You might have noticed (had you read the flightship pages) that one curious thing is that they describe it as for "tropical" locations. I suspect that's specifically to avoid the high waves you commmonly find in and around North Sea and English Channel, for instance. This would work well in the predominantly calm tropical waters. (And, FWIW, big storms interfere with shipping schedules, too, so what's the difference? Higher speed and modern satellite weather info should pretty much eliminate all significant surprises to any reasonably prudent flightship master.)
Actually, you can make an argument that OV is showing up to the party just as the police are chasing everyone off and no one cares for party tricks any more. (Maybe that part about the police is a little *too* good an analogy...)
In any case the reality is that I'm not sure the world needs another audio compression method. I'm not even sure the ones we've got will be anything but a curiosity in another year or two: With Moore's law still making networking faster and storage bigger and cheaper, it's possible that the preferred audio format may simply be raw WAV rips of the CD Audio tracks. The nice thing about that is that even the technologically illiterate can ow play high-quality audio without having to concern themselves with codecs or optimized encoding parameters to ensure good quality when ripping/encoding. The best way to make the encoding problem go away is simply to make encoding go away. With cheap 160 GB drives already on the market, this *will* happen: audiophiles first, then many of the rest of us as prices continue to fall...
The OSI 7-layer model is useless. It reflects no real-world network implementation other than X.25, which was designed to fit it rather than the other way around (despite the fact that it's a reference model, not an implementation model - anyone who's worked with X.25 knows what a pain some of these decisions created.)
True story: Dave Clark, of MIT and one of the fathers of TCP/IP, used to teach a class on "The art and engineering of protocol performance" at Interop (back when it only had one name and fit easily in the small San Jose convention center.) In one of these classes, someone asked why mapping the 7-layer OSI model to the real-world TCP/IP model was so awkward. He told this true story (he was a member of one of the committees nivolved at the time): When the OSI decided to study the networking problem, they formed seven study commitees, pretty much arbitrarily, to research the problem. They were NEVER intended to define boundaries. But in typical ISO "politics taking precedence over reality", when they got back together, they could not agree on how and where to create interface divisions, so to avoid deadlock, they finally decided to just use the ones they had for the study committees. So there you have it - there are seven layers because there were seven study committees, and that goes a long way to explaining the uselessness of the OSI model.
Personally, I think we should exorcise this horror of French bureaucracy from every network textbook on the planet. The number of man-hours that have been wasted trying to apply it to the real world (especially since TCP/IP trounced OSI permanently almost 10 years ago now) is staggering. It would be hard to identify a single more damaging and bone-headed idea in the world of networking than the OSI model. It should die, along with pretty nearly all other ISO standards relating to networking and communications. (And yes, I realize that is quite a strong statement from me, but it's true.)
Since about 1998, this has been taken care of by the "palm rejection" feature of the Synaptics touchpad driver. Of course it only works for Windows, and so far as I know, there's no equivalent for Unix/Linux, but then, laptops *still* aren't Linux-friendly, but at least we've progressed to the point that you can *load* Linux on most of them. (FWIW: I have still *never* seen properly functioning audio on *any* Linux machine, laptop or not. Sad, but true. I'm told it's possible, just not with any of the hardware I've ever owned (which is only about half oddball stuff like Librettos.))
As someone who used to be a program manager in Dell's portables group, I can tell you that Dell has a number of product marketing people that ensure that a special competition lab prepares machines for tests (they get tweaks the rest of us may never see, both hardware and software, mostly drivers), then carefully follow-up on the testing and results reporting/weighting to ensure that Dell is always at or near the top. (You'll notice that weight itself is seldom given much weight in notebook tests that include Dells, primarily because they have historically tended to be on the pudgy side - the Latitude CP family contains a half-pound of *screws* for cryin' out loud...)
This isn't meant as a slam at Dell - I suspect nearly all other OEMs do this as well, but this is an area in which Dell makes it a point to execute even better than usual. It's simply not possible to be important enough to get a "test machine" from Dell and not have some PM continually looking over your shoulder throughout the testing. Is that undue influence on the results? I guess that depends on your perspective...
As for the advertising stick, I've never seen it used, but then, it doesn't really have to be, now does it? When everyone *knows* you're carrying a big stick, you have the luxury of being able to speak softly. (TR was right about so many things...)
Your post is self-contradictory, primarily because you seem to have no understanding of derivative works under copyright law.
First, of course, you're wrong in stating that You cannot copyright the information in a phone book because all it contains is a list of facts.
Nowhere in copyright law is (or should) the distinction of whether or not the material is factual enter in to the picture. By your logic, only fiction would be protectable: clearly a ludicrous proposition. As a more practical reality check on my argument, I urge you to check the first few pages of your local phone books - you'll find that regardless of the fact that they're simply compilations of facts, they are indeed copyrighted, and such copyrights have been correctly upheld by the courts. (There have been some particularly interesting cases regarding yellow pages: not surprising, since that's where the money is in that business.)
Using the facts from a Microsoft specification document (patent considerations aside) may or may not be infringing, depending, as I understand it, on whether or not "a reasonable man" would conclude that the new work was in fact "derived" from the original MS work. In that case, it would infringe the MS copyright, regardless of what new form it is cast into. (So simply restating the MS information in a new form would still result in an infringing document.) If it was so different as to not be seen as clearly derivative, then it might be judged non-infringing. Unfortunately, the only way to know for sure is to go to court. (That's not necessarily a bad thing, either: we do not want the law to attempt to speculatively handle all possible eventualities.)
VarTec is indeed the chaeapest LD provider I've found. The vast majority of my LD is within Texas, where it's easy to get ripped by the intrastate rates. FYI, you can compare LD plans to find the cheapest for your particular situation at ratekeeper.com (I have no relationship with them, other than having found their service to be handy.) Note that it's pretty much *always* cheaper to do the 1010xyz bypass than to dial direct.
I have no idea why, but I can't touch VarTec's bypass with direct service from anyone. No big deal, I simply have the bypass # (1010811) programmed into my phone for one-touch dialing. After a week or two, you just naturally whack that button before dialing an LD call - For that kind of savings I'll put up with a very minor inconvenience of pressing one more button...
This is asinine. I'm sure the IEEE, the OSI, and other stadards bodies will be quite shocked to find that they can't keep soaking you for the big bucks in return for those "open" standards. Internet standards are a great model, and we should see more of them, but they are clearly not yet the norm everywhere. Most "standards" are indeed copyrighted and not freely available. Some of them are quite expensive, to boot. (The oppressive policies of OSI and the ITU were one reason the IETF decided to come up with its own standards processes...)
Of course, you're leaving out the most hideous, complex, and mine-ridden install of any OS on the planet. I'm convinced the Debian developers do this on purpose so that any Debian users have proven themselves "worthy" simply by virtue of jumping through the endless hoops required the get the piece of dung running. I for one have far better things to do with my time. (And for the record, the only time I have ever totally horked a Linux machine was Debian's fault after doing an apt-get update (yes, from stable): the machine froze instantly and was so botched it would not boot. It was easier to start over with Red Hat or Caldera (I don't recall which I used, now) than try to untangle the damage.)
I like some of the concepts in Debian, but it's really a pain in the butt to get up and running correctly. Somehow the Debian team thinks this situation is OK, as it's been that way for years now. If someone would munge the *concepts* behind e-smith and debian together, we'd really have something...
There's nothing in human experience compared to which a sendmail config file could be considered simple.
You've not been around long enough to have tried uucp, I can tell. I've mostly stopped having bad dreams that begin "tip cua0", and involve serial breakout boxes...;-)
The employer pays you to work, there are NO work reasons (cut the crap about tech support IRC and suchlike - i've heard it and seen what these guys talk about - there's no tech support going on at all - its chatting) for IM clients that i can see other than wasting time.
Two things:
1) It's the company's netowrk, they bought it, built it, and paid for it for the good of the company. They can allow or disallow anything they want, any time they want, for any reason or no reason.
2) I've never seen a "legitimate" use for chat/IM. Typically, it is used simply to continue one's social life (usually miserably dreary) through the workday without raising so many eyebrows as several calls an hour would. Chat and IM area blight upon the Internet - personally, I'm in favor of blocking this crap everywhere, not just at work. (I have observed a striking, nearly 1:1 correlation: People who use IM are more likely to claim things they cannot really do or deliver on - they will then use IM as a crutch to attempt to leech the knowledge of others rather than simply learning for themselves. Top producers (whether coders, designers, or marketing folks) NEVER use IM. The correlation is nearly perfect. I don't know why...
Lineo wound up owning DR-DOS after the split from Caldera, which just kep the Linux stuff. So what will happen to DR-DOS now? It was always a better DOS than Microsoft's. Here's hoping they turn it loose, along with thier DOS web browser (which, to be honest, was really doggy.)
It's too bad that Caldera caved in to Microsoft and settled instead of holding their feet to the fire - the DR-DOS suit evidence that was buried by that settlement would have made the antitrust case far stronger than it was anyway...
As I cited in a reply farther up the page, Rackspace [rackspace.com] has few local customers here in Cowtown, USA (aka San Antonio TX) but that hasn't stopped them from prospering.
Come on, if you've EVER seen a western movie (and there are lots of good ones, so start renting), you know that Cowtown is Fort Worth, not San Antonio. It shouldn't take a Texan to point this out...
San Antonio, by the way, is hardly a "cowtown", anyway: It's the 9th largest city in the US, ahead of Detroit(10), San Jose(11), Indianapolis(12), and San Francisco(13), and it's the center of a lot of interesting telecom activity, including, unfortunately, telemarketing. (Source: http://www.infoplease.com/ipa/A0763098.html)
One thing that's not often realized or acknowledged by many mech hackers on thier way to becoming gearheads is that the hackability of cars varies quite a bit. Accordingly (no Honda pun intended), all hacks involve tradeoffs. You may be able to squeeze another 15 HP out of that 4-banger, at the cost of shortening it's life considerably. (And grenading those expensive upgrade parts in the process...) Optimizing these tradeoffs is not easy.
Case in point: Really high performance cars are hard to hack. This just makes sense if you think about it. The simple fact is that what separates high performance cars from their more proletarian brethren is that they have *already* been "pre-hacked" to improve performance. The law of diminishing returns definitely applies here. Hot-rodding Ferraris, Porsches, Vipers, and Cobras is *hard* simply because all the easy, high bang-for-the-buck stuff has already been done, and then some. What's left is almost by definition in the "not worth the money" category. (Of course, that doesn't stop the legion that has more money than sense, especially if they can afford these cars in the first place. These people ruin a lot of good cars.) If you want to hack for hacking's sake, start with something other than the top end, unless you like low ROI. Otherwise, buy a fast car and enjoy it with no or minor tweaks.
There are cars out there that are designed with very conservative safety/durability margins that are eminently hackable. (RX-7s come to mind, as do Miatas, Mustangs and GM F-bodies. The new Focus is shaping up to be a really good candidate, perhaps becoming the Datsun 510 of this decade, but Ford, as usual, makes it hard to order all the right bits.) Also, don't forget the time-tested method of getting the big win: engine swaps. If you choose a bigger engine from the same manufacturer, these are often not even all that difficult, and usually (at least until you break out the "blue-tip wrench") offer a fall-back position, if things don't work.
Although little Japanese motors can be hacked, I don't think many of them are practical hacks, since you quickly have to start upgrading everythign else once you've started. Thisis why there's still a lot of truth in the old hot-rodder's saying, "There are two ways to win: cubic inches, or cubic dollars." V8s and engine swaps remain popular because they offer the former. I'd put most Asian hacks in the latter category, and if you're going to do that, why not just buy a fast car in the first place, as it will almost always produce superior results?
Remember one last thing: Power is your friend, weight is you enemy. Very few hot-rodders put much effort into weight reduction (and it's hard on many modern cars), but keep that in mind when buying a car - A Miata is relatively at least as hackable as a Mustang because of this. Starting light and going lighter can make a tremendous difference. If you really want light, buy a Lotus: Colin Chapman once said he designed Lotus race cars with the following in mind: "The perfect car disintegrates as it crosses the finish line." *That's* performance optimization, gentlemen.
RX-7s are GREAT cars, and if you don't do anything stupid, they're just about bulletproof. If you want a motor you can abuse with impunity, this is it. The main thing to remember is that rotaries are *designed* to burn oil (they must, to lube the apex seals), so you MUST check the oil frequently. Always remember that modding a rotary is DIFFERENT from modding a piston engine. Stay conservative, and you'll be fine. (Bill Buckley and I would say that's always good advice.)
Rotaries are simple, powerful and reliable. Without a muffler, they will make you ears bleed. And there's something to be said for a motor with only a few moving parts that can be rebuilt on your kitchen table. (While perhaps not for raw mechanical beginners, rebuilding a rotary is far easier than rebuilding a piston engine. There are a number of good books out there on the topic.)
ROFLMAO: My eyes are wet. THAT is the funniest thing I've seen in a long time...
Too true, but I'll take the brio and fury of an Italian car over the soulless, sterile Germans any day, and have more fun doing it. There's really no substitute for an Italian car. I'd argue that there are *some* good rice-burners, and that they're good *because* they have the soul of an Italian car: The WRX is clearly a Lancia in spirit, and the NSX is nothing less than a modern Ferrari Dino by Honda, a concept that has quite a bit of appeal. The Mazda rotaries aren't like anything else, but are clearly built by people that think much like the Italians do when building their cars. (If you're on a budget, try a second-generation RX-7, turbo if possible, although a Miata with a Jackson Racing supercharger woudl do in a pinch.)
(WARNING: Italian cars are an ADDICTIVE ILLNESS. Once you own an Alfa or Ferrari, you will lose touch with reality and your remaining friends and loved ones will think you've taken leave of your senses. Which you have. But you'll be having a ton of fun in a car that doesn't look like a slab of cheese and makes such wonderful mechanical noises that you won't care about the stereo. (Mine's been broken since 1990 - I don't care.) You'll laugh at BMW, Porsche, and Lexus owners that spend far more on thier cars, but get so much less in return. You will learn that although parts aren't expensive relative to those others (Ferrari parts are cheaper than Lexus parts!), getting them here quickly is another matter. You'll learn that reliablity is relative, and the electrical systems will reinforce your understanding of quantum uncertainty. You won't think too much of taking a lovely drive to another city to get the car worked on by a competent mechanic. You'll learn to do a lot yourself, and that "accensione" means "ignition". You'll go for a weekend drive and have silly grin on your face that will last until Wednesday. Once you're hooked, you won't rest until you have a Ferrari, so just go ahead and buy one. Mine's not for sale. You've been warned.)
Although Wiki webs are nice in concept, I have yet to see one that did not very soon degenerate into a morass of ill-connected thought fragments. In fact, I think Wikis are quite possibly the strongest argument ever encountered for site administrators and/or moderators.
For an example of what I mean, visit www.tuxscreen.net and just *try* to figure out what all you have to do to get the current version of the software installed and running. The information is there (mostly), but it's so fragmented as to make it pretty much impossible to find or use.
Still, there's no doubt that two-way webs are someting we should strive for, a la Ted Nelson's Xanadu concept.
Sadly, the fundamental problem is that information organization is what produces much of the value of information, and organizing information is something that's fundamentally hard and difficult to automate. (Not that the KM (Knowledge Management) guys aren't trying, but they've got a long way to go, and those are pretty rarefied tools, not available to most of us, either by reason of price or complexity/learning curve.)
Completely OT: Wouldn't WP's tagged formatting code method make it an ideal way to create low-end XML? It already has great word-processing features, and claims an XML format. WP could output SGML 8 yrs ago or more. Re: WP and XML, search google or see, for example: http://www.xml.com/pub/a/2000/05/31/wordperfect/
Hardly off-topic, since SO's file formats are a very valid dimension of it's readiness for the real world. WordPerfect's document capabilities are often underrated due to it's UI. I've had a love/hate relationship with WordPerfect over the years - it's positively user hostile at times and when things go wrong, it's as opaque as vi without the benefit of simplicity. Still, it's the only WP I've ever found that handles long documents well (which is why it's still the standard in much of the legal and real estate businesses), and the "Reveal Codes" feature is very nice for power users. In many ways it's the best mix out there of word processor and desktop publishing capabilities.
As an aside, I've used SO6beta for several months now, and *if* you put in the small effort to learn it's slightly different UI philosophy, you'll find that it's a VERY capable office suite. There are a few bugs (which I and others have filed) that keep me from making it my everyday tool, but overall, it's really quite good. Although there are some holes realtive to MS Office, it also covers some nice areas that Office doesn't. Sun's got a winner here. When it hits the shelves, I'll buy a boxed copy rather than upgrade the Office 97 I've been using for years - I really think it will be good enough to make that viable. The fact that it runs on non-Windows OSes is just icing on the cake. (C'mon BSD!)
Well, actually, this is pretty darn close to the truth for applications that are threaded. IIRC, Oracle ran about 28x as fast on a 30-way UE6000 as it did on a single processor of the same speed. (To be fair, there are darn few apps that are really written to take advantage of threading, Oracle is one of the best...)
I know this is hard for some Linux folks to swallow, but that's the reason some of us really love Solaris - it scales more linearly that anything else I've encountered. It's not exactly linear, but for well-written apps, it's pretty close...
Re:I'd recommend things like this to anyone.
on
Chase the Rabbits
·
· Score: 2
Reading this brought back memories... Peter Shankman is right that the most impressive thing about this sort of activity is learning that your real limits are FAR, FAR beyond where you think they are, even after you've passed the point of "can't go another step..." That's an important lesson, and one that sadly, most people never learn.
In my case, it wasn't SEAL training, but rather PT in the Texas Aggie Corps of Cadets, where "crap-outs" were a regular part of one's existence, especially in the freshman "fish" year. I didn't start out as a terribly motivated student, and I can honestly say that it was the Corps that kept me in college - I was already capable of making a pretty decent living in computers or experimental stress analysis before I got there, so the money/success angle of college wasn't really a motivator for me, but the comaraderie was.
This sort of PT is **invaluable**, but is all too often categorized as "hazing" in today's ridiculously whiny PC culture. (I somehow doubt that today's cadets have to snake-crawl through pig crap, but they should. It builds character - and I'm completely serious.) There really is quite a lot to the old-fashioned military training idea of tearing down the old man in order to build up the new - the experience is invaluable. Despite what the whiners say about never letting someone "abuse" them like that, it is paradoxically exactly that sort of humility and dedication to teamwork with one's classmates that builds the character required to truly act as an individual.
Interestingly, it's quite possible that the sort of "crap-out for hire" service described by Shankman may be the only thing short of real BUD/S training that will push you that far anymore. (I'm *sure* there are bulletproof waivers required in which participants give up rights to legal action for any and all abuse, physical or mental.)
Anywhow, if anyone is looking for a good, well-rounded educational experience, I highly recommend Texas A&M (damn good engineering/technical school anyway) and membership in the Corps of Cadets. The great General George S. Patton may have said it best: "Give me an army of West Point graduates and I'll win a battle. Give me a handful of Texas Aggies, and I'll win a war!"
Almost. There's one more difference in there, and it's a big one: quality control.
Actually, the real truth is rather surprising: most computer/storage system OEMs have only token quality requirements for disk drive reliability. Sun is one of the only vendors that really takes this seriously: when I worked there a few years ago, just after the introduction of the original SPARC Storage Array, it was judged that there were too many field failures of drives, and Sun worked with the drive vendors to come up with tighter drive specs. Then they started testing them by pulling a few at random out of each incoming truckload. They only had to reject a couple of 18-wheelers full of Seagate disks before all the vendors got the message that they needed to send the best drives to Sun if they wanted their business. Interestingly enough, I have some reason to believe that Sun was getting better quality drives from IBM during this period than they were able to get for themselves internally.
As an more on-topic observation, we build storage systems for a living, and I did quite a bit of research on the newest high performance IDE mechanisms for our flagship product. The upshot is that trouble with the IBM drives is somewhat spotty: some people have no trouble at all, others have horrendous failure rates. There seems to be no way to predict in advance how your particular drives will act.
Because we're usually building multi-TB systems, I couldn't take that risk, so we standardized on the Maxtors (80 and 100 GB, 5400 RPM), which have proven to be fast, solid, reliable workhorses. I need to look at the faster Maxtors and the new 160GB WD drive RSN, but will probably continue to stay away from the IBMs for a while, until the re=prove themselves.
Sad, because IBM was once the gold standard in this field. They do still seem to have the best 2.5" laptop disks, by far, though: I really don't think anyone but Fujitsu is even in the same ballpark with IBM for the little drives, but even Fujitsu is clearly back a bit in ruggedness and reliability.
If you want server drives, pay for them; if you want to pay for consumer drives, don't expect more than consumer quality. Enjoy it when you get it, but if it counts, pay for it.
This kind of flies in the face of the entire idea of RAID, where the "I" originally stood for "Inexpensive" (now represented as "Independent" in many cases, since modern high-po SCSI drives are anything but inexpensive...) The realities of volume manufacturing mean that there is and should be no real difference in these drives, and if there is, it's more likely to be in the other direction: I assure you that a PO'd consumer that wants to vent to support about crappy service and lost data will cost the Mfr. FAR more than a commercial account that routinely sends back a few boxes of disks every month as part of their operations. Nothing is so expensive to support as an individual consumer! (I know for sure it costs one major PC vendor over $30 every time they pick up the phone for a customer support call.)
(Clue: Love Field and Intercontinental both are _small_ airports, usually used for commuter flights in and out of Dallas and Houston- they normally don't rate a hand-held bomb detector unit...)
No, sorry, you're clueless. (Sorry for the ad hominem, but you have to admit you set yourself up for that one...)
Houston Intercontinental(IAH) is NOT a small airport (you probably meant Hobby (HOU), the "old" airport, still quite active, but much smaller than IAH): IAH is Continental Airlines home hub and the 17th busiest airport in the *world* by passenger count and 13th by traffic. Love Field in Dallas is no slouch, either: it's the home base for Southwest Airlines (the only airline making significant money at the moment) and would be far busier if it weren't for meddlesome federal rules limiting the use of this really convenient in-town airport to avoid "hurting" the regional behemoth. In fact, Love and Hobby are such success stories, and so well liked and used by travellers, that several years ago, the FAA started insisting that whenever a new regional behemoth is opened, the old airport must be *destroyed* and rendered unusable, even for general aviation (private planes). Denver and Austin are examples of cities that have suffered mightily because of this, and are forced to live with ridiculously expensive white elephant airports that might as well be in another county...
Point being, what you consider "fixable" depends largely on your knowledge base.
This is a VERY valid point. After many years of amassing a great deal of knowledge about Windows, Linux, and various Unixes, I have finally reached the conclusion that once they start to act flaky, it's ALMOST ALWAYS easier and quicker to simply reinstall the OS than to figure out what obscure setting, library, etc. got broken. (Not that one shouldn't find root causes, but let's face it, much of the time it's just not worth the effort.)
Linux and Windows are both quite difficult to administer in very different ways. Those of you that have never used a real production-quality OS (Solaris and DEC/Compaq Tru64 are, AIX, IRIX, & HPUX aren't, IMO) really don't know how stable and easy to run those systems can be.
I like and use Linux routinely, but it's not a panacea - my most recent Linux reinstall was caused by doing an upgrade via Debian's vaunted apt-get mechanism: I didn't bother to even figure out what apt-get did to completely scrog the OS - it was obviously very bad and quite clear that reinstalling from an old Mandrake CD I had lying around would be much quicker and easier. (No, I won't touch Debian again for another year or two.)
By the same token, bit rot munches my Win98 to the point that I have to start over from scratch about every six to eight months. Again, I *could* probably fix it given the time, but I've got beter things to do, so reinstalling makes sense, at least until we get OSes that deliver real stability.
Librettos are expensive even used, so I probably wouldn't buy one, but I had one given to me a few years ago (with the Japanese keyboard, which is a pain), so I turned it into a very nice little portable file server: When I'm travelling, I now find it much easier just to bring the Libretto along, knowing I can have instant Ethernet access to all the critical resources I might need. This is much easier than dragging along a regular laptop that needs its own luggage. (Rule for road warriors: NEVER buy a computer that requires luggage!)
It's only a 75 MHz Pentium w 16MB RAM, so it was never even suitable for Win95, but it runs a slim Mandrake just fine. Even KDE works, although it takes a while to start up.:-)
I will say that installing any OS on this thing is a major PITA, since it has no floppy and no CD-ROM. I'd like to replace Mandrake with e-smith, which is rapidly becoming my favorite general-purpose server distro, but e-smith only installs from CD - Ugh! I could work around it if I spent several hours, but that kind of defeats the purpose of e-smith.
I suppose there are smaller file servers, out there, but not by much, especially with a built-in UPS...:-)
Slashdot Mirror
I don't like H-Paq very much.
Shouldn't that be H-Pox?
You might have noticed (had you read the flightship pages) that one curious thing is that they describe it as for "tropical" locations. I suspect that's specifically to avoid the high waves you commmonly find in and around North Sea and English Channel, for instance. This would work well in the predominantly calm tropical waters. (And, FWIW, big storms interfere with shipping schedules, too, so what's the difference? Higher speed and modern satellite weather info should pretty much eliminate all significant surprises to any reasonably prudent flightship master.)
Actually, you can make an argument that OV is showing up to the party just as the police are chasing everyone off and no one cares for party tricks any more. (Maybe that part about the police is a little *too* good an analogy...)
In any case the reality is that I'm not sure the world needs another audio compression method. I'm not even sure the ones we've got will be anything but a curiosity in another year or two: With Moore's law still making networking faster and storage bigger and cheaper, it's possible that the preferred audio format may simply be raw WAV rips of the CD Audio tracks. The nice thing about that is that even the technologically illiterate can ow play high-quality audio without having to concern themselves with codecs or optimized encoding parameters to ensure good quality when ripping/encoding. The best way to make the encoding problem go away is simply to make encoding go away. With cheap 160 GB drives already on the market, this *will* happen: audiophiles first, then many of the rest of us as prices continue to fall...
The OSI 7-layer model is useless. It reflects no real-world network implementation other than X.25, which was designed to fit it rather than the other way around (despite the fact that it's a reference model, not an implementation model - anyone who's worked with X.25 knows what a pain some of these decisions created.)
True story: Dave Clark, of MIT and one of the fathers of TCP/IP, used to teach a class on "The art and engineering of protocol performance" at Interop (back when it only had one name and fit easily in the small San Jose convention center.)
In one of these classes, someone asked why mapping the 7-layer OSI model to the real-world TCP/IP model was so awkward. He told this true story (he was a member of one of the committees nivolved at the time): When the OSI decided to study the networking problem, they formed seven study commitees, pretty much arbitrarily, to research the problem. They were NEVER intended to define boundaries. But in typical ISO "politics taking precedence over reality", when they got back together, they could not agree on how and where to create interface divisions, so to avoid deadlock, they finally decided to just use the ones they had for the study committees. So there you have it - there are seven layers because there were seven study committees, and that goes a long way to explaining the uselessness of the OSI model.
Personally, I think we should exorcise this horror of French bureaucracy from every network textbook on the planet. The number of man-hours that have been wasted trying to apply it to the real world (especially since TCP/IP trounced OSI permanently almost 10 years ago now) is staggering. It would be hard to identify a single more damaging and bone-headed idea in the world of networking than the OSI model. It should die, along with pretty nearly all other ISO standards relating to networking and communications. (And yes, I realize that is quite a strong statement from me, but it's true.)
Since about 1998, this has been taken care of by the "palm rejection" feature of the Synaptics touchpad driver. Of course it only works for Windows, and so far as I know, there's no equivalent for Unix/Linux, but then, laptops *still* aren't Linux-friendly, but at least we've progressed to the point that you can *load* Linux on most of them. (FWIW: I have still *never* seen properly functioning audio on *any* Linux machine, laptop or not. Sad, but true. I'm told it's possible, just not with any of the hardware I've ever owned (which is only about half oddball stuff like Librettos.))
As someone who used to be a program manager in Dell's portables group, I can tell you that Dell has a number of product marketing people that ensure that a special competition lab prepares machines for tests (they get tweaks the rest of us may never see, both hardware and software, mostly drivers), then carefully follow-up on the testing and results reporting/weighting to ensure that Dell is always at or near the top. (You'll notice that weight itself is seldom given much weight in notebook tests that include Dells, primarily because they have historically tended to be on the pudgy side - the Latitude CP family contains a half-pound of *screws* for cryin' out loud...)
This isn't meant as a slam at Dell - I suspect nearly all other OEMs do this as well, but this is an area in which Dell makes it a point to execute even better than usual. It's simply not possible to be important enough to get a "test machine" from Dell and not have some PM continually looking over your shoulder throughout the testing. Is that undue influence on the results? I guess that depends on your perspective...
As for the advertising stick, I've never seen it used, but then, it doesn't really have to be, now does it? When everyone *knows* you're carrying a big stick, you have the luxury of being able to speak softly. (TR was right about so many things...)
Your post is self-contradictory, primarily because you seem to have no understanding of derivative works under copyright law.
First, of course, you're wrong in stating that You cannot copyright the information in a phone book because all it contains is a list of facts.
Nowhere in copyright law is (or should) the distinction of whether or not the material is factual enter in to the picture. By your logic, only fiction would be protectable: clearly a ludicrous proposition. As a more practical reality check on my argument, I urge you to check the first few pages of your local phone books - you'll find that regardless of the fact that they're simply compilations of facts, they are indeed copyrighted, and such copyrights have been correctly upheld by the courts. (There have been some particularly interesting cases regarding yellow pages: not surprising, since that's where the money is in that business.)
Using the facts from a Microsoft specification document (patent considerations aside) may or may not be infringing, depending, as I understand it, on whether or not "a reasonable man" would conclude that the new work was in fact "derived" from the original MS work. In that case, it would infringe the MS copyright, regardless of what new form it is cast into. (So simply restating the MS information in a new form would still result in an infringing document.) If it was so different as to not be seen as clearly derivative, then it might be judged non-infringing. Unfortunately, the only way to know for sure is to go to court. (That's not necessarily a bad thing, either: we do not want the law to attempt to speculatively handle all possible eventualities.)
VarTec is indeed the chaeapest LD provider I've found. The vast majority of my LD is within Texas, where it's easy to get ripped by the intrastate rates. FYI, you can compare LD plans to find the cheapest for your particular situation at ratekeeper.com (I have no relationship with them, other than having found their service to be handy.) Note that it's pretty much *always* cheaper to do the 1010xyz bypass than to dial direct.
I have no idea why, but I can't touch VarTec's bypass with direct service from anyone. No big deal, I simply have the bypass # (1010811) programmed into my phone for one-touch dialing. After a week or two, you just naturally whack that button before dialing an LD call - For that kind of savings I'll put up with a very minor inconvenience of pressing one more button...
Facts are facts, and are NOT copyrightable.
This is asinine. I'm sure the IEEE, the OSI, and other stadards bodies will be quite shocked to find that they can't keep soaking you for the big bucks in return for those "open" standards. Internet standards are a great model, and we should see more of them, but they are clearly not yet the norm everywhere. Most "standards" are indeed copyrighted and not freely available. Some of them are quite expensive, to boot. (The oppressive policies of OSI and the ITU were one reason the IETF decided to come up with its own standards processes...)
Maybe I'm just spoiled because I use debian.
Of course, you're leaving out the most hideous, complex, and mine-ridden install of any OS on the planet. I'm convinced the Debian developers do this on purpose so that any Debian users have proven themselves "worthy" simply by virtue of jumping through the endless hoops required the get the piece of dung running. I for one have far better things to do with my time. (And for the record, the only time I have ever totally horked a Linux machine was Debian's fault after doing an apt-get update (yes, from stable): the machine froze instantly and was so botched it would not boot. It was easier to start over with Red Hat or Caldera (I don't recall which I used, now) than try to untangle the damage.)
I like some of the concepts in Debian, but it's really a pain in the butt to get up and running correctly. Somehow the Debian team thinks this situation is OK, as it's been that way for years now. If someone would munge the *concepts* behind e-smith and debian together, we'd really have something...
There's nothing in human experience compared to which a sendmail config file could be considered simple.
;-)
You've not been around long enough to have tried uucp, I can tell. I've mostly stopped having bad dreams that begin "tip cua0", and involve serial breakout boxes...
The employer pays you to work, there are NO work reasons (cut the crap about tech support IRC and suchlike - i've heard it and seen what these guys talk about - there's no tech support going on at all - its chatting) for IM clients that i can see other than wasting time.
Two things:
1) It's the company's netowrk, they bought it, built it, and paid for it for the good of the company. They can allow or disallow anything they want, any time they want, for any reason or no reason.
2) I've never seen a "legitimate" use for chat/IM. Typically, it is used simply to continue one's social life (usually miserably dreary) through the workday without raising so many eyebrows as several calls an hour would. Chat and IM area blight upon the Internet - personally, I'm in favor of blocking this crap everywhere, not just at work. (I have observed a striking, nearly 1:1 correlation: People who use IM are more likely to claim things they cannot really do or deliver on - they will then use IM as a crutch to attempt to leech the knowledge of others rather than simply learning for themselves. Top producers (whether coders, designers, or marketing folks) NEVER use IM. The correlation is nearly perfect. I don't know why...
Lineo wound up owning DR-DOS after the split from Caldera, which just kep the Linux stuff. So what will happen to DR-DOS now? It was always a better DOS than Microsoft's. Here's hoping they turn it loose, along with thier DOS web browser (which, to be honest, was really doggy.)
It's too bad that Caldera caved in to Microsoft and settled instead of holding their feet to the fire - the DR-DOS suit evidence that was buried by that settlement would have made the antitrust case far stronger than it was anyway...
As I cited in a reply farther up the page, Rackspace [rackspace.com] has few local customers here in Cowtown, USA (aka San Antonio TX) but that hasn't stopped them from prospering.
Come on, if you've EVER seen a western movie (and there are lots of good ones, so start renting), you know that Cowtown is Fort Worth, not San Antonio. It shouldn't take a Texan to point this out...
San Antonio, by the way, is hardly a "cowtown", anyway: It's the 9th largest city in the US, ahead of Detroit(10), San Jose(11), Indianapolis(12), and San Francisco(13), and it's the center of a lot of interesting telecom activity, including, unfortunately, telemarketing. (Source: http://www.infoplease.com/ipa/A0763098.html)
One thing that's not often realized or acknowledged by many mech hackers on thier way to becoming gearheads is that the hackability of cars varies quite a bit. Accordingly (no Honda pun intended), all hacks involve tradeoffs. You may be able to squeeze another 15 HP out of that 4-banger, at the cost of shortening it's life considerably. (And grenading those expensive upgrade parts in the process...) Optimizing these tradeoffs is not easy.
Case in point: Really high performance cars are hard to hack. This just makes sense if you think about it. The simple fact is that what separates high performance cars from their more proletarian brethren is that they have *already* been "pre-hacked" to improve performance. The law of diminishing returns definitely applies here. Hot-rodding Ferraris, Porsches, Vipers, and Cobras is *hard* simply because all the easy, high bang-for-the-buck stuff has already been done, and then some. What's left is almost by definition in the "not worth the money" category. (Of course, that doesn't stop the legion that has more money than sense, especially if they can afford these cars in the first place. These people ruin a lot of good cars.) If you want to hack for hacking's sake, start with something other than the top end, unless you like low ROI. Otherwise, buy a fast car and enjoy it with no or minor tweaks.
There are cars out there that are designed with very conservative safety/durability margins that are eminently hackable. (RX-7s come to mind, as do Miatas, Mustangs and GM F-bodies. The new Focus is shaping up to be a really good candidate, perhaps becoming the Datsun 510 of this decade, but Ford, as usual, makes it hard to order all the right bits.) Also, don't forget the time-tested method of getting the big win: engine swaps. If you choose a bigger engine from the same manufacturer, these are often not even all that difficult, and usually (at least until you break out the "blue-tip wrench") offer a fall-back position, if things don't work.
Although little Japanese motors can be hacked, I don't think many of them are practical hacks, since you quickly have to start upgrading everythign else once you've started. Thisis why there's still a lot of truth in the old hot-rodder's saying, "There are two ways to win: cubic inches, or cubic dollars." V8s and engine swaps remain popular because they offer the former. I'd put most Asian hacks in the latter category, and if you're going to do that, why not just buy a fast car in the first place, as it will almost always produce superior results?
Remember one last thing: Power is your friend, weight is you enemy. Very few hot-rodders put much effort into weight reduction (and it's hard on many modern cars), but keep that in mind when buying a car - A Miata is relatively at least as hackable as a Mustang because of this. Starting light and going lighter can make a tremendous difference. If you really want light, buy a Lotus: Colin Chapman once said he designed Lotus race cars with the following in mind: "The perfect car disintegrates as it crosses the finish line." *That's* performance optimization, gentlemen.
RX-7s are GREAT cars, and if you don't do anything stupid, they're just about bulletproof. If you want a motor you can abuse with impunity, this is it. The main thing to remember is that rotaries are *designed* to burn oil (they must, to lube the apex seals), so you MUST check the oil frequently. Always remember that modding a rotary is DIFFERENT from modding a piston engine. Stay conservative, and you'll be fine. (Bill Buckley and I would say that's always good advice.)
Rotaries are simple, powerful and reliable. Without a muffler, they will make you ears bleed. And there's something to be said for a motor with only a few moving parts that can be rebuilt on your kitchen table. (While perhaps not for raw mechanical beginners, rebuilding a rotary is far easier than rebuilding a piston engine. There are a number of good books out there on the topic.)
ROFLMAO: My eyes are wet. THAT is the funniest thing I've seen in a long time...
Too true, but I'll take the brio and fury of an Italian car over the soulless, sterile Germans any day, and have more fun doing it. There's really no substitute for an Italian car. I'd argue that there are *some* good rice-burners, and that they're good *because* they have the soul of an Italian car: The WRX is clearly a Lancia in spirit, and the NSX is nothing less than a modern Ferrari Dino by Honda, a concept that has quite a bit of appeal. The Mazda rotaries aren't like anything else, but are clearly built by people that think much like the Italians do when building their cars. (If you're on a budget, try a second-generation RX-7, turbo if possible, although a Miata with a Jackson Racing supercharger woudl do in a pinch.)
(WARNING: Italian cars are an ADDICTIVE ILLNESS. Once you own an Alfa or Ferrari, you will lose touch with reality and your remaining friends and loved ones will think you've taken leave of your senses. Which you have. But you'll be having a ton of fun in a car that doesn't look like a slab of cheese and makes such wonderful mechanical noises that you won't care about the stereo. (Mine's been broken since 1990 - I don't care.) You'll laugh at BMW, Porsche, and Lexus owners that spend far more on thier cars, but get so much less in return. You will learn that although parts aren't expensive relative to those others (Ferrari parts are cheaper than Lexus parts!), getting them here quickly is another matter. You'll learn that reliablity is relative, and the electrical systems will reinforce your understanding of quantum uncertainty. You won't think too much of taking a lovely drive to another city to get the car worked on by a competent mechanic. You'll learn to do a lot yourself, and that "accensione" means "ignition". You'll go for a weekend drive and have silly grin on your face that will last until Wednesday. Once you're hooked, you won't rest until you have a Ferrari, so just go ahead and buy one. Mine's not for sale. You've been warned.)
Although Wiki webs are nice in concept, I have yet to see one that did not very soon degenerate into a morass of ill-connected thought fragments. In fact, I think Wikis are quite possibly the strongest argument ever encountered for site administrators and/or moderators.
For an example of what I mean, visit www.tuxscreen.net and just *try* to figure out what all you have to do to get the current version of the software installed and running. The information is there (mostly), but it's so fragmented as to make it pretty much impossible to find or use.
Still, there's no doubt that two-way webs are someting we should strive for, a la Ted Nelson's Xanadu concept.
Sadly, the fundamental problem is that information organization is what produces much of the value of information, and organizing information is something that's fundamentally hard and difficult to automate. (Not that the KM (Knowledge Management) guys aren't trying, but they've got a long way to go, and those are pretty rarefied tools, not available to most of us, either by reason of price or complexity/learning curve.)
Completely OT: Wouldn't WP's tagged formatting code method make it an ideal way to create low-end XML? It already has great word-processing features, and claims an XML format. WP could output SGML 8 yrs ago or more. Re: WP and XML, search google or see, for example: http://www.xml.com/pub/a/2000/05/31/wordperfect/
Hardly off-topic, since SO's file formats are a very valid dimension of it's readiness for the real world. WordPerfect's document capabilities are often underrated due to it's UI. I've had a love/hate relationship with WordPerfect over the years - it's positively user hostile at times and when things go wrong, it's as opaque as vi without the benefit of simplicity. Still, it's the only WP I've ever found that handles long documents well (which is why it's still the standard in much of the legal and real estate businesses), and the "Reveal Codes" feature is very nice for power users. In many ways it's the best mix out there of word processor and desktop publishing capabilities.
As an aside, I've used SO6beta for several months now, and *if* you put in the small effort to learn it's slightly different UI philosophy, you'll find that it's a VERY capable office suite. There are a few bugs (which I and others have filed) that keep me from making it my everyday tool, but overall, it's really quite good. Although there are some holes realtive to MS Office, it also covers some nice areas that Office doesn't. Sun's got a winner here. When it hits the shelves, I'll buy a boxed copy rather than upgrade the Office 97 I've been using for years - I really think it will be good enough to make that viable. The fact that it runs on non-Windows OSes is just icing on the cake. (C'mon BSD!)
Well, actually, this is pretty darn close to the truth for applications that are threaded. IIRC, Oracle ran about 28x as fast on a 30-way UE6000 as it did on a single processor of the same speed. (To be fair, there are darn few apps that are really written to take advantage of threading, Oracle is one of the best...)
I know this is hard for some Linux folks to swallow, but that's the reason some of us really love Solaris - it scales more linearly that anything else I've encountered. It's not exactly linear, but for well-written apps, it's pretty close...
Reading this brought back memories... Peter Shankman is right that the most impressive thing about this sort of activity is learning that your real limits are FAR, FAR beyond where you think they are, even after you've passed the point of "can't go another step..." That's an important lesson, and one that sadly, most people never learn.
In my case, it wasn't SEAL training, but rather PT in the Texas Aggie Corps of Cadets, where "crap-outs" were a regular part of one's existence, especially in the freshman "fish" year. I didn't start out as a terribly motivated student, and I can honestly say that it was the Corps that kept me in college - I was already capable of making a pretty decent living in computers or experimental stress analysis before I got there, so the money/success angle of college wasn't really a motivator for me, but the comaraderie was.
This sort of PT is **invaluable**, but is all too often categorized as "hazing" in today's ridiculously whiny PC culture. (I somehow doubt that today's cadets have to snake-crawl through pig crap, but they should. It builds character - and I'm completely serious.) There really is quite a lot to the old-fashioned military training idea of tearing down the old man in order to build up the new - the experience is invaluable. Despite what the whiners say about never letting someone "abuse" them like that, it is paradoxically exactly that sort of humility and dedication to teamwork with one's classmates that builds the character required to truly act as an individual.
Interestingly, it's quite possible that the sort of "crap-out for hire" service described by Shankman may be the only thing short of real BUD/S training that will push you that far anymore. (I'm *sure* there are bulletproof waivers required in which participants give up rights to legal action for any and all abuse, physical or mental.)
Anywhow, if anyone is looking for a good, well-rounded educational experience, I highly recommend Texas A&M (damn good engineering/technical school anyway) and membership in the Corps of Cadets. The great General George S. Patton may have said it best: "Give me an army of West Point graduates and I'll win a battle. Give me a handful of Texas Aggies, and I'll win a war!"
Almost. There's one more difference in there, and it's a big one: quality control.
Actually, the real truth is rather surprising: most computer/storage system OEMs have only token quality requirements for disk drive reliability. Sun is one of the only vendors that really takes this seriously: when I worked there a few years ago, just after the introduction of the original SPARC Storage Array, it was judged that there were too many field failures of drives, and Sun worked with the drive vendors to come up with tighter drive specs. Then they started testing them by pulling a few at random out of each incoming truckload. They only had to reject a couple of 18-wheelers full of Seagate disks before all the vendors got the message that they needed to send the best drives to Sun if they wanted their business. Interestingly enough, I have some reason to believe that Sun was getting better quality drives from IBM during this period than they were able to get for themselves internally.
As an more on-topic observation, we build storage systems for a living, and I did quite a bit of research on the newest high performance IDE mechanisms for our flagship product. The upshot is that trouble with the IBM drives is somewhat spotty: some people have no trouble at all, others have horrendous failure rates. There seems to be no way to predict in advance how your particular drives will act.
Because we're usually building multi-TB systems, I couldn't take that risk, so we standardized on the Maxtors (80 and 100 GB, 5400 RPM), which have proven to be fast, solid, reliable workhorses. I need to look at the faster Maxtors and the new 160GB WD drive RSN, but will probably continue to stay away from the IBMs for a while, until the re=prove themselves.
Sad, because IBM was once the gold standard in this field. They do still seem to have the best 2.5" laptop disks, by far, though: I really don't think anyone but Fujitsu is even in the same ballpark with IBM for the little drives, but even Fujitsu is clearly back a bit in ruggedness and reliability.
If you want server drives, pay for them; if you want to pay for consumer drives, don't expect more than consumer quality. Enjoy it when you get it, but if it counts, pay for it.
This kind of flies in the face of the entire idea of RAID, where the "I" originally stood for "Inexpensive" (now represented as "Independent" in many cases, since modern high-po SCSI drives are anything but inexpensive...) The realities of volume manufacturing mean that there is and should be no real difference in these drives, and if there is, it's more likely to be in the other direction: I assure you that a PO'd consumer that wants to vent to support about crappy service and lost data will cost the Mfr. FAR more than a commercial account that routinely sends back a few boxes of disks every month as part of their operations. Nothing is so expensive to support as an individual consumer! (I know for sure it costs one major PC vendor over $30 every time they pick up the phone for a customer support call.)
(Clue: Love Field and Intercontinental both are _small_ airports, usually used for commuter flights in and out of Dallas and Houston- they normally don't rate a hand-held bomb detector unit...)
No, sorry, you're clueless. (Sorry for the ad hominem, but you have to admit you set yourself up for that one...)
Houston Intercontinental(IAH) is NOT a small airport (you probably meant Hobby (HOU), the "old" airport, still quite active, but much smaller than IAH): IAH is Continental Airlines home hub and the 17th busiest airport in the *world* by passenger count and 13th by traffic. Love Field in Dallas is no slouch, either: it's the home base for Southwest Airlines (the only airline making significant money at the moment) and would be far busier if it weren't for meddlesome federal rules limiting the use of this really convenient in-town airport to avoid "hurting" the regional behemoth. In fact, Love and Hobby are such success stories, and so well liked and used by travellers, that several years ago, the FAA started insisting that whenever a new regional behemoth is opened, the old airport must be *destroyed* and rendered unusable, even for general aviation (private planes). Denver and Austin are examples of cities that have suffered mightily because of this, and are forced to live with ridiculously expensive white elephant airports that might as well be in another county...
Point being, what you consider "fixable" depends largely on your knowledge base.
This is a VERY valid point. After many years of amassing a great deal of knowledge about Windows, Linux, and various Unixes, I have finally reached the conclusion that once they start to act flaky, it's ALMOST ALWAYS easier and quicker to simply reinstall the OS than to figure out what obscure setting, library, etc. got broken. (Not that one shouldn't find root causes, but let's face it, much of the time it's just not worth the effort.)
Linux and Windows are both quite difficult to administer in very different ways. Those of you that have never used a real production-quality OS (Solaris and DEC/Compaq Tru64 are, AIX, IRIX, & HPUX aren't, IMO) really don't know how stable and easy to run those systems can be.
I like and use Linux routinely, but it's not a panacea - my most recent Linux reinstall was caused by doing an upgrade via Debian's vaunted apt-get mechanism: I didn't bother to even figure out what apt-get did to completely scrog the OS - it was obviously very bad and quite clear that reinstalling from an old Mandrake CD I had lying around would be much quicker and easier. (No, I won't touch Debian again for another year or two.)
By the same token, bit rot munches my Win98 to the point that I have to start over from scratch about every six to eight months. Again, I *could* probably fix it given the time, but I've got beter things to do, so reinstalling makes sense, at least until we get OSes that deliver real stability.
Librettos are expensive even used, so I probably wouldn't buy one, but I had one given to me a few years ago (with the Japanese keyboard, which is a pain), so I turned it into a very nice little portable file server: When I'm travelling, I now find it much easier just to bring the Libretto along, knowing I can have instant Ethernet access to all the critical resources I might need. This is much easier than dragging along a regular laptop that needs its own luggage. (Rule for road warriors: NEVER buy a computer that requires luggage!)
:-)
:-)
It's only a 75 MHz Pentium w 16MB RAM, so it was never even suitable for Win95, but it runs a slim Mandrake just fine. Even KDE works, although it takes a while to start up.
I will say that installing any OS on this thing is a major PITA, since it has no floppy and no CD-ROM. I'd like to replace Mandrake with e-smith, which is rapidly becoming my favorite general-purpose server distro, but e-smith only installs from CD - Ugh! I could work around it if I spent several hours, but that kind of defeats the purpose of e-smith.
I suppose there are smaller file servers, out there, but not by much, especially with a built-in UPS...