I have no more sympathy left for any of the dot-com a** holes. They are the reason the economy is on it's knees right now...
Actually, it's Worldcom's fault. B-)
The Internet was supposed to have a doubling time that amounted to growing at a factor of 10 per year. Turns out this was a myth. But it was a very believable myth. Even up to a week before the myth was exposed as false, well respected people (such as a certain Stanford professor) were promulgating it, in both media stories and learned papers.
Well, it turns out that, for about two years, the internet's traffic WAS growing at a factor of ten per year. This was just after it was opened to general exploitation, and was probably the result of pent-up demand.
But that was about 7 years to 5 years in the past. After that it slowed down to about a factor of two per year. (At least for the next three years, until the dot-com collapse. Last I looked the nubmers weren't all in for the following period, and maybe it was a bit lower.)
Now doubling every year is still phenomonal growth. (Beats Moore's Law by a bunch.) But it's still a shortfall from what was expected by a factor of five per year.
Well the network equipment manufacturers designed and built for the factor-of-ten. And the dot-com companies built business models based on the factor-of-ten. And then the demand failed to materialize. By the time of the dot-com collapse the shortfall was a factor of 625.
Now some of those dot-coms were crooks milking venture capitalists and 401(k) plans. And some of them had business models that amounted to "lose a little on each sale and make it back on the volume", or "if you build it they will buy". But don't you think that SOME of them might have been viable businesses IF their market was as big as they thought - 625 times what it really was - and doubling every hundred-ish days?
And by now the shortfall is about a factor of five THOUSAND from what people were building for. Any bets on why there's a telecom/internet equipment manufacturer collapse?
Now all this apparently happened because Worldcom's UUNET subsidiary was cooking their books. They kept saying that the net was growing at this mythological factor of ten - but what was ACTUALLY behind the stat (if anything at all) was an expansion of capacity, not traffic.
Much like saying "We built enough storm drains this year to handle ten times the water as we could handle last year. We'd better build ten times as many this year, or we'll be flooded when the rains come".
Unfortunately, nobody checked their numbers. (They were the biggest backbone - by their own numbers B-) - so they REALLY ought to know...) Other companies that weren't keeping up with the mythological growth rate blamed their sales operations rather than questioning the numbers.
Eventually the real numbers finally came to light as a result of the investigation of Worldcom's other accounting "innovations". B-)
So that's why I say that it's all Worldcom's fault.
I can't believe you're modded to 5, while showing an almost complete ignorance of how the internet actually works.
I'm afraid that it's your own ignorance that is showing.
(Unfortunately, at least four people with moderator points (and similar ignorance) believed you. So they moderated my posting "overrated", knocking it down below your own.)
ICANN only does domain names. IP addresses are handled by IANA.
Wrong. See below.
I've heard exactly zero complaints about IANA.
Yes, there were very few complaints about the actions of the IANA, and for good reason. The IANA is also known by his given name: Jon Postel. See his eulogy: RFC 2468, as in "Who do we appreciate?") Jon was one of the founders of the Internet, and did a fantastic job of handing out (and delegating the handing out) of unique identifiers.
Unfortunately, Jon died in October of 1998, and his benovolent dictatorship has been supplanted by his successors - a not-so-benevolent junta - the ICANN.
As to what ICANN does, here's the first two sentences from their ICANN Fact Sheet:
Formed in October 1998, the
Internet Corporation for Assigned Names and Numbers (ICANN) is a non-profit, private-sector corporation formed by a broad coalition of the Internet's business, technical, academic, and user communities. ICANN has been recognized by the U.S. and other governments as the global consensus entity to coordinate the technical management of the Internet's domain name system, the allocation of IP address space, the assignment of protocol parameters, and the management of the root server system.
- DNS technical issues
- allocation of IP address space
- assignment of protocol parameters (including, in particular, protocol identification numbers)
- management of the root server system.
Got that?
Continuing:
As a technical coordinating body, ICANN's mandate is not to "run the Internet." Rather, it is to oversee the management of only those specific technical managerial and policy development tasks that require central coordination: the assignment of the Internet's unique name and number identifiers.
The point of my posting was that assigning unique identifiers in a global name space is an indivisible transaction. A mechanism for unambiguously performing such indivisible transactions IS an authority - whether that "authority" is a database engine, a benevolent dictator, or the clerk who counts the votes of a committee or electorate.
What prevents us from ignoring ICANN when we feel like it and doing our own thing?
The need for unique identifiers.
How do you do an internet if:
- A particular IP address may map to different hosts. (Packets addressed to 64.28.67.150 go to www.slashdot.org according to ICANN, a Microsoft server according to Joe's Nameservice and Grill, the US Army Recruiter according to the MIL BGP servers, and the local kindergarden according to a router configured by a junior high student. Which authority - and thus which route - do you think a commercial ISP (PAID to deliver packets) will honor?)
- A particular domain name may map to different organizations. (joe_user@slashdot.org may go to joe at VALinux, joe at Microsoft world headquarters, joe at the draft board, joe at the local kindergarden,...)
- A particular protocol number may specify different protocols. (WHICH IPV4 are you talking about? Which SMTP? Which NFS?)
- A "well known port" may perform different functions. (Imagine a new Microsoft OS putting a webserver up on port 414, or whatever port is used for an important service in the latest competing OS, and configuring the next release of Internet Explorer to try that port first. No need to "embrace" and "extend" before getting to "extinguish".)
and so on, depending on which organization the owner of any particular machine is affiliated with?
The answer is: You don't. (It's like the street addresses, state names, and personal names being a matter of political debate and faction-fighting - while someone's trying to send you a letter.)
Assigning a unique name or number is an indivisible transaction. In the absense of a solution to the "distributed update" problem you HAVE to do that with a single-point mechanism - an "authority". The best solution yet found is delegation - which is what ICANN does with domain naming and selling blocks of IP numbers.
Which brings up the question: Why are domain names handled by ICANN, rather than the trademark/servicemark section of the Patent and Trademark office?
Yep. They did. See their web site. (Sorry, don't have time to copy the link from another posting...)
But when THEY make an opto-isolator they put the emitter(s) and the detector(s) on the same chip, getting the alignment between them by using the same set of masks. Then they convert the region between the two sides into silicon Dioxide (also known as "glass") to form a VERY GOOD transparent insulator without disturbing the alignment.
Hm, the physical layer of the DSL line can just be described as a MODEM kind of thing. But there are other layers in the DSL box.
Right. It includes a MODEM, and other stuff. The Alcatel ANTs, for instance, support packet-over-ATM with up to 8 virtual circuits, so there's an ATM layer. Then there's an Ethernet layer. And it has a whole internet stack and an HTML server to provide a virtual console for configuration.
That's a bit more than your typical MODEM, even in these days of peripherals smarter than the mainframes of a decade ago.
I say inform your local newspaper or television news station.
Your ISP and local TV station are very likely owned either by the same media conglomerate as your ISP, or another that is pulling the same stunt. If you live in anything larger than a village, the same is probably true of your local newspaper.
Blue Gene is NOT a shared-memory computer -- with a single kernel running all 64k processors -- but rather a cluster of 32k seperate computers (with two processors), each on one chip.
Looks like someone already imagined a Beowulf Cluster.
I guess now we have to imagine a Beowulf Cluster of Beowulf Clusters to imagine a Beowulf Cluster of these.
It seems to be the ideal system to run the next Microsoft operating system.
But think how much it would cost for the 65,000 licenses.
Not to mention the network time to activate them all.
Then heaven help them if they add a peripheral, say just before a high-profile chess match, and need to REactivate them all.
Now if the mean time to failure of a MS system gets up to, say, 30 days, and you have 65,000+ processors, that's a mean time to failure of about 40 seconds. It would be 2167 reboots per day if a reboot didn't take more than 40 seconds. B-)
Media Companies, Communications Companies, Oil Companies...they all yell about how new technologies will ruin their business models and how they need to be protected!
The Mafia has a very effective business model which the government has also tried to ruin (or at least CLAIMED to BE TRYING to ruin) from time to time. A big part of their business model consists of using the government to keep the competition from doing the same things that THEY do, while buying off its agents so THEY don't have to obey the laws imposed on others. Another part consists of lobbying FOR laws against entry into their product lines - drugs, "Adult Entertainment", extortion,...
Wait a minute... Didn't they run the music distribution business back in the crosby/ol'-blue'eyes/jukebox days? Like with "kneecapping" and extorting a Detroit radio station to kill the career of a local singer who sounded too much like their star?
And aren't a couple of the current media conglomerates direct descendants of those organizations? Without EVER having the organized crime members running their "business model" systematically extracted (other than one at a time, over outside activities) from their operations?
Presumably this also means that if you're using their "cell", they can charge you what they like. I can see their motivation...
Bingo!
Cell phones bypass the airphone, with its big bill of which the airline gets a cut.
Why should the airline take ANY risk of interference with the flight insturments when it's also costing them money? But they might accept a little when it's both under their control as to interference AND it's PAYING them money.
But I bet part of the impetus comes from the cellphone companies themselves. Using a cellphone in the air works. But on the ground a cellphone is "heard" by only a handfull of cells. In the air it is "heard" by MANY cells, chewing up bandwidth on each - and the SAME chunk of it, making the allocation of channels to calls on the ground difficlut. So even if you're paying for the call you're a net loss to your cell carrier, possibly forcing him to drop several calls by other customers.
Why the use of ANY electronic device is prohibited below a certain altitude, except when sitting still at the gate?
Because you can get away with just about any manouver in an airplane as long as you don't do it close to the ground. Continents have the right of way.
Takeoffs and landings require extreme precision, because going about a foot low means destroying the plane and possibly the cockpit crew and many in the cabin. There are a BUNCH of radio-based aids on a large number of frequencies and using a variety of methods - and if the one that's being used to guide the plane at a particular instant is suddenly interfered with, there may be no time to recognize that it's malfunctioning and switch to something else. So screwing up any one of them at a critical moment may result in a landing you don't walk away from, a mid-air collision, or some other mishap.
Similarly, the airport and the space immediately adjacent is a 3D traffic jam, coordinated by radio calls. Garbling even one radio message could result in a collision, in the air or on the ground. (As with highways they have a few even when they're NOT being interfered with. Now imagine highways with an occasional light going all-ways-green...)
Once the plane is AWAY from the space around the airport it has an ENORMOUS space to work in, and considerable time to work with. And there are "lanes" in airspace, as well as a rule that breaks it into stacks of altitude ranges where everything that isn't passing through in a well-known place is going in about the same direction. So if your laptop jams a navigational aid there's time to switch to another. (And if it somehow jams ALL of 'em the crew can run on internal nav and non-radio instruments and avoid other airplanes and mountains until the stew can get you to turn the bloody thing off.)
Are you talking about:
1) The core of a telephone carrier's network?
2) The core of the network of an ISP that is providing some telephony-related application (like POTS-emulation-over-cable, VoIP, or VoIP-related QoS enhancements)?
3) Commercial standalone PBXes?
4) PBX replacements (as a plugin card/driver/app for a PC)?
5) Modem-based answering machine/fax applications?
6) Desktop VoIP applications?
7) Server-room network VoIP servers?
8) Server-room VoIP/POTS bridges?
9) Voice menu hell servers - standalone or part of one of the above? or a host of other "Production Telephony" applications?
When I saw the question the first thing I thought was 1). But the text seems more directed to 3) and 9), while responders are all over the map.
Well, according to my SmartFilter, not only do they sell PBX stuff, but also oodles of sex.
Interesting.
Now I have no idea that SmartFilter, Microsoft, or anybody else is doing this. But...
Wouldn't it be a powerful marketing strategy to get your competitors listed as sex (or otherwise icky-poo) sites on as many censorware lists as possible?
It wouldn't be anywhere NEAR as obvious as getting them onto black hole lists. Email disruptions would be noticed right away. Censorware deletions are much more subtle - and less suspect.
Even better, when the US Feds come with guns and loudspeakers, it's always in your best interset to let them in; foreign governemtns and internal cults alike get smacked by the feds.
According to the Branch Davidians they TRIED to let 'em in - and Koresh got shot while trying. And a lot more later, which amounted to the Davidians believing that they wouldn't be allowed to surrender - and that they'd seen a number of people being shot for trying.
Similarly with Ruby Ridge. That standoff started with a government tresspasser shooting a dog and a teenage kid. Then it continued with a remote-controlled buggy with a phone and a gun coming up to the door and the phone ringing with the gun pointed at the house. And a sniper shooting a mother with babe-in-arms.
There are a few documentaries on this. You're welcome to believe them or dismiss them as nutcase propaganda. (But at least one of those on Waco is by a local TV news reporter who started out thinking they were a homicidal nut cult and ended up thinking they were a majority-non-white church group that had been systematically tortured to death by a jackbooted death-squad. And with work by Failure Analysis and infrared footage from the government itself to back up this interpretation.)
Bottom line is, when trying to get out from under the government's thumb, you have to be careful not to end up ground under its heel.
And once you start, stopping may not be under your control. It only takes ONE side to run a war.
Isn't this where the idea of digital logic really got started?
Yes.
At least its how it was taught when I was in school. We even did some design work in async. Cool stuff. Easy to do, fast as hell...
Never did figure out why it never caught on.
It DID catch on. But the chips kept getting bigger.
It's easier to design silicon compilers for synchronous designs than for asynchronous - and when you've got millions of gates per chip you REALLY want compiler assist, rather than to lay out all the circuit details by human effort.
It's also easier to make automated TEST program generators for synchronous designs, to run the machines that test the chips when they come out of fabrication and reject the ones that are broken. You NEVER get high-90s coverage with human-generated "functional" tests - but a compiler can get there easily:
- Add muxes in front of the flops to string 'em into "scan chains" - big shift registers connected either to the regular pins or a "JTAG" controller. Then on the tester you'll:
- "Scan in" a random starting state.
- Step the chip a few times.
- "Scan out" the result and see if it matches expected, simultaneously scanning in a different starting condition.
The test generation program becomes essentially a random-number generator, chip simulator, and fault-tested-so-far counter, with a few finesses for things like getting things reset properly, testing gates with big fanin, making sure busses aren't floating, rejecting patterns that don't test anything new, working around flops that weren't on the scan chain because they were on a critical path, avoiding logic loops that become implied RS flops or ring oscilators (depending on whether the loop has an even or odd number of inversions), identifying logic circuits that have untestable failure modes, and the like.
But full-scan and partial-scan don't work if the flops aren't tied to a small number of clock domains that can be tied together or otherwise controlled directly by the tester. Asynchronous logic elements (such as ripple counters or other circuitry where a flop's clock is driven from another flop's output, or other logic that's something other than a clock distribution and switching system) just don't scan well.
There IS a way to get the same sort of massive observability and controllability over asynchronous designs - the Cross Check array - along with automatic test program generation systems to work with it. (Think of DRAM- or active-matrix-LCD-style cross-point addressing of test-points and signal-injection points - about one for every four regular transistors on the chip.) It tests async designs just fine, and gives better coverage than full scan with about half the silicon overhead.
But it's patented. The company that made it never got much market penetration in the US fabs. It has since merged and the product may be completely gone at this point. Except for Sony, which had an unlimited license from funding them when they were a startup, their own software, and (as of a few years back at least) used it in all of their consumer chips.
That doesn't explain why the profound increase in Autism is present in CALIFORNIA, with no evidence (so far) of anything similar elsewhere.
So, assuming that further research doesn't come up with a similar rise elsewhere, that raises the question of "Why California". (Or why heavily in California and more lightly elsewhere.) Exposure to chlorinated carbon compounds is not particularly higher there - especially given Californian's aversion to chlorinated water.
Continuing to assume that causality implies correlation and it's a chemical issue, let's go down the list.
Right at the top: Gasoline oxygenation additives - in particular: MTBE, to which everybody in California who pumps their own gas - or drinks water - is exposed to in significant amounts, and has been for years. It's quite toxic, and has been implicated as a cause of asthma for a while now.
Certain illegal drugs exposure during and just before pregnancy is probably significantly higher in California - both because the drugs have been more prevalent here and because people interested in taking them have migrated here. But they're hardly unknown or unpopular elsewhere.
Dropping drugs for a bit: Another candidate is child rearing practices - in schools and home. California is the epicenter of the feel-good schools of child rearing. Systematically reward self-destructive behavior with attention and you can quickly teach a child to emit it continuously. And self-esteem based teaching systems will do precicely that.
California is also a hotbed of NON-vaccination. Side-effects of childhood or foetal viral infection immediately springs to mind. (Second trimester influenza has already been implicated in another mental syndrome - Schizophrenia.)
I could go on. But somehow I don't think the culprit will turn out to be Freon, PCB, Vynil Chloride, or chlorinated water byproducts.
This is what Nobel laureate Konrad Lorenz had to say about this way back in 1973-"Human culture, after enveloping and filling the whole globe, is in danger of being killed by its own excretion, of dying from an illness closely analogous to uraemia.
"Human Culture"? Yes, some of them will change. Some will die off or mutate, some will grow or shrink, some new ones will from. They do that from time to time - often on a scale of hundreds of years or less.
Human Beings, and extinction? Hardly. (Though the current enormous population is supported by farm, transport, and food preservation technologies - so a loss of this tech or an increase in its price, through economic collapse or regulation, means a significant die-off.)
Humans started out as a handfull of hunter-gatherers, before or during the last ice age. They expanded to inhabit essentially every bit of land area and floating ice except Antarctica BEFORE they developed industrial civilization and the scientific method. (Name another animal - other than human parasites - that managed that.)
Plains, deserts, steppes, mountains, ice caps... I doubt humanity could be wiped out by any climatic change that didn't boil or freeze ALL the water or eliminate all oxygen from the air.
The planet finally coming the rest of the way out of the last Ice Age - with the temperate zones shifting a couple hundred miles further from the equator and steaming jungles expanding beyond Brazil and central Africa - doesn't even qualify. (Heck: For raw biomass, suitably modified crops, or even CURRENT crops, it's probably a significant improvement.) And some of us would count the loss of the outer edge of certain seacoast cities to be a bonus. B-) Going back into a full Ice Age is more of a problem - though the greening of the equatorial deserts might make up for the loss of some more poleward land to glaciers.
Of course, if temperature shifts actually become a problem we can fix them directly, without screwing around with the CO2 level of the atmosphere. Just orbit a few hundred square miles of aluminized mylar, suitably located and oriented to provide a bit of shade if things are getting too hot, a bit of extra sunlight if they're getting too cold. Or whatever hack the rocket scientists come up with that's cheaper.
You want a robust space program anyhow - so you have something to spot and deflect the next incoming asteroid or comet fragment. Such an impact turning the whole planet into a broiler-oven for a day or so is the REAL threat of "global warming". THAT would once again reduce the ecosystem to plants with very robust seeds and resistance to PH variatioins and mouse-sized animals that happened to be underground at the time. (And maybe a few humans who had hung out in underground sites that didn't collapse and squirreled away a few years of supplies to last until they could grow something to eat.)
But I doubt temperature shifts (let alone the handfull of degrees that has the lefties drooling for more power and the media paniced) will be a problem for food production at all. Most of the food production of the world is now essentially an industrial operation, while the rest benefits from the tech. A few degrees of temperature change just means you change which crops - or which strains of a particular crop - you grow in a particular field. Shifts in weather patterns ditto, maybe with a change in irrigation or include the crops' water usage in the selection criteria, a few marginal plots going out of production, and new land becoming able to support crops.
Humanity will be forced to invent some sort of planetary kidney - or it will die from its own waste products."
Now that's true. But we've been doing EXACTLY THAT for quite a while now. When any given type of pollution becomes enough of a problem to bother with, we FIX it. Baby Boomers are old enough to remember Los Angeles smog before auto industry folk (including me) fixed up the engines. But that's NOTHING compared to, say, the killer fogs of London (driven by high-sulfur heating coal). Or just the indoor air of any human habitation in a cold climate before gas heat. And just think a moment about the streets of a city served only by horse- and ox-drawn vehicles. Talk about pollution...
Tech sometimes creates a new sort, or new amount, of pollution - "excretion" in Lorenz's vocabulary. But once it becomes a problem, more tech generally solves it (sometimes after quite a few years of griping by the people for whom it is a problem.)
Personally, I'm strongly of the opinion that both of these viewpoints are harmful. Over on the right there seems to be a lack of consideration for other very localized harm burning nasty stuff can cause. As a lifelong inhabitant of Los Angeles I've seen this first hand.
As someone married to an American Indian, who grew up on a west-coast reservation (of a different tribe - her mom was a teacher) with degree in history among her collection (and her dad was a history professor), let me tell you something about Los Angeles (that she brings up whenever it an air pollution are mentioned together B-) ).
Seems the local Indian name for the area translates to "Valley of the Smokes". The shape of the land and the wind patterns over much of the year trap airborne pollution - so badly that a single campfire would smoke it up for a day or more.
It's a testimony to US automobile technology (even if driven by legislation) that so many cars can now operate in that valley without photochemical smog being so thick that the light is blocked.
By the way: DON'T call them "Native Americans". It annoys them. (If you're born here YOU are a "Native American".) "American Indians", however, is a running "ignorant/stupid/crazy European invaders" joke: They were so dumb they thought they were in India - half a world away, Ho Ho! B-) (A poll of members of a large number of tribes showed the preference for "A. I." over "N. A." runs in the 80s-90s% range.)
... say node A is visible to base B and base C but base B and ase C are out of each others range. If B and C don't "hear" each other they can't work in unison without a third party. So when Node A (mobile presumably) hunts for a connection, both B and C try to talk at the same time to A, thus hampering the usable bandwidth.
But mobile node A can BE the "third party". If base stations B and C have unique (or at least non-colliding) identifiers then mobile node A can say, as part of its transmission, "I'm talking to B." or "I'm talking to C."
There are variants that work even if mobile node A is saying "Who's out there for me to talk to?". Look at the IP address resolution protocols for working examples. For starters, the a similar case arises on an Ethernet when a machine that wants to be booted uses RARP, broadcasting its Ethernet address and asking potentially redundant servers for its IP address. (Granted the servers COULD hear each other. But the protocol works even if they can't.)
You can also "flash" other areas by making a no-effect change (like inserting no characters). Like a blink comparitor when looking for star patterns.
Actually, you can't, since "u" only undoes the last change. I don't know where that came from. (That's what I get for posting when half-asleep. B-( )
There is a way to blink multiple changes that I can't recall just now - I think it's writing the file to/tmp/%, then repeated control-shift-"u". The multiple undo wouldn't break that.
Of course with multi-level undo you don't have to "blink" 'em to find 'em, so it really is an improvement.
But it IS an annoyance. What it really breaks is hammering on "u" to blink the last change while you decide which way you want it. Try that with vim and it merrily unwinds your edit session. Then you have to use control-R - a two-key-one-hand annoyance - to get your changes back.
Why not use Vim then... syntax highlighting, multi-level undo and all the other goodies, but with the efficient VI key bindings:-)
Actually, I do, generally when I'm on a linux platform. It's close enough to vi to be almost interchangable.
That multi-level undo has a downside, though. It kills a vi idiom: In vi "u" undoes the last change - but a "change" includes a previous undo. So hitting "u" repeatedly first causes the cursor to jump to the latest change, then "flashes" the change, making it even easier to spot. You can also "flash" other areas by making a no-effect change (like inserting no characters). Like a blink comparitor when looking for star patterns.
Best of both worlds would have been if vim had used a different keystroke for multi-level undo (like it added a stroke for multi-level redo), or had a switch to revert to vi behavior.
They can only do that with the code as-is or with purely in-house development added to it. They won't be able to say that about any GPL-sphere additions- if you want to nip this in the bud, do such good work on the GPLed version that it becomes the definitive one and completely leaves the in-house version in the dust.
On the other hand, if you want to have the best of both worlds, do some really good work in the GPLed branch (carefully avoiding the inclusion of other people's GPLed code in your work). Then offer to license it back to the foundation to also be included in the "in-house" version - for a fee. B-)
Hear hear! Vi has its points...
on
Blender Is GPL
·
· Score: 3, Interesting
I agree. In no way, shape, or form, is the "vi" interface a good one.
Huh? It's fast, it's efficient and it's easy on your fingers. How is that a bad thing? Just because you don't like it doesn't mean everyone has to agree.
Hear hear!
Back when I got my first unix box (FAR enough back that, when then entire list of email-connected sites fit on three pages, mine was there), I wanted to build and try emacs. But there was this little problem - the machine had only 2 megabytes and no demand paging. Emacs (even back then) wouldn't fit. (A tongue-in-cheek claim was circulating that the name was an acronym: Eight Megabytes And Constantly Swapping. B-) )
So I learned to use vi.
And then I was VERY active on a bulletin board for several years - using vi. And I got very fast with it.
Some time later I had access to a bigger machine and a colleague pointed out that emacs had a vi emulation mode - so I could ease in without having to learn new navigation keys right off the bat. I looked into it - and it had TWO distinct vi emulation modes. Oops. With one I might have tried it. But I didn't have time to find the better of the two. So I dropped it.
A little later a Netnews posting demoed a potential attack on those who used emacs as a news reader or mail reader. Seems that emacs had a little-know feature: You could include a snippet of lisp code in the comments in a program file, and emacs would run it. This was intended to set up tab stops, language editing modes, and the like. But this also worked in mail and netnews reading modes. The demo's lisp code would pop up a "See, I got you!" window and delete itself from the display of the item itself. But in principle it could do anything you could do from emacs - which is anything you can do from any shell, with a lisp interpreter handy to do complicated stuff. No clicking on attachments - just LOOK at the letter or news item and you're owned.
Windows macro virus vulnerability? Emacs had it first, and BETTER! B-) Imagine a lisp worm in netnews forging postings in your name, both replicating itself on "nice" groups and faking love letters on alt.binaries.pictures.child-molestation. Or dumping the contents of any "src" directory you can read to an alt.binaries group. (And heaven help you if you read news or mail when logged in as root...)
Of course this "feature" was on by default in the standard distribution. In those days, or days not too much earlier, RMS' approach to security was rumored to be having a blank password on root in his personal machine and letting this be known - in the belief that if there was no skill needed to break in, and thus no reputation to be gained, nobody would bother. (Apparently that worked with MIT students. But don't try it with the general population net-connected.)
Well, I had spent years doing classified research, which made me itch about security holes. So I decided to stick to vi for a while longer - along with the plethora of unix utilities that do essentially anything I need done that's beyond vi's power.
Since then I've occasionally seen an emacs-ism that has tempted me - like colored displays of comments vs. declarations vs. code. But every time I'm tempted I watch a colleague doing simple text editing with emacs, and count the keystrokes he has to use to do the simple stuff that constitutes the bulk of my editing work. And it always seems to take him a lot more strokes with emacs than it takes me with vi. So I'm generally not tempted for long.
Vi was designed for a very different world - the world of dumb character-based computer terminals in the days before ANSI standardized their behavior. There were literally HUNDREDS of different terminal designs, with a boggling array of differences in display geometry, control-character to cursor-motion mapping, and other odities. Vi (actually the "visual" mode of the "ex" editor) encapsulated these idiosyncrasies in a "termcap" (terminal-capability) definition file, thus letting the user do full-screen WYSIWYG editing on ANY of them using a common set of keystrokes - and letting the sysadmin add definitions for new terminals as they came out. This brought the user out of the dim world of command-line-only editors (such as "ed" and "teco") into the instant feedback of a screen display - halfway to the window systems that weren't available yet.
And - much to the surprise of its author - it did it very well. So well that people like me (who now have the vi commands "hard-wired" into our nervous systems from long use) still use it when we have serious text hacking to do.
It was 54 years ago in central Italy, driving at night on a desert mountain road. I saw a fiery fireball in the sky, moving slowly from left to right.
I had the time to: understand (maybe) what it was, wake up my wife, stop the car, get out an look. Total time maybe 20 seconds. The 'object' was moving slowly, spewing green flames and eaving a long lasting orange trail behind. Trajectory was more or less horizontal. It disapeared in a flash.
I (and a bunch of other people) saw something similar at Lake Tahoe a few years back. Distinctly purple, slow moving, leaving trail, no sound, no flash.
Turned out to be a space shuttle re-entering on its way to Cape Canaveral. The purple is due to the composition of the tiles. The final orbit and upper-atmosphere reentry is visible over Tahoe due to the inclination. (I think that's a side-effect of chosing an inclination that lets them switch to Vandenberg [on the next pass?] if the weather at Canaveral is too crummy.)
Well, for starters you could put an array of accoustic, microwave, or electrical transmitters & sensors in the pipe just BEHIND the drill and image the region ahead of the drill with radar and/or sonar. If you see a pocket of something that sounds/conducts/reflects like oil a bit off to one side, you can adjust the drill to curve in that direction (or send the NEXT one over that way).
10 BPS just doesn't cut it for uploading imaging information, even if you put most of the fancy processing down with the sensore. But T1 rates are just fine.
There's lots of other stuff you want to monitor - temperature, pressure, conductivity, etc. to find out what sort of stuff you're drilling through.
And it's important to know when to give up, stop pouring money down THIS hole and start over somewhere else. It costs a LOT to run the rig long enough to drill even another foot...
I recall, back in the early days, a company in Ann Arbor made a little board with a CMOS Z80-clone, a ROM with a BASIC interpreter, a serial port, and a few I/O ports - including some analog inputs. They sold a LOT of 'em to an oil company.
Seems that every now and then they would pull up the drill and send one of these down to measure some stuff. Then they would send the drill down behind it and grind it up. It was cheaper to buy a new one (and the associated cable) each time than to leave the rig idle long enough to pull the old one up. (And considering how fast a winch can crank, and how much custom computer stuff cost back in those days, that will tell you a lot about the per-minute cost of an oil rig and drilling team.)
So imagine how much they can save if they don't need to pull the DRILL up - disassembling it as they go - then reverse the whole process to put it back down, every time they want to take another reading.
Slashdot Mirror
I have no more sympathy left for any of the dot-com a** holes. They are the reason the economy is on it's knees right now...
Actually, it's Worldcom's fault. B-)
The Internet was supposed to have a doubling time that amounted to growing at a factor of 10 per year. Turns out this was a myth. But it was a very believable myth. Even up to a week before the myth was exposed as false, well respected people (such as a certain Stanford professor) were promulgating it, in both media stories and learned papers.
Well, it turns out that, for about two years, the internet's traffic WAS growing at a factor of ten per year. This was just after it was opened to general exploitation, and was probably the result of pent-up demand.
But that was about 7 years to 5 years in the past. After that it slowed down to about a factor of two per year. (At least for the next three years, until the dot-com collapse. Last I looked the nubmers weren't all in for the following period, and maybe it was a bit lower.)
Now doubling every year is still phenomonal growth. (Beats Moore's Law by a bunch.) But it's still a shortfall from what was expected by a factor of five per year.
Well the network equipment manufacturers designed and built for the factor-of-ten. And the dot-com companies built business models based on the factor-of-ten. And then the demand failed to materialize. By the time of the dot-com collapse the shortfall was a factor of 625.
Now some of those dot-coms were crooks milking venture capitalists and 401(k) plans. And some of them had business models that amounted to "lose a little on each sale and make it back on the volume", or "if you build it they will buy". But don't you think that SOME of them might have been viable businesses IF their market was as big as they thought - 625 times what it really was - and doubling every hundred-ish days?
And by now the shortfall is about a factor of five THOUSAND from what people were building for. Any bets on why there's a telecom/internet equipment manufacturer collapse?
Now all this apparently happened because Worldcom's UUNET subsidiary was cooking their books. They kept saying that the net was growing at this mythological factor of ten - but what was ACTUALLY behind the stat (if anything at all) was an expansion of capacity, not traffic.
Much like saying "We built enough storm drains this year to handle ten times the water as we could handle last year. We'd better build ten times as many this year, or we'll be flooded when the rains come".
Unfortunately, nobody checked their numbers. (They were the biggest backbone - by their own numbers B-) - so they REALLY ought to know...) Other companies that weren't keeping up with the mythological growth rate blamed their sales operations rather than questioning the numbers.
Eventually the real numbers finally came to light as a result of the investigation of Worldcom's other accounting "innovations". B-)
So that's why I say that it's all Worldcom's fault.
I'm afraid that it's your own ignorance that is showing.
(Unfortunately, at least four people with moderator points (and similar ignorance) believed you. So they moderated my posting "overrated", knocking it down below your own.)
ICANN only does domain names. IP addresses are handled by IANA.
Wrong. See below.
I've heard exactly zero complaints about IANA.
Yes, there were very few complaints about the actions of the IANA, and for good reason. The IANA is also known by his given name: Jon Postel. See his eulogy: RFC 2468, as in "Who do we appreciate?") Jon was one of the founders of the Internet, and did a fantastic job of handing out (and delegating the handing out) of unique identifiers.
Unfortunately, Jon died in October of 1998, and his benovolent dictatorship has been supplanted by his successors - a not-so-benevolent junta - the ICANN.
As to what ICANN does, here's the first two sentences from their ICANN Fact Sheet:
- DNS technical issues
- allocation of IP address space
- assignment of protocol parameters (including, in particular, protocol identification numbers)
- management of the root server system.
Got that?
Continuing:
The point of my posting was that assigning unique identifiers in a global name space is an indivisible transaction. A mechanism for unambiguously performing such indivisible transactions IS an authority - whether that "authority" is a database engine, a benevolent dictator, or the clerk who counts the votes of a committee or electorate.
What prevents us from ignoring ICANN when we feel like it and doing our own thing?
...)
The need for unique identifiers.
How do you do an internet if:
- A particular IP address may map to different hosts. (Packets addressed to 64.28.67.150 go to www.slashdot.org according to ICANN, a Microsoft server according to Joe's Nameservice and Grill, the US Army Recruiter according to the MIL BGP servers, and the local kindergarden according to a router configured by a junior high student. Which authority - and thus which route - do you think a commercial ISP (PAID to deliver packets) will honor?)
- A particular domain name may map to different organizations. (joe_user@slashdot.org may go to joe at VALinux, joe at Microsoft world headquarters, joe at the draft board, joe at the local kindergarden,
- A particular protocol number may specify different protocols. (WHICH IPV4 are you talking about? Which SMTP? Which NFS?)
- A "well known port" may perform different functions. (Imagine a new Microsoft OS putting a webserver up on port 414, or whatever port is used for an important service in the latest competing OS, and configuring the next release of Internet Explorer to try that port first. No need to "embrace" and "extend" before getting to "extinguish".)
and so on, depending on which organization the owner of any particular machine is affiliated with?
The answer is: You don't. (It's like the street addresses, state names, and personal names being a matter of political debate and faction-fighting - while someone's trying to send you a letter.)
Assigning a unique name or number is an indivisible transaction. In the absense of a solution to the "distributed update" problem you HAVE to do that with a single-point mechanism - an "authority". The best solution yet found is delegation - which is what ICANN does with domain naming and selling blocks of IP numbers.
Which brings up the question: Why are domain names handled by ICANN, rather than the trademark/servicemark section of the Patent and Trademark office?
Are they trying to re-invent opto-isolators?
Yep. They did. See their web site. (Sorry, don't have time to copy the link from another posting...)
But when THEY make an opto-isolator they put the emitter(s) and the detector(s) on the same chip, getting the alignment between them by using the same set of masks. Then they convert the region between the two sides into silicon Dioxide (also known as "glass") to form a VERY GOOD transparent insulator without disturbing the alignment.
Hm, the physical layer of the DSL line can just be described as a MODEM kind of thing. But there are other layers in the DSL box.
Right. It includes a MODEM, and other stuff. The Alcatel ANTs, for instance, support packet-over-ATM with up to 8 virtual circuits, so there's an ATM layer. Then there's an Ethernet layer. And it has a whole internet stack and an HTML server to provide a virtual console for configuration.
That's a bit more than your typical MODEM, even in these days of peripherals smarter than the mainframes of a decade ago.
I say inform your local newspaper or television news station.
Your ISP and local TV station are very likely owned either by the same media conglomerate as your ISP, or another that is pulling the same stunt. If you live in anything larger than a village, the same is probably true of your local newspaper.
Blue Gene is NOT a shared-memory computer -- with a single kernel running all 64k processors -- but rather a cluster of 32k seperate computers (with two processors), each on one chip.
Looks like someone already imagined a Beowulf Cluster.
I guess now we have to imagine a Beowulf Cluster of Beowulf Clusters to imagine a Beowulf Cluster of these.
The mind boggles.
It seems to be the ideal system to run the next Microsoft operating system.
But think how much it would cost for the 65,000 licenses.
Not to mention the network time to activate them all.
Then heaven help them if they add a peripheral, say just before a high-profile chess match, and need to REactivate them all.
Now if the mean time to failure of a MS system gets up to, say, 30 days, and you have 65,000+ processors, that's a mean time to failure of about 40 seconds. It would be 2167 reboots per day if a reboot didn't take more than 40 seconds. B-)
Media Companies, Communications Companies, Oil Companies...they all yell about how new technologies will ruin their business models and how they need to be protected!
...
... Didn't they run the music distribution business back in the crosby/ol'-blue'eyes/jukebox days? Like with "kneecapping" and extorting a Detroit radio station to kill the career of a local singer who sounded too much like their star?
The Mafia has a very effective business model which the government has also tried to ruin (or at least CLAIMED to BE TRYING to ruin) from time to time. A big part of their business model consists of using the government to keep the competition from doing the same things that THEY do, while buying off its agents so THEY don't have to obey the laws imposed on others. Another part consists of lobbying FOR laws against entry into their product lines - drugs, "Adult Entertainment", extortion,
Wait a minute
And aren't a couple of the current media conglomerates direct descendants of those organizations? Without EVER having the organized crime members running their "business model" systematically extracted (other than one at a time, over outside activities) from their operations?
Or am I confused?
Presumably this also means that if you're using their "cell", they can charge you what they like. I can see their motivation ...
Bingo!
Cell phones bypass the airphone, with its big bill of which the airline gets a cut.
Why should the airline take ANY risk of interference with the flight insturments when it's also costing them money? But they might accept a little when it's both under their control as to interference AND it's PAYING them money.
But I bet part of the impetus comes from the cellphone companies themselves. Using a cellphone in the air works. But on the ground a cellphone is "heard" by only a handfull of cells. In the air it is "heard" by MANY cells, chewing up bandwidth on each - and the SAME chunk of it, making the allocation of channels to calls on the ground difficlut. So even if you're paying for the call you're a net loss to your cell carrier, possibly forcing him to drop several calls by other customers.
Why the use of ANY electronic device is prohibited below a certain altitude, except when sitting still at the gate?
Because you can get away with just about any manouver in an airplane as long as you don't do it close to the ground. Continents have the right of way.
Takeoffs and landings require extreme precision, because going about a foot low means destroying the plane and possibly the cockpit crew and many in the cabin. There are a BUNCH of radio-based aids on a large number of frequencies and using a variety of methods - and if the one that's being used to guide the plane at a particular instant is suddenly interfered with, there may be no time to recognize that it's malfunctioning and switch to something else. So screwing up any one of them at a critical moment may result in a landing you don't walk away from, a mid-air collision, or some other mishap.
Similarly, the airport and the space immediately adjacent is a 3D traffic jam, coordinated by radio calls. Garbling even one radio message could result in a collision, in the air or on the ground. (As with highways they have a few even when they're NOT being interfered with. Now imagine highways with an occasional light going all-ways-green...)
Once the plane is AWAY from the space around the airport it has an ENORMOUS space to work in, and considerable time to work with. And there are "lanes" in airspace, as well as a rule that breaks it into stacks of altitude ranges where everything that isn't passing through in a well-known place is going in about the same direction. So if your laptop jams a navigational aid there's time to switch to another. (And if it somehow jams ALL of 'em the crew can run on internal nav and non-radio instruments and avoid other airplanes and mountains until the stew can get you to turn the bloody thing off.)
"Is Linux used in production telephony"?
What the HECK do you mean by that?
Are you talking about:
1) The core of a telephone carrier's network?
2) The core of the network of an ISP that is providing some telephony-related application (like POTS-emulation-over-cable, VoIP, or VoIP-related QoS enhancements)?
3) Commercial standalone PBXes?
4) PBX replacements (as a plugin card/driver/app for a PC)?
5) Modem-based answering machine/fax applications?
6) Desktop VoIP applications?
7) Server-room network VoIP servers?
8) Server-room VoIP/POTS bridges?
9) Voice menu hell servers - standalone or part of one of the above?
or a host of other "Production Telephony" applications?
When I saw the question the first thing I thought was 1). But the text seems more directed to 3) and 9), while responders are all over the map.
EACH of these seems to deserve its own item!
Well, according to my SmartFilter, not only do they sell PBX stuff, but also oodles of sex.
Interesting.
Now I have no idea that SmartFilter, Microsoft, or anybody else is doing this. But...
Wouldn't it be a powerful marketing strategy to get your competitors listed as sex (or otherwise icky-poo) sites on as many censorware lists as possible?
It wouldn't be anywhere NEAR as obvious as getting them onto black hole lists. Email disruptions would be noticed right away. Censorware deletions are much more subtle - and less suspect.
Even better, when the US Feds come with guns and loudspeakers, it's always in your best interset to let them in; foreign governemtns and internal cults alike get smacked by the feds.
According to the Branch Davidians they TRIED to let 'em in - and Koresh got shot while trying. And a lot more later, which amounted to the Davidians believing that they wouldn't be allowed to surrender - and that they'd seen a number of people being shot for trying.
Similarly with Ruby Ridge. That standoff started with a government tresspasser shooting a dog and a teenage kid. Then it continued with a remote-controlled buggy with a phone and a gun coming up to the door and the phone ringing with the gun pointed at the house. And a sniper shooting a mother with babe-in-arms.
There are a few documentaries on this. You're welcome to believe them or dismiss them as nutcase propaganda. (But at least one of those on Waco is by a local TV news reporter who started out thinking they were a homicidal nut cult and ended up thinking they were a majority-non-white church group that had been systematically tortured to death by a jackbooted death-squad. And with work by Failure Analysis and infrared footage from the government itself to back up this interpretation.)
Bottom line is, when trying to get out from under the government's thumb, you have to be careful not to end up ground under its heel.
And once you start, stopping may not be under your control. It only takes ONE side to run a war.
Isn't this where the idea of digital logic really got started?
Yes.
At least its how it was taught when I was in school. We even did some design work in async. Cool stuff. Easy to do, fast as hell...
Never did figure out why it never caught on.
It DID catch on. But the chips kept getting bigger.
It's easier to design silicon compilers for synchronous designs than for asynchronous - and when you've got millions of gates per chip you REALLY want compiler assist, rather than to lay out all the circuit details by human effort.
It's also easier to make automated TEST program generators for synchronous designs, to run the machines that test the chips when they come out of fabrication and reject the ones that are broken. You NEVER get high-90s coverage with human-generated "functional" tests - but a compiler can get there easily:
- Add muxes in front of the flops to string 'em into "scan chains" - big shift registers connected either to the regular pins or a "JTAG" controller. Then on the tester you'll:
- "Scan in" a random starting state.
- Step the chip a few times.
- "Scan out" the result and see if it matches expected, simultaneously scanning in a different starting condition.
The test generation program becomes essentially a random-number generator, chip simulator, and fault-tested-so-far counter, with a few finesses for things like getting things reset properly, testing gates with big fanin, making sure busses aren't floating, rejecting patterns that don't test anything new, working around flops that weren't on the scan chain because they were on a critical path, avoiding logic loops that become implied RS flops or ring oscilators (depending on whether the loop has an even or odd number of inversions), identifying logic circuits that have untestable failure modes, and the like.
But full-scan and partial-scan don't work if the flops aren't tied to a small number of clock domains that can be tied together or otherwise controlled directly by the tester. Asynchronous logic elements (such as ripple counters or other circuitry where a flop's clock is driven from another flop's output, or other logic that's something other than a clock distribution and switching system) just don't scan well.
There IS a way to get the same sort of massive observability and controllability over asynchronous designs - the Cross Check array - along with automatic test program generation systems to work with it. (Think of DRAM- or active-matrix-LCD-style cross-point addressing of test-points and signal-injection points - about one for every four regular transistors on the chip.) It tests async designs just fine, and gives better coverage than full scan with about half the silicon overhead.
But it's patented. The company that made it never got much market penetration in the US fabs. It has since merged and the product may be completely gone at this point. Except for Sony, which had an unlimited license from funding them when they were a startup, their own software, and (as of a few years back at least) used it in all of their consumer chips.
I blame the chlorinated carbon molecule.
That doesn't explain why the profound increase in Autism is present in CALIFORNIA, with no evidence (so far) of anything similar elsewhere.
So, assuming that further research doesn't come up with a similar rise elsewhere, that raises the question of "Why California". (Or why heavily in California and more lightly elsewhere.) Exposure to chlorinated carbon compounds is not particularly higher there - especially given Californian's aversion to chlorinated water.
Continuing to assume that causality implies correlation and it's a chemical issue, let's go down the list.
Right at the top: Gasoline oxygenation additives - in particular: MTBE, to which everybody in California who pumps their own gas - or drinks water - is exposed to in significant amounts, and has been for years. It's quite toxic, and has been implicated as a cause of asthma for a while now.
Certain illegal drugs exposure during and just before pregnancy is probably significantly higher in California - both because the drugs have been more prevalent here and because people interested in taking them have migrated here. But they're hardly unknown or unpopular elsewhere.
Dropping drugs for a bit: Another candidate is child rearing practices - in schools and home. California is the epicenter of the feel-good schools of child rearing. Systematically reward self-destructive behavior with attention and you can quickly teach a child to emit it continuously. And self-esteem based teaching systems will do precicely that.
California is also a hotbed of NON-vaccination. Side-effects of childhood or foetal viral infection immediately springs to mind. (Second trimester influenza has already been implicated in another mental syndrome - Schizophrenia.)
I could go on. But somehow I don't think the culprit will turn out to be Freon, PCB, Vynil Chloride, or chlorinated water byproducts.
This is what Nobel laureate Konrad Lorenz had to say about this way back in 1973-"Human culture, after enveloping and filling the whole globe, is in danger of being killed by its own excretion, of dying from an illness closely analogous to uraemia.
"Human Culture"? Yes, some of them will change. Some will die off or mutate, some will grow or shrink, some new ones will from. They do that from time to time - often on a scale of hundreds of years or less.
Human Beings, and extinction? Hardly. (Though the current enormous population is supported by farm, transport, and food preservation technologies - so a loss of this tech or an increase in its price, through economic collapse or regulation, means a significant die-off.)
Humans started out as a handfull of hunter-gatherers, before or during the last ice age. They expanded to inhabit essentially every bit of land area and floating ice except Antarctica BEFORE they developed industrial civilization and the scientific method. (Name another animal - other than human parasites - that managed that.)
Plains, deserts, steppes, mountains, ice caps... I doubt humanity could be wiped out by any climatic change that didn't boil or freeze ALL the water or eliminate all oxygen from the air.
The planet finally coming the rest of the way out of the last Ice Age - with the temperate zones shifting a couple hundred miles further from the equator and steaming jungles expanding beyond Brazil and central Africa - doesn't even qualify. (Heck: For raw biomass, suitably modified crops, or even CURRENT crops, it's probably a significant improvement.) And some of us would count the loss of the outer edge of certain seacoast cities to be a bonus. B-) Going back into a full Ice Age is more of a problem - though the greening of the equatorial deserts might make up for the loss of some more poleward land to glaciers.
Of course, if temperature shifts actually become a problem we can fix them directly, without screwing around with the CO2 level of the atmosphere. Just orbit a few hundred square miles of aluminized mylar, suitably located and oriented to provide a bit of shade if things are getting too hot, a bit of extra sunlight if they're getting too cold. Or whatever hack the rocket scientists come up with that's cheaper.
You want a robust space program anyhow - so you have something to spot and deflect the next incoming asteroid or comet fragment. Such an impact turning the whole planet into a broiler-oven for a day or so is the REAL threat of "global warming". THAT would once again reduce the ecosystem to plants with very robust seeds and resistance to PH variatioins and mouse-sized animals that happened to be underground at the time. (And maybe a few humans who had hung out in underground sites that didn't collapse and squirreled away a few years of supplies to last until they could grow something to eat.)
But I doubt temperature shifts (let alone the handfull of degrees that has the lefties drooling for more power and the media paniced) will be a problem for food production at all. Most of the food production of the world is now essentially an industrial operation, while the rest benefits from the tech. A few degrees of temperature change just means you change which crops - or which strains of a particular crop - you grow in a particular field. Shifts in weather patterns ditto, maybe with a change in irrigation or include the crops' water usage in the selection criteria, a few marginal plots going out of production, and new land becoming able to support crops.
Humanity will be forced to invent some sort of planetary kidney - or it will die from its own waste products."
Now that's true. But we've been doing EXACTLY THAT for quite a while now. When any given type of pollution becomes enough of a problem to bother with, we FIX it. Baby Boomers are old enough to remember Los Angeles smog before auto industry folk (including me) fixed up the engines. But that's NOTHING compared to, say, the killer fogs of London (driven by high-sulfur heating coal). Or just the indoor air of any human habitation in a cold climate before gas heat. And just think a moment about the streets of a city served only by horse- and ox-drawn vehicles. Talk about pollution...
Tech sometimes creates a new sort, or new amount, of pollution - "excretion" in Lorenz's vocabulary. But once it becomes a problem, more tech generally solves it (sometimes after quite a few years of griping by the people for whom it is a problem.)
Personally, I'm strongly of the opinion that both of these viewpoints are harmful. Over on the right there seems to be a lack of consideration for other very localized harm burning nasty stuff can cause. As a lifelong inhabitant of Los Angeles I've seen this first hand.
As someone married to an American Indian, who grew up on a west-coast reservation (of a different tribe - her mom was a teacher) with degree in history among her collection (and her dad was a history professor), let me tell you something about Los Angeles (that she brings up whenever it an air pollution are mentioned together B-) ).
Seems the local Indian name for the area translates to "Valley of the Smokes". The shape of the land and the wind patterns over much of the year trap airborne pollution - so badly that a single campfire would smoke it up for a day or more.
It's a testimony to US automobile technology (even if driven by legislation) that so many cars can now operate in that valley without photochemical smog being so thick that the light is blocked.
By the way: DON'T call them "Native Americans". It annoys them. (If you're born here YOU are a "Native American".) "American Indians", however, is a running "ignorant/stupid/crazy European invaders" joke: They were so dumb they thought they were in India - half a world away, Ho Ho! B-) (A poll of members of a large number of tribes showed the preference for "A. I." over "N. A." runs in the 80s-90s% range.)
... say node A is visible to base B and base C but base B and ase C are out of each others range. If B and C don't "hear" each other they can't work in unison without a third party. So when Node A (mobile presumably) hunts for a connection, both B and C try to talk at the same time to A, thus hampering the usable bandwidth.
But mobile node A can BE the "third party". If base stations B and C have unique (or at least non-colliding) identifiers then mobile node A can say, as part of its transmission, "I'm talking to B." or "I'm talking to C."
There are variants that work even if mobile node A is saying "Who's out there for me to talk to?". Look at the IP address resolution protocols for working examples. For starters, the a similar case arises on an Ethernet when a machine that wants to be booted uses RARP, broadcasting its Ethernet address and asking potentially redundant servers for its IP address. (Granted the servers COULD hear each other. But the protocol works even if they can't.)
You can also "flash" other areas by making a no-effect change (like inserting no characters). Like a blink comparitor when looking for star patterns.
/tmp/%, then repeated control-shift-"u". The multiple undo wouldn't break that.
Actually, you can't, since "u" only undoes the last change. I don't know where that came from. (That's what I get for posting when half-asleep. B-( )
There is a way to blink multiple changes that I can't recall just now - I think it's writing the file to
Of course with multi-level undo you don't have to "blink" 'em to find 'em, so it really is an improvement.
But it IS an annoyance. What it really breaks is hammering on "u" to blink the last change while you decide which way you want it. Try that with vim and it merrily unwinds your edit session. Then you have to use control-R - a two-key-one-hand annoyance - to get your changes back.
Why not use Vim then... syntax highlighting, multi-level undo and all the other goodies, but with the efficient VI key bindings :-)
Actually, I do, generally when I'm on a linux platform. It's close enough to vi to be almost interchangable.
That multi-level undo has a downside, though. It kills a vi idiom: In vi "u" undoes the last change - but a "change" includes a previous undo. So hitting "u" repeatedly first causes the cursor to jump to the latest change, then "flashes" the change, making it even easier to spot. You can also "flash" other areas by making a no-effect change (like inserting no characters). Like a blink comparitor when looking for star patterns.
Best of both worlds would have been if vim had used a different keystroke for multi-level undo (like it added a stroke for multi-level redo), or had a switch to revert to vi behavior.
They can only do that with the code as-is or with purely in-house development added to it. They won't be able to say that about any GPL-sphere additions- if you want to nip this in the bud, do such good work on the GPLed version that it becomes the definitive one and completely leaves the in-house version in the dust.
On the other hand, if you want to have the best of both worlds, do some really good work in the GPLed branch (carefully avoiding the inclusion of other people's GPLed code in your work). Then offer to license it back to the foundation to also be included in the "in-house" version - for a fee. B-)
I agree. In no way, shape, or form, is the "vi" interface a good one.
Huh? It's fast, it's efficient and it's easy on your fingers. How is that a bad thing? Just because you don't like it doesn't mean everyone has to agree.
Hear hear!
Back when I got my first unix box (FAR enough back that, when then entire list of email-connected sites fit on three pages, mine was there), I wanted to build and try emacs. But there was this little problem - the machine had only 2 megabytes and no demand paging. Emacs (even back then) wouldn't fit. (A tongue-in-cheek claim was circulating that the name was an acronym: Eight Megabytes And Constantly Swapping. B-) )
So I learned to use vi.
And then I was VERY active on a bulletin board for several years - using vi. And I got very fast with it.
Some time later I had access to a bigger machine and a colleague pointed out that emacs had a vi emulation mode - so I could ease in without having to learn new navigation keys right off the bat. I looked into it - and it had TWO distinct vi emulation modes. Oops. With one I might have tried it. But I didn't have time to find the better of the two. So I dropped it.
A little later a Netnews posting demoed a potential attack on those who used emacs as a news reader or mail reader. Seems that emacs had a little-know feature: You could include a snippet of lisp code in the comments in a program file, and emacs would run it. This was intended to set up tab stops, language editing modes, and the like. But this also worked in mail and netnews reading modes. The demo's lisp code would pop up a "See, I got you!" window and delete itself from the display of the item itself. But in principle it could do anything you could do from emacs - which is anything you can do from any shell, with a lisp interpreter handy to do complicated stuff. No clicking on attachments - just LOOK at the letter or news item and you're owned.
Windows macro virus vulnerability? Emacs had it first, and BETTER! B-) Imagine a lisp worm in netnews forging postings in your name, both replicating itself on "nice" groups and faking love letters on alt.binaries.pictures.child-molestation. Or dumping the contents of any "src" directory you can read to an alt.binaries group. (And heaven help you if you read news or mail when logged in as root...)
Of course this "feature" was on by default in the standard distribution. In those days, or days not too much earlier, RMS' approach to security was rumored to be having a blank password on root in his personal machine and letting this be known - in the belief that if there was no skill needed to break in, and thus no reputation to be gained, nobody would bother. (Apparently that worked with MIT students. But don't try it with the general population net-connected.)
Well, I had spent years doing classified research, which made me itch about security holes. So I decided to stick to vi for a while longer - along with the plethora of unix utilities that do essentially anything I need done that's beyond vi's power.
Since then I've occasionally seen an emacs-ism that has tempted me - like colored displays of comments vs. declarations vs. code. But every time I'm tempted I watch a colleague doing simple text editing with emacs, and count the keystrokes he has to use to do the simple stuff that constitutes the bulk of my editing work. And it always seems to take him a lot more strokes with emacs than it takes me with vi. So I'm generally not tempted for long.
Vi was designed for a very different world - the world of dumb character-based computer terminals in the days before ANSI standardized their behavior. There were literally HUNDREDS of different terminal designs, with a boggling array of differences in display geometry, control-character to cursor-motion mapping, and other odities. Vi (actually the "visual" mode of the "ex" editor) encapsulated these idiosyncrasies in a "termcap" (terminal-capability) definition file, thus letting the user do full-screen WYSIWYG editing on ANY of them using a common set of keystrokes - and letting the sysadmin add definitions for new terminals as they came out. This brought the user out of the dim world of command-line-only editors (such as "ed" and "teco") into the instant feedback of a screen display - halfway to the window systems that weren't available yet.
And - much to the surprise of its author - it did it very well. So well that people like me (who now have the vi commands "hard-wired" into our nervous systems from long use) still use it when we have serious text hacking to do.
It was 54 years ago in central Italy, driving at night on a desert mountain road. I saw a fiery fireball in the sky, moving slowly from left to right.
I had the time to: understand (maybe) what it was, wake up my wife, stop the car, get out an look. Total time maybe 20 seconds. The 'object' was moving slowly, spewing green flames and eaving a long lasting orange trail behind. Trajectory was more or less horizontal. It disapeared in a flash.
I (and a bunch of other people) saw something similar at Lake Tahoe a few years back. Distinctly purple, slow moving, leaving trail, no sound, no flash.
Turned out to be a space shuttle re-entering on its way to Cape Canaveral. The purple is due to the composition of the tiles. The final orbit and upper-atmosphere reentry is visible over Tahoe due to the inclination. (I think that's a side-effect of chosing an inclination that lets them switch to Vandenberg [on the next pass?] if the weather at Canaveral is too crummy.)
Why do you need 1 MB/s for a big honking DRILL?
Well, for starters you could put an array of accoustic, microwave, or electrical transmitters & sensors in the pipe just BEHIND the drill and image the region ahead of the drill with radar and/or sonar. If you see a pocket of something that sounds/conducts/reflects like oil a bit off to one side, you can adjust the drill to curve in that direction (or send the NEXT one over that way).
10 BPS just doesn't cut it for uploading imaging information, even if you put most of the fancy processing down with the sensore. But T1 rates are just fine.
There's lots of other stuff you want to monitor - temperature, pressure, conductivity, etc. to find out what sort of stuff you're drilling through.
And it's important to know when to give up, stop pouring money down THIS hole and start over somewhere else. It costs a LOT to run the rig long enough to drill even another foot...
I recall, back in the early days, a company in Ann Arbor made a little board with a CMOS Z80-clone, a ROM with a BASIC interpreter, a serial port, and a few I/O ports - including some analog inputs. They sold a LOT of 'em to an oil company.
Seems that every now and then they would pull up the drill and send one of these down to measure some stuff. Then they would send the drill down behind it and grind it up. It was cheaper to buy a new one (and the associated cable) each time than to leave the rig idle long enough to pull the old one up. (And considering how fast a winch can crank, and how much custom computer stuff cost back in those days, that will tell you a lot about the per-minute cost of an oil rig and drilling team.)
So imagine how much they can save if they don't need to pull the DRILL up - disassembling it as they go - then reverse the whole process to put it back down, every time they want to take another reading.