Can The Proprietary GPS Daemon Be Removed?

Richard Franks spontificus at gmail.com
Mon Dec 4 13:55:50 CET 2006


On 12/3/06, Dave Crossland <dave at lab6.com> wrote:
> On 03/12/06, Richard Franks <spontificus at gmail.com> wrote:
> > I'm happy to support open source development where appropriate. I
> > think in this case though, reverse engineering a proprietary driver
> > for something as single-purpose as a GPS chip is overly aggressive.
> > And to what end?
>
> The point of Free Software is to drive proprietary software out of
> existence. I'm sorry to hear that you've been told it was anything
> other than that. That is the reason GNU/Linux exists, and will
> continue to exist.
>
> If this seems strange, I recommend reading the essay "Why Software
> Should Not Have Owners" at
> http://www.gnu.org/philosophy/shouldbefree.html
>
> It explains why all proprietary software is wrong, and why we cannot
> tolerate it if we're to live in an ethical, sustainable, friendly way.

I am aware of the arguments - I simply, respectfully, disagree with
all-or-nothing idealism. I would love to live in a reality where the
goals of the FSF were fully reached, but we have to deal with the
reality we're dealt too!


> > If it is without the cooperation of Global Locate,
> > then surely they can switch off AGPS access if they choose?
>
> Please could you re-explain this - I don't understand enough about how
> AGPS works to understand what you mean.

It's my understanding that Global Locate undertake to supply
positional data for the AGPS systems they make.


> > Even with
> > their blessing, how long would we expect it to take to improve upon
> > the efforts of their expert full-time development and QA staff who
> > surely take into account balancing not only individual unit
> > performance but that of the overall network?
>
> Imagine someone arguing that creating a society where we have free
> elections, where everyone's allowed to vote, isn't worthwhile because
> it will take a long time to overthrow a tyrant dictator and set up a
> senate/congress/parliment.
>
> They would be missing the point.

Skipping dictators/ideals/voting - the point is how long would it take
to reverse-engineer and develop a new GPS driver, test it, obtain
blessing from Global Locate? If you think longer than a month, then
for the first release, it's a moot point as it's simply not going to
happen.


> > What if we used the open-source replacement and accidentally crashed
> > their servers in the process? What if someone else crashes their
> > servers on purpose?
>
> It is naieve to expect that if you offer a network service to the
> public, no one will act in bad faith. Spammers and crackers will come
> knocking. Therefore I would expect anyone offering network services to
> maintain good security practices.
>
> Efforts to create a good-faith Free Software client for a network
> service should therefore be unable to crash their servers.

FC6 download? ;-)


> If they need to write better servers, they need to write better servers.

Traditionally GPS devices exist in closed-embedded environments -
which mitigates security concerns somewhat. I think Global Locate
deserve credit for seeing the potential of what the
OpenMoko/Linux/GPS/Phone combo could acheive.

I know of a lot of small companies who live or die based upon 1-2
low-margin hardware products which they develop.


> Please explain how it is possible to stop someone reverse engineering
> a protocol and writing a Free Software implementation.
>
> I don't believe its theoretically possible.

The difference between theory and reality is that in theory there is
no difference between the two, in reality there is.

Otherwise we'd all be running open-source ATI drivers, no?


> > Supporting open source initiatives does not mean that one has to
> > methodologically seek to remove or replace all instances of closed
> > source.
>
> I'm sorry you've misunderstood what's going on here. That is precisely
> what the GNU/Linux operating system developers are doing.
>
> "The idea is not just to produce a scattering of free programs that
> were nice to use. Rather, the idea is to systematically build free
> software so that one can escape completely from non-free software.
> Non-free software is basically antisocial, it subjugates it users, and
> it should not exist."
> - http://www.zmag.org/content/showarticle.cfm?ItemID=9350

The point at which it starts approaching religious fundementalism, is
the point at which I jump off the train.

I support open source, and I'd love to see every bit of software free.
But at the end of the day I don't publicly upload all of my companies
proprietry software, because in our reality, that would be unethical.
When you start talking of ethical absolutes, regardless of social
consensus, you're talking religion.


> When a developer makes proprietary software available to the public,
> the code and algorithms are published. If a developer wishes for them
> to be secret, they are foolish to make them available to the public.

This is a belief that no company should be allowed to make money from
selling software that they create. There's a difference between the
NTP patent trolls, and a company whose bread and butter is the one or
two bits of hardware they sell.


> Now, its true that its wrong to break the terms of an agreement. But
> it is also wrong to refuse to share with your friends.

Your house? Your spouse?


> If Free Software doesn't exist for what you want to do, you can write
> your own if you know how, or pay someone to write some for you. If
> proprietary software is made available to the public, reverse
> engineering it to write Free Software replacements is inevitable.

Still waiting for those ATI drivers ;-)


> > "However, straight up, while the proprietary GPS daemon is included by
> > default, or in fact recommended/mentioned by OpenMoko, its not going
> > to be endorsed by the FSF:"
> >
> > What good is freedom of software, if it comes at the cost of freedom of choice?
>
> Some choices are ethical, some are not.

Is it an ethical choice to proactively restrict the free flow of
information, including that which relates to obtaining closed-source
software?


> > So any product with a FSF Endorsement is therefore unable to even
> > mention an alternative closed-source replacement which may provide me
> > with better functionality?
>
> The primary goal of Free Software is to use ethical, sustainable,
> friendly software. It is counterproductive to use non-free software,
> even if it provides the illusion of better functionality today,
> because it is unethical and thus unsustainable in the future.

Since there isn't an open-source GPS driver for the chipset in
question, it's not an illusion that the closed-source driver will
provide 'better functionality'.


> > I have to say, as far as PR in the market-place of ideas goes, it
> > would make me think twice about picking up a phone with a FSF
> > endorsement, as my first thought would be.. how hard do I have to hunt
> > if I want to use this phone to its full potential?
>
> If the phone contains non-free software, it cannot be used to its full
> potential.

If it didn't contain any non-free software then GSM/GPS wouldn't
work.. I don't think it's unreasonable to reach a goal or an ideal, by
incremental steps.

Cheers,
Richard




More information about the community mailing list