Re: (no subject)

2009-08-18 Thread David Liguori



Rahul Sundaram wrote:

On 08/18/2009 09:24 AM, gil...@altern.org wrote:

  

Unfortunately, the legal approach of Mr Sundaram still doesn't answer this
enigma: if proprietary document formatting is just defining an awfully
more complex and secret way of producing bold than , then what are
media codecs exactly?



Already answered in the FAQ. Read it carefully.

Rahul

  
Media formats almost always involve lossy compression  There are reasons 
for selecting one lossy compression scheme over another that are very 
different from considerations of document formatting, which do seem to 
be primarily a way of keeping things proprietary.  Lossy compression is 
still very much a topic of ongoing research.  If someone comes up with a 
clever scheme for lowering the bit count while minimizing the subjective 
loss, that person could reasonably claim it as intellectual property.  
I'm not plugging the proprietary software distribution paradigm, just 
saying.


Of course, media formats are also entangled with "digital rights 
management",  which has everything to do with the issues addressed in 
your FAQ.  Generally these work against usability, data efficiency and 
quality. 

I won't touch here the debate over music wanting to be free vs. 
musicians wanting to be paid and the extent to which DRM accomplishes 
the latter, other than to point out that it is frequently and plausibly 
argued that musicians are pretty much the last people DRM protects, if 
at all.


--
fedora-list mailing list
fedora-list@redhat.com
To unsubscribe: https://www.redhat.com/mailman/listinfo/fedora-list
Guidelines: http://fedoraproject.org/wiki/Communicate/MailingListGuidelines


Re: "Blinking lights of death" ? Netgear Switch GS108

2009-05-04 Thread David Liguori



Aldo Foot wrote:

On Wed, Apr 29, 2009 at 7:23 PM, Robert L Cochran
 wrote:
  

Has the device firmware been pharmed or simply partly flashed and then a
power failure struck?

If you would like to donate the unit to me, I'll try to find the time to
take a look at it in the next few months. I'm still very much an amateur,
and I'd like to try analyzing why the unit is not working and see if I can
fix it. This assumes physical damage to the circuit board of some sort.


<...snip>
  

The only way to be sure it is capacitor plague is to gut the unit and look
at the PC board. If you look at the capacitors in it, and one or two are
slightly bulged, or have black stuff that looks to be leaking from the
bottoms, the device is dead, and should be replaced, unless you are the
hacky type and change out the cap's or harvest the good parts off the corpse
of the old switch.


~Seann
  


The unit has a blown capacitor, bulged and brown matter around it. Also there is
a white-ish powder under the PCB. It looks dirty. There is corrosion.
Not sure it's
worth fixing. It went on for well over a year working perfectly in a
well ventilated
room. The unit is for a server at my workplace, so they will replace it.
Many years ago, I used to make and fix network cards and switches at some
company, so I'll give it a shot myself at fixing this one just for kicks.

I very much appreciate your offer to fix this thing --even though there was no
warranty implied. :-)

Now, I've got to get a more reliable unit to last longer.
~af

  
One of the more common mechanisms for failure of electrolytic capacitors 
is too high an ambient temperature over a period of time.  Usually the 
temperature rating is on the cap.  You say the room is well-ventilated 
but that doesn't rule out too high an ambient temperature in a room full 
of equipment, especially if it was sitting on top of or in a rack full 
of other equipment.  It's more likely to have open rather than short 
circuited.  An ESR (effective series resistance) meter will tell.


If you're pretty sure the capacitor is what's ailing it and it's 
through-hole rather than surface mounted, I would consider it well worth 
fixing, or even trying if there's greater than a 10% success 
probability.  Many 8-port switches aren't worth fixing below that.


--
fedora-list mailing list
fedora-list@redhat.com
To unsubscribe: https://www.redhat.com/mailman/listinfo/fedora-list
Guidelines: http://fedoraproject.org/wiki/Communicate/MailingListGuidelines


Re: Maple on Fedora

2009-03-23 Thread David Liguori



D. Hugh Redelmeier wrote:

| From: Hiisi 

| Maple is not open source (free). It was main argument for me to choose another
| symbolic arithmetic program - maxima ( maxima.sourceforge.net ). It's
| brilliant.

My daughter bought Mathmatica for a project perhaps five years ago.
She ran it on RHL9 or something like it.  Anyway, she would have to
relicence it to move it to a newer distro release or machine so she still
runs it on the same ancient machine.

If she had bought it for Windows XP, it would have had a much longer
lifetime.

Lesson: proprietary licensing models work even worse on open platforms
because open platform have a tradition of binary obsolescence.

Perhaps Maplesoft has a better upgrade model than Wolfram.

The only symbolic algebra package I ever bought was muMATH for the
z80: http://en.wikipedia.org/wiki/MuMATH
I never got it to work due to its kind of copy protection.  There's
a theme here.

PARI/GP worked well for my modest needs but it would not work for
yours.

  
This has been my experience as well.  Maple and Matlab both ostensibly 
can be installed on Linux but in practice it's difficult or impossible.  
I tried years ago, gave up and run them both in Windows.  Open source 
substitutes sometimes do what I need and sometimes not.  If I have time 
I play with them.  In principle I like Linux and open source, but am by 
no means ideological about it.


--
fedora-list mailing list
fedora-list@redhat.com
To unsubscribe: https://www.redhat.com/mailman/listinfo/fedora-list
Guidelines: http://fedoraproject.org/wiki/Communicate/MailingListGuidelines