Technology and Trends
 Linux Activists Mailing List Archives
  
From: papresco@cantor.math.uwaterloo.ca (Paul Prescod)
Subject: Linux Foundation
Date: Wed, 17 Mar 1993 12:01:34 GMT

A few people have been wondering about how to support Linux development.
Here's my proposal:

1.We all send money to "the Linux Foundation" care of Linus. (sorry for 
  dragging you into this, Linus, but you are the only "central figure").

2.We set up a mailing list "linuxfoundation" or something.  People mail in
  their nominations and votes for the best contributions to Linux
  that year.

3.The nominations would be auto-posted to this newsgroup (or a .misc newsgroup
  if we ever get around to splitting this newsgroup) and the votes would
  be broadcast around the mailling list.

4.Several people on the list would do an "official" count.

5.We divide up the money according to a predetermined algorithm (or give it
  all to one person in a lump sum).

6.Linus mails off some checks.

Probably nobody would be able to make a living off of Linux development, but
they could at least get some recompense.  The work to Linux isn't as much
as it sounds...he just has to keep the checks, take them to the bank every
so often, and write the checks to the winners.  Other people would do all
the vote administration.

From: goer@ellis.uchicago.edu (Richard L. Goerwitz)
Subject: Re: Linux Foundation
Date: 17 Mar 93 14:44:44 GMT
Reply-To: goer@midway.uchicago.edu

papresco@cantor.math.uwaterloo.ca (Paul Prescod) writes:
>A few people have been wondering about how to support Linux development.
>Here's my proposal:
>
>1.We all send money to "the Linux Foundation" care of Linus. (sorry for 
>  dragging you into this, Linus, but you are the only "central figure").

In this case, I shy away from telling anyone what to do.  I have a sug-
gestion, though:  Create a base Linux release that's stable enough for
commercial purposes, then offer service contracts for institutions that
want regular maintenance and updates, or who want special programming
jobs done.  A Cygnus type arrangement.

I don't think that much money would be generated if a Linux Foundation
were set up as a charitable organization to which contributions would be
made.  Administering such a beast would probably be a headache as well,
and considering that no one could make a living at it that way anyway,
I question whether it is a good idea in the first place.

-- 

   -Richard L. Goerwitz              goer%midway@uchicago.bitnet
   goer@midway.uchicago.edu          rutgers!oddjob!ellis!goer

From: torvalds@klaava.Helsinki.FI (Linus Torvalds)
Subject: Re: Linux Foundation
Date: 17 Mar 93 15:13:35 GMT

In article <C4182M.KLo@undergrad.math.waterloo.edu> 
papresco@cantor.math.uwaterloo.ca (Paul Prescod) writes:
>A few people have been wondering about how to support Linux development.
>Here's my proposal:
>
>1.We all send money to "the Linux Foundation" care of Linus. (sorry for 
>  dragging you into this, Linus, but you are the only "central figure").

No can do.  If people really want to set up some kind of "Linux
Foundation", it needs to be done some other way: I'm *way* too
disorganized to ever be able to re-distribute fair shares of the
proceedings to the rest of the linux people.  Even if I got a list of
names calculated by some kind of voting procedure. 

Also, from a purely economic standpoint, the contact person should
probably be in the US, as that's where most of the users are (as well as
people like tytso, hlu, obz..).  That way most things could be handled
by normal inter-bank drafts or whatever (no, I don't know what I'm
talking about), without bothering with currency changes back and forth. 

Something similar to what you propose (with the exception that I got all
the money :-) was done by hpa@nwu.edu (Peter Anvin) a couple of months
ago.  And it worked very well indeed, thank you, as I got my computer
paid off with it (about USD $750).  So you don't *need* to drag me into
this, just find some trustworthy suckerperson to handle it all. 

Finally: throwing money around can make people feel bad (some people
feel guilty for not sending any, others feel overlooked for not getting
it etc).  You can always try to find alternative ways of showing your
appreciation: I've gotten books, a T-shirt etc, and if you think some
program/feature is epsecially nifty, you can always surprise the person
responsible with something like that by snailmail.. 

                Linus

From: papresco@napier.uwaterloo.ca (Paul Prescod)
Subject: Linux Foundation
Date: Wed, 17 Mar 1993 15:42:29 GMT

A few people have been wondering about how to support Linux development.
Here's my proposal:

1.We all send money to "the Linux Foundation" care of Linus. (sorry for 
  dragging you into this, Linus, but you are the only "central figure").

2.We set up a mailing list "linuxfoundation" or something.  People mail in
  their nominations and votes for the best contributions to Linux
  that year.

3.The nominations would be auto-posted to this newsgroup (or a .misc newsgroup
  if we ever get around to splitting this newsgroup) and the votes would
  be broadcast around the mailling list.

4.Several people on the list would do an "official" count.

5.We divide up the money according to a predetermined algorithm (or give it
  all to one person in a lump sum).

6.Linus mails off some checks.

Probably nobody would be able to make a living off of Linux development, but
they could at least get some recompense.  The work to Linus isn't as much
as it sounds...he just has to keep the checks, take them to the bank every
so often, and write the checks to the winners.  Other people would do all
the vote administration.

NOTE: This is NOT an attempt to give money to everyone who deserves it. Just
      reward a few, creative, talented people with really neat products, like
Dosemu, tcp/ip etc.  

Hopefully in the future, Linux will have many cutting edge programs and
concepts other operating systems don't have.  This proposal is supposed to
encourage this.

From: adam@netcom.com (Adam J. Richter)
Subject: Re: Linux Foundation
Date: Wed, 17 Mar 1993 21:03:06 GMT

In article <1993Mar17.144444.2420@midway.uchicago.edu> 
goer@midway.uchicago.edu writes:
>In this case, I shy away from telling anyone what to do.  I have a sug-
>gestion, though:  Create a base Linux release that's stable enough for
>commercial purposes, then offer service contracts for institutions that
>want regular maintenance and updates, or who want special programming
>jobs done.  A Cygnus type arrangement.

        I've been asking a few people who are connected with potential
customers for this sort of thing what they think about commercial support
for the Yggdrasil Linux/GNU/X ("LGX") distribution from authors of the
software on the CDROM.  I've also asked a couple of Linux contributors
who shall for the moment remain nameless, and they were quite enthusiastic
about possibly being able to support themselves by working on Linux in
this way.  I'd be interested in hearing from other Linux developers
who would like to be involved in this sort of project if it gets off
the ground.

        As always, if you are the author of any software on the LGX beta
release, you can get a free copy.

        This is not to say that you will have to work for Yggdrasil
if you want to provide support for LGX.  On the contrary, I want to
encourage third party support vendors and I will even do things like
promoting them with a listing in the manual, selling them advertising,
and doing other types of cooperative marketting.  I think that a more
open market in support is an important selling point for a free operating
system in the business world.

-- 
Adam J. Richter                             Yggdrasil Computing, Incorporated
409 Evelyn Ave., Apt. 312, Albany CA 94706  PO Box 8418, Berkeley CA 94707-8418
(510) 528-3209                              (510) 526-7531, fax: (510) 528-8508
adam@netcom.com                             yggdrasil@netcom.com
Another member of the League for Programming Freedom (lpf@uunet.uu.net).

From: misch@eurom.rhein-main.de (Michaela Merz)
Subject: Re: Linux Foundation
Date: 19 Mar 93 11:56:30 GMT


On Wed, 17 Mar 1993 15:42:29 GMT,
papresco@napier.uwaterloo.ca (Paul Prescod) wrote:

> A few people have been wondering about how to support Linux development.
> Here's my proposal:
> 
> 1.We all send money to "the Linux Foundation" care of Linus. (sorry for 
> dragging you into this, Linus, but you are the only "central figure").

Why should we found a new organisation? I think it would be a lot smarter,
if (Linus?) and others would make an agreement with the existing organi-
sations like the FSF or/and the FSAG. We should not split into a lot
different groups. 

Michaela


=====
Free Software Association of Germany * Great software should be free software
misch@eurom.rhein-main.de Voice: ++49-69-6312083
misch@eurom.fsag.incom.de Fido 2:247/14 Data: ++49-69-6312934
================= infos via server@eurom.fsag.incom.de ======================

Date: Fri, 19 Mar 1993 17:53:47 CET
From: <K11111I@ALIJKU11.BITNET>
Subject: Linux isnt real OS / Linux Foundation


Contents:

Linux is the future organisation form for developing software!
Linux is not a real OS, many lacks!!!
Please no bureaucracy in Linux ==> better suggestion

The most thing on Linux development what surprises me
is that it is selforganized. And it works great. Of course their
could be some improvements, but it still will be self organized.

When it works with no leader on the top (and obvious, there is no need for
 a company or a CEO to write a UNIX), why do you want to install a
bureaucratic institution which will cost a lot of money and will be hard to
controll (there is always need for control if there is money which does|nt
belong to the handlers)?

Richard L. Goerwitz writes:

>In this case, I shy away from telling anyone what to do.  I have a sug-
>gestion, though:  Create a base Linux release that's stable enough for
>commercial purposes, then offer service contracts for institutions that
>want regular maintenance and updates, or who want special programming
>jobs done.  A Cygnus type arrangement.



I think it is a good idea that Linux must be used by people and institutions
**which earn money** with it. After that they can share the money they
earned with Linux with the Linux programmers.
If nobody earns any money by using Linux, there will be no money to
share!!!!  (No private **user** is in the position to pay any money just for
playing around with Linux or doing his homework). So the money has to
come from **real** users.

A operating system that does|nt have any use to the real world is just fun
for the programmers and hobby users.


But how must a OS (e.g. Linux) be that it is used in real business and
somebody will pay for it?



I guess it should be:

1- stable and nearby bugfree (proofed by official standards like ISO 900x)
2- up to date with standards (needs working within the standartisation
comitees e.g. DCE)
3- proofed against break in the system (proofed by official standards)
4- proofed  against data losses (proofed by official standard)
5- implemented on nearby every Hardware (also in the future)
6- efficient use of Hardware
7- there should be a lot of user and developer software
8- well documented
9- easy to handle and use
10- there should be a responsible person for all that what is delivered to a
user
11- competent and fast bug support
12- good and fast user support (by remote control)
13- good and fast developer support
14- source code should be available
15- supported by good training for users and programmers and admins
16- the security, that the OS will be supported in future and will fit future
standards
17- the security, that there will be improvements in the future
18- fair prices  (compared with the delivered service)
19- little switching costs (no installation cost for the new OS)
20- little maintenance costs
21- possibility to use cheap Hardware (not IBM)
22- possibility to integrate in current Software and Hardware
23- come with nice design
24- ???????????????

Linux delivers point 6, 11, 13, 14, 18, 21, and that is absolutely great!!!

Because of some lacks Linux is not a OS to use for real jobs and from my
point of view probably will never become more than a toy for freaks. But
maybe this is still the intention from most of you.

In my opinion, the way Linux is developed could be a new way of creating
and implementing and *selling* new and good software by independent
and self-responsible programmers, administrators, software-user-coaches.
But maybe Linux|er have total different intentions.

As I can see, Microsoft has not a ideal structure to develop SW,
but Linux has. Lets turn MS out of business.
(I have a lot of good reason|s for this short statement)


Some comments to the points:

1) Software will never be bugfree, so fast and competent bug support is
essential. The only person which can deliver this is mostly the person
which wrote the software. All others rely on her. She should be responsible
for the bug support. And she should be payed for that responsibility (if she
wants).

2) Software (also OS) is only interresting for *users* when it is improofed
all the time and up to date with standards. Software which stay|s on the
same level is worthless. Also Linux will die when it will not meet mayor
standards in the future because of the lack that Linus (or other students in
major Linux positions) will finish his study and switch to Microsoft (where
he will earn al lot of money).

3) this can be an argument for users to favor Linux

4) this can be an argument for users to favor Linux

5) an OS (and its enviroment) on just one limited Hardware is a big
argument against Linux (I don|t wanna start a discussion about portability)

7) Sorry, a lot of tools, no major products (Word processing, data bases..)
If Linux will become a real OS, a lot of software will be ported to Linux
from software companies. That software will be ported to Linux when
there are a lot of commercial users (which can pay for it). And a lot of
commercial users will be only there, if there is standard-software (so guys,
break the dead circle and start to program a multitasking, *modular* word-
processing, spreadsheed,....)

8) there is something going on

9) there are good concepts (like SLS), but even not simple enought.
    ** keep it as simple as possible, not more**

10) If I buy a car, the person which is responsible for the delivered thing is
the dealer.

So if Paul installs me a Linux, he is responsible for the thing, if something
does|nt work. Paul does|nt know the whole OS, so he relys on the guys
which wrote all the parts they delivered him. These guys are responsible
for what they delivered. If they are not responsible, Paul couldnt be sure
to get support from them if something goes wrong. And for that
responsibility Paul pay them on a volunteer basis.
Paul delivers me a UNIX persion, and I pay him. Without any contract. If I
pay him well, he will support me also in the future. If he did bad work, I
dont have to pay him anything. If I pay him to much, others will come and
offer me their service.

There could be hundreds of different versions of Linux supported from
somebody and availiable. Somebody can deliver and be responsible for a
more stable version (e.g. for commercial use), another one can deliver an
up to date version (for developing) (with in mind that there are a lot of
bugs).

It is up to Paul to assemble a stable version, out of parts were he probably
gets support, and also to meet my needs. And for that I ** as a user** will
pay him as long as he gives me my UN*X.

(Adam J. Richter) writes:
>Richard L. Goerwitz writes:
>>gestion, though:  Create a base Linux release that's stable enough for
>>commercial purposes, then offer service contracts for institutions that
>>want regular maintenance and updates, or who want special jobs done.  A
>>Cygnus type arrangement.

>        I've been asking a few people who are connected with potential
>customers for this sort of thing what they think about commercial support
>for the Yggdrasil Linux/GNU/X ("LGX") distribution from authors of the
>software on the CDROM.  I've also asked a couple of Linux contributors
>who shall for the moment remain nameless, and they were quite
>enthusiastic
>about possibly being able to support themselves by working on Linux in
>this way.  I'd be interested in hearing from other Linux developers
>who would like to be involved in this sort of project if it gets off
>the ground.

This could be one version of Linux, and one person which is responsible.


12) If there is a user/admin support, supporters should speak the language
of the users and understand their problems (insurance industry,
mechanical industry, electronicindustry, printmedias, health industry, they
all use different languages). And if you want to bring Linux to single users,
you also have to deliver admin support (not everybody is a Unix admin,
most people are not interrested in Unix, they are interrested in writing
letters, calculation budgets, writing emails, having the newest Software),
and you have to speak the profession language (engineer, secretary,
medics, professor, ...).

15) Look at Novell, they know that they have to deliver good training for a
complex product, and they suceed.

16) The bigger a Company, the more people use the OS, the more money a

company makes out of a product,  the higher is the probability that the company
will support me in the future.
If a company can|t rely on that, she will never use a OS from them.
Linux is not a Company, but a realy big Organisation. The thing it lacks, nobody
***uses*** the software as real OS for earning money. Major people of the
projekt can break down because of the need to earn money.

17) see 16)

19) When a company has to change any part of the software, there are
huge costs for training, data-restoring, recompiling.  If there will be costs to
+buy+ a new OS (new SW), nobody ever will switch.

****** Software is not a product, you can|t sell it  ************
**** Software is a stored service, you can pay for using the sevice ****

20) compared with ms.dos, compared with "no source av. OS".
e.g. 1 PC with Windows needs prox. 4000$ for maintenance a year, so if a
Workstation with Linux, X,... can deliver the same use (mainly Word
processing, Spreadsheed, Database, email), there is a lot of money for the
people they provide this (probably cheaper and much more efficient with a
real OS)


22) Source code av.!!!! If you need something very urgent, you can program
everything.

23) As you can see with Windows 3.x.   Design is more important than to be
error free. (MS invests a lot of money into its outfit, also does Next).
But I know, Unix has|nt to do anything with GUI. Just take care. A sign to
recognize is very important (see sun,sun,sun,sun).



In article < C4182M.KLo@undergrad.math.waterloo.edu>
papresco@cantor.math.uwaterloo.ca (Paul Prescod) writes:

>A few people have been wondering about how to support Linux
>development.
>Here's my proposal:
>
>1.We all send money to "the Linux Foundation" care of Linus. (sorry for
> dragging you into this, Linus, but you are the only "central figure").

Richard L. Goerwitz writes:

>I don't think that much money would be generated if a Linux Foundation
>were set up as a charitable organization to which contributions would be
>made.  Administering such a beast would probably be a headache as well,
>and considering that no one could make a living at it that way anyway,
>I question whether it is a good idea in the first place.


Nobody gives any money away when he does|nt know where the money
goes. If you spend your money to the red cross, you expect to be told what
will be payed with it.

I suggest something different:

If I use SLS, I can measure what I got (a perfect Unix with everything and,
and, and, and, and,..). For that I will give them a donation (because I want
them to continue their work). ==> no administration.

They got a lot of parts to produce the SLS, so they should give the major
part of the money to the people which delivered the parts (as GCC, Xfree,
Kernel,....). The SLS producer know them all, so they can send them a check
every second month or so. ==> nearby no admin.

If the guy who writes the kernel (I know its you) got a little help from FSF
he can send them a little donation.

and so on......


The important thing in this system is, that everbody knows what he got
from which person.

And off course, there is a absolutely minimum of admin.

And

******* no central money depot *******
******* no central administration ********


But: If Linux does|nt become a real OS with support and some of the other
things described above that a **real world user** can get out use of it,
there will be no money at all.

                                  Ed

From: barspi@wam.umd.edu (Barzilai Spinak)
Subject: Re: Linux isnt real OS / Linux Foundation
Date: Sat, 20 Mar 1993 04:19:27 GMT

In article < 93078.175347K11111I@ALIJKU11.BITNET> < K11111I@ALIJKU11.BITNET> writes:
[... some junk deleted ...]
>But how must a OS (e.g. Linux) be that it is used in real business and
>somebody will pay for it?
>
>I guess it should be:
>
>1- stable and nearby bugfree (proofed by official standards like ISO 900x)
>2- up to date with standards (needs working within the standartisation
>comitees e.g. DCE)
>3- proofed against break in the system (proofed by official standards)
>4- proofed  against data losses (proofed by official standard)
>5- implemented on nearby every Hardware (also in the future)
>6- efficient use of Hardware
>7- there should be a lot of user and developer software
>8- well documented
>9- easy to handle and use
>10- there should be a responsible person for all that what is delivered to a
>user
>11- competent and fast bug support
>12- good and fast user support (by remote control)
>13- good and fast developer support
>14- source code should be available
>15- supported by good training for users and programmers and admins
>16- the security, that the OS will be supported in future and will fit future
>standards
>17- the security, that there will be improvements in the future
>18- fair prices  (compared with the delivered service)
>19- little switching costs (no installation cost for the new OS)
>20- little maintenance costs
>21- possibility to use cheap Hardware (not IBM)
>22- possibility to integrate in current Software and Hardware
>23- come with nice design
>24- ???????????????
>
>Linux delivers point 6, 11, 13, 14, 18, 21, and that is absolutely great!!!
  I think it delivers and will deliver more than that.
And most of the OS that are out there (I don't have to name them) don't deliver
half of the above points and they are used every day by millions of people. 

And considering what I've learned of unix (I've been using it only for about 6 
months), all unix implementations have been and probably will remain mostly
a hacker's toy. Some people do real work with it every day but they are not
your regular dumb PC user who cannot tell a parity bit from a floppy disk.
(I have some stories with parity bits...) 

[... more crap deleted ...]
>7) Sorry, a lot of tools, no major products (Word processing, data bases..)
>If Linux will become a real OS, a lot of software will be ported to Linux
>from software companies. That software will be ported to Linux when
>there are a lot of commercial users (which can pay for it). And a lot of
>commercial users will be only there, if there is standard-software (so guys,
>break the dead circle and start to program a multitasking, *modular* word-
>processing, spreadsheed,....)

      A little what happened with OS/2, isn't it? And nobody in the industry
dared to say that it wasn't a real OS, anyway.

[... blah blah blah ...]
>20) compared with ms.dos, compared with "no source av. OS".
>e.g. 1 PC with Windows needs prox. 4000$ for maintenance a year, so if a
>Workstation with Linux, X,... can deliver the same use (mainly Word
>processing, Spreadsheed, Database, email), there is a lot of money for the
>people they provide this (probably cheaper and much more efficient with a
>real OS)

    $4K?!?!?!?!?!?!?!?!?!?!   Does your PC use electricity or cocaine?

[... rest of message deleted ...]

+------------------------------------\-------------------------------------+
| o     ____  /           _____/     /        Barzilai Spinak            o |
|      /     /           /           \            barspi@wam.umd.edu       | 
|     _____   /       _____  /       /            barspi@eng.umd.edu       | 
|    /       /              /        \              .....                  | 
|  _________/ ___/  _______/ ___/    /             (-O-O-)    Dale Manya!  | 
+------------------------------------\-----------nnn--U--nnn---------------+

From: papresco@napier.uwaterloo.ca (Paul Prescod)
Subject: A New Linux Foundation
Date: Sun, 21 Mar 1993 05:19:59 GMT


Fact 1:Many people want to support Linux, but don't know how.

Fact 2:Other people are frustrated with a lack of organization in the Linux
        community.

Earlier, I said we should set up a "Linux Foundation" and I outlined a
specific vision of a charitable organization geared towards paying talented
Linux developers for their work.  Some agreed with me, and some disagreed.  I
don't feel comfortable going ahead with something without the support of
the organization.

Therefore, I'd like to try again.  I propose a new Linux Foundation, unrelated
to the old.  It will have a much broader mandate.

1.What would the Linux Foundation be?  

An organization of users and developers dedicated to promoting Linux
and increasing and supporting the Linux  user base.

2.What would the Linux Foundation do?

Let me start by saying what it would not do:

Restrict Linux development in any way.
Make a profit for the members.
Restrict Linux distribution.
Decide who can and cannot use or sell Linux.

What it WOULD do:

Anything else that would promote Linux.

Some *POSSIBLE* ideas: organize databases of Linux developers, fund raise,
                       develop pamphlets and marketing ideas,  keep a 
database of current projects, promote corporate use of Linux etc. etc.

3.How would it be run?

Who knows!  That's the exciting part about it.  Everything about the Linux
Foundation is up in the air.  These issues have to be worked out in the 
first few months of it's establishment.  I think the first thing we would
do is set up a Linux Foundation mailing list.  From there we would go on
to a mandate and goals.  After that we would start on specific goals.

Opinions?

From: rick@ee.uwm.edu (Rick Miller, Linux Device Registrar)
Subject: The best way to "support Linux"!
Date: 21 Mar 1993 09:48:45 GMT

papresco@napier.uwaterloo.ca (Paul Prescod) writes:
>
>Fact 1:Many people want to support Linux, but don't know how.
>
>Fact 2:Other people are frustrated with a lack of organization in the Linux
>       community.
[...]
>Opinions?

Maybe I'm missing something, but it seems to me that the most logical
conclusion to be drawn from "Fact 1" and "Fact 2" is that these people
could contribute by *organizing* (archiving, packaging, distributing,
supporting, listing, meta-listing, ...) Linux!

THE BEST WAY TO SUPPORT LINUX IS TO CONTRIBUTE TO ITS DEVELOPMENT!

I'm not saying that Linus wouldn't like to see another half-dozen trinkets
surprise him in his mailbox, but "Linus" is not "Linux".  Wouldn't you much
rather make a contribution that *all* the folks developing Linux could have?

Look at *me*.  My skills in 'C' are limited to coding "C-ROBOTS" (a game like
core-wars, only with simulated robots running your code... and my 'bots will
beat your 'bots *any* day!  Nyah!  :-P  ).  So what have *I* done to "support
Linux"?

THE BEST WAY TO SUPPORT LINUX IS TO CONTRIBUTE TO ITS DEVELOPMENT!

I started the Linux Device List.  Someone suggested that it would be a good
idea, others agreed, so I filled the gap.  I've injected a little bit of order
into the Linux community at large.  (Actually, the laws of thermodynamics say
that Entropy is never lost, so I must be dissipating it as heat from my CPU.)

In the process of contributing, you'll learn more about your favorite OS and
you'll almost certainly find other places where your skills are needed.  My
next version of the Linux Device List will come complete with a new MAKEDEV
for example.  A logical next step, chu ne?  So I'll just say it once more:

THE BEST WAY TO SUPPORT LINUX IS TO CONTRIBUTE TO ITS DEVELOPMENT!

Rick Miller   <rick@ee.uwm.edu> | <rick@discus.mil.wi.us>   Ricxjo Muelisto
Occupation:  Husband, Father, WEPCo. WAN Mgr., Discus Sys0p, and  Linux fan

From: joem@netcom.com (Taos Mountain Software)
Subject: Re: Linux Foundation
Date: 21 Mar 93 09:49:28 GMT

I would like to see a professionally maintained bug tracking database,
moderation for some (sub)hierarchy of this group and new user questions
answered via E-mail.  

I think these things are more important to the success of LINUX than
monetary rewards for donated software.

-jgmarce

From: hlu@luke.eecs.wsu.edu (HJ Lu)
Subject: Re: Linux Foundation
Date: 21 Mar 93 11:35:32 GMT

In article < joemC48GMG.Hsw@netcom.com> 
joem@netcom.com (Taos Mountain Software) writes:
>I would like to see a professionally maintained bug tracking database,
>moderation for some (sub)hierarchy of this group and new user questions
>answered via E-mail.  
>
>I think these things are more important to the success of LINUX than
>monetary rewards for donated software.
>
>-jgmarce

The idea sounds good. I am in charge of the C library and the Linux
port of gcc. For the gcc bugs, I usually forward them to the gcc2
people. I think that is where they belong. As for the C library bugs, I
am fixing them as soon as I can. The problem is I have only limited
resources, 386sx-16MB with 4MB RAM and 100 MB HD. I really like to
have a bug tracking database system, like DejaGnu. But I cannot afford
it. 

Also I just finished my MS in CS and have been looking for a Unix/C
job. I cannot put all my time on Linux. I have applied quite a few
companies. But my lack of commercial experiences seems to be a
big obstacle for me to get a job offer. What I have been doing now is
when I see a bug report at school I will check the obvious. Usually
I have to go home and run gdb at late night and in early morning. It
is not as fast as I want to be. So far I have managed to

1. fix the bug in library, or
2. fix/find bug in application, or
3. find/fix bug in kernel

with 12 hours for most of bugs after a bug report is received.

That is only the part of the story. Since the FSF doesn't have all we
need, Linux uses utilties from all kinds of places. The problems are

1. There are not comprehensive, accurate and uptodate docs on
   the Linux C library and people are not used to read those
   soucre code. (BTW, I don't they should.)
2. Not all the utilties are well written for POSIX system, like Linux.

I have been fixing bugs in all kinds of system utilties while compiling
them under Linux. But unless the bug is very bad, usually I don't
bother to send my fix to author since I don't have the time to do that.
I wish there will be a good doc on the Linux C library and for each
system utilty there is someone who relays the Linux related bugs to
its author.

Last but not the least, Linux is on the road to be a complete Unix
system. But there is still no a real consensus about file structure. I
know there was a one. I am not sure how complete it was. But at that
time, we had no X11, no NFS, no shared library. I think it is the
right time now to make that decision. There are several different
Linux packages out there, like SLS, TAMU, LGX, ...... Unfortunately, I
cannot test either of them. Since the file structure is very closely
related to the C library, I really like to see some kind of standard
soon.


H.J.

Date: Sun, 21 Mar 1993 17:40:44 CET
From: <K11111I@ALIJKU11.BITNET>
Subject: Re: The best way to "support Linux"!

rick@ee.uwm.edu (Rick Miller, Linux Device Registrar) writes:

>papresco@napier.uwaterloo.ca (Paul Prescod) writes:
>>
>>Fact 1:Many people want to support Linux, but don't know how.
>>
>>Fact 2:Other people are frustrated with a lack of organization in the Linux
>>       community.
>>[...]
>>Opinions?

>Maybe I'm missing something, but it seems to me that the most logical
>conclusion to be drawn from "Fact 1" and "Fact 2" is that these people
>could contribute by *organizing* (archiving, packaging, distributing,
>supporting, listing, meta-listing, ...) Linux!

>THE BEST WAY TO SUPPORT LINUX IS TO CONTRIBUTE TO ITS >DEVELOPMENT!
>]xD{i]{{t({

Hey Boy, off course.

Microsoft: no mangers, no trainers, no financial experts, no strategists, no
support, no marketing, no lobby, just programmers?
Novell: no managers, no organisation, no custom service, just programming
freaks?

You are dreaming. A success of an OS is more then programming.
It also needs marketing, support, training, distribution...


>I'm not saying that Linus wouldn't like to see another half-dozen trinkets
>surprise him in his mailbox, but "Linus" is not "Linux".  Wouldn't you much
>rather make a contribution that *all* the folks developing Linux could >have?

>Look at *me*.  My skills in 'C' are limited to coding "C-ROBOTS" (a game >like
core-wars, only with simulated robots running your code... and my >'bots will
beat your 'bots *any* day!  Nyah!  :-P  ).  So what have *I* done >to "support
Linux"?#

I allready made a suggestion to this. See for K11111I.

                            ED         :-D

Date: Sun, 21 Mar 1993 17:51:17 CET
From: <K11111I@ALIJKU11.BITNET>
Subject: Re: Linux isnt real OS / Linux Foundation

========================================================================
In Article < 1993Mar20.041927.3668@wam.umd.edu>
Sender: usenet@wam.umd.edu (USENET News system) writes:

>In article <93078.175347K11111I@ALIJKU11.BITNET> ><K11111I@ALIJKU11.BITNET>
>writes:
>[... some junk deleted ...]

Please, if you dont understand things, dont judge them.

>>But how must a OS (e.g. Linux) be that it is used in real business and
>>somebody will pay for it?
>>
>>I guess it should be:
>>
>>1- stable and nearby bugfree (proofed by official standards like ISO 900x)
>>2- up to date with standards (needs working within the standartisation
>>comitees e.g. DCE)
>>3- proofed against break in the system (proofed by official standards)
>>4- proofed  against data losses (proofed by official standard)
>>5- implemented on nearby every Hardware (also in the future)
>>6- efficient use of Hardware
>>7- there should be a lot of user and developer software
>>8- well documented
>>9- easy to handle and use
>>10- there should be a responsible person for all that what is delivered to a
>>user
>>11- competent and fast bug support
>>12- good and fast user support (by remote control)
>>13- good and fast developer support
>>14- source code should be available
>>15- supported by good training for users and programmers and admins
>>16- the security, that the OS will be supported in future and will fit future
>>standards
>>17- the security, that there will be improvements in the future
>>18- fair prices  (compared with the delivered service)
>>19- little switching costs (no installation cost for the new OS)
>>20- little maintenance costs
>>21- possibility to use cheap Hardware (not IBM)
>>22- possibility to integrate in current Software and Hardware
>>23- come with nice design
>>24- ???????????????
>>
>>Linux delivers point 6, 11, 13, 14, 18, 21, and that is absolutely great!!!
>I think it delivers and will deliver more than that.
>And most of the OS that are out there (I don't have to name them) don't
>deliver half >of the above points and they are used every day by millions >of
people.

Boy you got it.
I wrote *should be* to get maximum use out of an OS, but as you pointed out,
there are some other aspects to become a major OS used by millions of people.

To become a major OS you need a perfect strategy, which includes not just a good
product, but also partners, hardware producers, support, dependent user,
distribution, publicity, power, applikations, developers,....

And the man which plays this best is Bill Gates. He can make gold out
of shit with strategy.

So Linux will fail. It is a good product (like many others), but has no
strategy (there are just programmers in the Linux-project, no strategists, and
if there were strategists, the hackers wouldnt believe them the need for a
strategy)

>And considering what I've learned of unix (I've been using it only for >about 6
>months), all unix implementations have been and probably will >remain mostly a
>hacker's toy. Some people do real work with it every day >but they are not your
>regular dumb
›PC user who cannot tell a parity bit >from a floppy disk.
>(I have some stories with parity bits...)

Unixers are only hackers, they complain that everybody uses MS-DOS and Windows,
but doesnt want to meet the needs from these dump PC users (see LaTex).
Microsoft has a strategy for these dump users, and makes out a lot of money. It
seems to me like Unixers
› dosnt like money because they doesnt believe in strategy.

>[... more crap deleted ...]
>>7) Sorry, a lot of tools, no major products (Word processing, data bases..)
>>If Linux will become a real OS, a lot of software will be ported to Linux
>>from software companies. That software will be ported to Linux when
>>there are a lot of commercial users (which can pay for it). And a lot of
>>commercial users will be only there, if there is standard-software (so guys,
>>break the dead circle and start to program a multitasking, *modular* word-
>>processing, spreadsheed,....)
>
>     A little what happened with OS/2, isn't it? And nobody in the industry
>dared to say that it wasn't a real OS, anyway.

Yes, and for that reason you need a damn good strategy, not just a good OS.

>[... blah blah blah ...]
>>20) compared with ms.dos, compared with "no source av. OS".
>>e.g. 1 PC with Windows needs prox. 4000$ for maintenance a year, so if a
>>Workstation with Linux, X,... can deliver the same use (mainly Word
>>processing, Spreadsheed, Database, email), there is a lot of money for the
>>people they provide this (probably cheaper and much more efficient with a
>>real OS)
>
>   $4K?!?!?!?!?!?!?!?!?!?!   Does your PC use electricity or cocaine?

Gartner Group/London made the research, not me.
It was presented by their company speaker Mr. Peter Sondergaerd.
(Der Standard, 2/19/1993 p.31)

In average a PC costs in his live 36.940 $.

administration = 14 percent
support = 12 percent
training for users = 57 percent
Hardware and Software = 17 percent

In average a DOS machine costs 7047$ a year.
In average a Windows machine costs 5891$ a year.
In average a Mac costs 4965$ a year.

From: Theodore Ts'o <tytso@athena.mit.edu>
Subject: Re: Linux isnt real OS / Linux Foundation
Date: 21 Mar 1993 16:53:05 -0500
Reply-To: tytso@athena.mit.edu

   Date: Sun, 21 Mar 1993 17:51:17 CET
   From: <K11111I@ALIJKU11.BITNET>

   To become a major OS you need a perfect strategy, which includes not
   just a good product, but also partners, hardware producers, support,
   dependent user, distribution, publicity, power, applikations,
   developers,.... 

   So Linux will fail. It is a good product (like many others), but has no
   strategy (there are just programmers in the Linux-project, no
   strategists, and if there were strategists, the hackers wouldnt
   believe them the need for a strategy)

Wait a minute!  In order to measure success or failure, you must first
have a goal in order to judge a project by.  Who said that Linux's goal
was to become a major OS?  That's certainly never been *my* goal --- and
despite that, I've put a lot of time in to Linux: the serial driver, the
c.o.l. mail digest, tsx-11.

I suspect that for a lot of the people who have put major amounts of
efforts into Linux, the object is to make Linux a *useful* OS.  But a
major OS?  Sorry, there is a lot of nonsense that you have to do in
order to become a major OS ("partners", "hardware producers", "support",
"publicity", "power") that I'm not going to waste time on --- especially
since I don't think there's anyway I'm going to get enough money out of
it to make it worth the headache.  For now, it means I can run a useful
Unix system on a 486 laptop, and that's enough.  (If someone will show
up on my doorstep with 2-3 million dollars, I'll reconsider about it.
:-)

But if there are people who are interested in making Linux a major OS,
that's great.  Right now, what that means is we need people to do the
integration, and the documentation, and the distribution.  And there are
people who are doing this, like Peter for example --- and they will go a
long way towards making Linux much more popular.  A "major" OS, though?
Probably not.  But so what?


I think a major problem with having a "Linux Foundation" is that
everybody has different goals as to what that foundation might do.  Some
people want Linux to be a major OS --- otherwise, they won't use it.
(Their loss.)  Other people want it to be a way to express their
gratitude.  Other people want it to be a central authority to dictate
a filesystem hierarchy.  Other people want to be able to hobnob with
other big important foundations like the OSF and UI.  

I'm not convinced that a Linux Foundation would be useful.  But if you
must create it, make sure you state up front what your goals and
objectives are, and share with all of the people that you're going to
hit up for money, and write it into your charter and make it your
mission statement.  Otherwise, I predict that it will easily be dragged
off course and have all sorts of internal conflicts and disagreements.

                                                - Ted

From: pmacdona@sanjuan (Peter MacDonald)
Subject: Re: New Linux Foundation, etc...
Date: 22 Mar 93 04:31:09 GMT

I will hate myself in the morning for getting sucked into this thread...

In article < 1993Mar22.000159.27508@ultb.isc.rit.edu> 
axi0349@ultb.isc.rit.edu (A.X. Ivasyuk) writes:
...
><K11111I@ALIJKU11.BITNET> writes:
...
>>You are dreaming. A success of an OS is more then programming.
>>It also needs marketing, support, training, distribution...

Aside from Ted's (valid) point about "success" != "commercial success",
there are a number of important downsides to hitting the big time.

For one thing, the larger the organization, the greater the inability
to produce meaningful S/W accomplishments.  Look at IBM, and MS for examples.  

Lots of reasons for this.  

        - Committee mentality bogs everything down 
        - many power groups vying for control and all neutralizing each other, 
        - the need to maintain backward compatibility even with braindead 
          things because "our customers demand it".
        - the Marketing dept starts dictating the technical direction, because
          after all, peoples livelyhood is at stake here.

Linux is a reasonable product, partly because it is free to evolve in a reasonable
fashion.  But also because it is shaped by reasonable people, who do on 
occasion exercise absolute authority and make crucial decisions, even when
that means breaking everything, for the common good.  Such luxury's may
not always be so easily afforded.


Peter 

From: papresco@napier.uwaterloo.ca (Paul Prescod)
Subject: Re: Linux isnt real OS / Linux Foundation
Date: Mon, 22 Mar 1993 10:34:06 GMT

In article <1oio41INN97m@senator-bedfellow.MIT.EDU> tytso@athena.mit.edu writes:
>   Date: Sun, 21 Mar 1993 17:51:17 CET

>Wait a minute!  In order to measure success or failure, you must first
>have a goal in order to judge a project by.  Who said that Linux's goal
>was to become a major OS?  That's certainly never been *my* goal --- and
>despite that, I've put a lot of time in to Linux: the serial driver, the
>c.o.l. mail digest, tsx-11.

I want Linus to be a major OS for two reasons:
1.I want to be able to make a living working on it.
2.I want there to be major development for it so I can run the "good" apps on
it.  TeX is nice, but it isn't the beall and endall.  I would like to run 
commercial type apps on it so I can delete my DOS/Windows partition.


>I think a major problem with having a "Linux Foundation" is that
>everybody has different goals as to what that foundation might do.  Some
>people want Linux to be a major OS --- otherwise, they won't use it.
>(Their loss.)  Other people want it to be a way to express their
>gratitude.  Other people want it to be a central authority to dictate
>a filesystem hierarchy.  Other people want to be able to hobnob with
>other big important foundations like the OSF and UI.  

Well, I'm not sure who is going to pay for someone to fly out and "hobnob" but
as far as the other goals go, isn't that like saying:

Some people will want to use Linux for development.
Some people will want to use Linux as a news server.
Some people will want to use Linux as a technical writing platform etc. etc.

Great!  Development people ported GCC, News people ported rn and others
ported TeX.

In the Linux Foundation, some may work on recompensing those who have given
us so much.  Some may work on integeration.  Some may work on lobbying
magazines.  Some may work on opening up distribution channels.  Some may
set up "databases." 

It will have the Linux spirit: Just do it!

There will be no infighting, because our mandate is broad enough to allow 
everyone to do their own pet project, and use the Foundation as a "front end"
a central point, a database server.  I will certainly never stop you from
doing anything...we will especially not restrict you technically.

From: Theodore Ts'o <tytso@athena.mit.edu>
Subject: Re: Linux isnt real OS / Linux Foundation
Date: 22 Mar 1993 17:20:31 -0500
Reply-To: tytso@athena.mit.edu

   From: papresco@napier.uwaterloo.ca (Paul Prescod)
   Date: Mon, 22 Mar 1993 10:34:06 GMT

   I want Linus to be a major OS for two reasons:
        .....

   In the Linux Foundation, some may work on recompensing those who have given
   us so much.  Some may work on integeration.  Some may work on lobbying
   magazines.  Some may work on opening up distribution channels.  Some may
   set up "databases." 

   There will be no infighting, because our mandate is broad enough to allow 
   everyone to do their own pet project, and use the Foundation as a
   "front end" a central point, a database server.  I will certainly
   never stop you from doing anything...we will especially not restrict
   you technically. 

These are such optimistic words.... on the other hand, the problem is
what happens when these goals conflict?  What if all of the people who
AREN'T doing the development want Linux to be a major OS, with all of
the polishing that entails, and start demanding that the volunteers
supplying the development effort do something they don't want to do.

For example, the very next news article after yours was the following:

|From: barspi@wam.umd.edu (Barzilai Spinak)
|Subject: Aiming at version 1.0
|Date: 22 Mar 93 16:25:09 GMT

|    I've been thinking (yes!), wouldn't it be better to stop adding new
|features to Linux for a while and concentrate on debugging what we already
|have?  What would it take to reach version 1.0?  If it's a matter of
|stability we won't ever reach it if new things get added to the
|kernel every other day. 
|
|I think that a decent Linux v1.0 would have a much greater psychological
|impact than an overstuffed and buggy Linux 0.99pl23.345-2ndpl4.

Here's a person who's trying to suggest that people who enjoy adding new
features to the Linux kernel STOP what they are doing so that polishing
for 1.0 can go on.  (What's this WE business?  Maybe I've missed
something, but I don't think I've seen any contributes from Barzilai.)

As long as there is this Official Linux Foundation that is Working To
Make Linux An Major OS, there is always going to be the temptation to
try to dictate to the developers What They Should Do in order to Make
The Users Happy.  In all of the "Major Operating Systems", the marketing
department always has control over the developers should do.

                                                - Ted

 From: sct@dcs.ed.ac.uk (Stephen Tweedie)
Subject: Re: Linux isnt real OS / Linux Foundation
Date: 22 Mar 93 23:29:55 GMT

In article <C4ADCv.2AB@undergrad.math.waterloo.edu>, 
papresco@napier.uwaterloo.ca (Paul Prescod) writes:

> In the Linux Foundation, some may work on recompensing those who
> have given us so much.  Some may work on integeration.  Some may
> work on lobbying magazines.  Some may work on opening up
> distribution channels.  Some may set up "databases."

> It will have the Linux spirit: Just do it!

Hark ye all to the wisest words spoken recently on comp.os.linux.

"Just do it" is the philosophy which has made Linux the success which
it is today.

Please do not misunderstand me.  I do not wish to damp down the calls
for a stable, mature and commercially-oriented Linux development.  I
just think that this cannot be the be-all and end-all of the Linux
effort.

Linux has been written by hackers - and I use the words a mark of
respect to those concerned.  The Linux development thrives on being at
the cutting edge of the technology.  The excitement of innovation is a
major force.

SLS is not on the bleeding edge of Linux development.  Peter does keep
SLS remarkably up to date, but he prefers to release stable and tested
distributions.  When there is an important new development, then once
it has proved itself stable it may also appear in SLS.  The ext2fs is
a prime example of this.  However, you won't see a pre-release alpha
kernel in an SLS distribution.

There _is_ a demand for stability.  There is at least a potential
market for commercial support of Linux.  Look at the Yggdrasil LGX
Linux CDROM release.  They are actually advertising an (admittedly
small, as yet) number of service vendors offering support related to
LGX.

However, I don't think that Linux development should be damped or
regulated just to meet the demands of stability.  There is room in the
wake of the cutting edge for stable releases to be packaged.  I
believe that the best way to ensure the future of Linux is to maintain
the current freedom of development; and the best way to assist the
more widespread acceptance of Linux is to support the package
maintainers.  Criticisms of Linux simply because it has no marketing
strategy are misplaced, because these are two separate aims, and I see
no conflict between them.

> There will be no infighting, because our mandate is broad enough to
> allow everyone to do their own pet project, and use the Foundation
> as a "front end" a central point, a database server.  I will
> certainly never stop you from doing anything...we will especially
> not restrict you technically.

Then I wish you all the best of luck.

Cheers,
 Stephen Tweedie.
---
Stephen Tweedie    (Internet: )
Department of Computer Science, Edinburgh University, Scotland.