Another example, the people telling us the world's weather is going to be a specific number of degrees hotter or colder in a decade. They can't even accuratly predict a week out.
Your ignorance of the scientific process doesn't invalidate it.
The example you mentioned is a straw man argument based on ignorance. There's a huge difference between climate and weather prediction. The former is like trying to predict the temperature at which a pot of water will boil, the latter is like predicting the location on the surface of the water where the steam bubbles will appear. One of those is predictable and depends on simple thermodynamics that is well understood, the other is full of randomness and uncertainty, and is difficult to predict far into the future.
Either way, none of that has anything to do with astronomy, which is about as precise, non-random, and rigorous as any science gets. The measurements these guys are making are just simple intensity measurements over time. There's no need to develop hugely complex models with trillions of unknowns and interacting nonlinear feedbacks and systems that require supercomputers to solve to a useful level of precision. The equations astronomers work with can be solved on the back of a napkin.
To give you an idea of the kind of precision that astronomers are used to working with, the Gaia Mission will create a star catalogue with position measurements as accurate as 20 microarcseconds. If you think of that as a fraction of a circle, that is 15 parts per trillion! The Kepler spacecraft has a rather pedestrian precision of only 20 parts per million, which is still orders of magnitude better than what any climatologist has ever had to work with.
On top of that, this mission is not making a prediction about the future, but making a straightforward measurement that can be trivially verified later. There's no uncertainty to speak of.
Technically, each conscious observer has a unique "observable universe" centred on their head.
Furthermore, since our brains have a finite size, each neuron has a slightly different observable universe centred around it, each with a slightly different horizon. In other words, a human observer can't even meaningfully speak about a single observable universe. What everyone observes is a superposition of a small 'volume' of parallel universes! 8)
You joke, but they've improved a lot over the last few years.
About the only bug that affects me is that if I alt-tab back to the desktop out of a full-screen game, then sometimes my performance plummets when I switch back.
It's a bug that's still present, despite years of complaints on forums, and has something to do with the power management switching the card to low-performance '2D' mode to reduce power usage, then failing to switch back to 3D mode.
If AMD got off their ass and fixed that bug, then it would be essentially bug free, at least from the perspective of an end-user.
Keep in mind that DirectX 11 is multi-threaded, which means that a single game can send graphics commands from multiple cores at once. Despite that substantial increase in complexity, Windows 7 graphics drivers are quite stable for most people. In comparison, OpenGL on Linux is still firmly rooted in the single-core era and crashes if you look at it wrong!
* Coal mining has directly killed 100,000 people [wikipedia.org] over the last century in the United States alone, and is killing thousands every year around the world, right now. It causes many more cases of lung disease in workers, at least tens of thousands a year worldwide.
That is ridiculously misleading. You make it sound like we're still loosing thousands of miners every year to mine collapses, like it's 1910.
Let me help you pull that head of yours out of the sand:
"However, in lesser developed countries and some developing countries, many miners continue to die annually, either through direct accidents in coal mines or through adverse health consequences from working under poor conditions. China, in particular, has the highest number of coal mining related deaths in the world, with an official statistic of 6,027 deaths in 2004."
Yeah, and nobody's pushing those CF bulbs (to the point of outlawing the cheaper alternative). With a plethora of dangerous chemicals and shorter lifespans and order of magnitude higher price than incandescents.
Cheaper only if you ignore the cost of electricity.
They're even more "cost effective" if you don't ignore the externalities, like damage to the environment.
And they last longer than incandescent bulbs too, usually 10x as long.
Here's a great quote: "A household that invested $90 in changing 30 fixtures to CFLs would save $440 to $1,500 over the five-year life of the bulbs, depending on your cost of electricity. Look at your utility bill and imagine a 12% discount to estimate the savings."
Nobody is working to run the price of oil up.
We've passed peak oil, so the price will go up in the long term, no matter what anybody does. Climate change or environmental science has nothing to do with it.
Moving away from oil is a good idea for that reason alone, even if you don't care about the future of the planet's environment.
Nobody forced NASA to use an inferior (but Greener, just ask 'em) material on the (6, count 'em, 6) SRB's of the Columbia leading to a failure that caused the loss of that shuttle and crew.
Huge pieces of foam falling off have been happening for a long time, it was a known engineering problem caused by a faulty design that compromised too much strength to reduce weight, and had nothing to do with 'saving the environment'.
Yeah, nobody's doing anything bad in the name of environmentalism right? "These aren't the droids we're looking for. Move on"....
If you think the crazy people trying to save the environment are putting people's "lives in danger", then lets put things into perspective, by comparing these "huge risks" you've listed compared to the way things are:
* Coal mining has directly killed 100,000 people over the last century in the United States alone, and is killing thousands every year around the world, right now. It causes many more cases of lung disease in workers, at least tens of thousands a year worldwide.
* The United States has gone to war in the Persian Gulf more than once to protect its interests there, which are mostly oil wells and pipelines. These conflicts have killed several hundred thousand locals, and over 10,000 american soldiers.
* Remember the oil spill in the gulf, and what that did to the environment? Exxon Valdez? Or these: List of Oil Spills
Compared to the thousands workers coughing their lungs up or dying kilometers underground, you'll excuse me if I ignore your cries of "oppression" by greenies because of the 1mg of mercury in a sealed glass tube that saves you money.
Do you think any network administrator will block the network port of an employee who's sick or on holidays? Or gone to a meeting for half a day?
You think they will talk to just any mac address?
Yes. Out of about a hundred customer sites I've been to, including many large government and financial institutions, I've only seen one that has implemented that kind of filtering.
By the way, the MAC address of a desktop PC is conveniently printed on the back. Doesn't exactly take cryptographic wizardry to figure out. Heck, even it wasn't, I'd just connect a cross-over cable to an unused PC, and wait for it to tell me what its MAC address is.
Repeat this ten times until it sinks in: "MAC address filtering is not security."
You think the IDS will not notice your ARP poisoning?
You mean the $100K piece of kit that few organisations can afford, and many who can, misconfigure or ignore? Most places I've seen that have IDS only have it for regulatory compliance reasons.
An IDS will do exactly zero to prevent passive attacks, like a hub stealthily inserted between a desktop PC and the floor switch.
Sure wired networks are a risk and there are ways around what I mentioned, but you are clearly talking about the follys of Windows Operators. Please do not call those folks System administrators.
Most large sites I've been to had Cisco Certified network engineers, and had purchased filthy expensive kit with more than enough technical capability to block any practical attack.
Also, most computers on wired networks are easily capable of end-to-end encryption, even with cheap unmanaged switching.
It's not that wired networks can't be secured, I'm saying that they usually aren't, due to ignorance, incompetence, or indifference.
Real enterprises treat it as a second class network, but all desktops are generally still on a wired network.
They also generally have you use an encrypted VPN even if you're on an internal WiFi.
The irony is that all but the most criminally negligent IT administrators would apply military-strength cryptography to their WiFi links, but allow data to traverse the wired connections in the clear, which means that the wireless link is substantially more secure!
One of the biggest vulnerabilities in any large office building is the wired network. It's trivial for an attacker dressed in a suit to simply walk in, sit down at an empty desk, plug in, and start doing packet captures. Switched networks provide minimal protection, thanks to DNS cache and ARP cache poisoning attacks and the like.
You'd be amazed at how ignorant typical IT administrators are of the risk. I've heard ridiculous things like:
"But you need to fill out a form to get network access!" - Only if I follow the rules. Nothing stops me from physically connecting.
"You need an AD account to connect to the network!" - They're thinking of network shares, but the exploitable vulnerabilities are at the IP network layer.
"Your computer is not a member of the domain, it can't connect!" - That's largely irrelevant, once you have a user account, practically everything is accessible even from a machine that's in an untrusted workgroup.
These aren't from rare isolated incidents either, I hear one of those three almost every time I sit down at a new customer as a consultant. System administrators live in a fantasy land of imagined security.
Can someone explain the disadvantages of SRBs? Is it just that they are more explosive?
They can't be turned off once ignited, can't be throttled, and they have high-pressure & high-temperature along the entire body of the booster instead of just in a relatively small "engine" at the bottom like liquid-fuelled rockets, which means they're a significant safety hazard if placed alongside liquid fuel tanks, like in most rocket designs.
What happened with the Challenger disaster is that a seal near the middle of one of the boosters failed, and the hot pressurised gasses escaped and cut into the main liquid tank like a welding torch. The same (or similar) risk will be present in the Ares V design.
Compare with the Saturn V, which had liquid-fuelled stages only, where a failure of a single engine could still result in a successful launch. This happened more than once during the Apollo missions, and no lives were lost.
Liquid fuelled rockets have their own issues too, like having to run turbo-pumps at enormous speeds and cryogenic temperatures. I found a scanned online version of the Saturn V Flight Manual recently. Here's a great quote:
The only substances used in the engine are the propellants and helium gas. The extremely low operating temperature of the engine prohibits the use of lubricants or other fluids.
I'm not sure if you've realized this, but even more obvious than that is that he's a breathless, hysterical idiot. Probably not much point in really carrying on too long a conversation with him other than to mock him.
I'm speaking from personal experience. Are you?
I have turned up at a customer site with NAP with a laptop running an "untrusted" OS, and was told it would simply not work.
I had to use one of their Windows desktops to work.*
If it wasn't for NAP, it would have connected to their network just fine, and I'm certain that it was far more secure and virus-free than their network.
This is reality now. If you want to keep your head in the sand, feel free.
*) Yes, I know, in principle one can simply turn off NAP on a switch port. In practice, that just won't happen in situations like this, particularly in big bureaucracies. The whole point of NAP is to prevent access, turning it off defeats the whole purpose, and most paper-pushers won't make an exception for some contractor or consultant. Also, yes, I know it's possible for "big corp" to add an open source kernel to their known trusted kernels list. Again, in practice, that doesn't happen, and definitely won't happen for some visitor with some random Linux or whatever on their laptop.
Exactly. If you're a corp who... y'know, actually has information you want to keep private (that you've won over hard-earned experience to achieve a competitive advantage), but maybe want to share with JV partners on a limited basis, the AD rights management allows you to achieve this.
That's the bait.
The hook is that after you've cryptographically ensured that it's physically impossible to extract your data out of the Microsoft-based DRM system, you've also dug yourself a hole down to vendor lock-in hell like you've never imagined.
A lot of this is about turning up the heat on the pot with a live lobster in it. Right now, it's merely pleasantly warm, but it's going to become uncomfortable soon!
It is about companies protected their trade secrets and confidential data. It isn't about stopping you from stealing something off of the piratebay.
Except that the data is not protected from employees, who can steal it all the same. Access control lists and transparent filesystem encryption already provide the necessary features for protecting data from employees. What it does do, is prevent open source applications from interfacing with the data in any way. It protects Microsoft's monopoly on your network, that is all.
It is about ensuring that the client is configured correctly and secure before connecting to a corporate network. I fail to see why this is a bad thing.
The phrase "configured correctly and secure before connecting to a corporate network" means exactly: "runs a trusted Windows kernel signed by Microsoft". That's not adding security in any shape, way, or form. It's not like insecure computers have an "evil bit" set on outgoing packets.There has never been a secure release of the Windows kernel, ever. There likely never will be. The machine is basically checked for a hash of the kernel. The checks can be made more complex, but it boils down to the same thing, to pass, the machine must be "one of a set of known and trusted versions of Windows".
Technologies like NAP are simply a method for locking out Linux, BSD, or any other OS that isn't made by a huge corporation. This may not sound bad if you're a "big corp" running "windows PCs", but it has a chilling effect. Developers or power users will no longer be able to run Linux, at all. Consultants and visitors will have to have a commercial OS, or they won't be able to get their job done, even if open source equivalents exist that would otherwise work just fine! Imagine a network with 100% Windows PCs and servers, with 100% enforced NAP. Some small vendor comes in, with a low-cost Linux offering... bzzt... can't play, sorry, try somewhere else.
And what is wrong with securing user data in a Medical Research Environment?
Step #1 along to path to requiring mandatory DRM on all medical data. Small pilot deployments are used as demonstrations to politicians. The vendor lock-in is not going to be obvious to anybody until it is far, far too late.
Sooner or later, if DRM hardware is 'everywhere', then a big corporation can simply make it mandatory for some file format or protocol... for... ahem... 'security'.
This will instantly lock out any possibility of an open source implementation of such a protocol, as most DRM schemes require code signed by a trusted central authority, which is a concept in diametric opposition to the 'open' part of the whole concept of open source.
Without open source, competition will be reduced, prices will go up, and your options as a customer will be restricted.
There are other abuses possible also, most of which you may never see coming until it is too late.
For example, if Microsoft can convince the idiots running most big bureaucracies that their network isn't safe from hackers unless there's an end-to-end DRM on everything, then this will effectively lock out their smaller competitiors from having any hope of even physically talking to any other machine on such a network. It probably won't do anything to increase safety from hackers, but it will certainly make Microsoft safe from their competition! This of course will increase costs for bureaucracies, which come out of your taxes.
You think I'm joking? Microsoft already tried this, it's called Active Directory Rights Management Services Role. Sounds innocent, right? It's horrifying! It's pure evil, the ultimate lock-in: using military grade cryptography to ensure that their customers stay locked in forever, and cannot possibly get their own data out of the walled garden of Microsoft software. They even tried to change low-level network protocols to prevent their competitors from competing on the 'corporate network' with their offerings by implementing open protocols: Network Access Protection. If you don't know what NAP is, it's a system that does nothing a firewall couldn't, except that to gain access, you must have a DRM-enabled computer running an OS kernel that's digitally signed by... a trusted authority.
Microsoft is pushing hard to have this technology become mandatory in some scenarios, like health data. Can you imagine if you couldn't obtain your own health records if you had one of those filthy 'untrusted' Linux computers? It's a very real possibility, and Microsoft wants it, bad.
To put it another way: This is not a feature Intel is including for free, out of the goodness of their hearts, just in case you want it. It's about increasing profits of the biggest corporations not just at your expense, but at the cost of your rights and freedoms. How does this not upset you?
The only people who hate DRM are people who want to be able to pirate. No one really cares about keeping those people happy, since they only way to do so is to give them free shit while getting nothing in return.
There's a lot more too it, which you simply do not get to see as a consumer, because it happens behind the scenes.
For example, the protected audio path introduced in Vista made many sound drivers much more complex to write, and resulted in poor performance and system stability problems. This was of course passed on to the customer as an additional expense, as well as yet another source of blue screens of death.
A much more chilling thought it that if DRM is available, it won't just be used for movies. Sooner or later, it will be mandatory for things other than entertainment. It's terrifying to think that a government could simply revoke the keys to, say, evidence stored on a citizen's personal computer because it's inconvenient for someone in a position of power. If you don't think that kind of abuse can happen, take a casual stroll through the material on Wikileaks!
Reminded my of stories where immortality is possible for future humans through high technology.
A real risk for such beings is the possibility of becoming marooned on a lifeless rocky planet, possibly forever, or close to it.
Could you get off such a planet, given essentially limitless time and no other resources other than yourself? (Assume that you cannot use your own body parts or fluids or whatever, due to the requirement of conserving the special high-tech / organic matter of your own body.) Could you ever build a spacecraft or even some sort of powerful beacon?
For these 'boot-strap' scenarios, we take a lot for granted. The Earth is extremely resource-rich, and not just in raw materials, but pre-processed ready-to-use items. Wood is actually a very complex fibre-reinforced composite with an excellent strength-to-weight ratio, and it also doubles as a convenient fuel. Animals have skins, bones, and tendons that can be made into clothes, tools, and fasteners. Even if one has no need to breathe, the atmosphere is still useful, because it is highly reactive. How far could we really get technologically starting from nothing without a reducing atmosphere? If there's no oxygen, there's no fire, and without that, most primitive technology is out the window. How would one refine ores into metals?
You've been listening to the announcements of politicians too much!
"Creating jobs" is often a synonym for "wasting money".
Jobs that do not produce anything useful are 100% pure waste. Does this produce anything useful? No. Hence, those jobs are wasting the time of people that could otherwise be gainfully employed.
At best, this could be construed as a publicly funded inefficient grant to the aerospace industry, but even then, a more productive use of that money would be to donate it directly to NASA earmarked for scientific exploration or something.
For example, it would be much more useful and interesting overall to fund a probe to go to Europa. The mission would generate new and interesting scientific results and NASA would have to develop new technologies like cryobots. Those have practical applications on Earth such as mineral exploration under ice sheets. There's a reasonable chance that life could be discovered, which would be the most amazing scientific discovery ever made by man. Worth a couple of billion? Probably. Sending a slab of metal to the moon? No.
I seem to recall a business plan back in the late 1990's to do something similar adjacent to the alaska pipeline; complete with a refinery.
...
I seem to recall they raised funds. Wonder what happened to them.
They probably 'discovered' well into the planning process that while it is possible to buy bandwidth, no amount of money can reduce network latency.
Delays would be ~10 ms minimum to California, and probably a lot higher due to switching delays and the lower speed of light in fibre optics. In practice, I'd expect 20ms or more, and 50ms or more to the east coast. That's too high for many applications.
Also, highly skilled technical workers aren't likely to accept a job opportunity in the middle of nowhere unless they're a paid a premium. In other words, staffing costs would be high, and some specialist roles may be impossible to fill.
Due to security concerns, many customers expect to have locked cages for their equipment and perform all installation and maintenance with their own staff. Can you imagine their costs if every time they had to replace a disk they'd have to fly an engineer out to the middle of the Alaskan tundra?
You're still using a mechanical hard-disk, right? That's the component that's bottle-necking your PC, not programmers!
Actually, if you're having performance issues on a modern computer that are solved by swapping out your system/application disk with an SSD chances are your real problem is that you're low on RAM.
I have 8 GB in my laptop, and 24 GB in my desktop. Both became massively faster with an SSD.
There's other IO activity that goes on that isn't effectively cached, even on a 64-bit operating system.
Cache does nothing for boot times, first-time application launch, opening a document for the first time, etc...
I'd have to wait an entire second to launch one of the most complex pieces of software on my computer.
I think you just refuted your own point. The most complex piece of software on your computer is a word processor. That's the problem. Things which are conceptually simple have become so monstrously bloated that they're now "complex software".
How is it "conceptually simple"?
Just because it's a common task that a lay person can understand doesn't make any of the required software complexity magically go away.
Consider just the capability of handling multiple languages. That is practically mandatory for 50% of the world population (maybe not you, but not the entire world is the United States). To handle Unicode alone, ignoring all the other legacy encodings, requires that the software do all of the following, in real-time:
- Handle 60,000+ code points - Load multiple fonts and map code points to glyphs. The mapping tables alone required several megabytes of memory. - Display text left-to-right and right-to-left, including mixed directions with justification, etc... - Load and handle the kerning information for all of those characters, and all their combinations. - Handle advanced typography features required by some languages like ligatures and context-sensitive glyph selection. - Handle input method editors to allow users to enter text where there are more characters than keys, like Chinese and Japanese. - Display all of that even with per-character styling, like colours and sizes.
Just think for a moment how insanely complex something as trivial as the "selection highlight" is when it has to handle anti-aliased formatted text that's a mix of left-to-right and right-to-left text using ligatures. Arabic text embedded in English text is a good example.
That's just text entry. Then there's spelling and grammar checking, localisation issues like date and number formatting, lexicographical sorting, and the list goes on.
A word processor is just about the most complex piece of software most normal users ever run. Practically everything else on an ordinary office PC is simpler.
As a comparison, most developer IDEs use fixed point text, and many do not support Unicode, none support justification, multiple columns, tables, or any advanced layout at all. You "think" of it as more complex because you use it for more complex tasks, but the software is much simpler internally. It's possible to write a simple compiler from scratch in a day, and I've written a simple real-time syntax highlighting code editor in under a week.
I bet you would struggle to write a text box control (let alone a word processor) with full internationalisation support, from scratch, in under a year. There's no practical way to do so with less than 5 MB of code & data, because the font and Unicode tables are that big alone. The input method editor lookup tables are also quite large, and the code behind them is very complex.
You are comparing software start-up time to code efficiency.
Efficient code would allow you to run the exact same software (from your point of view) on a 1GHz single-core CPU.
It's a valid comparison. The startup of Word is mostly single-threaded, and my CPU is 2.6 GHz, so on a 1 Ghz processor that measurement of 0.4 seconds would be... about 1 second. Oh no.. the horror! I'd have to wait an entire second to launch one of the most complex pieces of software on my computer.
Processors have been getting steadily faster, but most people's perception of their computer's speed has been completely dominated by the disk speed, so they haven't noticed.
When my customers complain about how "computers are so slow" I like to do a little demo for them using Mathematica, which is something I used to use at Uni, and still have installed in case I need it. I usually demonstrate it computing Pi to 1 million places in about a second, or symbolically integrating some complex equation instantly where the output is five screenfuls of hideously complex looking maths. Puts things in perspective for them.
I'll stop you right there. It has a mechanical drive, right?
Think about that for a second. You've got a solid state processor with hundreds of millions of parts switching billions of times per second waiting on... a single, moving, mechanical part that can't exceed about a hundred movements per second.
Your computer's mechanical hard disk has a latency about a million times higher than your computer's processor or memory!
Mechanical disk performance has not improved in over a decade! Ignore the benchmarks about streaming throughput, that's irrelevant to real-world performance. What hasn't changed is the latency and the random IOPS.
The SSD I have in my laptop now can do 60,000 random 4KB reads per second, with a latency under 100ms. If you think about it, that's still glacially slow compared to the processor. Ideally, the latency should be a few microseconds, which is what you'd expect for a 4KB transfer on a high-bandwidth link, so there's still a long way to go before disks have "caught up" to the rest of the PC.
You may as well be running software off floppies or tape, and whine about how it takes forever to spool back and forth when you use your software.
On my computer, practically nothing makes me wait, or if I have to wait, it's because real heavy-weight computation is taking place. Even then, if I switch to another program, the alt-tab transition is instant, and all the other software remains perfectly responsive while I wait.
Instead, because of bloat, we're stuck running yesterday's software with tomorrow's hardware.
Bullshit!
You're still using a mechanical hard-disk, right? That's the component that's bottle-necking your PC, not programmers!
Get an SSD to find out what your CPU and your "bloated" software is capable of.
On my old laptop, with an old 2.6 GHz Core 2 Duo CPU, which is several generations older than the CPU in the article, if I double-click the Word 2010 icon, it launches instantly. Not after a second or two, instantly. It's like notepad. I get the pretty transparent window borders, the ribbon, font-smoothing, everything. How is that "bloated"?
I checked it too: Starting up Word 2010 takes ~400ms of processor time, total. It doesn't even show up in the Task Manager, I had to use SysInternals Process Explorer!
Acrobat Reader, the "horribly bloated" application that most people hate for being slow? It also launches instantly, and uses a massive 180ms of processor time. It's eating up my CPU! Oh wait... it's not.
It doesn't work all that well on low-end hardware or virtual machines
It's been demonstrated to match XP performance on even quite low-end gear by several third-party tests. My experience is that's it's faster, particularly the 64-bit builds, which increase the file cache size from ~400MB max to "all of physical memory", which is a big improvement.
Every time you deploy an image you have to manually re-register the thing with Microsoft so it doesn't disable itself
You're Doing It Wrong. If you're supporting Windows 7 for businesses, you should be using KMS or MAK, and using the volume licensed Enterprise editions, not Windows 7 Home or whatever.
Still no decent backup system
It's the best ever - it has both file-level and image-based backups, it can take live snapshots of disks for both types, back up open files, it has a built-in scheduler, and a bunch of other features.
The VHD disk images created by Windows 7 can be mounted as virtual disks using a GUI or the command-line, can be used to boot from directly without having to be restored first, can be trivially converted into a virtual machine disk, and the install CD has a built-in restore wizard.
I haven't seen comparable features in any other operating system except OSX.
More importantly, if you're backing up desktops, You're Doing It Wrong. Laptops should use offline folders to sync with the master copy of the user data on a server, and shouldn't need backing up. Desktops should use folder redirection and/or roaming profiles. Back up your servers, not your desktops.
You can even do it the "Linux way" if you want to: I've seen sample scripts floating about that take a VSS snapshot of a disk, mount it as a folder or drive letter, and use rsync to incrementally update a backup, then release the snapshot automatically. I've done this myself for Windows Server 2003, about 6 years ago, it's nothing new.
XP Mode is buggy and compatibility in general is bad (especially in the 64-bit versions)
You shouldn't even need XP-mode most of the time, particularly on 32-bit editions of Windows 7. I've found that even the 64-bit editions will run just about anything if you simply set the "compatibility flags" on the main program executables. Just how bad are these applications that you have to support? Shouldn't you be blaming the app vendors instead of Microsoft?
Still no EXT3/EXT4 (or any Unix-type), Large FAT or GPT support
Are you kidding me? First, Windows has had GPT disk and boot support since Windows Server 2003 Service Pack 1, it has xFAT, NTFS on removable drives, and there's third-party EXT3 plugins.
If you think EXT3 on Windows is an important feature, again, You're Doing It Wrong. NTFS is a superior filesystem for Windows in practically every way. If you want to share data between Windows and Linux, use NTFS drivers on Linux, or a server with SAMBA.
Limit of 2 physical processors? Really? It's easy to get 4 processors in a box these days with 8 cores each especially in the academic world
That sucks, but 2 sockets is 12-16 cores these days. If you need more computing power than that, than you can afford a Windows Server 2008 R2 license, which gives you almost all the Windows 7 features, and more processor socket licenses. It's a commercial operating system, and it costs money.
Full Disk Encryption requires TPM chips which are missing in just about any system these days so you still have to go into a 3rd party solution.
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.
...to bring that experiment up to the level of Mr. Young's work, you would have found that F=ma broke down at some scales, and discovered quantum theory and relativity. Of course, that rigor took physics over 300 years of further effort!
But my point is that it does not break down at ordinary length scales, velocities, and energies. Given the energy levels and testing methods available to him at the time, Newton was essentially 100% correct. He is still, within the scope that his experiments cover, 100% correct today. His results are still taught, and will be taught forever, because they will never become invalid unless the laws of the physical universe change.
A lot of the research that is being done now in the 'softer' sciences is just flat out wrong, and new results totally invalidate previous findings.
Do you see the difference?
It's not about sample size, or the target of the experiment, it's the method and the lack of rigour. For example, Wikipedia has a great example about the Hawthorne Effect. It was discovered when doing 'experiments' on the working conditions of factory workers that the subjects reacted to being 'experimented on' more strongly than the result of the experiment itself. E.g.: varying the intensity level of illumination had a temporary boost on productivity, no matter what that change was.
That result should have never been an issue. The whole experiment is just shockingly bad. Mistake 1: The employees were told there was an experiment going on. Mistake 2: There was no control group. Mistake 3: Only one factory was used in the experiment.
I'm sure they made other errors too, but you get the idea.
This was a mere 60 years ago! Newton had been doing proper, rigorous physics for 325 years! The sad thing is that this lax approach to science is still going on today, and bad results are still published.
The problem with Psychology is not the subject matter, but the researchers and their sloppy techniques. There's a great quote by Richard Feynman in his essay about Cargo Cult Science:
[There] have been many experiments running rats through all kinds of mazes, and so on—with little clear result. But in 1937 a man named Young did a very interesting one. He had a long corridor with doors all along one side where the rats came in, and doors along the other side where the food was. He wanted to see if he could train the rats to go in at the third door down from wherever he started them off. No. The rats went immediately to the door where the food had been the time before.
The question was, how did the rats know, because the corridor was so beautifully built and so uniform, that this was the same door as before? Obviously there was something about the door that was different from the other doors. So he painted the doors very carefully, arranging the textures on the faces of the doors exactly the same. Still the rats could tell. Then he thought maybe the rats were smelling the food, so he used chemicals to change the smell after each run. Still the rats could tell. Then he realized the rats might be able to tell by seeing the lights and the arrangement in the laboratory like any commonsense person. So he covered the corridor, and still the rats could tell.
He finally found that they could tell by the way the floor sounded when they ran over it. And he could only fix that by putting his corridor in sand. So he covered one after another of all possible clues and finally was able to fool the rats so that they had to learn to go in the third door. If he relaxed any of his conditions, the rats could tell.
Now, from a scientific standpoint, that is an A-number-one experiment. That is the experiment that makes rat-running experiments sensible, because it uncovers the clues that the rat is really using—not what you think it’s using. And that is the experiment that tells exactly what conditions you have to use in order to be careful and control everything in an experiment with rat-running.
I looked into the subsequent history of this research. The next experiment, and the one after that, never referred to Mr. Young. They never used any of his criteria of putting the corridor on sand, or being very careful. They just went right on running rats in the same old way, and paid no attention to the great discoveries of Mr. Young, and his papers are not referred to, because he didn’t discover anything about the rats. In fact, he discovered all the things you have to do to discover something about rats. But not paying attention to experiments like that is a characteristic of cargo cult science.
(emphasis mine)
Most psychologists and social scientists are lazy, and simply don't do real science. Without reproducibility, there's no real result. It's worse than nothing, it's actively counter-productive, because introducing falsehoods into the set of knowledge on which other future science is based upon undermines those results also.
Newton's experiments are still reproducible today, to many digits of precision, after more than three centuries! We've built on them, sure, and come up with newer laws that cover a wider range of circumstances, but it wasn't like.. oh no.. due to only doing the experiment once due to lack of funding or some other excuse, it's not really "F=ma" but "F=ma^2"! Can you imagine if physics, chemistry, or engineering were done this way? H2O? Did I say that? I meant H3O! Or was it H5O2? I forget... never mind.
While I can (and have) reproduced Newton's results in a lab, psychologists still quote Freud and teach his lunacy at a tertiary level. The guy was a nutcase, and never did any real science, yet it's still part of standard introductory psych textbooks.
well, the modern life is about 10^6 times more pleasant than the hunter gatherer existence, so i will disagree with you there.
It's not more pleasant than the easiest hunter gatherer existence.
True, but that's only possible in a few places, like tropical island paradises. That lifestyle is no longer possible for the vast majority of the population. So the rest of us work, and live a life that is almost as comfortable. I have basically a zero risk of starvation, far lower risk of disease or illness than any hunter-gatherer, and I have an easy job that takes up only about 1/4 of my waking hours.
There's also strong evidence that the 'paradise' islands were regularly raided by the neighboring warrior tribes from the more ordinary islands to steal the food and rape the women.
Historically, humans lived a far more violent life than modern people. See this TED video: A brief history of violence.
Slashdot Mirror
Another example, the people telling us the world's weather is going to be a specific number of degrees hotter or colder in a decade. They can't even accuratly predict a week out.
Your ignorance of the scientific process doesn't invalidate it.
The example you mentioned is a straw man argument based on ignorance. There's a huge difference between climate and weather prediction. The former is like trying to predict the temperature at which a pot of water will boil, the latter is like predicting the location on the surface of the water where the steam bubbles will appear. One of those is predictable and depends on simple thermodynamics that is well understood, the other is full of randomness and uncertainty, and is difficult to predict far into the future.
Either way, none of that has anything to do with astronomy, which is about as precise, non-random, and rigorous as any science gets. The measurements these guys are making are just simple intensity measurements over time. There's no need to develop hugely complex models with trillions of unknowns and interacting nonlinear feedbacks and systems that require supercomputers to solve to a useful level of precision. The equations astronomers work with can be solved on the back of a napkin.
To give you an idea of the kind of precision that astronomers are used to working with, the Gaia Mission will create a star catalogue with position measurements as accurate as 20 microarcseconds. If you think of that as a fraction of a circle, that is 15 parts per trillion! The Kepler spacecraft has a rather pedestrian precision of only 20 parts per million, which is still orders of magnitude better than what any climatologist has ever had to work with.
On top of that, this mission is not making a prediction about the future, but making a straightforward measurement that can be trivially verified later. There's no uncertainty to speak of.
Technically, each conscious observer has a unique "observable universe" centred on their head.
Furthermore, since our brains have a finite size, each neuron has a slightly different observable universe centred around it, each with a slightly different horizon. In other words, a human observer can't even meaningfully speak about a single observable universe. What everyone observes is a superposition of a small 'volume' of parallel universes! 8)
AMD's closed source drivers on windows suck too.
You joke, but they've improved a lot over the last few years.
About the only bug that affects me is that if I alt-tab back to the desktop out of a full-screen game, then sometimes my performance plummets when I switch back.
It's a bug that's still present, despite years of complaints on forums, and has something to do with the power management switching the card to low-performance '2D' mode to reduce power usage, then failing to switch back to 3D mode.
If AMD got off their ass and fixed that bug, then it would be essentially bug free, at least from the perspective of an end-user.
Keep in mind that DirectX 11 is multi-threaded, which means that a single game can send graphics commands from multiple cores at once. Despite that substantial increase in complexity, Windows 7 graphics drivers are quite stable for most people. In comparison, OpenGL on Linux is still firmly rooted in the single-core era and crashes if you look at it wrong!
* Coal mining has directly killed 100,000 people [wikipedia.org] over the last century in the United States alone, and is killing thousands every year around the world, right now. It causes many more cases of lung disease in workers, at least tens of thousands a year worldwide.
That is ridiculously misleading. You make it sound like we're still loosing thousands of miners every year to mine collapses, like it's 1910.
Let me help you pull that head of yours out of the sand:
"However, in lesser developed countries and some developing countries, many miners continue to die annually, either through direct accidents in coal mines or through adverse health consequences from working under poor conditions. China, in particular, has the highest number of coal mining related deaths in the world, with an official statistic of 6,027 deaths in 2004."
That's from Coal Mining.
On TV, you only hear about a) coal mine collapses, when they b) involve lots of people at once, and c) when they happen in developed countries.
Even if you consider first world countries only, the total is hundreds per year.
Yeah, and nobody's pushing those CF bulbs (to the point of outlawing the cheaper alternative). With a plethora of dangerous chemicals and shorter lifespans and order of magnitude higher price than incandescents.
Cheaper only if you ignore the cost of electricity.
They're even more "cost effective" if you don't ignore the externalities, like damage to the environment.
And they last longer than incandescent bulbs too, usually 10x as long.
Here's a great quote: "A household that invested $90 in changing 30 fixtures to CFLs would save $440 to $1,500 over the five-year life of the bulbs, depending on your cost of electricity. Look at your utility bill and imagine a 12% discount to estimate the savings."
Nobody is working to run the price of oil up.
We've passed peak oil, so the price will go up in the long term, no matter what anybody does. Climate change or environmental science has nothing to do with it.
Moving away from oil is a good idea for that reason alone, even if you don't care about the future of the planet's environment.
Nobody forced NASA to use an inferior (but Greener, just ask 'em) material on the (6, count 'em, 6) SRB's of the Columbia leading to a failure that caused the loss of that shuttle and crew.
Huge pieces of foam falling off have been happening for a long time, it was a known engineering problem caused by a faulty design that compromised too much strength to reduce weight, and had nothing to do with 'saving the environment'.
Yeah, nobody's doing anything bad in the name of environmentalism right? "These aren't the droids we're looking for. Move on"....
If you think the crazy people trying to save the environment are putting people's "lives in danger", then lets put things into perspective, by comparing these "huge risks" you've listed compared to the way things are:
* Coal mining has directly killed 100,000 people over the last century in the United States alone, and is killing thousands every year around the world, right now. It causes many more cases of lung disease in workers, at least tens of thousands a year worldwide.
* The United States has gone to war in the Persian Gulf more than once to protect its interests there, which are mostly oil wells and pipelines. These conflicts have killed several hundred thousand locals, and over 10,000 american soldiers.
* Remember the oil spill in the gulf, and what that did to the environment? Exxon Valdez? Or these: List of Oil Spills
Compared to the thousands workers coughing their lungs up or dying kilometers underground, you'll excuse me if I ignore your cries of "oppression" by greenies because of the 1mg of mercury in a sealed glass tube that saves you money.
You think all the switch ports are on?
Do you think any network administrator will block the network port of an employee who's sick or on holidays? Or gone to a meeting for half a day?
You think they will talk to just any mac address?
Yes. Out of about a hundred customer sites I've been to, including many large government and financial institutions, I've only seen one that has implemented that kind of filtering.
By the way, the MAC address of a desktop PC is conveniently printed on the back. Doesn't exactly take cryptographic wizardry to figure out. Heck, even it wasn't, I'd just connect a cross-over cable to an unused PC, and wait for it to tell me what its MAC address is.
Repeat this ten times until it sinks in: "MAC address filtering is not security."
You think the IDS will not notice your ARP poisoning?
You mean the $100K piece of kit that few organisations can afford, and many who can, misconfigure or ignore? Most places I've seen that have IDS only have it for regulatory compliance reasons.
An IDS will do exactly zero to prevent passive attacks, like a hub stealthily inserted between a desktop PC and the floor switch.
Sure wired networks are a risk and there are ways around what I mentioned, but you are clearly talking about the follys of Windows Operators. Please do not call those folks System administrators.
That's called the No True Scotsman Fallacy.
Most large sites I've been to had Cisco Certified network engineers, and had purchased filthy expensive kit with more than enough technical capability to block any practical attack.
Also, most computers on wired networks are easily capable of end-to-end encryption, even with cheap unmanaged switching.
It's not that wired networks can't be secured, I'm saying that they usually aren't, due to ignorance, incompetence, or indifference.
So "Real enterprises" never use WiFi?
Real enterprises treat it as a second class network, but all desktops are generally still on a wired network.
They also generally have you use an encrypted VPN even if you're on an internal WiFi.
The irony is that all but the most criminally negligent IT administrators would apply military-strength cryptography to their WiFi links, but allow data to traverse the wired connections in the clear, which means that the wireless link is substantially more secure!
One of the biggest vulnerabilities in any large office building is the wired network. It's trivial for an attacker dressed in a suit to simply walk in, sit down at an empty desk, plug in, and start doing packet captures. Switched networks provide minimal protection, thanks to DNS cache and ARP cache poisoning attacks and the like.
You'd be amazed at how ignorant typical IT administrators are of the risk. I've heard ridiculous things like:
"But you need to fill out a form to get network access!"
- Only if I follow the rules. Nothing stops me from physically connecting.
"You need an AD account to connect to the network!"
- They're thinking of network shares, but the exploitable vulnerabilities are at the IP network layer.
"Your computer is not a member of the domain, it can't connect!"
- That's largely irrelevant, once you have a user account, practically everything is accessible even from a machine that's in an untrusted workgroup.
These aren't from rare isolated incidents either, I hear one of those three almost every time I sit down at a new customer as a consultant. System administrators live in a fantasy land of imagined security.
Can someone explain the disadvantages of SRBs? Is it just that they are more explosive?
They can't be turned off once ignited, can't be throttled, and they have high-pressure & high-temperature along the entire body of the booster instead of just in a relatively small "engine" at the bottom like liquid-fuelled rockets, which means they're a significant safety hazard if placed alongside liquid fuel tanks, like in most rocket designs.
What happened with the Challenger disaster is that a seal near the middle of one of the boosters failed, and the hot pressurised gasses escaped and cut into the main liquid tank like a welding torch. The same (or similar) risk will be present in the Ares V design.
Compare with the Saturn V, which had liquid-fuelled stages only, where a failure of a single engine could still result in a successful launch. This happened more than once during the Apollo missions, and no lives were lost.
Liquid fuelled rockets have their own issues too, like having to run turbo-pumps at enormous speeds and cryogenic temperatures. I found a scanned online version of the Saturn V Flight Manual recently. Here's a great quote:
The only substances used in the engine are the propellants and helium gas. The extremely low operating temperature of the engine prohibits the use of lubricants or other fluids.
Just... wow.
I'm not sure if you've realized this, but even more obvious than that is that he's a breathless, hysterical idiot. Probably not much point in really carrying on too long a conversation with him other than to mock him.
I'm speaking from personal experience. Are you?
I have turned up at a customer site with NAP with a laptop running an "untrusted" OS, and was told it would simply not work.
I had to use one of their Windows desktops to work.*
If it wasn't for NAP, it would have connected to their network just fine, and I'm certain that it was far more secure and virus-free than their network.
This is reality now. If you want to keep your head in the sand, feel free.
*) Yes, I know, in principle one can simply turn off NAP on a switch port. In practice, that just won't happen in situations like this, particularly in big bureaucracies. The whole point of NAP is to prevent access, turning it off defeats the whole purpose, and most paper-pushers won't make an exception for some contractor or consultant. Also, yes, I know it's possible for "big corp" to add an open source kernel to their known trusted kernels list. Again, in practice, that doesn't happen, and definitely won't happen for some visitor with some random Linux or whatever on their laptop.
Exactly. If you're a corp who... y'know, actually has information you want to keep private (that you've won over hard-earned experience to achieve a competitive advantage), but maybe want to share with JV partners on a limited basis, the AD rights management allows you to achieve this.
That's the bait.
The hook is that after you've cryptographically ensured that it's physically impossible to extract your data out of the Microsoft-based DRM system, you've also dug yourself a hole down to vendor lock-in hell like you've never imagined.
A lot of this is about turning up the heat on the pot with a live lobster in it. Right now, it's merely pleasantly warm, but it's going to become uncomfortable soon!
You have no idea what AD Rights management is for. See http://en.wikipedia.org/wiki/Rights_Management_Services
It is about companies protected their trade secrets and confidential data. It isn't about stopping you from stealing something off of the piratebay.
Except that the data is not protected from employees, who can steal it all the same. Access control lists and transparent filesystem encryption already provide the necessary features for protecting data from employees. What it does do, is prevent open source applications from interfacing with the data in any way. It protects Microsoft's monopoly on your network, that is all.
What NAP really is: http://en.wikipedia.org/wiki/Network_Access_Protection
It is about ensuring that the client is configured correctly and secure before connecting to a corporate network. I fail to see why this is a bad thing.
The phrase "configured correctly and secure before connecting to a corporate network" means exactly: "runs a trusted Windows kernel signed by Microsoft". That's not adding security in any shape, way, or form. It's not like insecure computers have an "evil bit" set on outgoing packets.There has never been a secure release of the Windows kernel, ever. There likely never will be. The machine is basically checked for a hash of the kernel. The checks can be made more complex, but it boils down to the same thing, to pass, the machine must be "one of a set of known and trusted versions of Windows".
Technologies like NAP are simply a method for locking out Linux, BSD, or any other OS that isn't made by a huge corporation. This may not sound bad if you're a "big corp" running "windows PCs", but it has a chilling effect. Developers or power users will no longer be able to run Linux, at all. Consultants and visitors will have to have a commercial OS, or they won't be able to get their job done, even if open source equivalents exist that would otherwise work just fine! Imagine a network with 100% Windows PCs and servers, with 100% enforced NAP. Some small vendor comes in, with a low-cost Linux offering... bzzt... can't play, sorry, try somewhere else.
And what is wrong with securing user data in a Medical Research Environment?
Step #1 along to path to requiring mandatory DRM on all medical data. Small pilot deployments are used as demonstrations to politicians. The vendor lock-in is not going to be obvious to anybody until it is far, far too late.
It doesn't work like that.
Sooner or later, if DRM hardware is 'everywhere', then a big corporation can simply make it mandatory for some file format or protocol... for... ahem... 'security'.
This will instantly lock out any possibility of an open source implementation of such a protocol, as most DRM schemes require code signed by a trusted central authority, which is a concept in diametric opposition to the 'open' part of the whole concept of open source.
Without open source, competition will be reduced, prices will go up, and your options as a customer will be restricted.
There are other abuses possible also, most of which you may never see coming until it is too late.
For example, if Microsoft can convince the idiots running most big bureaucracies that their network isn't safe from hackers unless there's an end-to-end DRM on everything, then this will effectively lock out their smaller competitiors from having any hope of even physically talking to any other machine on such a network. It probably won't do anything to increase safety from hackers, but it will certainly make Microsoft safe from their competition! This of course will increase costs for bureaucracies, which come out of your taxes.
You think I'm joking? Microsoft already tried this, it's called Active Directory Rights Management Services Role. Sounds innocent, right? It's horrifying! It's pure evil, the ultimate lock-in: using military grade cryptography to ensure that their customers stay locked in forever, and cannot possibly get their own data out of the walled garden of Microsoft software. They even tried to change low-level network protocols to prevent their competitors from competing on the 'corporate network' with their offerings by implementing open protocols: Network Access Protection. If you don't know what NAP is, it's a system that does nothing a firewall couldn't, except that to gain access, you must have a DRM-enabled computer running an OS kernel that's digitally signed by... a trusted authority.
Microsoft is pushing hard to have this technology become mandatory in some scenarios, like health data. Can you imagine if you couldn't obtain your own health records if you had one of those filthy 'untrusted' Linux computers? It's a very real possibility, and Microsoft wants it, bad.
I'm not making this up, check it out: Using Digital Rights Management for Securing Data in a Medical Research Environment.
To put it another way: This is not a feature Intel is including for free, out of the goodness of their hearts, just in case you want it. It's about increasing profits of the biggest corporations not just at your expense, but at the cost of your rights and freedoms. How does this not upset you?
The only people who hate DRM are people who want to be able to pirate. No one really cares about keeping those people happy, since they only way to do so is to give them free shit while getting nothing in return.
There's a lot more too it, which you simply do not get to see as a consumer, because it happens behind the scenes.
For example, the protected audio path introduced in Vista made many sound drivers much more complex to write, and resulted in poor performance and system stability problems. This was of course passed on to the customer as an additional expense, as well as yet another source of blue screens of death.
A much more chilling thought it that if DRM is available, it won't just be used for movies. Sooner or later, it will be mandatory for things other than entertainment. It's terrifying to think that a government could simply revoke the keys to, say, evidence stored on a citizen's personal computer because it's inconvenient for someone in a position of power. If you don't think that kind of abuse can happen, take a casual stroll through the material on Wikileaks!
Reminded my of stories where immortality is possible for future humans through high technology.
A real risk for such beings is the possibility of becoming marooned on a lifeless rocky planet, possibly forever, or close to it.
Could you get off such a planet, given essentially limitless time and no other resources other than yourself? (Assume that you cannot use your own body parts or fluids or whatever, due to the requirement of conserving the special high-tech / organic matter of your own body.) Could you ever build a spacecraft or even some sort of powerful beacon?
For these 'boot-strap' scenarios, we take a lot for granted. The Earth is extremely resource-rich, and not just in raw materials, but pre-processed ready-to-use items. Wood is actually a very complex fibre-reinforced composite with an excellent strength-to-weight ratio, and it also doubles as a convenient fuel. Animals have skins, bones, and tendons that can be made into clothes, tools, and fasteners. Even if one has no need to breathe, the atmosphere is still useful, because it is highly reactive. How far could we really get technologically starting from nothing without a reducing atmosphere? If there's no oxygen, there's no fire, and without that, most primitive technology is out the window. How would one refine ores into metals?
...could provide a fair number of jobs...
You've been listening to the announcements of politicians too much!
"Creating jobs" is often a synonym for "wasting money".
Jobs that do not produce anything useful are 100% pure waste. Does this produce anything useful? No. Hence, those jobs are wasting the time of people that could otherwise be gainfully employed.
At best, this could be construed as a publicly funded inefficient grant to the aerospace industry, but even then, a more productive use of that money would be to donate it directly to NASA earmarked for scientific exploration or something.
For example, it would be much more useful and interesting overall to fund a probe to go to Europa. The mission would generate new and interesting scientific results and NASA would have to develop new technologies like cryobots. Those have practical applications on Earth such as mineral exploration under ice sheets. There's a reasonable chance that life could be discovered, which would be the most amazing scientific discovery ever made by man. Worth a couple of billion? Probably. Sending a slab of metal to the moon? No.
I seem to recall a business plan back in the late 1990's to do something similar adjacent to the alaska pipeline; complete with a refinery.
...
I seem to recall they raised funds. Wonder what happened to them.
They probably 'discovered' well into the planning process that while it is possible to buy bandwidth, no amount of money can reduce network latency.
Delays would be ~10 ms minimum to California, and probably a lot higher due to switching delays and the lower speed of light in fibre optics. In practice, I'd expect 20ms or more, and 50ms or more to the east coast. That's too high for many applications.
Also, highly skilled technical workers aren't likely to accept a job opportunity in the middle of nowhere unless they're a paid a premium. In other words, staffing costs would be high, and some specialist roles may be impossible to fill.
Due to security concerns, many customers expect to have locked cages for their equipment and perform all installation and maintenance with their own staff. Can you imagine their costs if every time they had to replace a disk they'd have to fly an engineer out to the middle of the Alaskan tundra?
You're still using a mechanical hard-disk, right? That's the component that's bottle-necking your PC, not programmers!
Actually, if you're having performance issues on a modern computer that are solved by swapping out your system/application disk with an SSD chances are your real problem is that you're low on RAM.
I have 8 GB in my laptop, and 24 GB in my desktop. Both became massively faster with an SSD.
There's other IO activity that goes on that isn't effectively cached, even on a 64-bit operating system.
Cache does nothing for boot times, first-time application launch, opening a document for the first time, etc...
I think you just refuted your own point. The most complex piece of software on your computer is a word processor. That's the problem. Things which are conceptually simple have become so monstrously bloated that they're now "complex software".
How is it "conceptually simple"?
Just because it's a common task that a lay person can understand doesn't make any of the required software complexity magically go away.
Consider just the capability of handling multiple languages. That is practically mandatory for 50% of the world population (maybe not you, but not the entire world is the United States). To handle Unicode alone, ignoring all the other legacy encodings, requires that the software do all of the following, in real-time:
- Handle 60,000+ code points
- Load multiple fonts and map code points to glyphs. The mapping tables alone required several megabytes of memory.
- Display text left-to-right and right-to-left, including mixed directions with justification, etc...
- Load and handle the kerning information for all of those characters, and all their combinations.
- Handle advanced typography features required by some languages like ligatures and context-sensitive glyph selection.
- Handle input method editors to allow users to enter text where there are more characters than keys, like Chinese and Japanese.
- Display all of that even with per-character styling, like colours and sizes.
Just think for a moment how insanely complex something as trivial as the "selection highlight" is when it has to handle anti-aliased formatted text that's a mix of left-to-right and right-to-left text using ligatures. Arabic text embedded in English text is a good example.
That's just text entry. Then there's spelling and grammar checking, localisation issues like date and number formatting, lexicographical sorting, and the list goes on.
A word processor is just about the most complex piece of software most normal users ever run. Practically everything else on an ordinary office PC is simpler.
As a comparison, most developer IDEs use fixed point text, and many do not support Unicode, none support justification, multiple columns, tables, or any advanced layout at all. You "think" of it as more complex because you use it for more complex tasks, but the software is much simpler internally. It's possible to write a simple compiler from scratch in a day, and I've written a simple real-time syntax highlighting code editor in under a week.
I bet you would struggle to write a text box control (let alone a word processor) with full internationalisation support, from scratch, in under a year. There's no practical way to do so with less than 5 MB of code & data, because the font and Unicode tables are that big alone. The input method editor lookup tables are also quite large, and the code behind them is very complex.
You are comparing software start-up time to code efficiency.
Efficient code would allow you to run the exact same software (from your point of view) on a 1GHz single-core CPU.
It's a valid comparison. The startup of Word is mostly single-threaded, and my CPU is 2.6 GHz, so on a 1 Ghz processor that measurement of 0.4 seconds would be... about 1 second. Oh no.. the horror! I'd have to wait an entire second to launch one of the most complex pieces of software on my computer.
Processors have been getting steadily faster, but most people's perception of their computer's speed has been completely dominated by the disk speed, so they haven't noticed.
When my customers complain about how "computers are so slow" I like to do a little demo for them using Mathematica, which is something I used to use at Uni, and still have installed in case I need it. I usually demonstrate it computing Pi to 1 million places in about a second, or symbolically integrating some complex equation instantly where the output is five screenfuls of hideously complex looking maths. Puts things in perspective for them.
On my work computer...
I'll stop you right there. It has a mechanical drive, right?
Think about that for a second. You've got a solid state processor with hundreds of millions of parts switching billions of times per second waiting on... a single, moving, mechanical part that can't exceed about a hundred movements per second.
Your computer's mechanical hard disk has a latency about a million times higher than your computer's processor or memory!
Mechanical disk performance has not improved in over a decade! Ignore the benchmarks about streaming throughput, that's irrelevant to real-world performance. What hasn't changed is the latency and the random IOPS.
The SSD I have in my laptop now can do 60,000 random 4KB reads per second, with a latency under 100ms. If you think about it, that's still glacially slow compared to the processor. Ideally, the latency should be a few microseconds, which is what you'd expect for a 4KB transfer on a high-bandwidth link, so there's still a long way to go before disks have "caught up" to the rest of the PC.
You may as well be running software off floppies or tape, and whine about how it takes forever to spool back and forth when you use your software.
On my computer, practically nothing makes me wait, or if I have to wait, it's because real heavy-weight computation is taking place. Even then, if I switch to another program, the alt-tab transition is instant, and all the other software remains perfectly responsive while I wait.
Instead, because of bloat, we're stuck running yesterday's software with tomorrow's hardware.
Bullshit!
You're still using a mechanical hard-disk, right? That's the component that's bottle-necking your PC, not programmers!
Get an SSD to find out what your CPU and your "bloated" software is capable of.
On my old laptop, with an old 2.6 GHz Core 2 Duo CPU, which is several generations older than the CPU in the article, if I double-click the Word 2010 icon, it launches instantly. Not after a second or two, instantly. It's like notepad. I get the pretty transparent window borders, the ribbon, font-smoothing, everything. How is that "bloated"?
I checked it too: Starting up Word 2010 takes ~400ms of processor time, total. It doesn't even show up in the Task Manager, I had to use SysInternals Process Explorer!
Acrobat Reader, the "horribly bloated" application that most people hate for being slow? It also launches instantly, and uses a massive 180ms of processor time. It's eating up my CPU! Oh wait... it's not.
It doesn't work all that well on low-end hardware or virtual machines
It's been demonstrated to match XP performance on even quite low-end gear by several third-party tests. My experience is that's it's faster, particularly the 64-bit builds, which increase the file cache size from ~400MB max to "all of physical memory", which is a big improvement.
Every time you deploy an image you have to manually re-register the thing with Microsoft so it doesn't disable itself
You're Doing It Wrong. If you're supporting Windows 7 for businesses, you should be using KMS or MAK, and using the volume licensed Enterprise editions, not Windows 7 Home or whatever.
Still no decent backup system
It's the best ever - it has both file-level and image-based backups, it can take live snapshots of disks for both types, back up open files, it has a built-in scheduler, and a bunch of other features.
The VHD disk images created by Windows 7 can be mounted as virtual disks using a GUI or the command-line, can be used to boot from directly without having to be restored first, can be trivially converted into a virtual machine disk, and the install CD has a built-in restore wizard.
I haven't seen comparable features in any other operating system except OSX.
More importantly, if you're backing up desktops, You're Doing It Wrong. Laptops should use offline folders to sync with the master copy of the user data on a server, and shouldn't need backing up. Desktops should use folder redirection and/or roaming profiles. Back up your servers, not your desktops.
You can even do it the "Linux way" if you want to: I've seen sample scripts floating about that take a VSS snapshot of a disk, mount it as a folder or drive letter, and use rsync to incrementally update a backup, then release the snapshot automatically. I've done this myself for Windows Server 2003, about 6 years ago, it's nothing new.
XP Mode is buggy and compatibility in general is bad (especially in the 64-bit versions)
You shouldn't even need XP-mode most of the time, particularly on 32-bit editions of Windows 7. I've found that even the 64-bit editions will run just about anything if you simply set the "compatibility flags" on the main program executables. Just how bad are these applications that you have to support? Shouldn't you be blaming the app vendors instead of Microsoft?
Still no EXT3/EXT4 (or any Unix-type), Large FAT or GPT support
Are you kidding me? First, Windows has had GPT disk and boot support since Windows Server 2003 Service Pack 1, it has xFAT, NTFS on removable drives, and there's third-party EXT3 plugins.
If you think EXT3 on Windows is an important feature, again, You're Doing It Wrong. NTFS is a superior filesystem for Windows in practically every way. If you want to share data between Windows and Linux, use NTFS drivers on Linux, or a server with SAMBA.
Limit of 2 physical processors? Really? It's easy to get 4 processors in a box these days with 8 cores each especially in the academic world
That sucks, but 2 sockets is 12-16 cores these days. If you need more computing power than that, than you can afford a Windows Server 2008 R2 license, which gives you almost all the Windows 7 features, and more processor socket licenses. It's a commercial operating system, and it costs money.
Full Disk Encryption requires TPM chips which are missing in just about any system these days so you still have to go into a 3rd party solution.
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.
You still have to downlo
...to bring that experiment up to the level of Mr. Young's work, you would have found that F=ma broke down at some scales, and discovered quantum theory and relativity. Of course, that rigor took physics over 300 years of further effort!
But my point is that it does not break down at ordinary length scales, velocities, and energies. Given the energy levels and testing methods available to him at the time, Newton was essentially 100% correct. He is still, within the scope that his experiments cover, 100% correct today. His results are still taught, and will be taught forever, because they will never become invalid unless the laws of the physical universe change.
A lot of the research that is being done now in the 'softer' sciences is just flat out wrong, and new results totally invalidate previous findings.
Do you see the difference?
It's not about sample size, or the target of the experiment, it's the method and the lack of rigour. For example, Wikipedia has a great example about the Hawthorne Effect. It was discovered when doing 'experiments' on the working conditions of factory workers that the subjects reacted to being 'experimented on' more strongly than the result of the experiment itself. E.g.: varying the intensity level of illumination had a temporary boost on productivity, no matter what that change was.
That result should have never been an issue. The whole experiment is just shockingly bad. Mistake 1: The employees were told there was an experiment going on. Mistake 2: There was no control group. Mistake 3: Only one factory was used in the experiment.
I'm sure they made other errors too, but you get the idea.
This was a mere 60 years ago! Newton had been doing proper, rigorous physics for 325 years! The sad thing is that this lax approach to science is still going on today, and bad results are still published.
Excuses!
The problem with Psychology is not the subject matter, but the researchers and their sloppy techniques. There's a great quote by Richard Feynman in his essay about Cargo Cult Science:
[There] have been many experiments running rats through all kinds of mazes, and so on—with little clear result. But in 1937 a man named Young did a very interesting one. He had a long corridor with doors all along one side where the rats came in, and doors along the other side where the food was. He wanted to see if he could train the rats to go in at the third door down from wherever he started them off. No. The rats went immediately to the door where the food had been the time before.
The question was, how did the rats know, because the corridor was so beautifully built and so uniform, that this was the same door as before? Obviously there was something about the door that was different from the other doors. So he painted the doors very carefully, arranging the textures on the faces of the doors exactly the same. Still the rats could tell. Then he thought maybe the rats were smelling the food, so he used chemicals to change the smell after each run. Still the rats could tell. Then he realized the rats might be able to tell by seeing the lights and the arrangement in the laboratory like any commonsense person. So he covered the corridor, and still the rats could tell.
He finally found that they could tell by the way the floor sounded when they ran over it. And he could only fix that by putting his corridor in sand. So he covered one after another of all possible clues and finally was able to fool the rats so that they had to learn to go in the third door. If he relaxed any of his conditions, the rats could tell.
Now, from a scientific standpoint, that is an A-number-one experiment. That is the experiment that makes rat-running experiments sensible, because it uncovers the clues that the rat is really using—not what you think it’s using. And that is the experiment that tells exactly what conditions you have to use in order to be careful and control everything in an experiment with rat-running.
I looked into the subsequent history of this research. The next experiment, and the one after that, never referred to Mr. Young. They never used any of his criteria of putting the corridor on sand, or being very careful. They just went right on running rats in the same old way, and paid no attention to the great discoveries of Mr. Young, and his papers are not referred to, because he didn’t discover anything about the rats. In fact, he discovered all the things you have to do to discover something about rats. But not paying attention to experiments like that is a characteristic of cargo cult science.
(emphasis mine)
Most psychologists and social scientists are lazy, and simply don't do real science. Without reproducibility, there's no real result. It's worse than nothing, it's actively counter-productive, because introducing falsehoods into the set of knowledge on which other future science is based upon undermines those results also.
Newton's experiments are still reproducible today, to many digits of precision, after more than three centuries! We've built on them, sure, and come up with newer laws that cover a wider range of circumstances, but it wasn't like.. oh no.. due to only doing the experiment once due to lack of funding or some other excuse, it's not really "F=ma" but "F=ma^2"! Can you imagine if physics, chemistry, or engineering were done this way? H2O? Did I say that? I meant H3O! Or was it H5O2? I forget... never mind.
While I can (and have) reproduced Newton's results in a lab, psychologists still quote Freud and teach his lunacy at a tertiary level. The guy was a nutcase, and never did any real science, yet it's still part of standard introductory psych textbooks.
well, the modern life is about 10^6 times more pleasant than the hunter gatherer existence, so i will disagree with you there.
It's not more pleasant than the easiest hunter gatherer existence.
True, but that's only possible in a few places, like tropical island paradises. That lifestyle is no longer possible for the vast majority of the population. So the rest of us work, and live a life that is almost as comfortable. I have basically a zero risk of starvation, far lower risk of disease or illness than any hunter-gatherer, and I have an easy job that takes up only about 1/4 of my waking hours.
There's also strong evidence that the 'paradise' islands were regularly raided by the neighboring warrior tribes from the more ordinary islands to steal the food and rape the women.
Historically, humans lived a far more violent life than modern people. See this TED video: A brief history of violence.