I've just fixed an error in a Linux kernel for Android which affected
a customer based on the kernel crashdump output by comparing the
assembly code in the raw kernel binary I extracted out of the
corresponding boot image file with the source code of the affected
driver and patching the binary (for reasons I won't detail here
:-). Granted, this was a rather trivial bug but until Thursday last
week, I knew essentially nothing about Android except that it existed
and the person who used to work with this code was not available for
any assistance.

Fixing errors in code written (or maintained) by someone else, even
under fairly adverse circumstances, is a lot less 'impossible' than
some people are apt to assume ...

Rainer Weikusat <rweikusat@mssgmbh.com> writes:

> Fixing errors in code written (or maintained) by someone else, even
> under fairly adverse circumstances, is a lot less 'impossible' than
> some people are apt to assume ...

Or, to put it another way: Software freedom is something every user
needs in all software, even if they have no intention of exercising that
freedom at the time they receive the software.

If everyone has the freedom to improve software and share the
improvements, then a majority of people don't need to exercise that
freedom to benefit from it. When no-one with skill is prevented from
improving and sharing the software, then users only need to find someone
with the skill willing to help.

-- 
 \     “The Things to do are: the things that need doing, that you see |
  `\     need to be done, and that no one else seems to see need to be |
_o__)                   done.” —Richard Buckminster Fuller, 1970-02-16 |
Ben Finney

Ben Finney <ben+unix@benfinney.id.au> writes:

[...]

> Or, to put it another way: Software freedom is something every user
> needs in all software, even if they have no intention of exercising that
> freedom at the time they receive the software.
>
> If everyone has the freedom to improve software and share the
> improvements, then a majority of people don't need to exercise that
> freedom to benefit from it. When no-one with skill is prevented from
> improving and sharing the software, then users only need to find someone
> with the skill willing to help.

The text below the pagebreak is of an entirely non-technical nature.


The problem with this is that this 'software freedom' thingie greatly
impacts the chances that the "why would I read the manual" guy who
wrote the buggy code, who probably wouldn't even be willing to go on
strike for my working conditions, and who very probably got his job by
meeting someone he knew from university who knew someone who knew
someone who &c, can make a relatively comfortable living by dumping
some buggy code he wrote on some set of 'end-users' who can do little
but accept it with obedient gratitude, be it only because such are the
conditions in the walled stronghold supposed to protect them from an
evil outside world they voluntarily built for themselves.

Without education, freedom is nothing but the freedom to chose a
different, surely benevolent, dictator for life.

On Wed, 08 Aug 2012 11:29:13 +0100
Rainer Weikusat <rweikusat@mssgmbh.com> wrote:

> The problem with this is that this 'software freedom' thingie greatly
> impacts the chances that [a programmer] can make a relatively
> comfortable living by dumping some buggy code on ...
> 'end-users' who can do little but accept it

Open-source systems are no different from closed-source systems in that
way.  

I am using a 4g tether to write this, connected via bluetooth to an
iPhone.  Periodically, all inbound traffic ceases.  Why?  tcpdump shows
consecutive unanswered arp requests for the interfaces's default
gateway amidst many more unanswered DNS requests.  

Question: which link-level layer is failing to deliver?  Is it the
bluetooth link, or the 4g?  

Can I get IP statistics, or 4g statistics, or bluetooth statistics on
the iPhone?  No.  Can I see *anything* about the underlying
connections, even the ID of the tower I'm talking to? No.  No iOS API
function provides access to that information.  

Link-level statistics would show whether Apple or AT&T is
failing to provide the service I'm paying for.  Without them, I'm at
the mercy of their so-called customer service, both of which are free
to point to the other.  At least at level-2 support at AT&T, they don't
seem to have access to those statistics either, not that they'd be able
to interpret them if they did.  

If you've checked the price of those companies' stocks lately, you'll
note they're making a comfortable living indeed dumping mediocre
service -- and, probably, broken code, but who can tell? -- on
end-users who can do little but accept it.  

--jkl

"James K. Lowden" <jklowden@speakeasy.net> writes:
> On Wed, 08 Aug 2012 11:29:13 +0100
> Rainer Weikusat <rweikusat@mssgmbh.com> wrote:
>
>> The problem with this is that this 'software freedom' thingie greatly
>> impacts the chances that [a programmer] can make a relatively
>> comfortable living by dumping some buggy code on ...
>> 'end-users' who can do little but accept it
>
> Open-source systems are no different from closed-source systems in that
> way.

[complaint about "large company customer service"]

I don't quite understand how this would be relate to anything I
wrote. Also, I didn't write 'a programmer' in my text and don't think
that the term you inserted into it would be synonymous to the
descripton I gave.

On Wed, 08 Aug 2012 17:24:01 +0100
Rainer Weikusat <rweikusat@mssgmbh.com> wrote:

> "James K. Lowden" <jklowden@speakeasy.net> writes:
> > On Wed, 08 Aug 2012 11:29:13 +0100
> > Rainer Weikusat <rweikusat@mssgmbh.com> wrote:
> >
> >> The problem with this is that this 'software freedom' thingie
> >> greatly impacts the chances that [a programmer] can make a
> >> relatively comfortable living by dumping some buggy code on ...
> >> 'end-users' who can do little but accept it
> >
> > Open-source systems are no different from closed-source systems in
> > that way.
> 
> I don't quite understand how this would be relate to anything I
> wrote. Also, I didn't write 'a programmer' in my text and don't think
> that the term you inserted into it would be synonymous to the
> descripton I gave.

I thought your point was that open source code allows lazy programmers
to make a comfortable living while to most end users the source code
appears to be in Medival Klingon.  My answer is the commonplace
that terrible software is everywhere and often comes without source
code.  

> [complaint about "large company customer service"]

No, not exactly.  The complaint is that iOS, because it is proprietary,
can hide information, and that in so doing *enables* poor customer
service.  If I had the statistics to demonstrate the problem -- if we
had them, because there are thousands of us -- the company at fault
would be forced to correct the problem.  Only by walling that off can
both companies smugly shrug and blame the other.  

--jkl

James K. Lowden <jklowden@speakeasy.net> wrote:
> I thought your point was that open source code allows lazy programmers
> to make a comfortable living while to most end users the source code
> appears to be in Medival Klingon.  My answer is the commonplace
> that terrible software is everywhere and often comes without source
> code.  

FdI#0815:

Closed Source Software:
    binary only (unmaintainable because of lack of source code)

Open Source Software:
    binary only (unmaintainable because of source code)

SCNR,
VB.
-- 
“Isabell steht auf Cocktailschirmchen, Analverkehr und House
und hat 186 Freunde
ansonsten weiss ich nichts über sie”
    (Jenna Gesse)

Volker Birk <bumens@dingens.org> writes:
> James K. Lowden <jklowden@speakeasy.net> wrote:
>> I thought your point was that open source code allows lazy programmers
>> to make a comfortable living while to most end users the source code
>> appears to be in Medival Klingon.  My answer is the commonplace
>> that terrible software is everywhere and often comes without source
>> code.  
>
> FdI#0815:
>
> Closed Source Software:
>     binary only (unmaintainable because of lack of source code)
>
> Open Source Software:
>     binary only (unmaintainable because of source code)

Another nice example of someone confusing "I can't do it" with "it
can't be done" ...

Rainer Weikusat <rweikusat@mssgmbh.com> wrote:
>> FdI#0815:
>> Closed Source Software:
>>     binary only (unmaintainable because of lack of source code)
>> Open Source Software:
>>     binary only (unmaintainable because of source code)
> Another nice example of someone confusing "I can't do it" with "it
> can't be done" ...

Get a life. We call that "humor", outside, where the pizza man comes
from.

Yours,
VB.
-- 
“Isabell steht auf Cocktailschirmchen, Analverkehr und House
und hat 186 Freunde
ansonsten weiss ich nichts über sie”
    (Jenna Gesse)

Volker Birk <bumens@dingens.org> writes:
> Rainer Weikusat <rweikusat@mssgmbh.com> wrote:
>>> FdI#0815:
>>> Closed Source Software:
>>>     binary only (unmaintainable because of lack of source code)
>>> Open Source Software:
>>>     binary only (unmaintainable because of source code)
>> Another nice example of someone confusing "I can't do it" with "it
>> can't be done" ...
>
> Get a life. We call that "humor",

And I didn't think it was funny. Actually, since it serves to
reinforce the wrong notion that there's a special, magical connection
between 'the code' and 'the guy who wrote the code' and consequently,
any user of the code necessarily needs this guy, I'd rather put it
down as 'evil' (genauergesagt, als den Versuch, eine Truebnis
herbeizureden, in der sich fischen laesst :-).

"James K. Lowden" <jklowden@speakeasy.net> writes:
> On Wed, 08 Aug 2012 17:24:01 +0100
> Rainer Weikusat <rweikusat@mssgmbh.com> wrote:
>
>> "James K. Lowden" <jklowden@speakeasy.net> writes:
>> > On Wed, 08 Aug 2012 11:29:13 +0100
>> > Rainer Weikusat <rweikusat@mssgmbh.com> wrote:
>> >
>> >> The problem with this is that this 'software freedom' thingie
>> >> greatly impacts the chances that [a programmer] can make a
>> >> relatively comfortable living by dumping some buggy code on ...
>> >> 'end-users' who can do little but accept it
>> >
>> > Open-source systems are no different from closed-source systems in
>> > that way.
>> 
>> I don't quite understand how this would be relate to anything I
>> wrote. Also, I didn't write 'a programmer' in my text and don't think
>> that the term you inserted into it would be synonymous to the
>> descripton I gave.
>
> I thought your point was that open source code allows lazy programmers
> to make a comfortable living while to most end users the source code
> appears to be in Medival Klingon.  My answer is the commonplace
> that terrible software is everywhere and often comes without source
> code.

The point I was trying to make is that 'software freedom' on its own
is of little use to anyone who is not a developer because most people
have no idea how they could profit from it. Consequently, they will
voluntarily surrrender their freedom in this respect to some (circle of)
'experts who are henceforth responsible for dealing with that' and
these experts have a natural self-interest in re-establishing
'software' as a 'natural monopoly'.

Rainer Weikusat <rweikusat@mssgmbh.com> wrote:
>> Get a life. We call that "humor",
> And I didn't think it was funny.

Sincere condolences.

Yours,
VB.
-- 
“Isabell steht auf Cocktailschirmchen, Analverkehr und House
und hat 186 Freunde
ansonsten weiss ich nichts über sie”
    (Jenna Gesse)

Volker Birk <bumens@dingens.org> writes:
> Rainer Weikusat <rweikusat@mssgmbh.com> wrote:
>>> Get a life. We call that "humor",
>> And I didn't think it was funny.
>
> Sincere condolences.

.... and apparently, it also wasn't meant to be funny.

On Thu, 09 Aug 2012 19:55:07 +0100
Rainer Weikusat <rweikusat@mssgmbh.com> wrote:

> The point I was trying to make is that 'software freedom' on its own
> is of little use to anyone who is not a developer 

Quite so.  Although there are non-developers who do profit from it,
some handsomely.  

There are valuable free-rider effects, too: many non-developers benefit
from their freedom indirectly, because others were able to isolate a
problem or add a feature.  

> Consequently, they will voluntarily surrrender their freedom in this
> respect to some (circle of) 'experts who are henceforth responsible
> for dealing with that' 

Well, you say "surrender".  An economist would call that the rational
exercise of comparative advantage.  

> and these experts have a natural self-interest in re-establishing
> 'software' as a 'natural monopoly'.

No they don't, not collectively.  Each one might like to have a
monopoly, sure, natural or not.  It's not easy, particularly because
the necessary conditions of a natural monopoly are rare.  But that's
really OT here.  :-)

--jkl