Wait... you want new stuff, but you don't want to get new stuff? Isn't that what "upgrade" kinda, you know, means? "Stable" and "new" are very rarely coincidental... there's a reason they don't throw every new widget into a stable distro... because that would make it UN-stable.
I'm thinking of having two "layers", if you will: One would be the kernel, libc and other keys libraries, etc., the things that make my hardware a system. The other would be the apply goodness. The idea is that apply goodness could/would be supported on multiple cores, which would remain stable. Sure, more testing, more regression, perhaps impractical.
A stable core with a lifetime of more than 6 months, rolling releases of apply goodness.
I cannot upgrade my LTS server because I had the temerity to install non-Ubuntu packages (might have been hylafax, I simply have not investigated): The upgrade tool detects the non-Ubuntu package and stops dead. I am stuck with an LTS server that will eventually be unsupported and that will be a real pain to upgrade, because of what seems to me to be a lamo upgrade tool. Glad I got that off my chest.
ooo-kay, time to scrape egg from face. Replies to my post motivated me to investigate. Turns out it was a bug in the upgrade package, since fixed. Didn't catch it at the time because I couldn't figure out at the time that it was a bug (couldn't find the evidence), no one else had caught it yet (guess I jumped on the server upgrade much earlier than others) and the original message blamed the failure on the presence of a non-Ubuntu package.
Now, weeks later, it works, the bug having been found by more resourceful folk than me. Mea culpa
Umm... you could just uninstall the unsupported packages, then install new ones afterwards. Dependency tracking is the wonder and magic of APT, and you would get that with any sane distro
Good idea, and maybe that will work in my case...
...unless the package isn't part of the distro. Like I wrote, I haven't investigated, because having an operational server that does what I need is more important right now than upgrading. It's a priority thing. (Just for fun, I checked packages.ubuntu.com and hylafax is there, so it probably isn't hylafax blocking me.)
What else do I have on this box? Hmm, let's see, I set it up when Dapper was released, added a couple of things I wanted - and I'm pretty sure not all were APT-based - and haven't really touched it since. So, no, I don't remember. It works, I leave it alone, we get along great.
Sometimes loony ideas are exactly what the world requires - they're the ideas that result in the biggest, baddest, bestest change.
I can hardly stand Ubuntu because it has stale packages
I've been using Ubuntu since Dapper pre-betas, and I tend to agree with you: The "everything new goes into the next release" approach means that I have to wait for upgrades to get cool stuff ('cause there is always cool stuff 'round the corner) and it also means I have to upgrade even if I don't want to, because of the cool stuff!
No, I don't believe that I'm contradicting myself: I don't want to upgrade, because I like stable, but I want the cool stuff, so I have to upgrade. Sigh. It is the Ubuntu release cycle that forces me into this.
That said, I don't plan on switching distros, 'cause I love Ubuntu: Love the mindshare, the community, the approach, the distro itself, etc. It's good stuff. But I would like this pet peeve of mine improved.
And OK, while I'm at it: I wanted BackupPC 3.0 on my LTS server a looong time ago, and was really looking forward to 8.04 LTS, 'cause I knew the upgrade path would be smooth and BackupPC 3.0 and a lot of other goodness would be there! Yay!
Guess what? I cannot upgrade my LTS server because I had the temerity to install non-Ubuntu packages (might have been hylafax, I simply have not investigated): The upgrade tool detects the non-Ubuntu package and stops dead. I am stuck with an LTS server that will eventually be unsupported and that will be a real pain to upgrade, because of what seems to me to be a lamo upgrade tool.
Glad I got that off my chest.
I Love Ubuntu.
But I want a release management plan that combines a stable base with rolling releases of the cool stuff. Don't know how this would work, but it seems like a really good idea to me.
Westjet flies 737s, employs dual jetways at [some] airports, opens the front and rear doors and move stairs in [at other airports].... It really does work better.
Westjet has also experimented with various boarding orders, such as, in no particular order:
Back of the plane first
Front of the plane first
Odd number rows first
By shoe colour
By sock colour
Full general boarding without ordering
And more
IIRC, they determined that the best, fastest way to board a plane was (wait for it!)
Full general boarding without ordering
Primarily because the fastest, most motivated, most aggressive, most whatever people get up first, hustle into line, and hustle into their seats. Slower folks gravitate to the end of the line, meander onto the plane, etc. The fast and efficient folks are in front, the slow folks take their time and aren't in anyone's way, and the plane loads quickly.
The mathematician has an interesting idea but fails to account for the human element: The old lady in 15F is going to take her time getting to the in-terminal seating grid, take her time on the gangway, take her time in the plane, and need assistance from a flight attendant to stow her bag, check her seat, and ask the folks in 15D and 15E to move ('cause she's too reserved to do it herself). Best to leave her to last, so she can get the assistance she needs and not be in anyone's way.
And with nothing but general boarding, that's what happens on Westjet flights.
Of course, it helps greatly that Westjet attendants are helpful, friendly, warm, welcoming, and have great senses of humour.
No, I am not an owner. I'm a very satisfied customer who flies Westjet whenever I can, loathes Air Canada with every fibre of his being, and thinks the best thing that could happen to AC would be for Air Austria to buy it out, fire every single last AC attendant, and replace them with AA employees. Now there's an efficient, helpful, airline. But that's a story for another day.
Here I was, planning to mod this discussion, but I can't believe what you just wrote.
I... wonder why nobody has proposed some kind of govt subsidized antivirus program... why not buy out ESET or similar and allow all US residents (or the world, for good will) to get a good free antivirus?
You're kidding, right?
Please, tell me you are kidding!
Why in the name of GFSM or whatever deity you care to insert would anyone in their right mind do or propose this? It boggles my mind since what you propose is alreadyavailable! And has been for years.
To forestall certain trollish and flamish responses (oo, woe is me to think this might work), if this really was a good idea, one could promote these services far more cost effectively than actually funding/buying them.
But of course it is a terrible idea, for any of the following reasons:
The US government would not likely do any such thing, given that these services are in competition with American corporations.
The money would be better spent on improving existing alternatives to Windows. Personally, I like Ubuntu (YMMV), but I don't know that they need the money (sure, everyone would like more $$, but Canonical has pretty deep pockets, no?).
But for now it simply doesn't matter - too many users are ignorant of the fact their OS comes from a vendor who simply doesn't give a damn and/or wants to squeeze yet more $$$ from the pockets of its beloved customers, whom it loves and respects dearly (either for its own bank accounts or those of its incestuously intertwined corporate "partners")....
After over 40 years in space this is the first guy to bring this up?? Hm... Smells fishy if you ask me..
Reminds me of an anecdote from one the Apollo 17 astronauts: He noticed that moon dust smelled and wondered why no one had mentioned it before. Eventually he realized it was a cultural thing: In pilot culture, "out of the ordinary" can get you grounded, where "out of the ordinary" is what science culture is all about. And the early Apollo astronauts were all pilots, mostly test pilots.
It only takes one curious person to open a new door and most of us don't notice the door is there, even if we pass it by every day of our lives.
Lots of trailing off there......implies you can't complete a thought...maybe half-baked?
LOL! Almost ROFL!
Yes, much of my mind is half baked after weeks of thinking functional. Such a different way of viewing the world, it is. I find myself sorting all tasks into simple procedural (script), big procedural (C or perl), complex procedural (C), and really complex (Haskell). ():: my language choices, YMMV.
I've a few things in mind, I'm letting the concepts soak whilst I consider the different ways of representing and manipulating the data. It's an enriching experience, even before writing a line of code.
The biggest [Haskell] change was getting accustomed to passing functions around like any other parameter, and stuffing them into lists like any other data.
:->
I smile because that was an "ah ha" C moment for me: When I discovered I could do exactly that, just treat functions like pointers and pass them around. That shortened a lot of code, and got me more OO-like (as implied by another reply) without having to leave C's comfy playground.
Point-free is doing that for me in Haskell: Sure, I can pass around functions or parameters, but when writing the functions themselves, I don't need to! Cool!
Thanks for the Monads analogy. I was getting there, the validation helps assure me I'm on the right track. (Found a great paper on Monads by googling "monads considered harmful"......funny that.)
Languages are for the most part trivial. And universal.
How about lazy evaluation and currying......time to make another language, Haskell.... Do we want... a pointer?
Please do forgive me (hee hee hee: Please ==> forgive $ me) if I haven't quite gotten your point, but I cannot square your first and last paragraphs, specifically the parts I've quoted.
Perhaps if you'd written "Imperative languages are for the most part trivial and universal? Perhaps then could easily equate C and Fortran and Perl and sed, and leave Haskell and Lisp out of the mix. Or perhaps if you'd written "OO languages are for the most part trivial and universal? Perhaps then I could equate (more or less easily, don't think it's quite NP) Objective C and Java and C++. (But see below....)
But the bold unqualified "they're all languages, get over it" sort of assertion doesn't parse.
My intro was Fortran, then F77, then Pascal, then C (OMG! Pointers! The Bomb!), and it was evolution, a little more cool, a little more flexibility all along.
Then I learned C++ at work by day and Java for fun at night, and my head hurt, 'cause I liked the imperative style and OO was weirdly different but everyone was swilling kool aid so I stuck it out...
...and every night I'd discover something in Java that was THE BOMB that would solve that day's problem and every next day I'd find that C++ didn't have that feature (does today, AFAIK, but STL was busted then, so everyone rolled their own)......but I moved out of real programming before I got my head around OO.
And now I'm learning Haskell, just 'cause I've learned that making my head hurt from time to time is a great way of stretching myself, of getting better at everything....
And I don't understand - at least, I haven't wrapped my head around Monads yet, though I get what they're for. And you know what? No pain.
None. I can feel the approach of enlightenment. SYB and reflect, baby, introspection and lazy evaluation and side-effect free (or reliably and provably constrained side-effect management) via implicit state passing.
Whoa.
I feel like Neo after the roof but just before the hallway - I'm starting to believe.
I've read the "SYB in C++" paper. I get what they're doing. But they admit the gap:
SYB in Haskell depends on... features... not available in C++, most notably rank-2 types, higher-order functions, and polymorphic type extension....
Scrap++ is a great exercise, but how do you get to SYB without those? You don't.
There's a guy out there that/.ers love or hate, no middle ground, so I won't reference him directly, but he's right: Different programming models change how you think of problems, and the right model opens so many doors you didn't even know existed. Doors you couldn't even have described until you knew they were there, but you were unable to find the hallway until you squinted, looked sideways at the world, and watched it shift......and were freed from the imperative....
"Hello World" is a cool teaching aid in Fortran and C and even perl (do it 15 different ways, without string literals or character types, preferably with a program one column wide:->)
But Haskell? No. When you learn Haskell, think big. 'Cause its programming model is so way different you have no idea. I'm at the point I almost consider Monads harmful......but I'll get the other side of the koan soon.
And when I have a simple repetitive task that I need to automate, I'll stick to bash, 'cause it's clean and readable, and sufficiently fast and sufficiently limited that I have to force myself to be literate, which makes it so much easier 6 months later when I need to tweak $ remember script.
I won't use Haskell for that. No way. But that replacement for scrabble/scribble I've been thinking of? That tool for edit
If you're not doing anything wrong......why do you need privacy?
I am truly amazed that this canard of question is still asked in the age of the "Jedi Kid" and facebook and youtube and myspace....
Unless you lead a perfectly moral life 100% of the time and never do anything that anyone else would consider to be immoral, illegal, objectionable, or just plain hilarious, then you need privacy.
And guess what? There are enough different groups with enough different agendas to find something you do immoral, illegal, objectionable, or just plain hilarious.
Since you don't believe you need privacy, let me record your day, everyday. Then let me take one event out context, say, the one time you absolutely lost it, foaming mouth with rage at your 2 year old because you slept badly and you're worried about your Mom's operation and your boss is being a dick and your blood-sugar is low.
You remember that moment? Scared the crap out your kid? And you were glad your spouse wasn't there because they would have thought you a monster? But you immediately regretted it, calmed down, cried a little yourself, and hugged your kid until you both felt better, and there were never any lasting effects, except that little bit of guilt or shame you still secretly feel?
Now, I am going to take that video and I am going to publish it everywhere. I am going to spam the bricks and mortar world with printouts of you foaming at the mouth.
How will your spouse and friends and family and coworkers react? Think you'll ever get a table at your favourite restaurant again? What will your kid think when they see that picture years from now?
I agree, you've never done anything wrong. Human, yes. Wrong, no. Still feel you don't need privacy?
Once upon a time, if your reputation was ruined by an isolated event, you would have moved a few villages away, and all would have been well.
On the Internet, there are no isolated events, and there is but one village. Better hope for reincarnation, because that one moment - recorded and widely shared, because you had no need of privacy - has ruined your life. And your spouse's. And your kid's. Etc.
Hot things rise and are less dense. Cold things go down. Except this occurs with steroids on the sun.
But remember that there is no gravity at the centre of the sun, just as there is no gravity at the centre of the Earth. (OK, strictly speaking, there is micro-gravity.)
Assume that the sun is pretty much a fluid (and ignore viscosity, or anything viscosity-like, such as electromagnetic attraction or repulsion in a plasma). This means that denser bits will sink to a point because the closer to the centre they get the less force there is pulling them to the centre.
(As an aside, I can imagine that the centre of the sun would be a wild place with all these layers and bands and blobs of materials of different densities and electromagnetic properties bumping about, without gravity to order them......perhaps like a superheated liquid that has not yet begun to boil, but when it does, WATCH OUT!)
Finding the "no gravity at the center" part hard to imagine? Imagine that the centre of the Earth is a cavity (with a nice 20C temperature, 1 ATM pressure, etc., so that you can survive this thought experiment). Imagine that you are at the centre of this cavity. How much gravity is there on you? Well, the dominant factor would be the planet all around you. OK, so far so good. But every bit of that planet pulling you in one direction is almost entirely cancelled out by the bit 180 degrees opposite. Despite unevenness in the Earth, in the end, you would be simply free floating. At least to several decimal places.
At the surface, gravity is a big deal and the "dense things fall" part works. But the closer they get to the centre, the less they are pulled toward the centre. And as their fall is slowed by the all other material around them, they stop. Or, more to the point, gravity becomes less important than nuclear pressure, which will tend to push things toward the surface again.
Under GNU/Linux, you typically have better educated users.
This was true back in the day, that is, when virtually all Linux users were home-brew hacking DYIers who either loved all things CSish or hated all things M$ish and knew there were alternatives.
You know, the gentoo and sid crowds.
Then RedHat happened and Ubuntu happened and hell froze over and DELL and HP started shipping systems with an OS other than Windoze and what you say is no longer true.
It's probably still true that the majority of Linux users are "better educated" (or, perhaps, informed and intent hacker hobbyists) and that virtually all people running Linux servers fall into that crowd, but it is no longer true that "only the educated" run Linux.
There are enough people now running Linux because it just works for them, enough people who still aren't really clear on what OS is and DO NOT NEED TO BE!. Seriously, why should they give a damn, they just want their computer to work, just like they want their car and their alarm systems and the elevators downtown to work without having to know a ton of geeky crap or push 16 buttons in exactly the right sequence slap!......where was I?
There are enough people now running Linux because it just works that Linux needs to consider that these users may not always know what they are doing. Ubuntu does this pretty well, with the way things are hidden behind an extra password dialog, along with decent - adequate? - explanatory text. It should be enough to give sufficient pause to prevent serious damage.
There is no need to defend against those users, they are not attacking their own machine.
It's not a question of users attacking their own machines. It's a question of preventing accidental damage of the kind that Linux seemed to once revel in encouraging....
It is uneducated users that are tricked into executing malicious code, that allow outside attackers to control their machine.
Bollocks. Everybody makes mistakes. Windows - at least older versions - ensured that all mistakes were grave. Modern Linux - and modern Windows when properly configured and properly patched (is this an NP problem?:->) - make it so mistakes are less likely to be 100% fatal 100% of the time.
And to return to your first quote....
Under GNU/Linux, you typically have better educated users.
Under BSD, you typically have better educated users.
There, fixed that for you.
(I don't use BSD, never have, but I do recognise that Linux has, for whatever reason, taken off in non-geek circles in a way BSD has yet to, and may never want to. Don't get me wrong, some of the BSD products seem downright amazing, but the user bases of BSD and Linux have diverged considerably, and for the moment Linux is winning the popularity contest. Does that make it better? No. Worse? No. Just more popular.)
Encryption would not help protect EPC from fraudulent messages. Safeguarding against fraudulent messages is a problem in authenticity (is the message from an authoritative source?) and integrity (am I receiving the message that was sent, without modification?) and not a problem in confidentiality.
Encryption provides confidentiality protection, not integrity or authenticity. (Yes, MACs can be used for integrity protection, but a MAC doesn't encrypt the message, it just uses an encryption algorithm to provide integrity protection.)
Encryption would be counterproductive in EPC because a) the information should be available to all, and b) using encryption would require recipients to be able to decrypt the messages, which would require the decryption key, which would allow them to author fraudulent messages. Unless the decryption key was their private key (no use of symmetric at all) in which case the system would scale to 1, perhaps 2 users (since the sender would have to encrypt all information with the public key of each recipient).
This sort of key management nightmare also rules out the use of MACs for integrity.
The only practical way to provide this sort of integrity and authenticity is to digitally sign all messages.
Oh no, I seem to have opened the box clearly labelled "Practical PKI, Property of Pandora".
Let the flame fest begin....
vi, Linux, GPLv3, CLI, ST:TOS, Picard, social democracy, pro-choice, anti death penalty.
What? Oh, making my religious war choices clear. Shall I just call you a Bush-loving Moore-hating Nazi WMD maker now, and get it all over with?
Oh, yeah, while we're at it, I'm president of RPLCWADNWEFOTB (Rock-and-Punk Loving Canadians who Absolutely Despise Nickelback with Every Fibre of Their Being), so I'm with you on that one....
You didn't bet burned because you where walking and your feet where [sic] dry.
Uh, sorry, no. He didn't burn his feet because they were wet! Read Feynman's autobiography - can't remember which one - where he describes this.
Basically, this works because of nervousness - the participants are terrified that they will burn their feet, their feet sweat, they walk across the coals, and the sweat evaporates, forming a protective barrier.
No shit.
Feynman describes how he burned his feet the third or fourth time when he stepped on the coals in a state of great confidence - he'd done it before, it would fine! Oops....
Quite funny in an OMG that would hurt like hell glad it wasn't me kind of way.
the fact that energy and momentum are conserved has been verified experimentally in excruciating detail
Well, yes, as long as you do your relativistic integrals over a sufficiently long interval, e.g., much, much longer thant the length of the collision....
That was the day I sat slack jawed with my head on my desk in 3rd year Special Relativity, unwilling a) to accept just how weird the universe was, and b) that all previous physics teachers had lied to me about this conservation stuff....
For example, consider a small positively charged particle moving toward a much, much larger positively charged particle at relativistic speeds ("small" and "large" allow you to imagine that the effects of the former on the latter are negligible). If, at any time T you draw the force vectors, you will see that the force of the large object acting on the small object is not parallel to the line between their centres, because the force acting at T was transmitted at c at time T-t. So the force acting at T is along the line that existed at T-t.
So if you sum your vectors over too small an interval, you lose conservation. You need to make the integral interval long enough for the "error" to be "close enough to zero" to ignore, for whatever your purposes happen to be.
The more exact your requirements, the closer to infinity the required interval becomes. (Said he with a certain pedagological glibness.)
And all of this with classical physics, albeit in a relativistic setting.
Now go quantum, and watch out, conservation is tricky. Go quantum cosmological, e.g., the controversial idea that another universe lies at the other end of certain singularities, and conservation may be lost altogether.
Like a lot of other techniques and rules and laws in physics, you have to know how to apply conservation laws. Another good example of this is learning to ignore >c phase velocities in solutions to certain DE's and PDE's - sure, the result contains terms that appear to violate relativity, you learn to ignore them because they are physically meaningless - since no information is being conveyed at >c, you just drop the term and move on.
What's the point, indeed. We should have moved away from MD5 signatures years ago. It's only a matter of time before some maliciously inclined asshat starts forging MD5 signatures on FLOSS packages, just to prove a point.
Hey, why not mention them by name? I think it's worth supporting "the little guys" not only with my $$$ but by/vertisement!
Seriously, though, as geeks it behooves us all to support techcos that offer good service at good prices, deliver value, actually have a person answer the phone on the first or second ring, and aren't knobs.
I went to teksavvy in the first place for the price, the low cost fixed IP, and the "it's your connection, do with it as you will" philosophy (servers are OK, no traffic shaping - at least none that negatively impacts my downloads or uploads, regardless of protocol), and having been there a while, I'd be willing to pay a monthly "not a bunch of knobs like the artful dodger major cable provider and nationwide unsympathetic DSL provider, both of whom shall remain nameless" premium.
Your post advocates a
( ) technical (X) legislative ( ) market-based ( ) vigilante
approach to fighting spam.
Furthermore, your approach appears to require a level of international cooperation akin to
( ) Passing a meaningless UN resolution
( ) Negotiating a world wide free trade agreement
( ) private, i.e., commercial and civil, law
( ) Banning land mines
( ) Adding a permanent member to the UN Security Council
( ) Achieved balanced copyright reform
( ) Censuring Cowboy Neal
(X) Doing anything truly useful about climate change
( ) Eliminating Britney Spears
Your idea will not work. Here is why it won't work. (One or more of the following may apply to your particular idea, and it may have other flaws which used to vary from jurisdiction to jurisdiction before a useful treaty can be negotiated.)
( ) Spammers can easily use it to harvest email addresses
( ) Mailing lists and other legitimate email uses would be affected
( ) No one will be able to find the guy or collect the money
( ) It is defenseless against brute force attacks
(X) It will stop spam for two weeks and then we'll be stuck with it
(X) Users of email will not put up with it
( ) Microsoft will not put up with it
( ) The police will not put up with it
( ) Requires too much cooperation from spammers
(X) Requires immediate total cooperation from everybody at once
( ) Many email users cannot afford to lose business or alienate potential employers
( ) Spammers don't care about invalid addresses in their lists
( ) Anyone could anonymously destroy anyone else's career or business
Specifically, your plan fails to account for
( ) Laws expressly prohibiting it
(X) Lack of centrally controlling authority for email
(X) Open relays in foreign countries
( ) Ease of searching tiny alphanumeric address space of all email addresses
(X) Asshats
(X) Jurisdictional problems
(X) Unpopularity of weird new taxes
( ) Public reluctance to accept weird new forms of money
( ) Huge existing software investment in SMTP
(X) Susceptibility of protocols other than SMTP to attack
( ) Willingness of users to install OS patches received by email
(X) Armies of worm riddled broadband-connected Windows boxes
( ) Eternal arms race involved in all filtering approaches
( ) Extreme profitability of spam
( ) Joe jobs and/or identity theft
(X) Technically illiterate politicians
(X) Extreme stupidity on the part of people who do business with spammers
(X) Dishonesty on the part of spammers themselves
( ) Bandwidth costs that are unaffected by client filtering
( ) Outlook
and the following philosophical objections may also apply:
(X) Ideas similar to yours are easy to come up with, yet none have ever been shown practical
( ) Any scheme based on opt-out is unacceptable
( ) SMTP headers should not be the subject of legislation
(X) Blacklists suck
( ) Whitelists suck
( ) We should be able to talk about Viagra without being censored
( ) Countermeasures should not involve wire fraud or credit card fraud
( ) Countermeasures should not involve sabotage of public networks
(X) Countermeasures must work if phased in gradually
( ) Sending email should be free
(X) Why should we have to trust you and your servers?
( ) Incompatiblity with open source or open source licenses
(X) Feel-good measures do nothing to solve the problem
( ) Temporary/one-time email addresses are cumbersome
( ) I don't want the government reading my email
( ) Killing them that way is not slow and painful enough
Furthermore, this is what I think about you:
( ) Sorry dude, but I don't think it would work.
( ) This is a stupid idea, and you're a stupid person for suggesting it.
( ) Nice try, assh0le! I'm going to find out where you live and burn your house down!
(X) uh, come to think of it, I have no particular opinion of you nor any desire to form one.
This excellently written piece of satire obviously went right over the heads of quite a few Slashdotters
Methinks twas the length. It wasn't until the last paragraph that I began LOL, nay ROFL, after having a supercilious smirk pasted on my smug mug, wondering whither the writer was going.
Could this be serious? I asked myself. Surely not! I answered, more from desperate disbelief that the Rogue of Redmond could do something well, let along right. But this seems so earnest, so serious in phrase if not intent.... Surely not.... And yet....
And then I stumbled upon bloated assware that isn't worthy of the fifteen cockslapping gigabytes of hard disk it requires and that he should switch to Linux and verily, the coffee burst forth from my nasal passages and spattered my screen even as my diaphragm did heave and cramp.
Moisture condensation and damage can occur below the roofs and within outer walls of heated buildings if indoor air pressure is significantly greater than outdoor air pressure.
Of course, the next section of this page appears to contradict this....
I asked about maintaining a positive pressure differential when we had an ERV installed (for reasons similar to those suggested by the PP); the technician indicated that while a nice theory, it could cause the ERV to ice up. They had been instructed to create a slightly negative pressure differential for this reason.
So my modern, plastic sealed house has slightly negative pressure relative to the outside. Several years and counting, and no negative side effects as far as I can tell.
Oh, and that's in Ottawa: Summer highs in the 40s, and very humid, winter lows in the -30s, and very dry. Nothing too extreme....
It has to be much, much easier to overclock machines when you never have to worry about overheating. Who needs liquid cooling when you can have polar cooling?
Apparently not: The FA mentions that they are at 12,000 feet, so they have a real problem with computer fans not being able to move the thin air.
Other effects of the thin air include laptop disks that don't spin properly, because they are built to float on a layer of air and are designed for near-sea-level densities. The air is also very dry, leading to increased risk of fires and disk failures caused by static.
Fire is a huge problem in general, because in the winter they have no choice but to fight and extinguish. Relocation isn't an option. Very interesting article.
Slashdot Mirror
I'm thinking of having two "layers", if you will: One would be the kernel, libc and other keys libraries, etc., the things that make my hardware a system. The other would be the apply goodness. The idea is that apply goodness could/would be supported on multiple cores, which would remain stable. Sure, more testing, more regression, perhaps impractical.
A stable core with a lifetime of more than 6 months, rolling releases of apply goodness.
Pardon the half-baked-ness....
ooo-kay, time to scrape egg from face. Replies to my post motivated me to investigate. Turns out it was a bug in the upgrade package, since fixed. Didn't catch it at the time because I couldn't figure out at the time that it was a bug (couldn't find the evidence), no one else had caught it yet (guess I jumped on the server upgrade much earlier than others) and the original message blamed the failure on the presence of a non-Ubuntu package.
Now, weeks later, it works, the bug having been found by more resourceful folk than me. Mea culpa
The upgrade is proceeding happily right now....
Good idea, and maybe that will work in my case...
...unless the package isn't part of the distro. Like I wrote, I haven't investigated, because having an operational server that does what I need is more important right now than upgrading. It's a priority thing. (Just for fun, I checked packages.ubuntu.com and hylafax is there, so it probably isn't hylafax blocking me.)
What else do I have on this box? Hmm, let's see, I set it up when Dapper was released, added a couple of things I wanted - and I'm pretty sure not all were APT-based - and haven't really touched it since. So, no, I don't remember. It works, I leave it alone, we get along great.
:->
Sometimes loony ideas are exactly what the world requires - they're the ideas that result in the biggest, baddest, bestest change.
I've been using Ubuntu since Dapper pre-betas, and I tend to agree with you: The "everything new goes into the next release" approach means that I have to wait for upgrades to get cool stuff ('cause there is always cool stuff 'round the corner) and it also means I have to upgrade even if I don't want to, because of the cool stuff!
No, I don't believe that I'm contradicting myself: I don't want to upgrade, because I like stable, but I want the cool stuff, so I have to upgrade. Sigh. It is the Ubuntu release cycle that forces me into this.
That said, I don't plan on switching distros, 'cause I love Ubuntu: Love the mindshare, the community, the approach, the distro itself, etc. It's good stuff. But I would like this pet peeve of mine improved.
And OK, while I'm at it: I wanted BackupPC 3.0 on my LTS server a looong time ago, and was really looking forward to 8.04 LTS, 'cause I knew the upgrade path would be smooth and BackupPC 3.0 and a lot of other goodness would be there! Yay!
Guess what? I cannot upgrade my LTS server because I had the temerity to install non-Ubuntu packages (might have been hylafax, I simply have not investigated): The upgrade tool detects the non-Ubuntu package and stops dead. I am stuck with an LTS server that will eventually be unsupported and that will be a real pain to upgrade, because of what seems to me to be a lamo upgrade tool.
Glad I got that off my chest.
I Love Ubuntu.
But I want a release management plan that combines a stable base with rolling releases of the cool stuff. Don't know how this would work, but it seems like a really good idea to me.
Westjet flies 737s, employs dual jetways at [some] airports, opens the front and rear doors and move stairs in [at other airports].... It really does work better.
Westjet has also experimented with various boarding orders, such as, in no particular order:
IIRC, they determined that the best, fastest way to board a plane was (wait for it!)
Primarily because the fastest, most motivated, most aggressive, most whatever people get up first, hustle into line, and hustle into their seats. Slower folks gravitate to the end of the line, meander onto the plane, etc. The fast and efficient folks are in front, the slow folks take their time and aren't in anyone's way, and the plane loads quickly.
The mathematician has an interesting idea but fails to account for the human element: The old lady in 15F is going to take her time getting to the in-terminal seating grid, take her time on the gangway, take her time in the plane, and need assistance from a flight attendant to stow her bag, check her seat, and ask the folks in 15D and 15E to move ('cause she's too reserved to do it herself). Best to leave her to last, so she can get the assistance she needs and not be in anyone's way.
And with nothing but general boarding, that's what happens on Westjet flights.
Of course, it helps greatly that Westjet attendants are helpful, friendly, warm, welcoming, and have great senses of humour.
No, I am not an owner. I'm a very satisfied customer who flies Westjet whenever I can, loathes Air Canada with every fibre of his being, and thinks the best thing that could happen to AC would be for Air Austria to buy it out, fire every single last AC attendant, and replace them with AA employees. Now there's an efficient, helpful, airline. But that's a story for another day.
Here I was, planning to mod this discussion, but I can't believe what you just wrote.
You're kidding, right?
Please, tell me you are kidding!
Why in the name of GFSM or whatever deity you care to insert would anyone in their right mind do or propose this? It boggles my mind since what you propose is already available! And has been for years.
To forestall certain trollish and flamish responses (oo, woe is me to think this might work), if this really was a good idea, one could promote these services far more cost effectively than actually funding/buying them.
But of course it is a terrible idea, for any of the following reasons:
But for now it simply doesn't matter - too many users are ignorant of the fact their OS comes from a vendor who simply doesn't give a damn and/or wants to squeeze yet more $$$ from the pockets of its beloved customers, whom it loves and respects dearly (either for its own bank accounts or those of its incestuously intertwined corporate "partners")....
After over 40 years in space this is the first guy to bring this up?? Hm... Smells fishy if you ask me..
Reminds me of an anecdote from one the Apollo 17 astronauts: He noticed that moon dust smelled and wondered why no one had mentioned it before. Eventually he realized it was a cultural thing: In pilot culture, "out of the ordinary" can get you grounded, where "out of the ordinary" is what science culture is all about. And the early Apollo astronauts were all pilots, mostly test pilots.
It only takes one curious person to open a new door and most of us don't notice the door is there, even if we pass it by every day of our lives.
Lots of trailing off there... ...implies you can't complete a thought ...maybe half-baked?
LOL! Almost ROFL!
Yes, much of my mind is half baked after weeks of thinking functional. Such a different way of viewing the world, it is. I find myself sorting all tasks into simple procedural (script), big procedural (C or perl), complex procedural (C), and really complex (Haskell). () :: my language choices, YMMV.
Have you decided to do anything cool with it yet?
I've a few things in mind, I'm letting the concepts soak whilst I consider the different ways of representing and manipulating the data. It's an enriching experience, even before writing a line of code.
The biggest [Haskell] change was getting accustomed to passing functions around like any other parameter, and stuffing them into lists like any other data.
:->
I smile because that was an "ah ha" C moment for me: When I discovered I could do exactly that, just treat functions like pointers and pass them around. That shortened a lot of code, and got me more OO-like (as implied by another reply) without having to leave C's comfy playground.
Point-free is doing that for me in Haskell: Sure, I can pass around functions or parameters, but when writing the functions themselves, I don't need to! Cool!
Thanks for the Monads analogy. I was getting there, the validation helps assure me I'm on the right track. (Found a great paper on Monads by googling "monads considered harmful"... ...funny that.)
Languages are for the most part trivial. And universal.
How about lazy evaluation and currying... ...time to make another language, Haskell.... Do we want... a pointer?
Please do forgive me (hee hee hee: Please ==> forgive $ me) if I haven't quite gotten your point, but I cannot square your first and last paragraphs, specifically the parts I've quoted.
Perhaps if you'd written "Imperative languages are for the most part trivial and universal? Perhaps then could easily equate C and Fortran and Perl and sed, and leave Haskell and Lisp out of the mix. Or perhaps if you'd written "OO languages are for the most part trivial and universal? Perhaps then I could equate (more or less easily, don't think it's quite NP) Objective C and Java and C++. (But see below....)
But the bold unqualified "they're all languages, get over it" sort of assertion doesn't parse.
My intro was Fortran, then F77, then Pascal, then C (OMG! Pointers! The Bomb!), and it was evolution, a little more cool, a little more flexibility all along.
Then I learned C++ at work by day and Java for fun at night, and my head hurt, 'cause I liked the imperative style and OO was weirdly different but everyone was swilling kool aid so I stuck it out...
...and every night I'd discover something in Java that was THE BOMB that would solve that day's problem and every next day I'd find that C++ didn't have that feature (does today, AFAIK, but STL was busted then, so everyone rolled their own)... ...but I moved out of real programming before I got my head around OO.
And now I'm learning Haskell, just 'cause I've learned that making my head hurt from time to time is a great way of stretching myself, of getting better at everything....
And I don't understand - at least, I haven't wrapped my head around Monads yet, though I get what they're for. And you know what? No pain.
None. I can feel the approach of enlightenment. SYB and reflect, baby, introspection and lazy evaluation and side-effect free (or reliably and provably constrained side-effect management) via implicit state passing.
Whoa.
I feel like Neo after the roof but just before the hallway - I'm starting to believe.
I've read the "SYB in C++" paper. I get what they're doing. But they admit the gap:
Scrap++ is a great exercise, but how do you get to SYB without those? You don't.
There's a guy out there that /.ers love or hate, no middle ground, so I won't reference him directly, but he's right: Different programming models change how you think of problems, and the right model opens so many doors you didn't even know existed. Doors you couldn't even have described until you knew they were there, but you were unable to find the hallway until you squinted, looked sideways at the world, and watched it shift... ...and were freed from the imperative....
"Hello World" is a cool teaching aid in Fortran and C and even perl (do it 15 different ways, without string literals or character types, preferably with a program one column wide :->)
But Haskell? No. When you learn Haskell, think big. 'Cause its programming model is so way different you have no idea. I'm at the point I almost consider Monads harmful... ...but I'll get the other side of the koan soon.
And when I have a simple repetitive task that I need to automate, I'll stick to bash, 'cause it's clean and readable, and sufficiently fast and sufficiently limited that I have to force myself to be literate, which makes it so much easier 6 months later when I need to tweak $ remember script.
I won't use Haskell for that. No way. But that replacement for scrabble/scribble I've been thinking of? That tool for edit
If you're not doing anything wrong... ...why do you need privacy?
I am truly amazed that this canard of question is still asked in the age of the "Jedi Kid" and facebook and youtube and myspace....
Unless you lead a perfectly moral life 100% of the time and never do anything that anyone else would consider to be immoral, illegal, objectionable, or just plain hilarious, then you need privacy.
And guess what? There are enough different groups with enough different agendas to find something you do immoral, illegal, objectionable, or just plain hilarious.
Since you don't believe you need privacy, let me record your day, everyday. Then let me take one event out context, say, the one time you absolutely lost it, foaming mouth with rage at your 2 year old because you slept badly and you're worried about your Mom's operation and your boss is being a dick and your blood-sugar is low.
You remember that moment? Scared the crap out your kid? And you were glad your spouse wasn't there because they would have thought you a monster? But you immediately regretted it, calmed down, cried a little yourself, and hugged your kid until you both felt better, and there were never any lasting effects, except that little bit of guilt or shame you still secretly feel?
Now, I am going to take that video and I am going to publish it everywhere. I am going to spam the bricks and mortar world with printouts of you foaming at the mouth.
How will your spouse and friends and family and coworkers react? Think you'll ever get a table at your favourite restaurant again? What will your kid think when they see that picture years from now?
I agree, you've never done anything wrong. Human, yes. Wrong, no. Still feel you don't need privacy?
Once upon a time, if your reputation was ruined by an isolated event, you would have moved a few villages away, and all would have been well.
On the Internet, there are no isolated events, and there is but one village. Better hope for reincarnation, because that one moment - recorded and widely shared, because you had no need of privacy - has ruined your life. And your spouse's. And your kid's. Etc.
Still feel you don't need privacy?
Hot things rise and are less dense. Cold things go down. Except this occurs with steroids on the sun.
But remember that there is no gravity at the centre of the sun, just as there is no gravity at the centre of the Earth. (OK, strictly speaking, there is micro-gravity.)
Assume that the sun is pretty much a fluid (and ignore viscosity, or anything viscosity-like, such as electromagnetic attraction or repulsion in a plasma). This means that denser bits will sink to a point because the closer to the centre they get the less force there is pulling them to the centre.
(As an aside, I can imagine that the centre of the sun would be a wild place with all these layers and bands and blobs of materials of different densities and electromagnetic properties bumping about, without gravity to order them... ...perhaps like a superheated liquid that has not yet begun to boil, but when it does, WATCH OUT!)
Finding the "no gravity at the center" part hard to imagine? Imagine that the centre of the Earth is a cavity (with a nice 20C temperature, 1 ATM pressure, etc., so that you can survive this thought experiment). Imagine that you are at the centre of this cavity. How much gravity is there on you? Well, the dominant factor would be the planet all around you. OK, so far so good. But every bit of that planet pulling you in one direction is almost entirely cancelled out by the bit 180 degrees opposite. Despite unevenness in the Earth, in the end, you would be simply free floating. At least to several decimal places.
At the surface, gravity is a big deal and the "dense things fall" part works. But the closer they get to the centre, the less they are pulled toward the centre. And as their fall is slowed by the all other material around them, they stop. Or, more to the point, gravity becomes less important than nuclear pressure, which will tend to push things toward the surface again.
More or less. FWIW. YMMV. IANAL. IANAA. Etc.
Under GNU/Linux, you typically have better educated users.
This was true back in the day, that is, when virtually all Linux users were home-brew hacking DYIers who either loved all things CSish or hated all things M$ish and knew there were alternatives.
You know, the gentoo and sid crowds.
Then RedHat happened and Ubuntu happened and hell froze over and DELL and HP started shipping systems with an OS other than Windoze and what you say is no longer true.
It's probably still true that the majority of Linux users are "better educated" (or, perhaps, informed and intent hacker hobbyists) and that virtually all people running Linux servers fall into that crowd, but it is no longer true that "only the educated" run Linux.
There are enough people now running Linux because it just works for them, enough people who still aren't really clear on what OS is and DO NOT NEED TO BE!. Seriously, why should they give a damn, they just want their computer to work, just like they want their car and their alarm systems and the elevators downtown to work without having to know a ton of geeky crap or push 16 buttons in exactly the right sequence slap!... ...where was I?
There are enough people now running Linux because it just works that Linux needs to consider that these users may not always know what they are doing. Ubuntu does this pretty well, with the way things are hidden behind an extra password dialog, along with decent - adequate? - explanatory text. It should be enough to give sufficient pause to prevent serious damage.
There is no need to defend against those users, they are not attacking their own machine.
It's not a question of users attacking their own machines. It's a question of preventing accidental damage of the kind that Linux seemed to once revel in encouraging....
It is uneducated users that are tricked into executing malicious code, that allow outside attackers to control their machine.
Bollocks. Everybody makes mistakes. Windows - at least older versions - ensured that all mistakes were grave. Modern Linux - and modern Windows when properly configured and properly patched (is this an NP problem? :->) - make it so mistakes are less likely to be 100% fatal 100% of the time.
And to return to your first quote....
Under GNU/Linux, you typically have better educated users.
Under BSD, you typically have better educated users.
There, fixed that for you.
(I don't use BSD, never have, but I do recognise that Linux has, for whatever reason, taken off in non-geek circles in a way BSD has yet to, and may never want to. Don't get me wrong, some of the BSD products seem downright amazing, but the user bases of BSD and Linux have diverged considerably, and for the moment Linux is winning the popularity contest. Does that make it better? No. Worse? No. Just more popular.)
Encryption would not help protect EPC from fraudulent messages. Safeguarding against fraudulent messages is a problem in authenticity (is the message from an authoritative source?) and integrity (am I receiving the message that was sent, without modification?) and not a problem in confidentiality.
Encryption provides confidentiality protection, not integrity or authenticity. (Yes, MACs can be used for integrity protection, but a MAC doesn't encrypt the message, it just uses an encryption algorithm to provide integrity protection.)
Encryption would be counterproductive in EPC because a) the information should be available to all, and b) using encryption would require recipients to be able to decrypt the messages, which would require the decryption key, which would allow them to author fraudulent messages. Unless the decryption key was their private key (no use of symmetric at all) in which case the system would scale to 1, perhaps 2 users (since the sender would have to encrypt all information with the public key of each recipient).
This sort of key management nightmare also rules out the use of MACs for integrity.
The only practical way to provide this sort of integrity and authenticity is to digitally sign all messages.
Oh no, I seem to have opened the box clearly labelled "Practical PKI, Property of Pandora".
Let the flame fest begin....
vi, Linux, GPLv3, CLI, ST:TOS, Picard, social democracy, pro-choice, anti death penalty.
What? Oh, making my religious war choices clear. Shall I just call you a Bush-loving Moore-hating Nazi WMD maker now, and get it all over with?
Oh, yeah, while we're at it, I'm president of RPLCWADNWEFOTB (Rock-and-Punk Loving Canadians who Absolutely Despise Nickelback with Every Fibre of Their Being), so I'm with you on that one....
You didn't bet burned because you where walking and your feet where [sic] dry.
Uh, sorry, no. He didn't burn his feet because they were wet! Read Feynman's autobiography - can't remember which one - where he describes this.
Basically, this works because of nervousness - the participants are terrified that they will burn their feet, their feet sweat, they walk across the coals, and the sweat evaporates, forming a protective barrier.
No shit.
Feynman describes how he burned his feet the third or fourth time when he stepped on the coals in a state of great confidence - he'd done it before, it would fine! Oops....
Quite funny in an OMG that would hurt like hell glad it wasn't me kind of way.
Well, yes, as long as you do your relativistic integrals over a sufficiently long interval, e.g., much, much longer thant the length of the collision....
That was the day I sat slack jawed with my head on my desk in 3rd year Special Relativity, unwilling a) to accept just how weird the universe was, and b) that all previous physics teachers had lied to me about this conservation stuff....
For example, consider a small positively charged particle moving toward a much, much larger positively charged particle at relativistic speeds ("small" and "large" allow you to imagine that the effects of the former on the latter are negligible). If, at any time T you draw the force vectors, you will see that the force of the large object acting on the small object is not parallel to the line between their centres, because the force acting at T was transmitted at c at time T-t. So the force acting at T is along the line that existed at T-t.
So if you sum your vectors over too small an interval, you lose conservation. You need to make the integral interval long enough for the "error" to be "close enough to zero" to ignore, for whatever your purposes happen to be.
The more exact your requirements, the closer to infinity the required interval becomes. (Said he with a certain pedagological glibness.)
And all of this with classical physics, albeit in a relativistic setting.
Now go quantum, and watch out, conservation is tricky. Go quantum cosmological, e.g., the controversial idea that another universe lies at the other end of certain singularities, and conservation may be lost altogether.
Like a lot of other techniques and rules and laws in physics, you have to know how to apply conservation laws. Another good example of this is learning to ignore >c phase velocities in solutions to certain DE's and PDE's - sure, the result contains terms that appear to violate relativity, you learn to ignore them because they are physically meaningless - since no information is being conveyed at >c, you just drop the term and move on.
What's the point, indeed. We should have moved away from MD5 signatures years ago. It's only a matter of time before some maliciously inclined asshat starts forging MD5 signatures on FLOSS packages, just to prove a point.
MD5 is broken and should not be used. It's time the FLOSS world went to at least SHA-224, if not SHA-512 (for future proofing, lots of bits). And just for reference, there is an open call for a new secure hash.
Hey, why not mention them by name? I think it's worth supporting "the little guys" not only with my $$$ but by /vertisement!
Seriously, though, as geeks it behooves us all to support techcos that offer good service at good prices, deliver value, actually have a person answer the phone on the first or second ring, and aren't knobs.
I went to teksavvy in the first place for the price, the low cost fixed IP, and the "it's your connection, do with it as you will" philosophy (servers are OK, no traffic shaping - at least none that negatively impacts my downloads or uploads, regardless of protocol), and having been there a while, I'd be willing to pay a monthly "not a bunch of knobs like the artful dodger major cable provider and nationwide unsympathetic DSL provider, both of whom shall remain nameless" premium.
Your post advocates a
( ) technical (X) legislative ( ) market-based ( ) vigilante
approach to fighting spam.
Furthermore, your approach appears to require a level of international cooperation akin to
( ) Passing a meaningless UN resolution
( ) Negotiating a world wide free trade agreement
( ) private, i.e., commercial and civil, law
( ) Banning land mines
( ) Adding a permanent member to the UN Security Council
( ) Achieved balanced copyright reform
( ) Censuring Cowboy Neal
(X) Doing anything truly useful about climate change
( ) Eliminating Britney Spears
Your idea will not work. Here is why it won't work. (One or more of the following may apply to your particular idea, and it may have other flaws which used to vary from jurisdiction to jurisdiction before a useful treaty can be negotiated.)
( ) Spammers can easily use it to harvest email addresses
( ) Mailing lists and other legitimate email uses would be affected
( ) No one will be able to find the guy or collect the money
( ) It is defenseless against brute force attacks
(X) It will stop spam for two weeks and then we'll be stuck with it
(X) Users of email will not put up with it
( ) Microsoft will not put up with it
( ) The police will not put up with it
( ) Requires too much cooperation from spammers
(X) Requires immediate total cooperation from everybody at once
( ) Many email users cannot afford to lose business or alienate potential employers
( ) Spammers don't care about invalid addresses in their lists
( ) Anyone could anonymously destroy anyone else's career or business
Specifically, your plan fails to account for
( ) Laws expressly prohibiting it
(X) Lack of centrally controlling authority for email
(X) Open relays in foreign countries
( ) Ease of searching tiny alphanumeric address space of all email addresses
(X) Asshats
(X) Jurisdictional problems
(X) Unpopularity of weird new taxes
( ) Public reluctance to accept weird new forms of money
( ) Huge existing software investment in SMTP
(X) Susceptibility of protocols other than SMTP to attack
( ) Willingness of users to install OS patches received by email
(X) Armies of worm riddled broadband-connected Windows boxes
( ) Eternal arms race involved in all filtering approaches
( ) Extreme profitability of spam
( ) Joe jobs and/or identity theft
(X) Technically illiterate politicians
(X) Extreme stupidity on the part of people who do business with spammers
(X) Dishonesty on the part of spammers themselves
( ) Bandwidth costs that are unaffected by client filtering
( ) Outlook
and the following philosophical objections may also apply:
(X) Ideas similar to yours are easy to come up with, yet none have ever been shown practical
( ) Any scheme based on opt-out is unacceptable
( ) SMTP headers should not be the subject of legislation
(X) Blacklists suck
( ) Whitelists suck
( ) We should be able to talk about Viagra without being censored
( ) Countermeasures should not involve wire fraud or credit card fraud
( ) Countermeasures should not involve sabotage of public networks
(X) Countermeasures must work if phased in gradually
( ) Sending email should be free
(X) Why should we have to trust you and your servers?
( ) Incompatiblity with open source or open source licenses
(X) Feel-good measures do nothing to solve the problem
( ) Temporary/one-time email addresses are cumbersome
( ) I don't want the government reading my email
( ) Killing them that way is not slow and painful enough
Furthermore, this is what I think about you:
( ) Sorry dude, but I don't think it would work.
( ) This is a stupid idea, and you're a stupid person for suggesting it.
( ) Nice try, assh0le! I'm going to find out where you live and burn your house down!
(X) uh, come to think of it, I have no particular opinion of you nor any desire to form one.
How am I going to download an Internet browser if my Operating System has no way of browsing the Internet?
apt-get install firefox
This excellently written piece of satire obviously went right over the heads of quite a few Slashdotters
Methinks twas the length. It wasn't until the last paragraph that I began LOL, nay ROFL, after having a supercilious smirk pasted on my smug mug, wondering whither the writer was going.
Could this be serious? I asked myself. Surely not! I answered, more from desperate disbelief that the Rogue of Redmond could do something well, let along right. But this seems so earnest, so serious in phrase if not intent.... Surely not.... And yet....
And then I stumbled upon bloated assware that isn't worthy of the fifteen cockslapping gigabytes of hard disk it requires and that he should switch to Linux and verily, the coffee burst forth from my nasal passages and spattered my screen even as my diaphragm did heave and cramp.
I don't know, but it might not be a good idea. According to http://healthandenergy.com/suggested_indoor_air_pressure.htm,
Of course, the next section of this page appears to contradict this....
http://www.trane.com/commercial/library/vol31_2/index.asp#control has more on this. Summary: It's complicated, man.
I asked about maintaining a positive pressure differential when we had an ERV installed (for reasons similar to those suggested by the PP); the technician indicated that while a nice theory, it could cause the ERV to ice up. They had been instructed to create a slightly negative pressure differential for this reason.
So my modern, plastic sealed house has slightly negative pressure relative to the outside. Several years and counting, and no negative side effects as far as I can tell.
Oh, and that's in Ottawa: Summer highs in the 40s, and very humid, winter lows in the -30s, and very dry. Nothing too extreme....
It has to be much, much easier to overclock machines when you never have to worry about overheating. Who needs liquid cooling when you can have polar cooling?
Apparently not: The FA mentions that they are at 12,000 feet, so they have a real problem with computer fans not being able to move the thin air.
Other effects of the thin air include laptop disks that don't spin properly, because they are built to float on a layer of air and are designed for near-sea-level densities. The air is also very dry, leading to increased risk of fires and disk failures caused by static.
Fire is a huge problem in general, because in the winter they have no choice but to fight and extinguish. Relocation isn't an option. Very interesting article.
Verizon DSL for $34.95/mo for 3.0/768
In parts of Ontario, Alberta, BC, and Quebec, we have this little independent, $29.95 for 5M/800K, $4 for a static IP.
Oh, and they're cool with servers.
Gotta love the little guys. Glad I dropped Rogers....