.
they are all done by the same referense shematic, but how to trick some cpu
makers to give that info?

I agree -- since a big part of my plan is to configure a tool chain as
part of the complete kit, I don't see that it matters much
either. It's dreadfully important that the development tool chain not
mix with the hosted system! (This stuff as a build engineer was
critical to me.) You really don't want to use the generic built in
compiler and have it wander around pulling in #include files from
hither and yon!

If I move ahead with this I will have the full docs for the board so
smuggling them out of the factory in my lunch pail is not needed.

So much for grandiosity.

Meanwhile, I am fussing with my reflashed usr2450 wondering how anyone
manages to do anything useful with one... :-) I guess I have to dig
through email archives? Docs on openap site end with "welcome to
linux" and are totally missing on linuxap site. Sort want details like...
is it doing DHCP on the WLAN?? Is default config a P2P bridge? etc...

Brian

[Bruno Lopes F. Cabral]

Hi

Post by Brian Wilson
Meanwhile, I am fussing with my reflashed usr2450 wondering how anyone

reflashed with linuxap?

Post by Brian Wilson
manages to do anything useful with one... :-) I guess I have to dig
through email archives? Docs on openap site end with "welcome to
linux" and are totally missing on linuxap site. Sort want details like...
is it doing DHCP on the WLAN?? Is default config a P2P bridge? etc...

I wrote a portuguese FAQ to linuxap some time ago... I think I can
translate it to english, if there is enough interest...

BTW, in linuxap, check /etc/rw and /etc/rc.d/init.d directories
and you'll find what it is capable of. on-line docs are inexistent
or targeted to compile/not to use

[]s, !3runo

[Bruno Lopes F. Cabral]

Hi there

Post by Scott Serr
5. Basically say, HERE is the demand. "We've identified you out of 9
potential corporate sponsors." Sponsorship costs no money. It only
requires technical information so our Linux engineers can utilize your
hardware. And potentially a small amount of email correspondence with
one of your hardware engineers.

seems pretty much what was done with Actiontec Dual PC Modem Router.
the problem is ... actiontec seems to be unable to deliver a simple
expansion to their routers, a daughterboard that would let anyone
recover for a bad flash/DB that have serial and USB expansion

Cisco has a story of supporting GPL and opensource, where it can.
I hope, with their Linksys acquisition, that what you "dream" can
became true ;-)

Cheers
!3runo