The amount of backpedaling among Apple proponents amuses me.
When Android was a blip on the radar, Apple fans said that it wouldn't amount to much.
When Android devices outsold the iPhone for the first time in Q1 2010, Apple fans said that it was just a fluke because everyone was waiting for the iPhone 4.
Now we're at a point where Android mobile phones are outselling all iOS devices combined. You can throw tablets and portable handhelds in there if you want, and you're still short.
But throwing in Apple TV? That's just desperate. Apple TV may nominally run iOS, but it's a different product in a different segment without an app store.
And then we hear about how it's all going to change with the Verizon iPhone. The gap between Android smartphone sales and iPhone sales is now more than 64 million devices per year. That's way more than the total number of iPhone users in the US today, and it's probably more than the total number of iPhone devices sold in the USA. Ever. It's also about 2/3 of Verizon's entire customer base.
The Verizon iPhone is probably going to be a blockbuster success. But that's not going to close the gap.
There were predictions last year that Android would be the #2 smartphone OS by 2012 behind Symbian. Then there were predictions that Android would be #1 in 2012. Now we find out that it's #1 in 2010.
615% growth in one year. From zero to #1 in under 30 months.
So, yeah, Apple is keeping pace. Barely. If you include iOS tablets. And iOS portable handhelds that aren't phones. And you don't include any of that for Android.
This is not a game you want to play. If you want to start counting things that aren't phones, the numbers are going to get very ugly very fast for Apple. Android is going to be in everything. It's already in e-readers, tablets, phones, and TVs. It's soon going to be in everything from car entertainment systems to noise-cancelling headphones.
So keep moving the goal posts. It doesn't matter any more.
I can tell you my biggest problem with modern 3D cinema: 24fps.
For whatever reason all 3D cinema is at 48Hz (24fps per eye). This creates a huge amount of motion judder, which (in my opinion) is much more noticeable in 3D.
The 60p football game I watched on a 3DTV at Best Buy resulted in significantly less eyestrain and was much more pleasant.
I used to dismiss "120Hz" TVs as a gimmick but since I purchased one (240Hz actually - Sony XBR9) I have decided that judder sucks. Motion interpolation like my XBR9 has is not as good as real higher-framerate content, but it still does a fantastic job of getting rid of judder in film. Yes, it makes film look like video. No, I don't care anymore.
We don't watch many black and white films anymore, we don't watch films with crappy sound anymore, and we don't watch 4:3 films anymore. 24 frames per second is no more a part of the cinema experience than black and white was. It's time to move on and acknowledge that smooth motion is better.
Don't forget, Apple is not really a computer company any more. It's an consumer electronics company. They sell to consumers, not geeks. (Geeks buy anyway, because the engineering is great, but again that's clearly the minority case.)
Except that the engineering isn't great. Apple's engineering optimizes some factors (like weight and appearance) at the expense of other factors (like durability, serviceability, and expandability).
Take the new MacBook Air 13. It's super-thin and super-light, which is cool from an engineering standpoint. But it also has an inadequate CPU cooling solution that's noisy and that causes the CPU to throttle under high sustained CPU/GPU load. There are PCs with the same problem, but they're almost universally seen as defective (for example, look at the complaints about throttling on the Dell E6400) whereas almost everyone just gives Apple a pass on this. Part of that is that Apple's throttling is more sophisticated (it throttles less because it's designed to compensate for inadequate cooling, whereas thermal monitor on a PC is mostly a "don't melt the CPU" last-resort feature) but it makes benchmarking very hard because you don't get consistent results from brief and sustained load.
Or take MagSafe. It's cool that it can save your laptop from being yanked off the table, but the earlier MagSafe cords had inadequate stress relief and many eventually shorted out and melted. Then there's the fact that the adapter is $70 and you can only get it from Apple, whereas PC adapters are widely available online for $30 or less - and that's for the genuine product (buying no-name, non-UL crap power adapters is a bad idea).
Or take the new SSD. Good luck finding a replacement for that if you want more disk space - at best you get a couple of manufacturers and high prices. My ThinkPad T400 takes regular 2.5" drives and you can swap them out in a couple of seconds.
Spill Diet Coke on your MacBook? You're in for a long and costly repair. Spill Diet Coke on your ThinkPad? You'll likely only need to replace the keyboard, which costs under $50 and can be replaced in a couple of minutes.
My ThinkPad T400 was cooler, quieter, more ergonomic, easier to service, easier to upgrade, more durable, and cheaper than the MacBook Pro that was available when I bought it. Those are engineering criteria that I care about. Size, weight, and looks are important, but to say that Apple's engineering is "great" tells only half the story. Apple is great at compromising everything in the name of a few factors - like size and appearance. Fortunately for Apple, those are the factors that sell machines to individuals. But if you want to keep your machine for years and if you care about it still being useful, you might want to look at something that's more durable, more serviceable, and more upgradeable.
My T400 is a better machine than the MacBook Pro. Not just cheaper, but better engineered.
Don't teach them Linux unless that's what the class is about.
My 1st-year CS intro course (CS 1300) was taught in Python, but that's not particularly important because virtually any decent programming language is acceptable.
An intro CS course should be focused more on things like algorithms, data structures, time/space complexity, and other theory. It's not the place to teach system administration and it's not the place to teach software engineering.
Pragmatically speaking, people are going to have to compile and run code. So you might end up having to teach some basic Linux skills. At my university we ran RHEL 4.x on the lab machines at the time (now they're on Ubuntu), and there was a weekly lab section where the TAs could provide 1-on-1 instruction on how to get things working. A substantial fraction of the class was already familiar with Linux, but there were also a good number of students who weren't.
The point is that the mechanics (e.g. "type make") should be addressed during the labs if possible, or in the instructions for the assignments, or during office hours, or during an (optional) concurrent course. The intent should be to provide the necessary skills for students who have no Linux background without devoting much of your (very limited) lecture time to it. If you spend a week or two trying to teach Linux, you'll find that 1/3 of your students don't show up.
The TPM requirement can be turned off using a group policy setting, but then it's not transparent to users, they have to enter a pass-phrase on every boot. External disk encryption doesn't require a TPM chip by default, I use that feature on my rather old laptop that doesn't have a TPM chip.
More than that, if you're supporting Windows 7 in a business context, why the hell are you buying crappy consumer-level PCs? Enterprises buy machines like HP EliteBooks, Lenovo ThinkPads, or Latitudes, all of which are better built and have longer support life-cycles than a consumer PC. And of course they have TPMs, with the exception of some lower-end ThinkPad models.
These machines aren't even particularly expensive. You can get a ThinkPad T410i for $750 right now.
This is absolutely correct. CS is not software engineering, it's not programming, and it's not computer engineering.
CS = Theory and research into computation. What is computable, on what sort of machine, in what time bounds. Research into new applications for computation (such as machine vision or natural language processing).
SWE = The engineering process of delivering software that is functional and reasonably defect-free on time and on budget.
CE = The design and engineering of computer systems. CPUs, GPUs, buses, storage systems, interconnects, etc.
Programming = The act of creating code, which (when done correctly) requires skills from CS, SWE, and CE.
The bottom line is that you can't be a good coder unless you have at least some of all three skills. Algorithms and time complexity matter. So does writing code that actually performs on real hardware. So does writing code that is maintainable and reasonably defect-free.
I am fortunate that my "CS" program was actually more of a CS+CE+SWE program. I am not an expert in any of those fields but I do know enough to work effectively on a team to solve problems and write good code.
Well you have some good points and some bad points.
First of all, you're absolutely right that safer vehicles breed less safe driving. It's a well-known, well-measured effect that needs to be considered every time a new safety device is considered.
That said, the idea that all accidents can be avoided with better driving is just crap. People screw up. I generally consider myself a pretty careful driver and I have had my share of mistakes. Anyone who is a safe driver should be able to identify many instances where they 'got away' with a lapse in concentration. Maybe you didn't see a car in your blind spot and almost hit them. Maybe you pushed that yellow light way further than you should have. Maybe you drove when you knew you were tired.
I have personally crashed a brand-new Prius into an SUV and done $20k worth of damage. Was I driving recklessly? No. Was I tired, on the phone, or drunk? No. But I was looking at the stopped car in the far lane instead of the moving car in the near lane, and by the time I pulled out to make the turn it was too late.
Chances are good that you'll screw up at some point.
Does this mean that we all need to drive tanks? No. But it does mean that you need to consider the risks of a supermini vehicle. A 'safe' supermini cannot protect you as well as a 'safe' larger vehicle. It doesn't need to be an SUV and it doesn't need to have crappy mileage.
The fact that people have anecdotes about how a Smart or another tiny car survived doesn't mean shit. In some kinds of crashes a supermini will do perfectly fine. In others it will be absolutely screwed.
The data, on the other hand, shows that heavier vehicles are safer. Yes, there are unsafe heavy vehicles, and yes, there is a spectrum of crash-worthiness among light vehicles from "total crap" to "decent". But in any given crash - be it single vehicle or multi-vehicle - the best smallest vehicles do not come close to the best heavier vehicles.
There is a risk trade-off here. The lowest risk option is not to drive at all. The highest risk option involves vehicles that provide little or no protection (like a motorcycle). And there is a spectrum of options in-between.
People need to be made aware of the negative consequences of their choices. In a world where gas is effectively free (relative to the cost of a vehicle) many people are going to choose inefficient vehicles. When the true cost of the fuel (including carbon emissions) is factored into the price, people will choose more efficient vehicles, drive less, or elect not to drive at all.
But you have to be careful not to throw stones from your glass house. There are many factors that have as much of an impact on your carbon emissions as the type of vehicle that you drive - like whether you eat meat and whether you use air-conditioning.
The most important thing is that you can't guess about how much energy something uses, because you're almost always wrong. Switching to washing your clothes in cold water (vs. hot) saves more energy than line-drying (vs. using a dryer). For me, in the summer months, using a fan to cool my apartment rather than the A/C saves more energy than not driving. Manufacturing and disposal are only a small part of the energy requirements in a car's life-cycle. Unplugging 'vampire' electronics is not going to save a whole ton of energy. Transit is not always a slam-dunk in terms of energy consumption, especially when it frequently runs at low utilization. Driving to the store to buy an item can often require more energy than having it shipped to you.
There are plenty of 10-inch Android tablets, almost all of which are cheaper than the iPad. The Archos 101, for example, is $299 for the 8GB version.
Whether it was Apple's doing or not, the rumors of the $999 iPad did a lot to make people think that the iPad was "cheap" at $499 and that somehow it was Apple's "sales volume" that allowed them to reach that "revolutionary price".
Of course all of that is BS. The 16GB iPad has the same ARM CPU and the same flash memory as the 16GB iPod Touch, which runs $275. And the 10" screen doesn't cost anywhere near $225, even if it is IPS. iSuppli estimates that the 16GB iPad costs $219 to make, which is not hard to believe considering that you can get a full netbook for around the same price.
The Prius ECU integrates with the climate control system, so if you turn on the heat the engine will run until there's enough coolant to give you heat.
Actually, if you put the heater on "auto" then the heat will stay off and the engine will run until the engine coolant gets up to a useful temperature (usually takes about 3 minutes), then the heater fan gradually spins up. When the engine coolant temperature finally gets hot enough (takes about 10 mins, or less if you're actively driving) then the engine will stop.
This is one of the advantages of building a car less like a series of pieces and more like an integrated system. The Prius is effectively a series of independent modules (ECUs) that work together through a common network (CAN bus, among others) for a single goal. The power steering can have different assist forces at different speeds because it gets speed data from the combi-display (speedometer). The engine can shut down when it's appropriate based on input from the hybrid system and other systems like the climate control computer. The hybrid system can avoid taking too much current to the battery when it's hot or cold because it communicates with the battery computer. The braking system can use less friction braking because it communicates with the hybrid system to provide regenerative braking (which, in turn, communicates with the battery and other systems).
The advantages to this integration are better efficiency, lower cost, reduced wiring complexity (since sensors and actuators are wired only to their nearby ECU, rather than all over the car), and other improvements. Something like cruise control is trivial to implement because you already have the data and control elements you need - it's just a matter of software.
This is why you absolutely, positively should not try to write code without an IDE - or at least an editor that offers a wide range of completion options (unfortunately I find that the people who use Emacs or Vim don't actually use the completion features they have - perhaps because they aren't as good as a good IDE like Eclipse).
Yeah, Eclipse takes 400MB of memory when I'm working on a medium-sized project. But it enables me to do what is right rather than having to do what is easy, because doing what's right is easy. I can use long identifier names when they're appropriate because they will be completed. I can use an interface if it's appropriate because Eclipse will auto-generate the method stubs anyway, so I don't have to type them twice. I can rename something if the name is inconsistent or confusing because it's easy to do. I can have my imports in the correct order and include only what's being used, because Eclipse does that for me. I can write Javadoc comments because Eclipse generates the right boilerplate.
You can do all of this without any code completion or refactoring tools. But it's more likely that you'll make a mistake at some point and it's more likely that you will get lazy and develop bad habits like using undescriptive variable names.
Right, I've used NX over a WAN link (albeit a very fast one) and it's able to handle non-intensive tasks like running Eclipse or general GNOME apps on a 2560x1600 display with reasonable performance. You can forget trying to do much of anything at that resolution with VNC over a WAN.
That said, NX totally dies if you try to do anything with animation or video. Protocols like PCoIP or HP RGS do a lot better here since they compress more.
It seems like what we call "real computer science" (like algorithms or theory of computation) is actually math. I don't see anything scientific about it at all.
Anyone who thinks CS is just about algorithms or computational theory doesn't have a broad base in CS. There is a huge amount of research in fields like computer vision, natural language processing (my focus), computer graphics, networking theory, and other topics that are unquestionably (in my opinion) CS but also not direct analogs of anything in mathematics.
It's true that CS is math-heavy, but so are many sciences. Theoretical Physics is absolute a science, but it could also be described as basically just the mathematical modeling of physical processes. Which is exactly what a lot of CS is.
I think the confusion comes from the fact that there are basically three different fields (CS, CE, and SWE) that are typically intermingled at the university level. Arguably, I think that's a good thing since the fields are interdependent enough that a student with a "CS" degree should have a background in all three.
Computer Science (CS) is basically the study of the theory and applications of computing. Designing an algorithm to sort or search or calculate faster fits here, but so does the study of the applications of algorithms to different tasks. Questions like "can we use a SVM classifier to identify the parts of speech in a sentence", or "which algorithms are best for converting a stereo image into a depth map", or "is there a better way to route in a wireless mesh network" fit here. CS is about expanding the range of problems that computing can solve and experimenting to find what works best in a particular situation. There is very little focus on the hardware as actually implemented (as opposed to a simplified theoretical model) or on the process of designing and building software. CS is not software engineering and it's not computer engineering.
Computer Engineering (CE, sometimes ECEN) is basically between electrical engineering and computer science, but it's really neither. CE is about engineering better computers, and it generally includes topics like microprocessor design, architectural design, low-level networking (e.g. QuickPath or HyperTransport), and compiler design. Some of these fields cross paths with CS (e.g. compiler design) and some cross paths with EE (e.g. IC design). But unlike a EE, a CE is not really concerned about the gate-level design or electrical properties of a CPU (they work with EEs on that) and unlike a computer scientist they actually care about the details of the hardware.
Software Engineering (SWE) is neither CS nor CE. It's also not programming. SWE is about the process of producing software: specification, design, testing, maintenance, and management. SWE has streaks of business management in it too: keeping projects on-time and on-budget is as much a management task as it is an engineering task. Software engineers exist to manage the challenge of putting together fantastically complex systems with very little time and very little money.
I can tell you what's not in any of these fields: programming. Yes, programming is an integral part of software engineering, computer engineering (usually an HDL) and nearly all computer science. Yeah, you can work out algorithms without ever writing a line of code or do an entire CPU design on paper, but in practice everyone wants to see their ideas actually implemented. But programming isn't what CS, SWE, or CE is about: programming is just the most common means used by those fields to express their ideas. You wouldn't say that literature study is about learning the English language, and by the same token CS/CE/SWE isn't about learning to program.
Most of this comes from the fact that programming is actually pretty easy. And, no, that's not me being arrogant: almost any first-year college student can be taught to write code in a semester. The majority of mathematicians, scientists, and engineers who graduate fr
Maybe you didn't like Universe. I certainly didn't like every episode. I'm not a fan of the soap-opeara dynamics either. But it seems like everyone stopped watching from the second episode and decided to hate it based on that. Universe has a solid story arc (even if it's a bit derivative), interesting characters, and good production values. That's more than you can say for just about any other sci-fi show that's on right now.
All these comments talking about the multitude of good sci-fi shows puzzle me. What the hell is on right now? SyFy is now loaded with ghost investigation shows and wrestling, BSG is over, Caprica got cancelled, Terminator got cancelled, Dollhouse got cancelled.
Maybe you didn't like Universe. But it's not going to be replaced with another sci-fi show when SyFy can get better ratings at a lower cost with wrestling.
I do that too, but sometimes it bites you in the ass - my credit card bank, for example, occasionally asks one of those questions in addition to the password.
The big lesson here is not that you should never get breached, or that you should use some super-secure password, or that you should use a different password on every site (you should).
No, the real lesson is that passwords themselves are faulty. No one is going to select and memorize a strong password for every website they use. They're going to either re-use passwords, or choose weak passwords, or write their passwords down (or use a password manager).
None of these are good answers. The expectation is that users are going to choose strong passwords, that they will never re-use passwords, that hashes (even with salt) are an effective way to protect passwords, and that users will never be tricked into revealing their password.
It's bullshit. It's always been bullshit. Users aren't careful with passwords, and why would we expect them to be - 99.9% of the time they get away with it. Humans are bad at evaluating the risk of things that are low frequency but high impact.
The other thing that's bullshit is password reset. It doesn't make any sense: how can someone who forgot their password remember "security questions" that are actually secure. No, 99 times out of 100 these systems use some crap like "Where were you born", which is pretty damn trivial to find out for any attacker. My brokerage account has a secure password that I only use there, but resetting the password requires only my username, SSN, ZIP code, and last name. And there are far, far more people who know that stuff than people who know my password.
It's time to get serious about replacing passwords. That's the lesson here.
After the 1984 incident, Amazon was sued by a customer and settled for $150,000. They also agreed not to remove books from customer's devices - not just in a wishy-washy statement but in their court settlement:
For copies of Works purchased pursuant to TOS granting "the non-exclusive right to keep a permanent copy" of each purchased Work and to "view, use and display [such Works] an unlimited number of times, solely on the [Devices] . . . and solely for [the purchasers'] personal, non-commercial use," Amazon will not remotely delete or modify such Works from Devices purchased and being used in the United States unless (a) the user consents to such deletion or modification; (b) the user requests a refund for the Work or otherwise fails to pay for the Work (e.g., if a credit or debit card issuer declines to remit payment); (c) a judicial or regulatory order requires such deletion or modification; or (d) deletion or modification is reasonably necessary to protect the consumer or the operation of a Device or network through which the Device communicates (e.g., to remove harmful code embedded within a copy of a Work downloaded to a Device). This paragraph does not apply to (a) applications (whether developed or offered by Amazon or by third parties), software or other code; (b) transient content such as blogs; or (c) content that the publisher intends to be updated and replaced with newer content as newer content becomes available. With respect to newspaper and magazine subscriptions, nothing in this paragraph prohibits the current operational practice pursuant to which older issues are automatically deleted from the Device to make room for newer issues, absent affirmative action by the Device user to save older issues.
If Amazon did this again, then they may be in for another lawsuit. I can believe that they removed the books from their service. But it doesn't make sense for them to pull the books from devices. Until we see more evidence than a couple of random unnamed sources in a blog post, I don't buy it.
Comcast is also doing stupid things with their Internet routing. For example, to get from Denver to anywhere else in Denver, you go through Dallas. This adds at least 30 ms to each ping. This is actually one of the more efficient routes they have now; google on CRAN and traceroute and you'll see.
That's just wrong. I'm in Boulder and I have a virtual server in Denver which is connected to Level3. My traceroutes go directly on Comcast's network to Level3 in Denver. Typical RTTs are in the 15ms range.
Depending upon your definition of power user, that's almost always been the case. I remember there was that brief period when AMD beat Intel to the 1ghz mark, but apart from that the high end stuff from Intel has typically been faster.
AMD was faster from when Athlon 64 was launched (December 2003) to when Core 2 was released (July 2006). But even today AMD is just about at parity with Core 2, and Intel is about to launch Sandy Bridge. Hopefully Bulldozer can change that.
Yes, there is. The only winning move is not to play.
I played WoW for over five years, including throughout my entire undergraduate education.
I have played the game for over 3500 hours, which is enough to get an air transport pilot license if I were flying a jet instead (to be fair, WoW is considerably cheaper).
I was on a 2450-rated 3v3 Arena team, which in our class/spec comp was #1 in the US and #2 in the world (although we played a weird comp).
I don't hate WoW. I would never begrudge someone for playing. But I do despise Greg Street "Ghostcrawler" ("GC") for what he did to the community and to the game. I despise the fact that 2v2s - which I loved playing - were effectively deprecated because GC said they couldn't be balanced. I despise the fact that damage was amped way up in WoLK, so much that people were dying in three hits at launch. It took almost a year for PvP to get back to a place where it was based on skill rather than getting lucky with crits. I despise the fact that mana was effectively removed from the game as a factor for healers, that threat was effectively removed as a factor for DPS, and that the release raid content was a too-easy warmed-over version of an instance from vanilla. And, most of all, I despise that GC has to constantly screw with things, as if a title that has stood as the #1 paid MMO for the past 5 years needs it.
Eventually I just got tired of it. I got tired of reading patch notes and checking the PTR to stay ahead of the curve. I got tired of adapting to each new flavor-of-the-month combo. Maybe if you never played PvP it was better, but in my world it was a constant cycle of getting screwed by some comp for 2 months, finally seeing it get fixed with a patch, and then having to deal with the next flavor comp.
So I'm not playing along anymore. I've found other games to play, like StarCraft II. There is a world of options out there and I eventually decided that I can live without WoW.
If you think that Microsoft somehow has the power to take down any Windows server at will, you're insane.
And if they somehow had some sort of super-secret back-door that no one has found and that they've managed to keep secret from most of their employees (who would undoubtedly leak it), they can't just go and use it against one of the most high-profile organizations in the world.
You think environmental impacts studies and lawsuits are what's keeping nuclear from being profitable enough to build? They're a fucking rounding error
The cost does not come from defending the lawsuit or doing the studies. The cost comes because such lawsuits can create delays. Delays mean that an operator may need to buy power from someone else, build other types of plants (like natural gas) or delay the retirement of obsolete plants. All of these actions carry significant costs.
Safety does have a cost. That doesn't mean that it's not absolutely the right thing to do when you're dealing with a technology that's intrinsically hazardous.
Why is IPv6 not based on MAC adresses? I've never understood this.
Well, first of all, it sort of is. The typical way to get an address on an IPv6 network is stateless auto-configuration, which basically allows your client to combine an advertised route prefix with the EUI-64 (basically a longer version of a MAC address that can be generated from a MAC address) to determine its IP. You don't need any configuration for new clients and they always get the same IP address. Note that Windows Vista/7 use a hashing function with random data and the MAC address so that you can't track a single machine based on its IPv6 address, which solves privacy concerns.
Second, you can't just use the MAC address because it's not easy to route traffic that way. Routing works today because networks are assigned contiguous blocks of addresses, so it's easy to tell where to route traffic based on the address prefix. If we just had MAC addresses (which contain no information about which devices are connected to which networks), routing would require huge tables that would frequently change. This works OK for a small to medium sized network (e.g. switched Ethernet) but it doesn't work at all for the Internet. Even medium-large organizations need to use subnets to effectively manage traffic, which aren't possible without network prefixes.
But if, as appears to be the case, Comcast threatens to resolve this by targeting video traffic specifically (which in practice means netflix), then they're in the wrong.
Except that it isn't the case, as far as I can tell. It seems that Level3 is playing us all for suckers - their press release doesn't indicate that Comcast is specifically targeting Internet video, it only refers to "Internet online movies and other content", which basically means "packets".
Level3 is trying to spin this as a network neutrality debate, when it's pretty clearly not. This is a peering dispute.
Level3 DOES NOT have to interconnect with Comcast to access Comcast subscribers. Comcast is not Tier 1, which means that there are alternate indirect routes between Comcast and Level3. However, not interconnecting with Comcast would create huge traffic imbalances that could jeopardize Level3's other peering relationships.
But getting back to your coke story, did your T-61 survive the incident? Did the drain holes work?
Unfortunately, the keyboard drains do nothing when you spill it in the vent holes. I've never tested the keyboard drains but I've seen videos of them working. I would imagine that the keyboard is screwed, although it's pretty easy and cheap to fix.
As for for the lack of a trackstick vs. gesturing on the touchpad, it is simply retraining the brain.
I use the touchpad actually, not the trackstick. And I do like certain gestures, like two finger scrolling. I use a utility called EnvyTouchPad, which ironically was designed to work around the awful clickpad on the HP Envy series (which is far, far worse than the Mac clickpads).
I have three major problems with the Mac touchpads:
1. It's noisy - considerably more so than even the HP clickpads. Some PC laptops have this problem too, but the ThinkPad is actually pretty quiet. It doesn't seem like a huge issue but it is socially awkward when I'm in class, especially considering that my courses are recorded for remote students and the microphones pick up everything. 2. It makes dragging much harder. Attempting to drag with one finger is problematic because of friction and the deadzone at the top of the touchpad. Instead, you have to use two fingers, which is wierd and error prone. Trying to do something like the right-mouse-button drag (which never appears in OS X but does appear in Windows under Boot Camp) is futile. 3. It's more error prone. If you want to right click, you can either use multiple fingers or assign a touch zone. Neither is as consistent as hitting a different button. The touch zone is not demarcated on the pad and even if it were (as it is on HP clickpads) there is no tactile feel. Multiple fingers work great except sometimes you mess up and rest part of your hand on the pad, causing misclicks.
The bottom line is that I just don't know why this is a good design. The only advantage I can think of is that you get slightly more touch room, but I have never found my T400 touchpad to be too small. Gestures are nice but they do not replace the need for buttons in my opinion.
I agree and disagree on the magnetic power adaptor. Right after diet coke spills, the second biggest cause of laptop destruction is the power adapter yanking the laptop to its death or cracking the solder joints in the motherboard. The Apple design solves both of those problems.
The T400 (as with most ThinkPads) doesn't have the power jack soldered to the motherboard - it's a separate part that's connected via a wire. The part runs about $12 on eBay and takes about 10 minutes to replace.
I have never actually yanked a laptop off of a table due to the power adapter. I'm not saying that it doesn't happen, but it just doesn't happen often enough to warrant an $80 MagSafe adapter. I have no issue with MagSafe in particular, what I have an issue with is that the adapters are so expensive and the MagSafe patent prevents anyone else from making compatible replacements.
However, while the Thinkpad has the Thinklight, the MAC has the backlit keyboard.
Actually, the new Air (the one I had) doesn't have a backlit keyboard - it was one of the features that was cut, along with the sleep LED, the IR sensor, and the ambient light sensor. None of these things really bug me - I rarely use the Thinklight on my T400 anyway, since I know the keyboard layout by memory.
You just brought up another thing I hate about most Macs (and to be fair, most PCS) - the sleep light. The Air I had didn't have this issue, but a MacBook/MacBook Pro would - LED indicators should not pulsate or blink. Most of the time, I sleep in the same room as my laptop, which makes blinking (or pulsating) LEDs very annoying. I had to disconnect my (custom-built) desktop's power LED to fix this issue, but it's not quite as easy wit
Slashdot Mirror
The amount of backpedaling among Apple proponents amuses me.
When Android was a blip on the radar, Apple fans said that it wouldn't amount to much.
When Android devices outsold the iPhone for the first time in Q1 2010, Apple fans said that it was just a fluke because everyone was waiting for the iPhone 4.
Now we're at a point where Android mobile phones are outselling all iOS devices combined. You can throw tablets and portable handhelds in there if you want, and you're still short.
But throwing in Apple TV? That's just desperate. Apple TV may nominally run iOS, but it's a different product in a different segment without an app store.
And then we hear about how it's all going to change with the Verizon iPhone. The gap between Android smartphone sales and iPhone sales is now more than 64 million devices per year. That's way more than the total number of iPhone users in the US today, and it's probably more than the total number of iPhone devices sold in the USA. Ever. It's also about 2/3 of Verizon's entire customer base.
The Verizon iPhone is probably going to be a blockbuster success. But that's not going to close the gap.
There were predictions last year that Android would be the #2 smartphone OS by 2012 behind Symbian. Then there were predictions that Android would be #1 in 2012. Now we find out that it's #1 in 2010.
615% growth in one year. From zero to #1 in under 30 months.
So, yeah, Apple is keeping pace. Barely. If you include iOS tablets. And iOS portable handhelds that aren't phones. And you don't include any of that for Android.
This is not a game you want to play. If you want to start counting things that aren't phones, the numbers are going to get very ugly very fast for Apple. Android is going to be in everything. It's already in e-readers, tablets, phones, and TVs. It's soon going to be in everything from car entertainment systems to noise-cancelling headphones.
So keep moving the goal posts. It doesn't matter any more.
I can tell you my biggest problem with modern 3D cinema: 24fps.
For whatever reason all 3D cinema is at 48Hz (24fps per eye). This creates a huge amount of motion judder, which (in my opinion) is much more noticeable in 3D.
The 60p football game I watched on a 3DTV at Best Buy resulted in significantly less eyestrain and was much more pleasant.
I used to dismiss "120Hz" TVs as a gimmick but since I purchased one (240Hz actually - Sony XBR9) I have decided that judder sucks. Motion interpolation like my XBR9 has is not as good as real higher-framerate content, but it still does a fantastic job of getting rid of judder in film. Yes, it makes film look like video. No, I don't care anymore.
We don't watch many black and white films anymore, we don't watch films with crappy sound anymore, and we don't watch 4:3 films anymore. 24 frames per second is no more a part of the cinema experience than black and white was. It's time to move on and acknowledge that smooth motion is better.
Except that the engineering isn't great. Apple's engineering optimizes some factors (like weight and appearance) at the expense of other factors (like durability, serviceability, and expandability).
Take the new MacBook Air 13. It's super-thin and super-light, which is cool from an engineering standpoint. But it also has an inadequate CPU cooling solution that's noisy and that causes the CPU to throttle under high sustained CPU/GPU load. There are PCs with the same problem, but they're almost universally seen as defective (for example, look at the complaints about throttling on the Dell E6400) whereas almost everyone just gives Apple a pass on this. Part of that is that Apple's throttling is more sophisticated (it throttles less because it's designed to compensate for inadequate cooling, whereas thermal monitor on a PC is mostly a "don't melt the CPU" last-resort feature) but it makes benchmarking very hard because you don't get consistent results from brief and sustained load.
Or take MagSafe. It's cool that it can save your laptop from being yanked off the table, but the earlier MagSafe cords had inadequate stress relief and many eventually shorted out and melted. Then there's the fact that the adapter is $70 and you can only get it from Apple, whereas PC adapters are widely available online for $30 or less - and that's for the genuine product (buying no-name, non-UL crap power adapters is a bad idea).
Or take the new SSD. Good luck finding a replacement for that if you want more disk space - at best you get a couple of manufacturers and high prices. My ThinkPad T400 takes regular 2.5" drives and you can swap them out in a couple of seconds.
Spill Diet Coke on your MacBook? You're in for a long and costly repair. Spill Diet Coke on your ThinkPad? You'll likely only need to replace the keyboard, which costs under $50 and can be replaced in a couple of minutes.
My ThinkPad T400 was cooler, quieter, more ergonomic, easier to service, easier to upgrade, more durable, and cheaper than the MacBook Pro that was available when I bought it. Those are engineering criteria that I care about. Size, weight, and looks are important, but to say that Apple's engineering is "great" tells only half the story. Apple is great at compromising everything in the name of a few factors - like size and appearance. Fortunately for Apple, those are the factors that sell machines to individuals. But if you want to keep your machine for years and if you care about it still being useful, you might want to look at something that's more durable, more serviceable, and more upgradeable.
My T400 is a better machine than the MacBook Pro. Not just cheaper, but better engineered.
Don't teach them Linux unless that's what the class is about.
My 1st-year CS intro course (CS 1300) was taught in Python, but that's not particularly important because virtually any decent programming language is acceptable.
An intro CS course should be focused more on things like algorithms, data structures, time/space complexity, and other theory. It's not the place to teach system administration and it's not the place to teach software engineering.
Pragmatically speaking, people are going to have to compile and run code. So you might end up having to teach some basic Linux skills. At my university we ran RHEL 4.x on the lab machines at the time (now they're on Ubuntu), and there was a weekly lab section where the TAs could provide 1-on-1 instruction on how to get things working. A substantial fraction of the class was already familiar with Linux, but there were also a good number of students who weren't.
The point is that the mechanics (e.g. "type make") should be addressed during the labs if possible, or in the instructions for the assignments, or during office hours, or during an (optional) concurrent course. The intent should be to provide the necessary skills for students who have no Linux background without devoting much of your (very limited) lecture time to it. If you spend a week or two trying to teach Linux, you'll find that 1/3 of your students don't show up.
More than that, if you're supporting Windows 7 in a business context, why the hell are you buying crappy consumer-level PCs? Enterprises buy machines like HP EliteBooks, Lenovo ThinkPads, or Latitudes, all of which are better built and have longer support life-cycles than a consumer PC. And of course they have TPMs, with the exception of some lower-end ThinkPad models.
These machines aren't even particularly expensive. You can get a ThinkPad T410i for $750 right now.
This is absolutely correct. CS is not software engineering, it's not programming, and it's not computer engineering.
CS = Theory and research into computation. What is computable, on what sort of machine, in what time bounds. Research into new applications for computation (such as machine vision or natural language processing).
SWE = The engineering process of delivering software that is functional and reasonably defect-free on time and on budget.
CE = The design and engineering of computer systems. CPUs, GPUs, buses, storage systems, interconnects, etc.
Programming = The act of creating code, which (when done correctly) requires skills from CS, SWE, and CE.
The bottom line is that you can't be a good coder unless you have at least some of all three skills. Algorithms and time complexity matter. So does writing code that actually performs on real hardware. So does writing code that is maintainable and reasonably defect-free.
I am fortunate that my "CS" program was actually more of a CS+CE+SWE program. I am not an expert in any of those fields but I do know enough to work effectively on a team to solve problems and write good code.
Well you have some good points and some bad points.
First of all, you're absolutely right that safer vehicles breed less safe driving. It's a well-known, well-measured effect that needs to be considered every time a new safety device is considered.
That said, the idea that all accidents can be avoided with better driving is just crap. People screw up. I generally consider myself a pretty careful driver and I have had my share of mistakes. Anyone who is a safe driver should be able to identify many instances where they 'got away' with a lapse in concentration. Maybe you didn't see a car in your blind spot and almost hit them. Maybe you pushed that yellow light way further than you should have. Maybe you drove when you knew you were tired.
I have personally crashed a brand-new Prius into an SUV and done $20k worth of damage. Was I driving recklessly? No. Was I tired, on the phone, or drunk? No. But I was looking at the stopped car in the far lane instead of the moving car in the near lane, and by the time I pulled out to make the turn it was too late.
Chances are good that you'll screw up at some point.
Does this mean that we all need to drive tanks? No. But it does mean that you need to consider the risks of a supermini vehicle. A 'safe' supermini cannot protect you as well as a 'safe' larger vehicle. It doesn't need to be an SUV and it doesn't need to have crappy mileage.
The fact that people have anecdotes about how a Smart or another tiny car survived doesn't mean shit. In some kinds of crashes a supermini will do perfectly fine. In others it will be absolutely screwed.
The data, on the other hand, shows that heavier vehicles are safer. Yes, there are unsafe heavy vehicles, and yes, there is a spectrum of crash-worthiness among light vehicles from "total crap" to "decent". But in any given crash - be it single vehicle or multi-vehicle - the best smallest vehicles do not come close to the best heavier vehicles.
There is a risk trade-off here. The lowest risk option is not to drive at all. The highest risk option involves vehicles that provide little or no protection (like a motorcycle). And there is a spectrum of options in-between.
People need to be made aware of the negative consequences of their choices. In a world where gas is effectively free (relative to the cost of a vehicle) many people are going to choose inefficient vehicles. When the true cost of the fuel (including carbon emissions) is factored into the price, people will choose more efficient vehicles, drive less, or elect not to drive at all.
But you have to be careful not to throw stones from your glass house. There are many factors that have as much of an impact on your carbon emissions as the type of vehicle that you drive - like whether you eat meat and whether you use air-conditioning.
The most important thing is that you can't guess about how much energy something uses, because you're almost always wrong. Switching to washing your clothes in cold water (vs. hot) saves more energy than line-drying (vs. using a dryer). For me, in the summer months, using a fan to cool my apartment rather than the A/C saves more energy than not driving. Manufacturing and disposal are only a small part of the energy requirements in a car's life-cycle. Unplugging 'vampire' electronics is not going to save a whole ton of energy. Transit is not always a slam-dunk in terms of energy consumption, especially when it frequently runs at low utilization. Driving to the store to buy an item can often require more energy than having it shipped to you.
Believe it or not, you're wrong.
There are plenty of 10-inch Android tablets, almost all of which are cheaper than the iPad. The Archos 101, for example, is $299 for the 8GB version.
Whether it was Apple's doing or not, the rumors of the $999 iPad did a lot to make people think that the iPad was "cheap" at $499 and that somehow it was Apple's "sales volume" that allowed them to reach that "revolutionary price".
Of course all of that is BS. The 16GB iPad has the same ARM CPU and the same flash memory as the 16GB iPod Touch, which runs $275. And the 10" screen doesn't cost anywhere near $225, even if it is IPS. iSuppli estimates that the 16GB iPad costs $219 to make, which is not hard to believe considering that you can get a full netbook for around the same price.
It's not even the first one from this site:
A Real World HTML 5 Benchmark
That story references the exact same website.
The Prius ECU integrates with the climate control system, so if you turn on the heat the engine will run until there's enough coolant to give you heat.
Actually, if you put the heater on "auto" then the heat will stay off and the engine will run until the engine coolant gets up to a useful temperature (usually takes about 3 minutes), then the heater fan gradually spins up. When the engine coolant temperature finally gets hot enough (takes about 10 mins, or less if you're actively driving) then the engine will stop.
This is one of the advantages of building a car less like a series of pieces and more like an integrated system. The Prius is effectively a series of independent modules (ECUs) that work together through a common network (CAN bus, among others) for a single goal. The power steering can have different assist forces at different speeds because it gets speed data from the combi-display (speedometer). The engine can shut down when it's appropriate based on input from the hybrid system and other systems like the climate control computer. The hybrid system can avoid taking too much current to the battery when it's hot or cold because it communicates with the battery computer. The braking system can use less friction braking because it communicates with the hybrid system to provide regenerative braking (which, in turn, communicates with the battery and other systems).
The advantages to this integration are better efficiency, lower cost, reduced wiring complexity (since sensors and actuators are wired only to their nearby ECU, rather than all over the car), and other improvements. Something like cruise control is trivial to implement because you already have the data and control elements you need - it's just a matter of software.
This is why you absolutely, positively should not try to write code without an IDE - or at least an editor that offers a wide range of completion options (unfortunately I find that the people who use Emacs or Vim don't actually use the completion features they have - perhaps because they aren't as good as a good IDE like Eclipse).
Yeah, Eclipse takes 400MB of memory when I'm working on a medium-sized project. But it enables me to do what is right rather than having to do what is easy, because doing what's right is easy. I can use long identifier names when they're appropriate because they will be completed. I can use an interface if it's appropriate because Eclipse will auto-generate the method stubs anyway, so I don't have to type them twice. I can rename something if the name is inconsistent or confusing because it's easy to do. I can have my imports in the correct order and include only what's being used, because Eclipse does that for me. I can write Javadoc comments because Eclipse generates the right boilerplate.
You can do all of this without any code completion or refactoring tools. But it's more likely that you'll make a mistake at some point and it's more likely that you will get lazy and develop bad habits like using undescriptive variable names.
Right, I've used NX over a WAN link (albeit a very fast one) and it's able to handle non-intensive tasks like running Eclipse or general GNOME apps on a 2560x1600 display with reasonable performance. You can forget trying to do much of anything at that resolution with VNC over a WAN.
That said, NX totally dies if you try to do anything with animation or video. Protocols like PCoIP or HP RGS do a lot better here since they compress more.
Anyone who thinks CS is just about algorithms or computational theory doesn't have a broad base in CS. There is a huge amount of research in fields like computer vision, natural language processing (my focus), computer graphics, networking theory, and other topics that are unquestionably (in my opinion) CS but also not direct analogs of anything in mathematics.
It's true that CS is math-heavy, but so are many sciences. Theoretical Physics is absolute a science, but it could also be described as basically just the mathematical modeling of physical processes. Which is exactly what a lot of CS is.
I think the confusion comes from the fact that there are basically three different fields (CS, CE, and SWE) that are typically intermingled at the university level. Arguably, I think that's a good thing since the fields are interdependent enough that a student with a "CS" degree should have a background in all three.
Computer Science (CS) is basically the study of the theory and applications of computing. Designing an algorithm to sort or search or calculate faster fits here, but so does the study of the applications of algorithms to different tasks. Questions like "can we use a SVM classifier to identify the parts of speech in a sentence", or "which algorithms are best for converting a stereo image into a depth map", or "is there a better way to route in a wireless mesh network" fit here. CS is about expanding the range of problems that computing can solve and experimenting to find what works best in a particular situation. There is very little focus on the hardware as actually implemented (as opposed to a simplified theoretical model) or on the process of designing and building software. CS is not software engineering and it's not computer engineering.
Computer Engineering (CE, sometimes ECEN) is basically between electrical engineering and computer science, but it's really neither. CE is about engineering better computers, and it generally includes topics like microprocessor design, architectural design, low-level networking (e.g. QuickPath or HyperTransport), and compiler design. Some of these fields cross paths with CS (e.g. compiler design) and some cross paths with EE (e.g. IC design). But unlike a EE, a CE is not really concerned about the gate-level design or electrical properties of a CPU (they work with EEs on that) and unlike a computer scientist they actually care about the details of the hardware.
Software Engineering (SWE) is neither CS nor CE. It's also not programming. SWE is about the process of producing software: specification, design, testing, maintenance, and management. SWE has streaks of business management in it too: keeping projects on-time and on-budget is as much a management task as it is an engineering task. Software engineers exist to manage the challenge of putting together fantastically complex systems with very little time and very little money.
I can tell you what's not in any of these fields: programming. Yes, programming is an integral part of software engineering, computer engineering (usually an HDL) and nearly all computer science. Yeah, you can work out algorithms without ever writing a line of code or do an entire CPU design on paper, but in practice everyone wants to see their ideas actually implemented. But programming isn't what CS, SWE, or CE is about: programming is just the most common means used by those fields to express their ideas. You wouldn't say that literature study is about learning the English language, and by the same token CS/CE/SWE isn't about learning to program.
Most of this comes from the fact that programming is actually pretty easy. And, no, that's not me being arrogant: almost any first-year college student can be taught to write code in a semester. The majority of mathematicians, scientists, and engineers who graduate fr
I don't get the comments here.
Maybe you didn't like Universe. I certainly didn't like every episode. I'm not a fan of the soap-opeara dynamics either. But it seems like everyone stopped watching from the second episode and decided to hate it based on that. Universe has a solid story arc (even if it's a bit derivative), interesting characters, and good production values. That's more than you can say for just about any other sci-fi show that's on right now.
All these comments talking about the multitude of good sci-fi shows puzzle me. What the hell is on right now? SyFy is now loaded with ghost investigation shows and wrestling, BSG is over, Caprica got cancelled, Terminator got cancelled, Dollhouse got cancelled.
Maybe you didn't like Universe. But it's not going to be replaced with another sci-fi show when SyFy can get better ratings at a lower cost with wrestling.
I do that too, but sometimes it bites you in the ass - my credit card bank, for example, occasionally asks one of those questions in addition to the password.
The big lesson here is not that you should never get breached, or that you should use some super-secure password, or that you should use a different password on every site (you should).
No, the real lesson is that passwords themselves are faulty. No one is going to select and memorize a strong password for every website they use. They're going to either re-use passwords, or choose weak passwords, or write their passwords down (or use a password manager).
None of these are good answers. The expectation is that users are going to choose strong passwords, that they will never re-use passwords, that hashes (even with salt) are an effective way to protect passwords, and that users will never be tricked into revealing their password.
It's bullshit. It's always been bullshit. Users aren't careful with passwords, and why would we expect them to be - 99.9% of the time they get away with it. Humans are bad at evaluating the risk of things that are low frequency but high impact.
The other thing that's bullshit is password reset. It doesn't make any sense: how can someone who forgot their password remember "security questions" that are actually secure. No, 99 times out of 100 these systems use some crap like "Where were you born", which is pretty damn trivial to find out for any attacker. My brokerage account has a secure password that I only use there, but resetting the password requires only my username, SSN, ZIP code, and last name. And there are far, far more people who know that stuff than people who know my password.
It's time to get serious about replacing passwords. That's the lesson here.
After the 1984 incident, Amazon was sued by a customer and settled for $150,000. They also agreed not to remove books from customer's devices - not just in a wishy-washy statement but in their court settlement:
http://online.wsj.com/public/resources/documents/amazon20091001.pdf
If Amazon did this again, then they may be in for another lawsuit. I can believe that they removed the books from their service. But it doesn't make sense for them to pull the books from devices. Until we see more evidence than a couple of random unnamed sources in a blog post, I don't buy it.
That's just wrong. I'm in Boulder and I have a virtual server in Denver which is connected to Level3. My traceroutes go directly on Comcast's network to Level3 in Denver. Typical RTTs are in the 15ms range.
AMD was faster from when Athlon 64 was launched (December 2003) to when Core 2 was released (July 2006). But even today AMD is just about at parity with Core 2, and Intel is about to launch Sandy Bridge. Hopefully Bulldozer can change that.
Yes, there is. The only winning move is not to play.
I played WoW for over five years, including throughout my entire undergraduate education.
I have played the game for over 3500 hours, which is enough to get an air transport pilot license if I were flying a jet instead (to be fair, WoW is considerably cheaper).
I was on a 2450-rated 3v3 Arena team, which in our class/spec comp was #1 in the US and #2 in the world (although we played a weird comp).
I don't hate WoW. I would never begrudge someone for playing. But I do despise Greg Street "Ghostcrawler" ("GC") for what he did to the community and to the game. I despise the fact that 2v2s - which I loved playing - were effectively deprecated because GC said they couldn't be balanced. I despise the fact that damage was amped way up in WoLK, so much that people were dying in three hits at launch. It took almost a year for PvP to get back to a place where it was based on skill rather than getting lucky with crits. I despise the fact that mana was effectively removed from the game as a factor for healers, that threat was effectively removed as a factor for DPS, and that the release raid content was a too-easy warmed-over version of an instance from vanilla. And, most of all, I despise that GC has to constantly screw with things, as if a title that has stood as the #1 paid MMO for the past 5 years needs it.
Eventually I just got tired of it. I got tired of reading patch notes and checking the PTR to stay ahead of the curve. I got tired of adapting to each new flavor-of-the-month combo. Maybe if you never played PvP it was better, but in my world it was a constant cycle of getting screwed by some comp for 2 months, finally seeing it get fixed with a patch, and then having to deal with the next flavor comp.
So I'm not playing along anymore. I've found other games to play, like StarCraft II. There is a world of options out there and I eventually decided that I can live without WoW.
If you think that Microsoft somehow has the power to take down any Windows server at will, you're insane.
And if they somehow had some sort of super-secret back-door that no one has found and that they've managed to keep secret from most of their employees (who would undoubtedly leak it), they can't just go and use it against one of the most high-profile organizations in the world.
The cost does not come from defending the lawsuit or doing the studies. The cost comes because such lawsuits can create delays. Delays mean that an operator may need to buy power from someone else, build other types of plants (like natural gas) or delay the retirement of obsolete plants. All of these actions carry significant costs.
Safety does have a cost. That doesn't mean that it's not absolutely the right thing to do when you're dealing with a technology that's intrinsically hazardous.
Well, first of all, it sort of is. The typical way to get an address on an IPv6 network is stateless auto-configuration, which basically allows your client to combine an advertised route prefix with the EUI-64 (basically a longer version of a MAC address that can be generated from a MAC address) to determine its IP. You don't need any configuration for new clients and they always get the same IP address. Note that Windows Vista/7 use a hashing function with random data and the MAC address so that you can't track a single machine based on its IPv6 address, which solves privacy concerns.
Second, you can't just use the MAC address because it's not easy to route traffic that way. Routing works today because networks are assigned contiguous blocks of addresses, so it's easy to tell where to route traffic based on the address prefix. If we just had MAC addresses (which contain no information about which devices are connected to which networks), routing would require huge tables that would frequently change. This works OK for a small to medium sized network (e.g. switched Ethernet) but it doesn't work at all for the Internet. Even medium-large organizations need to use subnets to effectively manage traffic, which aren't possible without network prefixes.
Except that it isn't the case, as far as I can tell. It seems that Level3 is playing us all for suckers - their press release doesn't indicate that Comcast is specifically targeting Internet video, it only refers to "Internet online movies and other content", which basically means "packets".
Level3 is trying to spin this as a network neutrality debate, when it's pretty clearly not. This is a peering dispute.
Level3 DOES NOT have to interconnect with Comcast to access Comcast subscribers. Comcast is not Tier 1, which means that there are alternate indirect routes between Comcast and Level3. However, not interconnecting with Comcast would create huge traffic imbalances that could jeopardize Level3's other peering relationships.
Unfortunately, the keyboard drains do nothing when you spill it in the vent holes. I've never tested the keyboard drains but I've seen videos of them working. I would imagine that the keyboard is screwed, although it's pretty easy and cheap to fix.
I use the touchpad actually, not the trackstick. And I do like certain gestures, like two finger scrolling. I use a utility called EnvyTouchPad, which ironically was designed to work around the awful clickpad on the HP Envy series (which is far, far worse than the Mac clickpads).
I have three major problems with the Mac touchpads:
1. It's noisy - considerably more so than even the HP clickpads. Some PC laptops have this problem too, but the ThinkPad is actually pretty quiet. It doesn't seem like a huge issue but it is socially awkward when I'm in class, especially considering that my courses are recorded for remote students and the microphones pick up everything.
2. It makes dragging much harder. Attempting to drag with one finger is problematic because of friction and the deadzone at the top of the touchpad. Instead, you have to use two fingers, which is wierd and error prone. Trying to do something like the right-mouse-button drag (which never appears in OS X but does appear in Windows under Boot Camp) is futile.
3. It's more error prone. If you want to right click, you can either use multiple fingers or assign a touch zone. Neither is as consistent as hitting a different button. The touch zone is not demarcated on the pad and even if it were (as it is on HP clickpads) there is no tactile feel. Multiple fingers work great except sometimes you mess up and rest part of your hand on the pad, causing misclicks.
The bottom line is that I just don't know why this is a good design. The only advantage I can think of is that you get slightly more touch room, but I have never found my T400 touchpad to be too small. Gestures are nice but they do not replace the need for buttons in my opinion.
The T400 (as with most ThinkPads) doesn't have the power jack soldered to the motherboard - it's a separate part that's connected via a wire. The part runs about $12 on eBay and takes about 10 minutes to replace.
I have never actually yanked a laptop off of a table due to the power adapter. I'm not saying that it doesn't happen, but it just doesn't happen often enough to warrant an $80 MagSafe adapter. I have no issue with MagSafe in particular, what I have an issue with is that the adapters are so expensive and the MagSafe patent prevents anyone else from making compatible replacements.
Actually, the new Air (the one I had) doesn't have a backlit keyboard - it was one of the features that was cut, along with the sleep LED, the IR sensor, and the ambient light sensor. None of these things really bug me - I rarely use the Thinklight on my T400 anyway, since I know the keyboard layout by memory.
You just brought up another thing I hate about most Macs (and to be fair, most PCS) - the sleep light. The Air I had didn't have this issue, but a MacBook/MacBook Pro would - LED indicators should not pulsate or blink. Most of the time, I sleep in the same room as my laptop, which makes blinking (or pulsating) LEDs very annoying. I had to disconnect my (custom-built) desktop's power LED to fix this issue, but it's not quite as easy wit