Re: brother printer/scanners
On 01/04/2017 11:27 AM, Brian wrote:
On Mon 02 Jan 2017 at 22:11:26 -0500, Jape Person wrote:
On 01/02/2017 06:50 PM, Brian wrote:
Could be. The license for printer and scanner drivers says you can
modify, alter, translate, reproduce and distribute their software.
Sounds good. Doesn't Debian say that too?
Yeah, I think I saw that scrolling by, too. The problem I had was trying to
figure out if they really were referring to every piece of the software, or
just to little bits. There were a lot of downloads with accompanying EULAs
in the text presented by the scripted install.
I imagine it's the binary-only libraries (or executables) used for
printing and scanning. Other programs are either shell scripts or
readable configuration files. Allowing modification of these binaries
is pretty useless without the source code, however. The license prior
to the present one didn't even allow this (except for debugging
purposes) or distribution of an altered binary.
I really don't understand what Brother et. al. expect to gain from this
obstinance. It's certain that the source or close relatives thereof is
"out there" and, one presumes, available. It's not the crown jewels or
trade secrets. It's just a tool that makes their hardware work.
Maybe I've gone about this the wrong way. I can see that an overworked
support division only looks at something like this as "more work"
instead of seeing it as an opportunity. But if Brother has folks with
technical expertise in their marketing division, surely those people
would see an advantage to having tech aficionados in the FOSS sector
examining, improving, and redistributing the firmware and software to
make the product work better. It would be as much a win for Brother as
it would be for FOSS.
Eh. It may be a waste of time for me to write them, but it'll keep me
out of my wife's hair for a few hours.
I stopped reading when I got to the Yoyodyne, Inc. bit. I have always glazed
over when reading legalese, suspecting that it is written by people who like
to confuse and make fun of those of us with no training in the finer points
of law. When I hit Yoyodyne (reference Buckaroo Bonzai) I knew for certain
that *someone* was being kidded. Since I was focused on testing, I let the
joke slide for another time.
Most people probably don't even get as far as you did. A Debian packager
would have to be more conscientious and careful. Which reminds me: the
so-called .deb packages aren't; they are converted rpms. Talk about
doing a job well!
Yep. Give the packing job to a couple of under-trained, over-worked
schmucks in the support group, and that's what you get.
The license also says there is no warranty to the software. They are
not liable for anything. Anything breaks and you get to keep both
pieces. Doesn't Debian say that too?
Yes, they do. However, whenever I break something from the Debian repos, I
always find a way to glue the pieces back together. I'm thinking that
restoring the Brother drivers or whatever they might break in the remainder
of the system might not be such an easy proposition. Their docs, such as
they are, would definitely not help in the process. The real reason I
preferred the scripted installer was that I didn't have to read Brother's
installation instructions. The installer guided me by the hand every step.
To its credit, the result was a device which worked with all of its bells
and whistles. Accomplishing that with the piecemeal package and policy
installations with their instructions was not so easy. There were many
mistakes encoded in just a few steps of those instructions. I had to
actually crank up my brain and reason that I really didn't have to connect a
usb cable during driver installation if I was preparing to use the device
solely as a networked multifunction printer.
Working (even in a substandard way) is not a problem. It is when
non-free software stops working that is a problem. Very often, the
finger is pointed at the Linux distribution, which is expected to fix
it somehow. A bug reporter insisting (usually without any evidence of
the nature of the interaction between free and non-free software) that
a change in Debian is the cause of printing or scanning failure is not
unknown. Vendors not keeping their offerings up-to-date wrt Debian is
not the norm, I would say.
Yes. I've actually always appreciated the approach many (most) Gnu/Linux
distros and the kernel team take when dealing with bug reports
concerning "tainted" systems.
It's not our problem is a perfectly reasonable response. They make their
source code available, after all. The purveyors of blobs want to win the
poker hand without showing their cards. In the real world you have to
put up or shut up. (Show me or bl*w me was the way the old boat chief on
the Skipjack used to put it.)
Some manufacturers, HP and Epson, for example, integrate their scanner
software and co-operate with SANE developers; Brother don't seem to,
although they do have drivers for some scanners which claim to be under
the GPL. No source code though.
Yeah, what's up with that? Don't the various GPL versions require the
source code to be made available if you're going to alter the software
and re-distribute it?
Furthermore, the license informs anyone who wants the source code for
the drivers to bugger off and not make stupid requests. Debian doesn't
do that.
The only reason I can imagine why they wouldn't want to make the source
available is that they don't want everyone reading it, pointing at them, and
laughing. But maybe I'm wrong. Perhaps it's written better than the
instructions. Scratch that. It has to be better written, or it wouldn't work
at all.
Anything can only be a guess. I suppose the software is written to do
what Brother wants it to do, so they are happy with it. SANE developers
have talked about a probable rewrite of any released source being needed
to fit in with sane, if only to accomodate non-Linux code. The same
would be the case for printing.
I would really like to use the device with xsane, but not enough to
install brscan* and its brethren.
For now, my wife's kooky little Android devices will be my scanning
interface. That works well enough.
Reply to: