Yeah, you're right. A Debian admin can agree to legal terms but whether or not they do so as "Joe Coder" or "Joe Coder, duly authorized representative of Debian/SPI" is not clear. Since it isn't clear, SPI is dragged in regardless which way it turns out to be.
Really this breaks down to Townes not understanding where his project authority overlaps with his personal and SPI's corporate legal liability. Townes' actions impact a very wide circle of people, could bankrupt SPI/Debian and/or result in someone spending jail time (unlikely but possible if he continues to willfully ignore legal council in legal matters like copyright & licensing. Violating the DMCA & Patriot Act could be easier than he knows).
In a way, you could say Debian and SPI are married. What one does has an impact on the other. SPI shouldn't meddle too much in Debian's internal activities but Debian needs to not buy 5,000 acres of swampland in Florida.
Debian is a *project* not an entity or person. Projects cannot be held legally accountable or own things.
SPI is not a legal firm full of lawyers. SPI is a corporation that provides the legal entity that can own property & purchase services for the Debian developers. SPI is the entity that OWNS the name "Debian", the servers the files are hosted on, and that contracts the attorneys that protect Debian developers. (Without knowing the twisty history, I wouldn't be surprised if Debian wasn't the project that caused SPI to be created.)
In human terms, SPI is Debian's legal guardian. SPI is legally responsible for Debian's debts, obligations, and will be the one against the wall if Debian does something bad. However Debian admins can agree to legal terms and contracts which put SPI on the spot.
I quote one of the posts: SPI projects shouldn't be taking advice from Sun's attorneys. We should be taking advice from SPI's attorneys.
In other words: "don't take legal advice from the attorneys who may be suing you tomorrow, especially when those attorneys may be suing you, me, and two dozen other people in the process."
Palm screwed the pooch on the Cingular Treos. The cellphone companies do not rewrite the firmware. Palm does to the specifications of each carrier.Palm flubbed the firmware and issued a silent recall.The only diffrence between unlocked palm ones is branding and a few other bells and whistles you can activate on them.
It might be the bells and whistles that Cingular demands be disabled that is causing the snafus. Imagine a Ford dealer requesting all Tauruses be shipped with a carbureator instead of a fuel injector. I can't say, it's conjecture on my part. I just know Cingular Treos have been only slightly better than putting rabid weasels up to your head.
And if there is/was a silent Palm1 recall, Cingular was dragging their feet about doing device swaps. They could have cheerfully offered to replace handsets on PalmOne's dime instead of making people jump through hoops to do so.
I frequent TreoCentral and MyTreo and the unlocked/international crowd don't seem to have the same quantities of complaints or it could be that PalmOne's unlocked firmware is released more often than Cingular will allow. For instance, the unlocked firmware (1.2) has the revised Blazer while Cingular's firmware (1.17) does not.
Cingular support does leave a bit wanting, but call and ask for the PMC data group, front line CSRs do not know to much about the Palm.
I ask to speak with a CSR/department who is familiar with Palms and network services each time I'm forced to call. They pretty much refused to transfer me anywhere. Nothing like having someone tell me that a "relaying denied" message on a Palm when using the Cingular mail server was a problem with "my corporate mail server" to demonstrate I'm talking to experts. I'd be fine with waiting for someone trained if they'd just send me there.
I have my own issues with Cingular. But I know a Cingular GSM phone will get your more coverage than a sprint one across the nation, I own a Sprint phone as well. Girlfriend is on sprint. Curious what areas does Cingular not service?
Since the AT&T buyout, probably none. I'm still a bit bitter about the AT&T buyout since they were my carrier. Quality dropped during the migration, they making noise about dismantling the TDMA towers (which way outnumbered GSM towers in my region of Kentucky) getting a Cingular rep to do anything but repeat herself like a parrot when they misbilled me was like pulling teeth. I know *how* I got misbilled (they changed my billing date and swapped how minutes were used between "standard" and "bonus") but the stupid woman didn't seem to understand that if I have a 400minute plan and I only use 225 minutes in the billing cycle that there's no way I am (or should be) paying for 75 minutes of "out of plan usage."
It's Cingular, not palm. Each of the carriers rewrites the stock Palm firmware to add/remove features. You'll note that each of the carriers has a different firmware version.
Three executives where I work have Cingular 650s. Two executives, myself, and another IT guy have Sprint 650s. The Cingulars have had to be replaced due to problems with audio vanishing, dropped calls, horrible data transfer rates, and frequent crashes on completely stock devices.
The biggest problem we have with the Sprint Treos is getting the right email username & passwords. I'm probably the harshest abuser of my 650 and it's pretty much rock solid. I've been a palm user for years and some of my favorite apps are abandonware that don't get along well with OS5 which did cause some crashing for the first month or so as I figured out what was 650-friendly and what was a valid replacement.
The last reboot I had occurred when I hit the "web browser" button at the exact same moment someone called me. Other than such freakish incidents, the device is wonderful.
If you want a 650 or even a 700w for Cingular, see if you can find one of the GSM-unlocked Treos on Ebay.
Blame the carrier, not Palm. Each carrier hacked up the Treo's firmware to varying degrees (eliminate BT networking, custom ringtone downloaders, etc). If a carrier can't turn off features, they won't carry the phone. Whee, isn't it wonderful to have carrier-locked phones?
Cingular Treo 650s are absolutely horrible. The half-dozen I've had to work with have been totally nightmarish. Audio stops mid call, freaky volume controls, and horrible data transfers. Worse, Cingular seems to have no clue how to support it and pushes you to use their add-on business email product. If you are going with Cingular, get the unlocked GSM Treos. I wish I'd known that before half our executives went with Cingular Treos.
Sprint 650's are pretty much rock solid with great data transfer speeds. Yeah, mine's crashed a few times but each time the internal log pointed at a 3rd party app I'd installed; I find an alternative product or report a bug and my life goes on. On average the Blazer web page download + render comes out to 150Kbit while FTP and email downloads is closer to 250kbit. The Sprint Blazer still has the limit on large web pages but it doesn't crash, just stops downloading additional data. They are better about most common data issues but their sales staff still get confused about the difference between SMTP/POP and crackberry push email.
It's a relief that Cingular doesn't serve all the areas we've got offices so that Sprint 650s can counter the bad press of the Cingular ones.
Can't talk about Verizon or T-Mobile Treos so they could be angels or devils.
And yeah, I dislike WinCE not because of how the devices function but because of the problems ActiveSync seems to engender. I can't count the number of email accounts that have gone over their limit due to the massive ActiveSync error log. I personally prefer PalmOS over WinCE but I can understand why some people are drawn to the WinCE devices.
This is why I'd predict that the "smartphone" will win over the "PDA". The gadgets that are being marketed as PDAs now mostly are physically too large for the typical shirt pocket. My wife even has a Treo, but she mostly leaves it home on the desk, because it's "too big", and carries a tiny cell phone that's just a phone. The Treo doesn't get used much, except for the few games she has loaded.
Not everyone needs a multi-function cell phone. I carried an excellent Siemens worldphone for quite a while. The sound was great, the screen was B&W so the battery life was 5-7 days (or ~6 hours talk time), and it was solid as a rock. I stopped using it when AT&T was purchased by the devil Cingular at about the same time my Visor Pro started wonking out.
But it, and every other non-PDA smartphone I've seen, sucks at internet access due to the small screen real estate. They are fine for SMS but don't even try to send email or read a web page.
And my treo650 fits in my shirt pocket without any problems, even in its leather flip-cover case. I generally don't keep anything bigger than an inkpen or my BT headset in my shirt pocket so I wouldn't do that.
Oh, and My treo is recognized by PCs with BT as a network device with no significant amount of fiddling required. I've only made sure it can synch up and get online as an experiment so I can't say what the speed is however my Treo generally gets somewhere between 50-150Kbit/s. I could get wifi if I wanted to use the SDIO slot but I don't see the need with always-on data service.
I've got a Treo650 (which I love) and it's pretty obvious that this is the form factor things will use. By this, I mean it has a screen as small as you can get away with in PDA/internet applications but is as big as the masses are willing to lug around. 2.5" x 4.5" x 1" is about the limits engineers have to work with from a market standpoint.
I figure the next generation Treo, call it the Treo800, will have all the required checkbox features: bluetooth (headset, modem, keyboard, stereo audio), 3G wireless data, wifi+VoIP, Expansion port (SDIO probably), >128MB of usable static memory, >2Mp digicam and hopefully a bog standard mini-USB port with a multiplatform driver so it acts as an A/V device.
Ultimately, there will be 4 flavors based on two hardware choices and OS: keyboard + small screen vs. large touchscreen + pen input and PalmOS vs Windows. Theoretically we could see a Unix-based product (PalmOS isn't Unix until I can run bash!), maybe from Apple, but I think the user-interface hurdle and the low margins will keep that unlikely until the Chinese or Indians are forced to perfect it in an attempt to avoid licensing an OS.
But whatever you expect to see, expect to see it in a 2.5" x 4.5" x 1" package. Maybe smaller but I doubt it. Clamshell & flip-screen designs will continue to fail in the market until they can fit that form factor without being delicate. Even then I expect them to fail b/c the keyboard will still be too small for the way most people type and the expense of the additonal components will be more than the market will bear.
I find this ironic since the professor who taught engineering ethics at my university pointed out the students paid the tuition and that the instructors are contracted to provide a service. He was at the time discussing professors who refused to keep office hours and that did not provided a reasonably detailed syllabus, a pretty fundamental lapse in the student-instructor contract. He discussed the student responsibilities as well (not disrupt class, keep questions relevant to the course, etc) which were really the responsibilities of the individual student to the rest of the paying class.
I can see that laptops could be disruptive to teachers and other students. I also see how a person who went to one of the increasing number of colleges that REQUIRE the use of a laptop would be quite upset.
These cases are going to increase in number as more and more people graduate from the "wired" schools and the tide is going to start turning against the instructors. This will likely cause several of them to leave their positions, unable to work in that kind of environment. I'm on the fence on this matter since a) I'm a PDA-addict who keeps most of my information digitally and it seems hypocritical to deny someone else their tech toy and b) I have trouble hearing over the sound of keyboards clicking and have trouble focusing if I can see a TV or other monitor in my line of sight.
I'm a long-time client of Webscriptions/Baen so that's the basis for my responses. I've used a few other ebook sources but not with any regularity or frequency to speak of.
* price (seems only fair the prices should be competitive à la less expensive than the same back in hardback (they're NOT!)).
Webscription books ARE cheaper than hardback. Typically they cost $4-6 and are available the same day as the hardback ships. Compare that to your $15+ hardback.
* compatibility - until and unless I know I can move my book around to different readers/computers to read without being fingered a criminal and with minimal fuss, I'm not interested.
Webscription books are available in various formats, including HTML. You buy the book and you can download whatever format you want, including going back and downloading again with a different format.
* convenience - related to previous comment -
Webscription files are DRM free.
* quality - I still haven't seen a device or reader that approaches the quality of print (even the e-ink doesn't), nor have I found something that comes close the the ambience and ergonomics of a book.... heck none of the e-readers even come close to smelling like a book.
I can't argue ambience or smell but some devices do have very high resolution. Windows Mobile apparently has a limit since the Treo 700w has a lower resolution than the Treo 650 or the upcoming Treo 700p. My treo 650's resolution is pretty darn good.
* portability - I don't know the state-of-the-art for things like printing portions of an e-book,
Printing HTML should be pretty close to page layout. Not sure about the other formats; I generally don't print out the ebooks since I have multiple PalmOS devices.
* selection - I don't want to make my decision on e-book reading based on what's available and what I have to do to get it.
Unfortunately this is a chicken/egg issue. Publishers won't make it available until it makes sense and it doesn't make sense if there's no buyers. I can only recommend you notify your favorite publishers that you'd like to purchase their books via Webscription or a similarly DRM-free system.
* price - did I mention price?
Did I mention I'm scottish? $4-$6. And then there are the free books. (Typically Book 1 in a series is released for free once Book 2 goes paperback or Book 3 arrives in hardback)
They aren't cancelling the existing lifetime subscriptions, just refusing to sell new ones as of next week.
Which means this week I switch from monthly to lifetime. Why, you ask, did I ever go monthly? My TiVo (SDH-400) has the lifetime TiVo Basic and I wasn't sure at Christmas if I needed all the wishlist and extra stuff. But I do. Oh, I do.
Combine graffiti anywhere (http://www.freewarepalm.com/utilities/graffitiany where.shtml) with the Graffiti-1 alphabet (yahm.palmoid.com/G14OS54.zip) and you go back to a nice, comfortable Palm. I use the keyboard at times and I'm fairly quick with it but I'm partial to putting data into the Treo while on my deskphone at work; using the thumboard while holding a regular phone up to your ear sucks!
Get something like Butler or Central to manage the side buttons (stupid, stupid Palm for not having reconfigurable buttons by default!) and Phone Technician for decent mp3 ringtones.
I agree though, there's no point in NOT including pen input when the device ships with the graffiti2 libraries by default! I will try the new ALP os when it comes out to give me a longer migration away from palm b/c I just plain don't like the WindowsCE/Mobile programs. I'm a simple (but heavy) user and they just seem clunky and overly complex with zero gain.
I know Peltiers have lower efficiency but looking at a steam system you get 50% heat-> mechanical going to a 75% efficiency alternator gives you a total conversion efficiency of 37%. Reduce this efficiency by the heat lost by the plumbing, probably no more than 5% so say 32%.
But then you have to factor in mass of the >50' of high temperature piping, fluids, the steam engine, and the alternator on the vehicle. Finally, existing vehicle designs aren't exactly bursting with big empty spaces. Installing a peltier shroud for the exhaust system would be far more straightforward.
So for immediate adoption (1-2 years) or even as a retrofit for existing models I believe a Peltier system would be more feasible. (Imagine the radio "boomers" using a peltier system to provide extra juice!) Of course cost could tank this completely (how many watts of Peltier can you buy for the cost of a steam system?).
Inarguably, steam/stirling would be the superior choice for the next generation of hybrids.
One thing people don't seem to be grasping here is that this technology is essentially orthogonal to conventional combustion-electric hybrids. There's no reason (aside from not owning the tech, of course) why Toyota couldn't add this to the Prius IV....could you really have room for both the steam system and the paraphenalia of a hybrid car, and could you afford to add both?
I can't speak of the efficiency but from a straightforward adoption concept, it seems like a Peltier system would be more appropriate as they are incredibly light weight heat->electricity conversion devices. The reduced weight could offset the Peltier's lower kW/$ ratio.
Re:I am not a Photo Pro, but I play one on TV..
on
Apple's Aperture Reviewed
·
· Score: 2, Insightful
If Aperature helped photographers take better pictures by, I don't know, suggesting aperature settings, then the reviewer's credentials would be suspect.
Given than Aperature is a workflow product more than an image manipulation product, a photo retoucher and art director should have an excellent idea of what a commercial photographer needs Aperature to do. Both deal with large numbers of images, are concerned about image quality, relevance, building libraries, searching by metadata, color data, or thumbnails.
The electricity->hydrogen->electricity cycle is only about 50% efficient using utility-scale 100MW plants
True, it is not the most efficient. However it means you can get energy (cracked hydrogen) to inland facilities without building a huge power infrastructure. Trucking the fuel in is not as efficient but sometimes you accept less power later in return for some power now. Remember, "perfection is the enemy of the good."
Second, I believe that using a floating platform with very tall (~400 feet or so) structures is asking for trouble. Something floating is far more vulnerable to storms than a securely grounded pile. There must be a good reason it's not being done now.
Oil rigs are already several hundred feet tall and they have much more mass at altitude than a couple of turbines. As for storms, the power pods can move at 20 knots and can get to a dock in just a few hours. During the storm there wouldn't be any pow.... wait.... Oh yeah! They have stored hydrogen that they can use to provide power during emergencies! Wow, what foresight!
Thirdly, why have the things so far from shore. Transmission losses (if undersea cables are employed) are large over such distances, and it does take quite a bit of aluminum to make such long wires
We're only talking about a mile or so of shore. Trading a mile of undersea power cable for buying fuel every year starts getting pretty appealing.
As to why, if you RTFA wind power is much greater once you get about a mile off-shore. Soil causes a lot of drag, interfering with wind patterns. Anyone who's been on a very large lake can tell you there is more wind at the center than at the edge. More wind = more power. Sufficiently more wind than you would have for the power loss for a measly mile of cabling.
If a ship must come to load the hydrogen every once in a while, then you just added a large operating expense (and one of the nice things about wind and solar is very low operating expenses).
Hmmm. Let me think. I can a)buy fuel oil or coal with the cost of materials and shipping or I can b)send a tanker no more than 10 miles round trip (1 mile out to sea, about 4 miles out of the sea lanes) to pick up fuel my turbines made for free. Hmmmmm......
So why not stick to tried and true near-shore and land based wind turbines?
Better power efficiencies - more wind a mile out to sea
Reduced risk - fixed turbines cannot be towed out of the way of storms
Loss of property - developing nations' only tourism may be beaches
risk to animals - many ocean/land transitions are nesting grounds. Migratory birds are evolved enough to be lazy, using major wind currents to boost their efficiency and aiming them right at many turbines
Reusability - if a nation uses a set of these to get their economy going enough to build permanent power facilities they can tow them to another region. Alternately they could be resold or moved to areas suffering from natural disasters. You can't resell a wind turbine easily.
Reduced military targets - a lot of regions are prone to violence and infrastructure is a big target. A mile of ocean can provide a surprising amount of defense. (Yes, the US military would SEAL right in there but I'm talking "freedom fighter" types who probably don't even have access to a Zodiac.)
Safety - Hydrogen production may be seen as a risk to some groups. Putting it a mile out to sea pretty much cuts loss of life down to any onboard staff.
Unfortunately, you can't reliably use outside air. While moisture is not typically a factor during winter, pollution is. Many data centers have particulate sensors in their fire system which will go crazy if a bus goes by. And it just dirties up the place, possibly voiding the customer's contracts.
Second, you need the pressurized cooling system. Yeah, your window AC may keep the room at 60F but if the cabinets are expecting cool air to be pumped up through the floor to be vented out the top you can write off that rack of bladeservers. Even in centers with wire-mesh racks, local heat deltas can be boggling and a lack of adequate airflow makes it worse.
....by reducing the cost of fast switching. There's plenty of dark fiber http://en.wikipedia.org/wiki/Dark_fiber out there for anyone who can afford the hardware and this may take OC12 fiber cards from ~$6000US to a couple of hundred.
At the very least, it will make it possible for gigabit ethernet switches to use an optical brain to handle much larger total loads and likely at lower costs. (No, I don't know if this is cheaper to make but I figure the low grade parts that don't run at full speed will be sold cost-competitively to recoup losses, ala the original Celeron)
It could help reinforce the imagery of the Woody Allen character. I live in the south which isn't exactly a mecca of judaism (which is wonderfully mixed metaphor) so I can only assume that Woody Allen's "Jew in New York" is based on what was, at least in the 60's and 70's, a fact-based stereotype. Given the way Germany went only 20 years earlier, it would make sense for Jews of the time to be very anti-authoritorian.
After all, most of the stereotypes you've heard about the south are based on fact, if only for a select group of people. A guy in my high school was known as Bubba, the first day of each hunting season was an excused absence if you showed your tag/license, and I knew people who drank 'shine from mason jars.
I'm not a Big Iron guy but it's my understanding that main frame hardware does NOT "go to hell" unless someone hits it with a hammer. Repeatedly. And with forethought. They are highly redundant devices, hence the expense.
The fact that you are running VMs is indicative you are not a common Windows user.:)
I'd imagine if a VM could expect to see multiple threads you'd see a decent speed boost. Each of those virtual environments has to be active on a pretty regular basis just to generate the clock timing, so there's 1 thread per VM. If you are actively *using* the virtual machines you could dedicate one thread per virtual CPU.
I'm not a big VMWare user but last time I played with it, you could emulate machines with clockspeeds of CPU Mhz/(n+1) where n is the number of virtual machines. So a 1Ghz machine could have 2VMs of around 333Mhz and a 3Ghz PC could have two 1Ghz VMs. Now pretend that a 5Ghz 1-thread CPU and a 2Ghz 8-thread CPU both cost $1,000. The 5Ghz would let you have 4 1Ghz VMs while the 2Ghz would let you have 8. That makes the 8-threader twice as cost effective.
If I'm wrong about the ratio of CPU to VM clockspeeds it only changes the break point. The fact that everyone has switched to a multicore/multithread standard indicates that we have already hit that breakpoint.
I think you misunderstood. I merely provided an example where the 32+ thread CPUs discussed in the articles prove useful to the common person. You have just provided an arguement why current CPU designs shouldn't have large numbers of threads.
IANA chip designer but I can see solutions to several of the problems you pointed out. Fast, large caches. Scalable I/O busses with QoS (e.g. HyperTransport). Large address spaces. Multiple memory controllers.
Of course, it's easy for me to say "just do it" but I'm assuming that if Sun is bothering to build a CPU intended to run large numbers CMT that they've addressed these issues and have some form of solution in place.
The simplest example is OS runs on one, the game another. But it's really not that simple. Let's take a typical Windows box since it's the bulk of the market.
Thread 1: OS kernel Thread 2: firewall Thread 3: GUI Thread 4: print server Thread 5-7: various services (update, power, etc) Thread 8: antivirus Thread 9: antivirus manager/keep-alive Thread 10-16: spyware (I said a typical Windows box) Thread 17+: applications
Yeah, CMT will be handy out of box as long as the OS is aware. I expect it will be wasteful the first couple of iterations but I can't count the number of times I've had to disable antivirus and yank the ethernet while running computationally intense applications.
Cell is a multipurpose system. It's main claim to fame is a low-level logic that allows it to farm tasks out to other Cells it connects to dynamically. One Cell is pretty potent and will likely be able to handle the needs of a typical HDTV so IBM hopes every TV, TiVo, and stereo system has a Cell.
The cell system workload sharing system is apparently accessible through the general bus so it can theoretically farm tasks out to any Cell on the same network. So if you've got a WiFi network between your PS4, HDTV, TiVo, Stereo, and cell-powered PDA your video games (or PDA) could take advantage of those other devices' unused clock cycles.
6 x 10^23 * 6500 Ev * 1.6 x10^-19 J/eV = 624 MJ/mole
That translates into 173KW-hr per mole or 57.6KW-hr/g of decay.
Out of 1g half decays (28.8kWhr) over 12 years (105192 hours) =~ 0.27W sustained. An average laptop runs ~25W so you'd need 10g of tritium and a small battery to buffer the power (cursed unpredictable quantum events!)
That's at 100% efficiency. One source indicated current betavoltaics are at 25% efficiency so you'd need 40g of tritium. If this news p-n capture grid really is the shizzle, it might be closer to 50% efficiency, meaning 20g tritium.
The same source (http://peswiki.com/index.php/PowerPedia:Beta_volt aic) that quoted the 25% efficiency said total product power:weight ratio is 25W/kg. Dunno how much is, fuel, capture material, shielding, or power cleaning but being optimistic this new material might drop it to 50W/kg or roughly 22W/lb which is spot on for laptop usage.
Slashdot Mirror
Yeah, you're right. A Debian admin can agree to legal terms but whether or not they do so as "Joe Coder" or "Joe Coder, duly authorized representative of Debian/SPI" is not clear. Since it isn't clear, SPI is dragged in regardless which way it turns out to be.
Really this breaks down to Townes not understanding where his project authority overlaps with his personal and SPI's corporate legal liability. Townes' actions impact a very wide circle of people, could bankrupt SPI/Debian and/or result in someone spending jail time (unlikely but possible if he continues to willfully ignore legal council in legal matters like copyright & licensing. Violating the DMCA & Patriot Act could be easier than he knows).
In a way, you could say Debian and SPI are married. What one does has an impact on the other. SPI shouldn't meddle too much in Debian's internal activities but Debian needs to not buy 5,000 acres of swampland in Florida.
Debian is a *project* not an entity or person. Projects cannot be held legally accountable or own things.
SPI is not a legal firm full of lawyers. SPI is a corporation that provides the legal entity that can own property & purchase services for the Debian developers. SPI is the entity that OWNS the name "Debian", the servers the files are hosted on, and that contracts the attorneys that protect Debian developers. (Without knowing the twisty history, I wouldn't be surprised if Debian wasn't the project that caused SPI to be created.)
In human terms, SPI is Debian's legal guardian. SPI is legally responsible for Debian's debts, obligations, and will be the one against the wall if Debian does something bad. However Debian admins can agree to legal terms and contracts which put SPI on the spot.
I quote one of the posts: SPI projects shouldn't be taking advice from Sun's attorneys. We should be taking advice from SPI's attorneys.
In other words: "don't take legal advice from the attorneys who may be suing you tomorrow, especially when those attorneys may be suing you, me, and two dozen other people in the process."
Darn good advice.
It might be the bells and whistles that Cingular demands be disabled that is causing the snafus. Imagine a Ford dealer requesting all Tauruses be shipped with a carbureator instead of a fuel injector. I can't say, it's conjecture on my part. I just know Cingular Treos have been only slightly better than putting rabid weasels up to your head.
And if there is/was a silent Palm1 recall, Cingular was dragging their feet about doing device swaps. They could have cheerfully offered to replace handsets on PalmOne's dime instead of making people jump through hoops to do so.
I frequent TreoCentral and MyTreo and the unlocked/international crowd don't seem to have the same quantities of complaints or it could be that PalmOne's unlocked firmware is released more often than Cingular will allow. For instance, the unlocked firmware (1.2) has the revised Blazer while Cingular's firmware (1.17) does not.
I ask to speak with a CSR/department who is familiar with Palms and network services each time I'm forced to call. They pretty much refused to transfer me anywhere. Nothing like having someone tell me that a "relaying denied" message on a Palm when using the Cingular mail server was a problem with "my corporate mail server" to demonstrate I'm talking to experts. I'd be fine with waiting for someone trained if they'd just send me there.
Since the AT&T buyout, probably none. I'm still a bit bitter about the AT&T buyout since they were my carrier. Quality dropped during the migration, they making noise about dismantling the TDMA towers (which way outnumbered GSM towers in my region of Kentucky) getting a Cingular rep to do anything but repeat herself like a parrot when they misbilled me was like pulling teeth. I know *how* I got misbilled (they changed my billing date and swapped how minutes were used between "standard" and "bonus") but the stupid woman didn't seem to understand that if I have a 400minute plan and I only use 225 minutes in the billing cycle that there's no way I am (or should be) paying for 75 minutes of "out of plan usage."
It's Cingular, not palm. Each of the carriers rewrites the stock Palm firmware to add/remove features. You'll note that each of the carriers has a different firmware version.
Three executives where I work have Cingular 650s. Two executives, myself, and another IT guy have Sprint 650s. The Cingulars have had to be replaced due to problems with audio vanishing, dropped calls, horrible data transfer rates, and frequent crashes on completely stock devices.
The biggest problem we have with the Sprint Treos is getting the right email username & passwords. I'm probably the harshest abuser of my 650 and it's pretty much rock solid. I've been a palm user for years and some of my favorite apps are abandonware that don't get along well with OS5 which did cause some crashing for the first month or so as I figured out what was 650-friendly and what was a valid replacement.
The last reboot I had occurred when I hit the "web browser" button at the exact same moment someone called me. Other than such freakish incidents, the device is wonderful.
If you want a 650 or even a 700w for Cingular, see if you can find one of the GSM-unlocked Treos on Ebay.
Blame the carrier, not Palm. Each carrier hacked up the Treo's firmware to varying degrees (eliminate BT networking, custom ringtone downloaders, etc). If a carrier can't turn off features, they won't carry the phone. Whee, isn't it wonderful to have carrier-locked phones?
Cingular Treo 650s are absolutely horrible. The half-dozen I've had to work with have been totally nightmarish. Audio stops mid call, freaky volume controls, and horrible data transfers. Worse, Cingular seems to have no clue how to support it and pushes you to use their add-on business email product. If you are going with Cingular, get the unlocked GSM Treos. I wish I'd known that before half our executives went with Cingular Treos.
Sprint 650's are pretty much rock solid with great data transfer speeds. Yeah, mine's crashed a few times but each time the internal log pointed at a 3rd party app I'd installed; I find an alternative product or report a bug and my life goes on. On average the Blazer web page download + render comes out to 150Kbit while FTP and email downloads is closer to 250kbit. The Sprint Blazer still has the limit on large web pages but it doesn't crash, just stops downloading additional data. They are better about most common data issues but their sales staff still get confused about the difference between SMTP/POP and crackberry push email.
It's a relief that Cingular doesn't serve all the areas we've got offices so that Sprint 650s can counter the bad press of the Cingular ones.
Can't talk about Verizon or T-Mobile Treos so they could be angels or devils.
And yeah, I dislike WinCE not because of how the devices function but because of the problems ActiveSync seems to engender. I can't count the number of email accounts that have gone over their limit due to the massive ActiveSync error log. I personally prefer PalmOS over WinCE but I can understand why some people are drawn to the WinCE devices.
Not everyone needs a multi-function cell phone. I carried an excellent Siemens worldphone for quite a while. The sound was great, the screen was B&W so the battery life was 5-7 days (or ~6 hours talk time), and it was solid as a rock. I stopped using it when AT&T was purchased by the devil Cingular at about the same time my Visor Pro started wonking out.
But it, and every other non-PDA smartphone I've seen, sucks at internet access due to the small screen real estate. They are fine for SMS but don't even try to send email or read a web page.
And my treo650 fits in my shirt pocket without any problems, even in its leather flip-cover case. I generally don't keep anything bigger than an inkpen or my BT headset in my shirt pocket so I wouldn't do that.
Oh, and My treo is recognized by PCs with BT as a network device with no significant amount of fiddling required. I've only made sure it can synch up and get online as an experiment so I can't say what the speed is however my Treo generally gets somewhere between 50-150Kbit/s. I could get wifi if I wanted to use the SDIO slot but I don't see the need with always-on data service.
I've got a Treo650 (which I love) and it's pretty obvious that this is the form factor things will use. By this, I mean it has a screen as small as you can get away with in PDA/internet applications but is as big as the masses are willing to lug around. 2.5" x 4.5" x 1" is about the limits engineers have to work with from a market standpoint.
I figure the next generation Treo, call it the Treo800, will have all the required checkbox features: bluetooth (headset, modem, keyboard, stereo audio), 3G wireless data, wifi+VoIP, Expansion port (SDIO probably), >128MB of usable static memory, >2Mp digicam and hopefully a bog standard mini-USB port with a multiplatform driver so it acts as an A/V device.
Ultimately, there will be 4 flavors based on two hardware choices and OS: keyboard + small screen vs. large touchscreen + pen input and PalmOS vs Windows. Theoretically we could see a Unix-based product (PalmOS isn't Unix until I can run bash!), maybe from Apple, but I think the user-interface hurdle and the low margins will keep that unlikely until the Chinese or Indians are forced to perfect it in an attempt to avoid licensing an OS.
But whatever you expect to see, expect to see it in a 2.5" x 4.5" x 1" package. Maybe smaller but I doubt it. Clamshell & flip-screen designs will continue to fail in the market until they can fit that form factor without being delicate. Even then I expect them to fail b/c the keyboard will still be too small for the way most people type and the expense of the additonal components will be more than the market will bear.
I find this ironic since the professor who taught engineering ethics at my university pointed out the students paid the tuition and that the instructors are contracted to provide a service. He was at the time discussing professors who refused to keep office hours and that did not provided a reasonably detailed syllabus, a pretty fundamental lapse in the student-instructor contract. He discussed the student responsibilities as well (not disrupt class, keep questions relevant to the course, etc) which were really the responsibilities of the individual student to the rest of the paying class.
I can see that laptops could be disruptive to teachers and other students. I also see how a person who went to one of the increasing number of colleges that REQUIRE the use of a laptop would be quite upset.
These cases are going to increase in number as more and more people graduate from the "wired" schools and the tide is going to start turning against the instructors. This will likely cause several of them to leave their positions, unable to work in that kind of environment. I'm on the fence on this matter since a) I'm a PDA-addict who keeps most of my information digitally and it seems hypocritical to deny someone else their tech toy and b) I have trouble hearing over the sound of keyboards clicking and have trouble focusing if I can see a TV or other monitor in my line of sight.
I'm a long-time client of Webscriptions/Baen so that's the basis for my responses. I've used a few other ebook sources but not with any regularity or frequency to speak of.
* price (seems only fair the prices should be competitive à la less expensive than the same back in hardback (they're NOT!)).
Webscription books ARE cheaper than hardback. Typically they cost $4-6 and are available the same day as the hardback ships. Compare that to your $15+ hardback.
* compatibility - until and unless I know I can move my book around to different readers/computers to read without being fingered a criminal and with minimal fuss, I'm not interested.
Webscription books are available in various formats, including HTML. You buy the book and you can download whatever format you want, including going back and downloading again with a different format.
* convenience - related to previous comment -
Webscription files are DRM free.
* quality - I still haven't seen a device or reader that approaches the quality of print (even the e-ink doesn't), nor have I found something that comes close the the ambience and ergonomics of a book.... heck none of the e-readers even come close to smelling like a book.
I can't argue ambience or smell but some devices do have very high resolution. Windows Mobile apparently has a limit since the Treo 700w has a lower resolution than the Treo 650 or the upcoming Treo 700p. My treo 650's resolution is pretty darn good.
* portability - I don't know the state-of-the-art for things like printing portions of an e-book,
Printing HTML should be pretty close to page layout. Not sure about the other formats; I generally don't print out the ebooks since I have multiple PalmOS devices.
* selection - I don't want to make my decision on e-book reading based on what's available and what I have to do to get it.
Unfortunately this is a chicken/egg issue. Publishers won't make it available until it makes sense and it doesn't make sense if there's no buyers. I can only recommend you notify your favorite publishers that you'd like to purchase their books via Webscription or a similarly DRM-free system.
* price - did I mention price?
Did I mention I'm scottish? $4-$6. And then there are the free books. (Typically Book 1 in a series is released for free once Book 2 goes paperback or Book 3 arrives in hardback)
They aren't cancelling the existing lifetime subscriptions, just refusing to sell new ones as of next week.
Which means this week I switch from monthly to lifetime. Why, you ask, did I ever go monthly? My TiVo (SDH-400) has the lifetime TiVo Basic and I wasn't sure at Christmas if I needed all the wishlist and extra stuff. But I do. Oh, I do.
Combine graffiti anywhere (http://www.freewarepalm.com/utilities/graffitiany where.shtml) with the Graffiti-1 alphabet (yahm.palmoid.com/G14OS54.zip) and you go back to a nice, comfortable Palm. I use the keyboard at times and I'm fairly quick with it but I'm partial to putting data into the Treo while on my deskphone at work; using the thumboard while holding a regular phone up to your ear sucks!
Get something like Butler or Central to manage the side buttons (stupid, stupid Palm for not having reconfigurable buttons by default!) and Phone Technician for decent mp3 ringtones.
I agree though, there's no point in NOT including pen input when the device ships with the graffiti2 libraries by default! I will try the new ALP os when it comes out to give me a longer migration away from palm b/c I just plain don't like the WindowsCE/Mobile programs. I'm a simple (but heavy) user and they just seem clunky and overly complex with zero gain.
I know Peltiers have lower efficiency but looking at a steam system you get 50% heat-> mechanical going to a 75% efficiency alternator gives you a total conversion efficiency of 37%. Reduce this efficiency by the heat lost by the plumbing, probably no more than 5% so say 32%.
But then you have to factor in mass of the >50' of high temperature piping, fluids, the steam engine, and the alternator on the vehicle. Finally, existing vehicle designs aren't exactly bursting with big empty spaces. Installing a peltier shroud for the exhaust system would be far more straightforward.
So for immediate adoption (1-2 years) or even as a retrofit for existing models I believe a Peltier system would be more feasible. (Imagine the radio "boomers" using a peltier system to provide extra juice!) Of course cost could tank this completely (how many watts of Peltier can you buy for the cost of a steam system?).
Inarguably, steam/stirling would be the superior choice for the next generation of hybrids.
I can't speak of the efficiency but from a straightforward adoption concept, it seems like a Peltier system would be more appropriate as they are incredibly light weight heat->electricity conversion devices. The reduced weight could offset the Peltier's lower kW/$ ratio.
If Aperature helped photographers take better pictures by, I don't know, suggesting aperature settings, then the reviewer's credentials would be suspect.
Given than Aperature is a workflow product more than an image manipulation product, a photo retoucher and art director should have an excellent idea of what a commercial photographer needs Aperature to do. Both deal with large numbers of images, are concerned about image quality, relevance, building libraries, searching by metadata, color data, or thumbnails.
True, it is not the most efficient. However it means you can get energy (cracked hydrogen) to inland facilities without building a huge power infrastructure. Trucking the fuel in is not as efficient but sometimes you accept less power later in return for some power now. Remember, "perfection is the enemy of the good."
Second, I believe that using a floating platform with very tall (~400 feet or so) structures is asking for trouble. Something floating is far more vulnerable to storms than a securely grounded pile. There must be a good reason it's not being done now.
Oil rigs are already several hundred feet tall and they have much more mass at altitude than a couple of turbines. As for storms, the power pods can move at 20 knots and can get to a dock in just a few hours. During the storm there wouldn't be any pow.... wait.... Oh yeah! They have stored hydrogen that they can use to provide power during emergencies! Wow, what foresight!
Thirdly, why have the things so far from shore. Transmission losses (if undersea cables are employed) are large over such distances, and it does take quite a bit of aluminum to make such long wires
We're only talking about a mile or so of shore. Trading a mile of undersea power cable for buying fuel every year starts getting pretty appealing.
As to why, if you RTFA wind power is much greater once you get about a mile off-shore. Soil causes a lot of drag, interfering with wind patterns. Anyone who's been on a very large lake can tell you there is more wind at the center than at the edge. More wind = more power. Sufficiently more wind than you would have for the power loss for a measly mile of cabling.
If a ship must come to load the hydrogen every once in a while, then you just added a large operating expense (and one of the nice things about wind and solar is very low operating expenses).
Hmmm. Let me think. I can a)buy fuel oil or coal with the cost of materials and shipping or I can b)send a tanker no more than 10 miles round trip (1 mile out to sea, about 4 miles out of the sea lanes) to pick up fuel my turbines made for free. Hmmmmm......
So why not stick to tried and true near-shore and land based wind turbines?
Unfortunately, you can't reliably use outside air. While moisture is not typically a factor during winter, pollution is. Many data centers have particulate sensors in their fire system which will go crazy if a bus goes by. And it just dirties up the place, possibly voiding the customer's contracts.
Second, you need the pressurized cooling system. Yeah, your window AC may keep the room at 60F but if the cabinets are expecting cool air to be pumped up through the floor to be vented out the top you can write off that rack of bladeservers. Even in centers with wire-mesh racks, local heat deltas can be boggling and a lack of adequate airflow makes it worse.
....by reducing the cost of fast switching. There's plenty of dark fiber http://en.wikipedia.org/wiki/Dark_fiber out there for anyone who can afford the hardware and this may take OC12 fiber cards from ~$6000US to a couple of hundred.
At the very least, it will make it possible for gigabit ethernet switches to use an optical brain to handle much larger total loads and likely at lower costs. (No, I don't know if this is cheaper to make but I figure the low grade parts that don't run at full speed will be sold cost-competitively to recoup losses, ala the original Celeron)
It could help reinforce the imagery of the Woody Allen character. I live in the south which isn't exactly a mecca of judaism (which is wonderfully mixed metaphor) so I can only assume that Woody Allen's "Jew in New York" is based on what was, at least in the 60's and 70's, a fact-based stereotype. Given the way Germany went only 20 years earlier, it would make sense for Jews of the time to be very anti-authoritorian.
After all, most of the stereotypes you've heard about the south are based on fact, if only for a select group of people. A guy in my high school was known as Bubba, the first day of each hunting season was an excused absence if you showed your tag/license, and I knew people who drank 'shine from mason jars.
I'm not a Big Iron guy but it's my understanding that main frame hardware does NOT "go to hell" unless someone hits it with a hammer. Repeatedly. And with forethought. They are highly redundant devices, hence the expense.
The fact that you are running VMs is indicative you are not a common Windows user. :)
/multithread standard indicates that we have already hit that breakpoint.
I'd imagine if a VM could expect to see multiple threads you'd see a decent speed boost. Each of those virtual environments has to be active on a pretty regular basis just to generate the clock timing, so there's 1 thread per VM. If you are actively *using* the virtual machines you could dedicate one thread per virtual CPU.
I'm not a big VMWare user but last time I played with it, you could emulate machines with clockspeeds of CPU Mhz/(n+1) where n is the number of virtual machines. So a 1Ghz machine could have 2VMs of around 333Mhz and a 3Ghz PC could have two
1Ghz VMs. Now pretend that a 5Ghz 1-thread CPU and a 2Ghz 8-thread CPU both cost $1,000. The 5Ghz would let you have 4 1Ghz VMs while the 2Ghz would let you have 8. That makes the 8-threader twice as cost effective.
If I'm wrong about the ratio of CPU to VM clockspeeds it only changes the break point. The fact that everyone has switched to a multicore
I think you misunderstood. I merely provided an example where the 32+ thread CPUs discussed in the articles prove useful to the common person. You have just provided an arguement why current CPU designs shouldn't have large numbers of threads.
IANA chip designer but I can see solutions to several of the problems you pointed out. Fast, large caches. Scalable I/O busses with QoS (e.g. HyperTransport). Large address spaces. Multiple memory controllers.
Of course, it's easy for me to say "just do it" but I'm assuming that if Sun is bothering to build a CPU intended to run large numbers CMT that they've addressed these issues and have some form of solution in place.
The simplest example is OS runs on one, the game another. But it's really not that simple. Let's take a typical Windows box since it's the bulk of the market.
Thread 1: OS kernel
Thread 2: firewall
Thread 3: GUI
Thread 4: print server
Thread 5-7: various services (update, power, etc)
Thread 8: antivirus
Thread 9: antivirus manager/keep-alive
Thread 10-16: spyware (I said a typical Windows box)
Thread 17+: applications
Yeah, CMT will be handy out of box as long as the OS is aware. I expect it will be wasteful the first couple of iterations but I can't count the number of times I've had to disable antivirus and yank the ethernet while running computationally intense applications.
Cell is a multipurpose system. It's main claim to fame is a low-level logic that allows it to farm tasks out to other Cells it connects to dynamically. One Cell is pretty potent and will likely be able to handle the needs of a typical HDTV so IBM hopes every TV, TiVo, and stereo system has a Cell.
. ars . ars
The cell system workload sharing system is apparently accessible through the general bus so it can theoretically farm tasks out to any Cell on the same network. So if you've got a WiFi network between your PS4, HDTV, TiVo, Stereo, and cell-powered PDA your video games (or PDA) could take advantage of those other devices' unused clock cycles.
Here's some A to RTF.
http://arstechnica.com/articles/paedia/cpu/cell-1
http://arstechnica.com/articles/paedia/cpu/cell-2
Avagdro's is 6x10^23 and it's 1.6x10^-19 J/eV.
t aic) that quoted the 25% efficiency said total product power:weight ratio is 25W/kg. Dunno how much is, fuel, capture material, shielding, or power cleaning but being optimistic this new material might drop it to 50W/kg or roughly 22W/lb which is spot on for laptop usage.
6 x 10^23 * 6500 Ev * 1.6 x10^-19 J/eV = 624 MJ/mole
That translates into 173KW-hr per mole or 57.6KW-hr/g of decay.
Out of 1g half decays (28.8kWhr) over 12 years (105192 hours) =~ 0.27W sustained. An average laptop runs ~25W so you'd need 10g of tritium and a small battery to buffer the power (cursed unpredictable quantum events!)
That's at 100% efficiency. One source indicated current betavoltaics are at 25% efficiency so you'd need 40g of tritium. If this news p-n capture grid really is the shizzle, it might be closer to 50% efficiency, meaning 20g tritium.
The same source (http://peswiki.com/index.php/PowerPedia:Beta_vol
Dude, it was walmart. Take them back.