[geeks] bridging networks with wireless

[Geoffrey S. Mendelson]

On Wed, Oct 18, 2006 at 10:38:21PM -0400, Charles Shannon Hendrix wrote:

I said:
> > In the U.S. that might violate the DMCA. 
> 
> I can find my chipset by looking at the driver disk or looking at my
> system information (in most cases), or I can use a PCI tools utility.
> 
> It would be hard to file a DMCA lawsuit to protect information you
> freely published yourself.

I did not say that it made sense. What I said is that it may violate the DMCA.
It may also prevent you from publishing that information and the manufacturer
could sue you for that. I don't know of any that would, but who knows?

> Well, if *I* were going to prevent that, then I'd make sure the user
> could not trivially get the chipset information.

They might. Will someone get on M/S case and then when you click on device
properties see "D-LINK chipset" instead of what you see now?
 
> All I'm questioning is painting the chipset and leaving the chipset out
> of the documentation, mainly because it makes buying the cards a pain in
> the ass, even for Windows use.

That's true. I see that all the time. 


> In other words, this stuff they do only prevents a user from knowing
> what is in the card *BEFORE* he buys it. After that, it serves no
> purpose, which is why I question it.

No, it prevents the average user from knowing what the chipset is and in
some cases the reverse engineer from easily seeing it. Some chipsets are
not easily found out.

> The information obfuscation I'm talking about only lasts until you buy
> the card.

Assuming you can get the chip IDs. As I said, some chips you can tell
via the hardware display, some you can not. RAMDACs are hard to find and
critical to a display card.

> Even if I'm a Windows user, I might want to avoid a particular chipset,
> and it is maddening that you can't get the information before buying the
> card.

All the more reason to hide it. :-)

> This is often true of other cards too, but is far worse with wireless,
> and I still say it serves no purpose.

It permits them to change chipsets as they go along without anyone returning
a card because the picture says chipset A, and now it's chipset B.

> Even in Windows, I often load alternate drivers because they are better
> than the manufacturer drivers.

The best example of that is Epson drivers, where Epson was very open about
the specs of their printers. One of the people on the GIMP team wrote drivers
so good for the Epson printers, they abandoned theirs and use his.

> Not everyone can choose their hardware, and not everyone is even aware
> there are things like "drivers" under the hood. All they know is that
> things either don't work or their printer is damaged.

Then why would they care? They would install the CD that came with the printer,
or use the driver included in the operating system. They would not download
third party drivers, install them and hope for something better. More likely,
if they bought the printer with their computer, the driver would be preinstalled
by the packager. 


> I can't always just switch brands because of a stupid policy like this
> in all cases. 

You can't always choose your vendors. If you want something that they have,
you may be stuck with it. I don't think there are any U.S. printer vendors
without chipped printers.


> In the inkjet market in particular, there really aren't many alternative
> choices.

I don't know. I buy HP because they have good support and the supplies are
readily available. If I were in the U.S., it might be a different story.

My wife does most of the printing here, and she just wants it to work. At
work they have a problem with a laser printer and could not get it to stop
hanging. Eventually they started running it with CUPS and had some problems
that were solved with a web search.

What was frightening is that their sysadmin (who works for a contractor),
found the answers. In every case the author of the posting was I. :-)

> Case in point: many (most?) of the Linux and BSD drivers are *BETTER*
> than the Windows drivers.

Some, especialy the Epson drivers. Most HP drivers are "great" because they
use Ghostcript and HP's driver. HP put a lot of effort into developing that
add-on for Ghostscript, the Ghostscript developers put effort into creating
the API, and due to a lot of internal lobbying, HP opensourced the drivers.

> In other words, just like almost all software.

Yes.


> Basically you are stating that software has bugs, and the industry is
> not very good at fixing or avoiding the problem.

Yes.

> 
> Wether it is opened or closed is rarely relevant with one exception: if
> the code or the specs are open, there is at least a chance someone else
> can fix it.

That may or may not be true. Probably 99% of the users can't. My point was
that open source does not make something good. It may enhance something,
or it may stifle development because no one will fund it. 

A friend of mine has an idea for  "killer app" for Open Office. It will never
go anywhere because no one will fund it. It's too big a project for people
to tackle at home.

I keep saying he should pitch it to SUN for inclusion in Star Office.

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/