Indeed. Most monitors also have similar adjustments.
What I'm talking about by saying BlurryVision is, say, a 1280x1024 LCD monitor running at 800x600 -- none of the pixels line up, and the aspect ratio is slightly wrong. Folks do this either out of ignorance or on purpose (to make things appear larger on-screen), but never with the goal of having a correct display - it's all about making things as big (or as easy) as possible. Turning off scaling would make the pixels line up, but it would look even worse to those people who are frequently affected by BlurryVision.
Just like FatPersonVision, where (for example) a normal-aspect picture gets stretched to fill a wide-aspect display: People have either done it out of ignorance or on purpose, but having their newfangled 16x9 monitor or television have black bars along each side -- even if it is the most-correct way to display the content -- seems deemed largely unacceptable by the masses. They seem far more tolerant of watching FatPersonVision than of having any portion of their "investment" gone to waste, and so disabling scaling doesn't work in that application either.
Like sound cards with lousy inputs, and mountains of important data with no backup, people don't care that it's wrong or that it could easily be improved.
The aforementioned Creative USB kit may fit your needs, depending on what ALSA does with it. The poster mentioned that the switch prevents the driver from resampling; however, the switch itself is just a USB Human Interface Device. It's up to the software to decide what needs done based on its position. The box itself a dumb device without any sort of internal DSP; internal resampling is quite far beyond its means.
I wish I could help more, but I don't have any gear which can report the output sampling rate of my own Soundblaster MP3+ (which apparently ended production a couple of years ago, though Ebay is sure to have some). It does at least produce good audio with ALSA in a plug-and-play fashion, where it presents itself as a bog-standard USB audio device.
The next step for you is probably to check the ALSA documentation or source to see if it ever supports sending 44.1KHz audio to USB devices without resampling at all. You'll either find a method that works, or discover that your holy grail does not exist within a reasonable budget.
Asbestos is a real danger to humans because it gets stuck in lungs. Fish, lacking lungs, shouldn't have this problem. Stuff goes *through* the gills, not into and out of them. My experience keeping fish shows that solids regularly pass through the gills of fish in the process of eating.
The asbestos is probably safer down there than anywhere else, I'd guess.
No - it's not hard. But it's also not very popular.
I'd wager that most folks, these days, never do any serious recording of audio. It's just not something that there are very many practical applications for in a modern world. And even when they do want to bring analog audio into a computer, it's probably only as a part of a video capture or VOIP rig, and they're just not paying much attention to the fidelity. And even when they do have a need to do serious recording and are paying attention, only the most glaring amounts of audible noise and distortion are likely to be noticed. People are generally pretty tolerant of relatively bad-sounding audio.
If the need were more common or they were paying more attention, cheap sound cards would commonly have the same huge number of reasonably good inputs as they currently do outputs, because that's what the market would demand..
Myself, I've been looking for a decent, cheap 4-channel sound input into the PC for years -- I've got a few old quad recordings of various rock music on 1/4" reels which I really want to listen to, but I will only do so in the presence of something with which to archive it with (the tapes are so old that it's not unlikely that playing them even once will destroy them).
Lately, the additional need for 4 or 8 (though preferably 12 or 16) inputs has risen as I'd like to begin making some live recordings of a band that I've been working with.
It's not hard to find sound card or external Firewire/USB box which can do these things -- it's just hard to justify the expense.
But it's not the expense which is keeping people away from recording on a PC, but rather just the fact that these sorts of tasks are esoteric enough that most people will never do them. Therefore, the market is, and is likely to remain, very thin.
Like RAID storage, backup devices, SAS drives, DVI-connected LCD monitors configured with 1:1 pixel ratios (instead of BlurryVision and/or FatPersonVision), most folks just don't have any reason to care about this aspect of computing.
If you're happy enough to have all of your sound processing done in software instead of with dedicated DSPs, just make sure that your next motherboard has some manner of digital output -- either optical or coaxial. Plug this straight into your home theater receiver, or a dedicated DAC[1], and the quality of the sound card generally just ceases to matter (at least for stereo material).
[1]: Myself, I use an Audio Alchemy DDE v1.1 which I picked up on Ebay 6 or 8 years ago for a song. It does a good job of converting 44.1 or 48KHz PCM audio into a very nice pair of analog outputs on RCA jacks.
So, Mr. Tarmail, would you care to answer the following question: Can I easily use tarmail in front of my existing postfix/amavis/clamav/f-prot rig? I don't mind processing mail twice (or more, really) -- I've got plenty of CPU to spare. If your MTA is really as slick as you say, I would to make a somewhat easy transition away from my current, complicated arrangement and onto yours.
(I'd research this myself, but I'm on my own time right now and would rather be looking into a strange issue with my car's parking brake than do pro-bono work for the company.)
Vista installed cleanly on my (then) 2-year-old Dell laptop. All of the hardware worked. I've upgraded a plethora of drivers since then, but that's just the way I like to do things -- it worked fine without help.
Vista wants a pile of extra memory vs XP (though it is nothing an extra $20 won't just positively cure at current prices), but then XP wanted a pile of extra memory compared to 98.
You do yourself a disservice by viewing your memories through rose-colored glasses.
Alienware is owned by Dell, but that doesn't mean that they act like Dell.
My wife recently bought a nice (though low-end, by Alienware standards) desktop computer from them. Though the ordering screens are similar (as well they should be - Dell's web-based ordering is rather slick), and credit for both companies is through Dell Financial Services, the similarities ends there.
The Alienware case is a regular ATX case, with a regular ATX backplate and regular ATX mounting holes, and is large enough to accept bloody any motherboard, whereas Dell uses a strange-ish quasi-Micro ATX design without a removable backplate. The motherboard itself is an off-the-shelf model (Foxconn, in this case), not some weird Dell special. The front panel connectors (including those for the large number of fancy LEDs) are compatible with regular ATX boards, instead of Dell's non-standard monolithic connector. There's a plethora of drive bays, with all of the hardware needed to use them included, whereas Dell seems to take great joy in including only as much hardware as is needed to assemble that particular system (on the low end of things, at least - Dimension 2350 and 2400 machines have provision to hold a number of 3.5" hard drives, but there's only enough hardware included to mount exactly one. The other bays are physically absent.). The price was very reasonable - about $100 more than equivalent parts from Newegg.
We had weird issues with the Alienware's extra LEDs on day 1. Called tech support, and without waiting in queue got a real human (in America!), who spoke real American English, had a real name, and who actually had at least half a clue. They sent a new part, which didn't fix the problem. Called back, again immediately got a real human, who dispatched both more parts and a warm body to install them. Problem solved.
And, sure, it'd have been better if the system didn't need any service, but I did feel pretty good about the whole process. It seemed that Alienware wanted to solve my problem, instead of just force me to jump through hoops.
Meanwhile, I loathe to call Dell support. One of the hinges on my laptop broke (which was reasonable enough after 2 years of hard use), and I had to wait for 20 minutes before some girl in Bangalore came on the line who only wanted to talk to me about reinstalling Windows XP. I had to fight with her for about 15 more minutes in order to get transferred to someone with enough clue to understand the simple problem and dispatch parts. And this with their premium support package!
So, yeah: They're the same company in that they're owned by the same people. But that heterogeneous ownership doesn't mean that they're at all similar in operation or quality.
I don't live in Parmer, so my view of things may be different from yours.
That said: My Dell Inspiron 6000d laptop was built and tested in Malaysia, flown to a FedEx facility in (IIRC) Kentucky, and was then shipped by truck to my home in Ohio. It at no point in its life has ever been in a Dell facility in the United States.
I understood, at the time, that this was all par for the course for this model of laptop.
FYI: I have a 1TB WD Mybook World Edition network-attached drive. It included a 3-year warranty, which was about 2 years longer than I expected it to be.
So far, so good, but the thing is slow like molasses in January. It's really an odd thing: Gigabit ethernet, but with a slow CPU, a slow network driver, and a slow SATA port. It is lucky to be able to sustain rates of 4.3 megabytes per second. But SSH was easy to turn on, and WD included a full development environment, which makes up for a (very small) bit of the pain involved in using it.
Somehow I'm thinking that before the 3 year warranty is up, I'll have yanked the drive and installed it in a real computer.
Why? Because a few people had a problem with Ford using Firestone's Rollove^H^H^H^H^H^H^HDiscoverer tires on their Explorer SUV.
Except, at the end of the day, the problem was really all about people driving with overloaded vehicles on underinflated tires. And however foolish this is (it's all because people don't fucking pay attention to the manner in which they operate and maintain their death machines), the guffaw cost Ford, Firestone, and the government (and thus the People) a lot of money.
In direct response to this, recommended tire inflation pressures have gone up (door jamb placard in a typical American sedan used to say 30 or 32PSI; it is now at least 35PSI). And, it seems, the car is required now to let the driver know when a tire is 25 or 30% underinflated.
And why? Simple: For the same reason that we have seat belts, air bags, and crash tests: It's cheaper this way. Sure, the tires wear out faster. And it's not as comfortable. And it doesn't handle as well.
Not exactly. Adolf's Second Law states the following: Any information, no matter how common or true, may escape the reigns of Google by virtue of newness alone.
And the Fifth Law proclaims that none of this matters, anyway: There is no Law in this Universe which is devoid of exception; the Fifth Law is no exception to this.
Personally, I'm shocked that you didn't know these things.
Now, if you'll excuse me, Godwin's Law says that I must now tend to the the swastika on the BMW, which has needed polishing for some time.
Any time an individual searches the Internet hoping to find negativity on any topic, no matter how innocent, they will not be disappointed in that effort.
I miss finger. Back when the net was much more naive -- when everyone's mail server had port port 25 open and would gladly relay mail for anyone -- one could find some hapless nit on IRC, finger their ISP's terminal server, snag their user name, derive from that their real name, find their home address in a telephone directory, and fire up Mapblast or Terraserver and spook the hell out them by saying things like "That's a nice lake that you live next to. The water is very pretty this time of night, isn't it?"
Now that I think of it, it's really surprising that I didn't wind up in jail when I was a kid.
I think I'll install fingerd on my WRT54G and stuff some random information into it, just for old time's sake.
The portable hardware probably isn't the difficult part, here. For fuck's sake: I've got more than enough horsepower in my 4-year-old Palm Zire 71 to organize, contain and transmit the output of one census worker.
Which is cool, I guess. But all of the overpowered portable hardware in the world will not change the fact that the software and back end required to make it useful DOES NOT EXIST.
No, they won't -- at least not much, if they were using a subdomain for their RBL (as is the only sane method of doing so).
They could abandon this subdomain (which would be silly), or just set up its SOA to have a huge TTL, and have an NS line in the right spot pointing to localhost.
Requests from end-user mail servers would still happen, perhaps thousands of them per minute, but they'll only be met with references to a nameserver known as 127.0.0.1. The DNS hierarchy will then cache this bogus nameserver for TTL seconds.
They'd still see some traffic, particularly from poorly-behaved DNS servers which don't honor TTL, but it ought to be pretty easy to limit their traffic to no more than one request, per server, no more frequently than every few days (at least on average).
Which, I'd think, would be good enough. But even if it's not: It's nowhere near as bad as you seem to make it appear.
Indeed. And as I remodel a house that I hope to move into soon, this is exactly the sort of thing that has so far prevented me from installing general-purpose LED lighting. (The geek cave will have, of course, several arrays of high-efficiency red LEDs, which should last longer than I do. But that's not exactly general-purpose.)
I'd wager that your LED projects are driving the diodes at near peak current. With less current, the phosphor will obviously degrade more slowly.
I don't have details on the curves involved, but just to throw some numbers out: Suppose that a regular white LED sustains reasonably good color for about 2.5 years of use at peak rated current.
I'd be perfectly happy to buy a bunch of pre-fab 10W LED fixtures, with diodes driven at 1/4 of their rated current, and 10 years of life. But such a thing doesn't seem to exist -- longevity doesn't seem to enter the equation in a world dominated by lumens-per-dollar.
We've only got two wide-spread (read: county-wide) networks that we maintain. On one, the access points are mounted near the middle of a 90-foot tower which is atop a 12-story office building, which is by far the tallest structure in the vicinity, but it sits near the bottom of a natural valley within an otherwise-flat landscape. The APs are only up around 120 feet, with topology taken into consideration.
The other network (which covers another entire county) has access points mounted at around 180 feet on a free-standing tower.
The subscriber modules are, as a rule for our 5.7GHz stuff, mounted up high. Most of them are on grain elevators at 120 to 160 feet.
Of course, we cheat a lot, though we're not doing anything illegal: Most of the access points also have reflectors. As you probably know, this cuts the width down to around 6 degrees, instead of 60. To get full 360 degree coverage and similar range, we'd at least need a whole bunch of APs and very careful frequency planning, if it is even possible.
We're able to get away with this limited coverage because we're using it to provide WAN service for a few very specific organizations, and not Internet access to random Joe Averages. I don't think we have more than 3 or 4 SMs on any one AP.
Height, at least in our experience, has been the key. We had one 5.7GHz links fail at 3 miles, with a reflector at both ends, when the SM was mounted to the side of a 3-story building at around only 35 feet. It was January, we were directly in-line with AP which was already talking talking to an elevator-mounted SM at 10 miles, and had visual line of sight. RSSI was around 900, things looked good. Then spring sprung, and the trees leafed, things turned to shit and there was much senseless (and expensive) moving about of the SM trying to improve things.
In that particular case, we eventually moved it to 2.4GHz, and it has been fine for a couple of years despite reporting rather high jitter (probably due to interference from all of the other 2.4GHz stuff near the downtown access point).
On the other hand, our longest link is (again) around 17 miles, and is dead solid at 5.7.
That said, it's honestly been somewhat of a mixed bag. I only say that your experience is weird because you seem to be so unsuccessful at getting any long-distance links up at all. It's not been a trouble-free ride for us, either.
I can't imagine deploying Canopy in a high-volume, home-subscriber ISP environment, as you seem to be doing. The parts are too expensive and the profit too small, and Joe Average doesn't want to buy a 150' tower and probably can't get access to any tall structures nearby. It's just worked fairly well for us doing our build-to-order networking stuff, where the margins allow to us to invest as much time and resources as needed in order to make it work. To that end, we've been 100% successful with every task to which we've applied Canopy, and though I'm quite certain that we've lost money on a few links, we're still rather pleased with the product.
I'd be willing to try to help with any specific applications you might have trouble with. Just drop me an email.
I live in one of only several counties in Ohio for which Aloha was successfully bid against, several years ago, and where the license is still held by a small company or individual.
So I, for one, will not be welcoming my deathstar overlords.
This could happen just as easily with linux. A new kernel version could uncover a flaw with an older driver even if that driver was opensource. And that older driver package would be flagged as incompatible with the new kernel, and the updaters would prevent you from mixxing them. Its not exactly as if this has never happened before.
That's not entirely likely to be true. What seems to happen over here in reality, is this:
Kernel (or X) update shows up, and the user is prompted to install it. After installing it, the system doesn't fucking work anymore, because nothing ever checked to see if the update would break with the installed hardware.
One of two things then occurs: If the user has another computer handy they get to research and attempt to solve the problem in the comfort of a web browser. If the user does not have another computer handy, they get to try to puzzle it out themselves using arcane command line tools that they've never seen before, with text-mode documentation they've never had to find before. If they're really clever, and can get to the network, they might install and use Links and be able to search forums for help, but chances are they've never used that before either.
In both cases, after pissing away several hours or evenings trying to make it work, they either succeed and their computer again works just as well as it did last week, or they get frustrated and abandon Linux altogether. (Maybe they'll try it again some day, maybe not.)
I've got many Canopy links at 5.7GHz which have been working at ranges averaging in the realm of 12 miles (the longest is 17 miles) for about four years without adjustment. These are mounted to things like the handrail on top of a grain elevator, or on non-penetrating mounts on the top of tall industrial buildings -- places I was sure that either vibration or ice accumulation would push things around in no time. It's been fine.
I've also got a handful of 2.4GHz links which are not quite line-of-sight, at ranges of 6 to 7 miles, all in populated areas, which also work without episode.
Really, I've only experienced a handful of problems with Canopy hardware: One time, after a nearby substation blew up, a subscriber module died and started eating power supplies. Another time, one of the two carriage bolts which hold the reflector to its mount snapped off and disappeared at 140 feet in the air. The strangest and most bothersome thing I've had trouble with is a number of repeat issues with some first-gen 5.2GHz gear where the RJ45 connector falls off of the PCB (this hasn't happened with later gear).
But, aside from those few failure modes, the stuff has been totally solid. In my experience, if the link shows good RSSI and low jitter, and works for more than a day or two, it will continue to work indefinitely.
That said, I have a lot issues with Canopy as a business unit. It took them more than a month to get around to sending a simple e-mail with some keys to upgrade a handful of access points to Advantage, and more than once I've just given up on waiting for Motorola to ship products.
And because of this, we've been putting plain-old 802.11g Engenius stuff up instead. The configuration tools aren't as polished as Canopy, but not too horrible. The units themselves use regular PoE, have integrated antennas, seem to be well-constructed. They cost less than $200 from Newegg -- a whole multi-site network can happen for the price of a single Canopy unit. There hasn't been much opportunity to move many of them yet, and none of the links have been particularly long distance, nor particularly busy, but so far the handful of them we have installed are working perfectly for the customers that have them.
And I, for one, absolutely fucking love being able to just open the laptop and access the WLAN. The less time I spend running an inverter to power a Canopy SM which is propped up on the dashboard, or running a 200' Ethernet cable up a tower, the happier I am.
SLA == Service Level Agreement. Most Slashdotters seem to think that an SLA means something like "OMG! Wonderful fat pipe all for me!!!", but it's just a contract, much like the TOS that the same Slashdotters seem to blame for everything including world hunger.
In fact, since they're both just contracts, either one can be good for the customer, or bad for the customer. The only innate differences are three words at the top of the page, which is about as insignificant a distinction as I can think of.
Defend? Hardly. I hate Microsoft with a certain passion normally reserved for childhood bullies, insurance adjusters, bankers, and pizza delivery people who show up an hour late.
But one thing I hate more is when people portray their illogical conclusions as fact. Most of the Microsoft bashing on here these days seems to revolve around warped and impractical expectations, coupled with a firm belief that anything that they've done lately which is different is absolutely wrong -- even if that differing wrongness is corrective of things which they've been previously damned for.
I have no particular adoration for Vista. I find that it suits my daily (ie: both business and personal) needs better than Linux, but that doesn't mean I like it. Instead, it just means that I hate it less on my everyday laptop than Ubuntu or Gentoo, or even XP.
Computers these days suck, you know. I used to run a half-dozen big-ish applications under DOS with 640k of RAM, and got everything done just fine. Later, I ran a half-dozen applications under Slackware 2 and OS/2 Warp with 20 megabytes of RAM, and got everything done just fine. Lately, I still only run a half-dozen applications, but it takes at least a gig or two of RAM to make start to suck less than 1995 - no matter what [modern] OS is in use.
(I'm seriously considering becoming a plumber instead -- at least gravity is a constant.)
Slashdot Mirror
Indeed. Most monitors also have similar adjustments.
What I'm talking about by saying BlurryVision is, say, a 1280x1024 LCD monitor running at 800x600 -- none of the pixels line up, and the aspect ratio is slightly wrong. Folks do this either out of ignorance or on purpose (to make things appear larger on-screen), but never with the goal of having a correct display - it's all about making things as big (or as easy) as possible. Turning off scaling would make the pixels line up, but it would look even worse to those people who are frequently affected by BlurryVision.
Just like FatPersonVision, where (for example) a normal-aspect picture gets stretched to fill a wide-aspect display: People have either done it out of ignorance or on purpose, but having their newfangled 16x9 monitor or television have black bars along each side -- even if it is the most-correct way to display the content -- seems deemed largely unacceptable by the masses. They seem far more tolerant of watching FatPersonVision than of having any portion of their "investment" gone to waste, and so disabling scaling doesn't work in that application either.
Like sound cards with lousy inputs, and mountains of important data with no backup, people don't care that it's wrong or that it could easily be improved.
The aforementioned Creative USB kit may fit your needs, depending on what ALSA does with it. The poster mentioned that the switch prevents the driver from resampling; however, the switch itself is just a USB Human Interface Device. It's up to the software to decide what needs done based on its position. The box itself a dumb device without any sort of internal DSP; internal resampling is quite far beyond its means.
I wish I could help more, but I don't have any gear which can report the output sampling rate of my own Soundblaster MP3+ (which apparently ended production a couple of years ago, though Ebay is sure to have some). It does at least produce good audio with ALSA in a plug-and-play fashion, where it presents itself as a bog-standard USB audio device.
The next step for you is probably to check the ALSA documentation or source to see if it ever supports sending 44.1KHz audio to USB devices without resampling at all. You'll either find a method that works, or discover that your holy grail does not exist within a reasonable budget.
Good luck.
Asbestos is a real danger to humans because it gets stuck in lungs. Fish, lacking lungs, shouldn't have this problem. Stuff goes *through* the gills, not into and out of them. My experience keeping fish shows that solids regularly pass through the gills of fish in the process of eating.
The asbestos is probably safer down there than anywhere else, I'd guess.
No - it's not hard. But it's also not very popular.
I'd wager that most folks, these days, never do any serious recording of audio. It's just not something that there are very many practical applications for in a modern world. And even when they do want to bring analog audio into a computer, it's probably only as a part of a video capture or VOIP rig, and they're just not paying much attention to the fidelity. And even when they do have a need to do serious recording and are paying attention, only the most glaring amounts of audible noise and distortion are likely to be noticed. People are generally pretty tolerant of relatively bad-sounding audio.
If the need were more common or they were paying more attention, cheap sound cards would commonly have the same huge number of reasonably good inputs as they currently do outputs, because that's what the market would demand..
Myself, I've been looking for a decent, cheap 4-channel sound input into the PC for years -- I've got a few old quad recordings of various rock music on 1/4" reels which I really want to listen to, but I will only do so in the presence of something with which to archive it with (the tapes are so old that it's not unlikely that playing them even once will destroy them).
Lately, the additional need for 4 or 8 (though preferably 12 or 16) inputs has risen as I'd like to begin making some live recordings of a band that I've been working with.
It's not hard to find sound card or external Firewire/USB box which can do these things -- it's just hard to justify the expense.
But it's not the expense which is keeping people away from recording on a PC, but rather just the fact that these sorts of tasks are esoteric enough that most people will never do them. Therefore, the market is, and is likely to remain, very thin.
Like RAID storage, backup devices, SAS drives, DVI-connected LCD monitors configured with 1:1 pixel ratios (instead of BlurryVision and/or FatPersonVision), most folks just don't have any reason to care about this aspect of computing.
It doesn't matter.
If you're happy enough to have all of your sound processing done in software instead of with dedicated DSPs, just make sure that your next motherboard has some manner of digital output -- either optical or coaxial. Plug this straight into your home theater receiver, or a dedicated DAC[1], and the quality of the sound card generally just ceases to matter (at least for stereo material).
[1]: Myself, I use an Audio Alchemy DDE v1.1 which I picked up on Ebay 6 or 8 years ago for a song. It does a good job of converting 44.1 or 48KHz PCM audio into a very nice pair of analog outputs on RCA jacks.
So, Mr. Tarmail, would you care to answer the following question: Can I easily use tarmail in front of my existing postfix/amavis/clamav/f-prot rig? I don't mind processing mail twice (or more, really) -- I've got plenty of CPU to spare. If your MTA is really as slick as you say, I would to make a somewhat easy transition away from my current, complicated arrangement and onto yours.
(I'd research this myself, but I'm on my own time right now and would rather be looking into a strange issue with my car's parking brake than do pro-bono work for the company.)
You're an odd one.
Vista installed cleanly on my (then) 2-year-old Dell laptop. All of the hardware worked. I've upgraded a plethora of drivers since then, but that's just the way I like to do things -- it worked fine without help.
Vista wants a pile of extra memory vs XP (though it is nothing an extra $20 won't just positively cure at current prices), but then XP wanted a pile of extra memory compared to 98.
You do yourself a disservice by viewing your memories through rose-colored glasses.
Alienware is owned by Dell, but that doesn't mean that they act like Dell.
My wife recently bought a nice (though low-end, by Alienware standards) desktop computer from them. Though the ordering screens are similar (as well they should be - Dell's web-based ordering is rather slick), and credit for both companies is through Dell Financial Services, the similarities ends there.
The Alienware case is a regular ATX case, with a regular ATX backplate and regular ATX mounting holes, and is large enough to accept bloody any motherboard, whereas Dell uses a strange-ish quasi-Micro ATX design without a removable backplate. The motherboard itself is an off-the-shelf model (Foxconn, in this case), not some weird Dell special. The front panel connectors (including those for the large number of fancy LEDs) are compatible with regular ATX boards, instead of Dell's non-standard monolithic connector. There's a plethora of drive bays, with all of the hardware needed to use them included, whereas Dell seems to take great joy in including only as much hardware as is needed to assemble that particular system (on the low end of things, at least - Dimension 2350 and 2400 machines have provision to hold a number of 3.5" hard drives, but there's only enough hardware included to mount exactly one. The other bays are physically absent.). The price was very reasonable - about $100 more than equivalent parts from Newegg.
We had weird issues with the Alienware's extra LEDs on day 1. Called tech support, and without waiting in queue got a real human (in America!), who spoke real American English, had a real name, and who actually had at least half a clue. They sent a new part, which didn't fix the problem. Called back, again immediately got a real human, who dispatched both more parts and a warm body to install them. Problem solved.
And, sure, it'd have been better if the system didn't need any service, but I did feel pretty good about the whole process. It seemed that Alienware wanted to solve my problem, instead of just force me to jump through hoops.
Meanwhile, I loathe to call Dell support. One of the hinges on my laptop broke (which was reasonable enough after 2 years of hard use), and I had to wait for 20 minutes before some girl in Bangalore came on the line who only wanted to talk to me about reinstalling Windows XP. I had to fight with her for about 15 more minutes in order to get transferred to someone with enough clue to understand the simple problem and dispatch parts. And this with their premium support package!
So, yeah: They're the same company in that they're owned by the same people. But that heterogeneous ownership doesn't mean that they're at all similar in operation or quality.
I don't live in Parmer, so my view of things may be different from yours.
That said: My Dell Inspiron 6000d laptop was built and tested in Malaysia, flown to a FedEx facility in (IIRC) Kentucky, and was then shipped by truck to my home in Ohio. It at no point in its life has ever been in a Dell facility in the United States.
I understood, at the time, that this was all par for the course for this model of laptop.
What about them?
FYI: I have a 1TB WD Mybook World Edition network-attached drive. It included a 3-year warranty, which was about 2 years longer than I expected it to be.
So far, so good, but the thing is slow like molasses in January. It's really an odd thing: Gigabit ethernet, but with a slow CPU, a slow network driver, and a slow SATA port. It is lucky to be able to sustain rates of 4.3 megabytes per second. But SSH was easy to turn on, and WD included a full development environment, which makes up for a (very small) bit of the pain involved in using it.
Somehow I'm thinking that before the 3 year warranty is up, I'll have yanked the drive and installed it in a real computer.
Why? Because a few people had a problem with Ford using Firestone's Rollove^H^H^H^H^H^H^HDiscoverer tires on their Explorer SUV.
Except, at the end of the day, the problem was really all about people driving with overloaded vehicles on underinflated tires. And however foolish this is (it's all because people don't fucking pay attention to the manner in which they operate and maintain their death machines), the guffaw cost Ford, Firestone, and the government (and thus the People) a lot of money.
In direct response to this, recommended tire inflation pressures have gone up (door jamb placard in a typical American sedan used to say 30 or 32PSI; it is now at least 35PSI). And, it seems, the car is required now to let the driver know when a tire is 25 or 30% underinflated.
And why? Simple: For the same reason that we have seat belts, air bags, and crash tests: It's cheaper this way. Sure, the tires wear out faster. And it's not as comfortable. And it doesn't handle as well.
But it saves money, and that's the bottom line,
Not exactly. Adolf's Second Law states the following: Any information, no matter how common or true, may escape the reigns of Google by virtue of newness alone.
And the Fifth Law proclaims that none of this matters, anyway: There is no Law in this Universe which is devoid of exception; the Fifth Law is no exception to this.
Personally, I'm shocked that you didn't know these things.
Now, if you'll excuse me, Godwin's Law says that I must now tend to the the swastika on the BMW, which has needed polishing for some time.
I would like to state the following:
Any time an individual searches the Internet hoping to find negativity on any topic, no matter how innocent, they will not be disappointed in that effort.
To wit:
Linux sucks. Windows sucks. Dell sucks. HP sucks. Driving sucks. Mercedes-Benz sucks. Kia sucks. Harley Davidson sucks. Furries suck. Google sucks. Indoor plumbing sucks.
I'd go on, but Adolf's Third Law states that I don't have to.
I miss finger. Back when the net was much more naive -- when everyone's mail server had port port 25 open and would gladly relay mail for anyone -- one could find some hapless nit on IRC, finger their ISP's terminal server, snag their user name, derive from that their real name, find their home address in a telephone directory, and fire up Mapblast or Terraserver and spook the hell out them by saying things like "That's a nice lake that you live next to. The water is very pretty this time of night, isn't it?"
Now that I think of it, it's really surprising that I didn't wind up in jail when I was a kid.
I think I'll install fingerd on my WRT54G and stuff some random information into it, just for old time's sake.
Um. Uh.
Hey, kids:
The portable hardware probably isn't the difficult part, here. For fuck's sake: I've got more than enough horsepower in my 4-year-old Palm Zire 71 to organize, contain and transmit the output of one census worker.
Which is cool, I guess. But all of the overpowered portable hardware in the world will not change the fact that the software and back end required to make it useful DOES NOT EXIST.
No, they won't -- at least not much, if they were using a subdomain for their RBL (as is the only sane method of doing so).
They could abandon this subdomain (which would be silly), or just set up its SOA to have a huge TTL, and have an NS line in the right spot pointing to localhost.
Requests from end-user mail servers would still happen, perhaps thousands of them per minute, but they'll only be met with references to a nameserver known as 127.0.0.1. The DNS hierarchy will then cache this bogus nameserver for TTL seconds.
They'd still see some traffic, particularly from poorly-behaved DNS servers which don't honor TTL, but it ought to be pretty easy to limit their traffic to no more than one request, per server, no more frequently than every few days (at least on average).
Which, I'd think, would be good enough. But even if it's not: It's nowhere near as bad as you seem to make it appear.
Indeed. And as I remodel a house that I hope to move into soon, this is exactly the sort of thing that has so far prevented me from installing general-purpose LED lighting. (The geek cave will have, of course, several arrays of high-efficiency red LEDs, which should last longer than I do. But that's not exactly general-purpose.)
I'd wager that your LED projects are driving the diodes at near peak current. With less current, the phosphor will obviously degrade more slowly.
I don't have details on the curves involved, but just to throw some numbers out: Suppose that a regular white LED sustains reasonably good color for about 2.5 years of use at peak rated current.
I'd be perfectly happy to buy a bunch of pre-fab 10W LED fixtures, with diodes driven at 1/4 of their rated current, and 10 years of life. But such a thing doesn't seem to exist -- longevity doesn't seem to enter the equation in a world dominated by lumens-per-dollar.
*sigh*
We just -- uh -- did it. :) Some more information:
We've only got two wide-spread (read: county-wide) networks that we maintain. On one, the access points are mounted near the middle of a 90-foot tower which is atop a 12-story office building, which is by far the tallest structure in the vicinity, but it sits near the bottom of a natural valley within an otherwise-flat landscape. The APs are only up around 120 feet, with topology taken into consideration.
The other network (which covers another entire county) has access points mounted at around 180 feet on a free-standing tower.
The subscriber modules are, as a rule for our 5.7GHz stuff, mounted up high. Most of them are on grain elevators at 120 to 160 feet.
Of course, we cheat a lot, though we're not doing anything illegal: Most of the access points also have reflectors. As you probably know, this cuts the width down to around 6 degrees, instead of 60. To get full 360 degree coverage and similar range, we'd at least need a whole bunch of APs and very careful frequency planning, if it is even possible.
We're able to get away with this limited coverage because we're using it to provide WAN service for a few very specific organizations, and not Internet access to random Joe Averages. I don't think we have more than 3 or 4 SMs on any one AP.
Height, at least in our experience, has been the key. We had one 5.7GHz links fail at 3 miles, with a reflector at both ends, when the SM was mounted to the side of a 3-story building at around only 35 feet. It was January, we were directly in-line with AP which was already talking talking to an elevator-mounted SM at 10 miles, and had visual line of sight. RSSI was around 900, things looked good. Then spring sprung, and the trees leafed, things turned to shit and there was much senseless (and expensive) moving about of the SM trying to improve things.
In that particular case, we eventually moved it to 2.4GHz, and it has been fine for a couple of years despite reporting rather high jitter (probably due to interference from all of the other 2.4GHz stuff near the downtown access point).
On the other hand, our longest link is (again) around 17 miles, and is dead solid at 5.7.
That said, it's honestly been somewhat of a mixed bag. I only say that your experience is weird because you seem to be so unsuccessful at getting any long-distance links up at all. It's not been a trouble-free ride for us, either.
I can't imagine deploying Canopy in a high-volume, home-subscriber ISP environment, as you seem to be doing. The parts are too expensive and the profit too small, and Joe Average doesn't want to buy a 150' tower and probably can't get access to any tall structures nearby. It's just worked fairly well for us doing our build-to-order networking stuff, where the margins allow to us to invest as much time and resources as needed in order to make it work. To that end, we've been 100% successful with every task to which we've applied Canopy, and though I'm quite certain that we've lost money on a few links, we're still rather pleased with the product.
I'd be willing to try to help with any specific applications you might have trouble with. Just drop me an email.
Not in my back yard!
I live in one of only several counties in Ohio for which Aloha was successfully bid against, several years ago, and where the license is still held by a small company or individual.
So I, for one, will not be welcoming my deathstar overlords.
This could happen just as easily with linux. A new kernel version could uncover a flaw with an older driver even if that driver was opensource. And that older driver package would be flagged as incompatible with the new kernel, and the updaters would prevent you from mixxing them. Its not exactly as if this has never happened before.
That's not entirely likely to be true. What seems to happen over here in reality, is this:
Kernel (or X) update shows up, and the user is prompted to install it.
After installing it, the system doesn't fucking work anymore, because nothing ever checked to see if the update would break with the installed hardware.
One of two things then occurs:
If the user has another computer handy they get to research and attempt to solve the problem in the comfort of a web browser.
If the user does not have another computer handy, they get to try to puzzle it out themselves using arcane command line tools that they've never seen before, with text-mode documentation they've never had to find before. If they're really clever, and can get to the network, they might install and use Links and be able to search forums for help, but chances are they've never used that before either.
In both cases, after pissing away several hours or evenings trying to make it work, they either succeed and their computer again works just as well as it did last week, or they get frustrated and abandon Linux altogether. (Maybe they'll try it again some day, maybe not.)
Weird.
I've got many Canopy links at 5.7GHz which have been working at ranges averaging in the realm of 12 miles (the longest is 17 miles) for about four years without adjustment. These are mounted to things like the handrail on top of a grain elevator, or on non-penetrating mounts on the top of tall industrial buildings -- places I was sure that either vibration or ice accumulation would push things around in no time. It's been fine.
I've also got a handful of 2.4GHz links which are not quite line-of-sight, at ranges of 6 to 7 miles, all in populated areas, which also work without episode.
Really, I've only experienced a handful of problems with Canopy hardware: One time, after a nearby substation blew up, a subscriber module died and started eating power supplies. Another time, one of the two carriage bolts which hold the reflector to its mount snapped off and disappeared at 140 feet in the air. The strangest and most bothersome thing I've had trouble with is a number of repeat issues with some first-gen 5.2GHz gear where the RJ45 connector falls off of the PCB (this hasn't happened with later gear).
But, aside from those few failure modes, the stuff has been totally solid. In my experience, if the link shows good RSSI and low jitter, and works for more than a day or two, it will continue to work indefinitely.
That said, I have a lot issues with Canopy as a business unit. It took them more than a month to get around to sending a simple e-mail with some keys to upgrade a handful of access points to Advantage, and more than once I've just given up on waiting for Motorola to ship products.
And because of this, we've been putting plain-old 802.11g Engenius stuff up instead. The configuration tools aren't as polished as Canopy, but not too horrible. The units themselves use regular PoE, have integrated antennas, seem to be well-constructed. They cost less than $200 from Newegg -- a whole multi-site network can happen for the price of a single Canopy unit. There hasn't been much opportunity to move many of them yet, and none of the links have been particularly long distance, nor particularly busy, but so far the handful of them we have installed are working perfectly for the customers that have them.
And I, for one, absolutely fucking love being able to just open the laptop and access the WLAN. The less time I spend running an inverter to power a Canopy SM which is propped up on the dashboard, or running a 200' Ethernet cable up a tower, the happier I am.
SLA == Service Level Agreement. Most Slashdotters seem to think that an SLA means something like "OMG! Wonderful fat pipe all for me!!!", but it's just a contract, much like the TOS that the same Slashdotters seem to blame for everything including world hunger.
In fact, since they're both just contracts, either one can be good for the customer, or bad for the customer. The only innate differences are three words at the top of the page, which is about as insignificant a distinction as I can think of.
Defend? Hardly. I hate Microsoft with a certain passion normally reserved for childhood bullies, insurance adjusters, bankers, and pizza delivery people who show up an hour late.
But one thing I hate more is when people portray their illogical conclusions as fact. Most of the Microsoft bashing on here these days seems to revolve around warped and impractical expectations, coupled with a firm belief that anything that they've done lately which is different is absolutely wrong -- even if that differing wrongness is corrective of things which they've been previously damned for.
I have no particular adoration for Vista. I find that it suits my daily (ie: both business and personal) needs better than Linux, but that doesn't mean I like it. Instead, it just means that I hate it less on my everyday laptop than Ubuntu or Gentoo, or even XP.
Computers these days suck, you know. I used to run a half-dozen big-ish applications under DOS with 640k of RAM, and got everything done just fine. Later, I ran a half-dozen applications under Slackware 2 and OS/2 Warp with 20 megabytes of RAM, and got everything done just fine. Lately, I still only run a half-dozen applications, but it takes at least a gig or two of RAM to make start to suck less than 1995 - no matter what [modern] OS is in use.
(I'm seriously considering becoming a plumber instead -- at least gravity is a constant.)
I'm not sure that you fully understand just how TiVo interacts with digital cable in these modern times, at least in the US.
Here is some information on the topic. Another article goes on about actual, real-live abuse of the Broadcast Flag on TiVo.
(And if it were an old TiVo with analog NTSC inputs, the Broadcast Flag wouldn't be a factor, anyway.)