[geeks] bridging networks with wireless

[Geoffrey S. Mendelson]

On Wed, Oct 18, 2006 at 03:03:21PM -0400, Charles Shannon Hendrix wrote:
> That makes absolutely no sense.
> 
> For one thing, the mpii interface allows you to do some illegal things
> without knowing anything about the chipset.
> 
> Otherwise, hiding the chipset doesn't stop you from finding out just
> by snooping your hardware or the driver that is running.

In the U.S. that might violate the DMCA. 

The restrictions are there to prevent the illegal useage of the device.
If you "hack" a driver to operate on illegal frequencies, use illegal power,
or transmit a different type of signal, then it's on you, not the
manufacturer.

> It's trivial to find signal boosting drivers, so it certainly hasn't
> stopped abuse.

I did not say it would stop the abuse, it prevents a normal person from
violating the law using normally available software. As for signal boosting,
I think that the "boosted" signals are still within U.S. specs. They are
illegal here, but that's another issue.


> Finally, if they were really worried about this, then why aren't the
> hardware limits stored in a ROM which the chipset references, rendering
> the card incapable of operating illegally in its given market?

So that they can sell the same card anywhere. It prevents someone from buying
a card over the Internet, or in the U.S. on vacation and then taking it
to somewhere else and operating it illegally. The driver can check the system's
location, if they even bother.

> > were capable of much hire resolution, but HP was afraid that they
> > would be damaged by poorly written drivers.
> 
> Yeah, right.
> 
> Oh, and don't forget, out of their deep concern for customer print
> quality and print head damage, they also chip their cartridges.
> 
> Surely you don't believe such obvious lies as this?

Obvious lie? How can you say that without knowing the structure of the 
printer? The point was it did not print a simple bit map directly, there
was some sort of "fudge" to cause the print head to move a half step and then
print a partial dot. 

It would not surprise me that the driver developer's had a pile of broken
printheads and HP decided that it was better to restrict the drivers than
to expect people to write good ones. 
 
> If you want to see real damage, look at all the real problems that have
> been caused by the lack of hardware information and chipped cartridges.

How is that damage? No one was forced to buy an HP printer. If you did not
like the support policies of HP, then you could buy a different brand.

If you have an issue with chipped cartridges and closed source drivers,
there are other printers out there. I expect that your average Windows
user really does not care. They are not going to use open source drivers,
they are not going to use refilled cartridges and they are not going
to refill their own.

Note that chipped printers are legal in the U.S. and it is illegal to
reverse engineer the chips. In the E.U., as of 1 Jan 2008, chipped printers
will be illegal.

As for open source drivers, that does not make them good. There is a lot
of really bad source code out there. Some is open source, some is closed
source. While Linus Torvalds exerts some control over what eventualy goes
into the Linux kernel, some things just get stuck in other programs 
with no testing at all.

I upgraded one of my systems to a new release of something and found out that
it had been released with an untested patch included at the last minute.
I applied the "fix" which did not work, and then went in and fixed the source
code myself. 

The website for bug reports did not work that day, and an email to the author
went unanswered. It may also have been because I asked about a much bigger
bug, which is not being fixed and is too much work for me to bother with.

I have a bug fix in for Mozilla Calendar that is almost two years old. 
It's a simple fix, they have to not write carriage returns on exported
iCal files. With the carriage returns (generated by the Windows version)
the files can be imported into another Mozilla calendar, but not iCal.
What is the use of an iCal format file that iCal won't read?

On the other hand I once had to read from the Postscript developer's manual
to a Vice President of Xerox to explain to him why his printer would not
print Postscript files of scanned images properly. This was not a $3k
printer, it was a $150k printer.

He did something about it, and I was told that it would be fixed in the
next release, about six months away. I was a consultant and left the
client before then. Meanwhile we found it cheap and easier to print the
files on HP printers, and let the Xerox printer work as a copy machine.

Geoff.
-- 
Geoffrey S. Mendelson, Jerusalem, Israel gsm at mendelson.com  N3OWJ/4X1GM
IL Voice: (07)-7424-1667  Fax ONLY: 972-2-648-1443 U.S. Voice: 1-215-821-1838 
Visit my 'blog at http://geoffstechno.livejournal.com/