[linux-elitists] BBC: Computers 'must become greener' (fwd from eugen@leitl.org)

Eugen Leitl eugen@leitl.org
Mon Mar 8 03:48:33 PST 2004


----- Forwarded message from Eugen Leitl <eugen@leitl.org> -----

From: Eugen Leitl <eugen@leitl.org>
Date: Mon, 8 Mar 2004 12:46:28 +0100
To: Joakim Ziegler <joakim@avmaria.com>, 31337 <linux-elitists@zgp.org>
Subject: Re: [linux-elitists] BBC: Computers 'must become greener'
User-Agent: Mutt/1.4i

On Mon, Mar 08, 2004 at 01:10:11AM -0600, Joakim Ziegler wrote:

> Of course, refrigerators and (especially) cars weigh a lot more than
> computers, and there are a lot more of them in the world than there are

It's not the weight. The bulk of resources are needed for semiconductor
fabbing. It's there where you can make the biggest impact by going to
molecular nanoelectronics (precursors you'll see in OLED/polymer printable,
etc).

Most cars now come with multiple computers, ditto househould appliances (not
refrigerators, yet, at least not most of them).

> computers. Also, sure, some areas in the world are short on water, but
> others are not, so why is "1.5 tonnes of water" significant? And what's

Because it's potable water. You need high-quality water for the reverse
osmosis/microfilter plant, and it typically generates a lot of wastewater.

Potable water is scarce about *everywhere*.

> with the "chemicals"? Wow, the scary, generic chemicals, they must all

What is up with programmers? Have they been flunking their chemistry in high
school? It can't take anyone more than half an hour to research the
production pathways for the basic components in a computer, and their
environmental impact.

I mean, really. Especially elitists.

> be bad! Also, who knows that the power you use is produced by fossil
> fuels?

I don't understand what you're getting at in that sentence. Most power today
is produced from fossils, or nuke, or similiar unsavory sources. Which is why
green computers burn less W/ops.
 
> I'm not saying that there's no environmental impact from computer
> production (although, as I understand it, the major environmental impact

Well, duh.

> from computers comes when they're scrapped), but this article is pretty

Hint: environment impact is *always* calculated over entire lifetime,
scrapping and operation costs included. Total cost of ownership, and all that.

> bogus, and seemingly designed to stir up environmentalists against the
> evils of computing. They even throw a random SUV reference in there,

I'm surprised you aren't aware that semiconductor production is not
environmentally neutral. Quite the opposite, in fact.

> just because everyone knows SUVs are bad.

Yeah, that was lame. I guess they just attempted to come through even those
with borderline flat EEG.
 
> > I don't see why any normal user would need a new computer these days.  
> > You can easily buy used brand name computers on online from the
> > manufacturer for as little 10% of the cost of a new machine.
> 
> There's also too little information here about what we can do to avoid
> this. "Never buying a new computer" is not an option for me, nor is it
> for most people, I imagine. The most long-term reusable part of a

*nix elitists typically don't use the bleeding edge, and have abnormally long
ownership cycles. You certainly can't prevent gaming addicts always buy
latest in greatest, so you have to regulate environmental production (which
is all outsorced to piss-poor countries with no or little environmental
accountability, anyway), and uses.

Recycling old-generation Windows PCs by installing open source *nixen on them
is actually a very good way of limiting environmental impact. It's not the
juice burned during operation, it's the production and scraping costs. 
So the big lesson's there's value in prolonging a life time of a PC. It's a
good idea to use these 1 GHz VIA fanless processors on miniITX boards, 
and 12 V external power supplies in new machines, where raw performance is
not an absolute requirement. Etc.

> computer is the case, and that's probably not the most environmentally
> unfriendly one to produce.

You're accurate. Notice that power supplies tend to have icky flame
retardants in there (and some power semiconductors to boot).
 
> So, does anyone have data on what parts of a computer pollute the most?
> What should one avoid? I know there's a lot of bad crap in CRT monitors;
> are LCDs better? And so on.

There's oodles of lead and semiconductors/flame retardants in a CRT, and CRTs
burn a lot of juice over the operation.

I would have to look LCD production (active matrix means the entire panel is
an array of transistors), but it's on glass, and large-structure size, so
it's not nearly as bad as an 300 mm wafer full of amds. The
backlit display contains some few mg of mercury vapor (next-generation 17"
LCDs will be LED-lit).

I suspect that OLED displays will be very very benign environmentally, even
if they're to use semiconductor nanoparticle (CdSe or CdTe typically) inks.

-- Eugen* Leitl <a href="http://leitl.org">leitl</a>
______________________________________________________________
ICBM: 48.07078, 11.61144            http://www.leitl.org
8B29F6BE: 099D 78BA 2FD3 B014 B08A  7779 75B0 2443 8B29 F6BE
http://moleculardevices.org         http://nanomachines.net



----- End forwarded message -----
-- Eugen* Leitl <a href="http://leitl.org">leitl</a>
______________________________________________________________
ICBM: 48.07078, 11.61144            http://www.leitl.org
8B29F6BE: 099D 78BA 2FD3 B014 B08A  7779 75B0 2443 8B29 F6BE
http://moleculardevices.org         http://nanomachines.net
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
Url : http://allium.zgp.org/pipermail/linux-elitists/attachments/20040308/6280c255/attachment.pgp 


More information about the linux-elitists mailing list