My group maintains and enhances an operating system. Obviously, we need full access on the machines we debug on. We also have separate "production" machines used for builds and source control where developers don't default to having admin privileges (and admin privs are generally reserved for the people less likely to break things). We used to give all the new developers admin privs from day 1, but that almost led to a few disasters (new people with full admin privs on an unfamiliar OS is not a good idea).

We generally try to let the admins take care of the production systems and only take over when they aren't around (it's only two people). And we let them know what we fixed because we appreciate the fact that they are normally dealing with it for us...

OTOH, one doesn't need any sort of special access to develop simple applications on decent operating systems like Unix or Max OS. One only needs special access when one starts installing shared libraries, doing kernel work, or setting up shared source control systems (although, it's generally not a good idea to let all the developers have uncontrolled access to the source control system, either).

I started using SPF because the backscatter from spammers forging my domain was getting to be 5-10 times more than the amount of spam I was getting. The backscatter stopped almost completely and it stopped immediately. Every once in a while I get a small burst of backscatter, but it doesn't last long.

I don't know this for sure, but I suspect that the spammers are checking for SPF before using a domain for forgeries. It would make sense, because using a domain with SPF records for spam makes it possible for anybody to determine it's spam. In particular, if any tier one suppliers are using SPF combined with mail volume to identify spam, they could spot the spam almost instantly -- no wait for complaints to come in. In particular, the spam could be spotted quickly enough to shut down the sender. It probably doesn't happen that much, but if one was sending spam, why would one forge a domain with an SPF record when there are so many others out there with no SPF record.

Ummh. I've been using hard drives for about 30 years, now. And the manufacturers have always used base 10 for the capacity. I've never seen them use base 2. Same for tape.

Hate to break this to you, but computers originally used base 10. Mainframes support base 10 and base 2 arithmetic. Legacy operating systems tended to print disk and memory sizes in base 10. You don't see "ls -l" in Unix printing the number of blocks in a file in Octal or hexadecimal.

This whole concept of 1K = 1024 didn't come about until microcomputers hit the scene. And it didn't happen because of some grand intellectual revelation. It came about because multiply was hard to do in early microprocessor assembly code.

10KW is kinda high -- unless everybody live in mansions. I figured I could fairly easily cover the A/C in my house with panels on half the roof (the half the points south). It required the good panels (16-18% efficient). One thing to remember is that peak A/C demand coincides with the max amount of sunlight.

I know all about the breeder reactors and other technology we have no plans to ever build. It is totally irrelevant because we have NO PLANS TO EVER BUILD IT. And it's not because of politics, it's because of cost and risk (read your own Wiki reference). Nuclear plants already cost more to build than most other technologies (PV being the exception) and that's without taking into account the true cost of fuel disposal/reprocessing or taking into account they get free liability insurance from the government (which all the other technologies have to pay for). If they had to actually reprocess the waste, the cost might be much higher.

I would hardly call the amount of nuclear waste being generated as minuscule. The entire (huge) Yucca mountain facility is already over subscribed! The total amount we are generating with existing plants is more than the planned Yucca mountain facility is capable of holding (DOE info). Your own references states there are "thousands of tons".

And that's if you buy into the fantasy that something that is dangerous for more than a million years can stay safely buried for that period of time. The "design" of the Yucca facility included estimating how long it would take for the containers to break down and how long after that it would take for the nuclear waste to leach into the groundwater. Given the amount of experience we have at tracking pollutants leeching into groundwater (a few decades), I would expect those estimates to be about as reliable as a Ouija board.

The fundamental problem here is that the entire nuclear industry and the nuclear part of DOE are still operating like it was 1970. They pretend the waste problem doesn't exist (hence, no reprocessing technology deployment) and use secrecy to hide what they are doing to avoid accountability. As long as that's the way they are going to operate, they have to be stopped.

Nuclear waste remains dangerous for a million years (DOE information). The only person who would make the statement it can be "easily contained and controlled" either hasn't researched the problem or has never actually implemented anything that works.

You are actually underestimating how long nuclear waste remains dangerous: The Yucca mountain repository license application requires them to consider the nuclear waste dangerous for a million years.

At this point, all the nukespeak people will jump in with talk of breeder reactors and all kinds of new technology. Unfortunately, nobody is even thinking about deploying new technology.

All the DOE and the nuclear industry want to make are the same old 70s reactors they have been building (or trying to build) for the last 40 years. The industry approach to nuclear waste is still "Bury it and forget about it." It probably won't leak soon enough for them to catch me.

And absolutely nothing has been done to solve the nuclear waste problem. My claim is that's because it can't be solved.

And again, we pretent nuclear waste doesn't exist!

Only if you take the classic Nukespeak tactic and forget about the waste disposal site... Perhaps we can store it all in your house? Oh, wait a minute, do you live on the same continent that I live on?

But what to you do with the nuclear waste that is dangerous for a million years?

This engineer doesn't think that disposing of nuclear waste is a solvable problem.

For starters, it isn't dangerous for tens of thousands of years. It's dangerous for more than a million years (source -- the licensing background documents for Yucca mountain produced by the DOE).

Even if it was only dangerous for tens of millions of years, we don't have a clue how to build anything that lasts that long. Remember that engineering is applying proven technologies to solve problems. Right now, the longest lived structures in the world are around 5000 years old (the pyramids) and they have already failed to serve there intended purpose (protect the graves of the builders). It stretches one's credibility to claim we can store something safely for 1000 years, let alone a million!

The best we can do is make an educated guess and I wouldn't trust it to be very good...

I have a BSEE from Cornell (class of 74) and I am employed maintaining the kernel TCP/IP stack for a commercial operating system (and have been for about 10 years). I have modified the Ethernet drivers and the TCP/IP protocol layers for enhanced performance and I am currently starting to work in drivers for 10G Ethernet. I am quite familiar with the entire data path from the hardware right on up to the socket API.

Perhaps it's your "certs" are lacking. I don't have "certs" I have 30 years of real world experience.

These new lawyers are clueless about the technical issues around the trap and trace claim (of course it looks like the RIAA lawyers are , too). It's illegal to record traffic on a network you don't own (i.e., the Internet). It's not illegal to monitor the network traffic within your own machine or even on a network you own. By the complaint's own description, the traffic was to (and therefore eventually within) the receiving party's computer. Look at this another way: If what Media Century did is illegal, then it's illegal to use TCP on the Internet PERIOD, because there is no significant technical difference between formatting the downloaded traffic into an ASCII dump and formatting the downloaded data into a music file. It all happens on the receiving party's computer. Oh, and on top of that, the TCP software in the receiving computer needs to know and use the IP addresses from both ends of the connection or the connection won't even work. It's pretty clear that what the trap and trace law applies to is a third party capturing traffic they aren't a party to...

Another implication of these claims is that it would be illegal for ISPs to log any IP addresses for later use by law enforcement.

The detective issue is a little more interesting. It could be thrown out because it's involved with Interstate Commerce, but if not, it means it will be all but impossible for anybody other than law enforcement to investigate spammers and other kinds of computer fraud. And we know how ineffective law enforcement is at solving computer crime. It will be a really really bad thing if this sticks.

Oh, and in case you didn't notice, they are also arguing that the RIAA should be legally bound by ZaZaA's shrinkwrap notice!

And all of this to protect some shlub who believes musicians should starve to death rather than get paid for the fruits of their labor.