(5) Isn't chock-full of functionality bugs and compatibility/interoperability problems (in Bluetooth, NFC, WiFi / WiFi Direct, wireless display (Miracast or WiDi or...)
(6) Adopts a "make it work well or don't ship it" policy - a lot of the gimmicky features in recent smartphones don't work all that well, and are really buggy or highly dependent upon individual users' use case (or biology, or...) meeting some limited design expectations. The best smartphones simply exclude gimmicks that aren't properly refined yet.
(7) Stop sacrificing battery life at the altar of thinness. Not to mention thinness also makes the phone harder to hold in the hand...
(8) Improve UI responsiveness and performance to the level of Apple. Reign in bad-behaving apps that try to stay awake all the time (or let the user control this - best option). Can't be that hard if Apple can do it.
(9) Bring back MicroSD.
(10) Bring back removable batteries.
(11) Better software support for using the phone as a get-me-over workstation with bluetooth / HDMI / Miracast as possible venues for keyboard/video/mouse
My first "new-age" smartphone (discounting those horrid old 3G Windows Mobile phones with a stylus in the mid-2000s) was a Motorola Droid 2. For a number of years, Motorola was well-known and respected among smartphone users for:
- Shipping fairly high-end kit, though perhaps not always the latest and greatest
- Very good power efficiency (for Android)
- A lack of the excessive amount of crapware you get on most phones; only the bare minimum the carrier forces on you
- A close-to-vanilla Android experience
- Great build quality and premium feel
- Reasonable prices - they were never the most expensive in the marketplace
- Generous battery capacity -- which, when combined with the power efficiency of their tuned SoCs, led to awesome battery life without any external batteries or extended batteries
- One of the less-hyped smartphone manufacturers (compared to Apple and Samsung) that still churned out well-engineered products and listened to their customers
Unfortunately these virtues seem to have fallen by the wayside to an extent, and the dominance of Samsung, (LG?), and Apple has pushed them out of the market it seems.
The only effect Lenovo could possibly have on them is to force them to cheapen their build. Everything Lenovo touches turns to cheap plastic.
Actually, I get the impression that Wheeler at some point might have been backstabbed or financially or personally harmed by one of his colleagues or associates while he was in the cable industry. Now that he's a regulator, he seems to be standing in the way of letting cable companies run amok with plans to further and further monetize their customer base.
I think, if we had no regulation, Comcast would gladly charge $10 per GB just like mobile carriers do. I'm not saying we have an awesome regulator who always goes to bat for the little guy -- don't make me laugh -- but I'm saying it could be worse. And it probably will get worse in the future. We will wistfully look back at 2016 as one of "the good years".
First, it's already a done deal. Samsung Knox, TouchWiz, and on and on -- and different, incompatible versions from each IHV -- are considered by these companies' executives to be their "crown jewels". They're "distinguishing" factors that set them apart from other handset makers. You would actually break a fair number of Android apps by going back and retroactively removing these incompatible subsystems, since a number of apps actually use them.
Second, the partners have the source code. Almost all of it save for perhaps the Google-branded applications suite. The code is also liberally licensed. If Google tried to put their foot down, the manufacturers would just ignore them and fork Android. See: CyanogenMod, Amazon FireOS.
Third, Google has no direct relationship with carriers. Carriers interface with the IHVs, not with Google (except, perhaps, for Nexus devices). As long as the IHVs are okay with slow security updates and love to make money by dumping their own crapware on the phones (and letting the carriers do the same), they have no motivation to change their ways.
Naming and shaming won't do much to change this momentum. It's a very poorly engineered, overly bureaucratic process created by tight competition and a tendency (by both IHVs and Google) to want to continually tighten up the ecosystem and make things more proprietary and less open.
Apple has really cornered the market on rapid OTAs and addressing security issues quickly. Google has a chance of matching them with first-party phones and Nexus devices, but that's it. Outside Nexus, Android is the wild wild west, kinda like the old Windows 3.1 / 95 days, but even worse in some respects.
There is no happy medium that's both open and secure because the market doesn't demand it. People keep buying Android's mostly-closed, buggy, insecure crap and Apple's completely-closed, less buggy, probably-more-secure-but-we-don't-really-know walled garden crap.
I'm trying to deduce what kind of belief system might be behind such a post, and I concluded (charitably) that you are being sarcastic with the intent to imply that we *shouldn't* be displacing workers with automation.
Great, OK.
So let's just completely do away with all automated information systems. We can keep computers because human beings can manually input data into them for public benefit (e.g. Wikipedia), but oh no, we can't automate anything, because that takes *jobs*.
Want to buy a new piece of electronic equipment? Head to your local Best Buy. We don't need no stinkin' automated Amazon warehouse taking all our jerbs! Oh, and by the way, that piece of electronics you just bought is hand-soldered, hand-machined, and completely unique. It's "Artisan". It's also anyone's guess as to whether the thing actually works. Why is my coworker's laptop 20% faster and much more stable than mine? Why, because a more experienced Artisan crafted it, of course.
Want to file for retirement? Head to your local Social Security office. We don't need no stinkin' online form taking all our jerbs!
Want to safely operate an airplane? Better bring all those Flight Engineers out of retirement and re-expand cockpits with a third seat and a console full of analog dials and switches. Hey, at least if he flips the wrong switch, the computer won't be there to automatically warn him that he's a stupid idiot about to kill 220 people, right?
Want to buy a train ticket? Great. Go talk to the lady at the counter, rather than sliding your credit card into the machine. At least she'll smile at you, possibly share interesting and engaging stories about her life, and help you out with a free ticket on that one day when you forgot your wallet. (This one is sincere; I really do like ticket agents and the personal service quality they bring to the experience of public transportation.)
Here's the thing, though. There are some jobs that computers and machines (or machines controlled by computers) do better than humans. Repetitive, mindless jobs. Mechanical jobs.
The reason we're able to produce so much stuff so efficiently today is because we use all these machines and all this automation to increase yield without having to pay umpteen workers to do it.
Heck, I think it would be nigh impossible to feed the world's population today without modern agriculture. Think of how many more jobs it would take to till fields by hand with scythes and manure, compared to tractors and artificial chemicals! We have an incredible capacity for converting petroleum into food. But that's suddenly a bad thing because of the potential jobs that could be handed out if we didn't do that? (BTW, it's worth mentioning that nobody *needs* to go hungry in the world today with our current food production. Hunger is a distribution (of wealth, and food) problem, not a capacity problem. It would definitely be a capacity problem if we resorted to fully manual farming, though.)
If you acknowledge these points, then you must acknowledge that industrial efficiency is beneficial. And once you accept that, the whole shaky house of cards built around "dem machines taken our jerbs!" falls down. It's only "your" job as long as it's needed.
But, as I emphasized in my first post, I don't think we can just sit back and not care about all the displaced workers. We need to have a good solution in place to ensure that these people can lead happy, fulfilled lives and also present them with opportunities to change to careers that have not been automated (yet), or indeed, careers that might never be automatable.
I write a post expressing concern for the welfare of the people who are losing jobs to automation (the "99%" in your undirected diatribe of rage against me).
You come back then accusing me of basically being a consumer (which, by the way, would have similar attitudes and behaviors in ANY advanced industrialized country; the US just happens to be marginally better at throwing larger amounts of cheaper crap at its citizens to buy, due to favorable foreign trade laws.)
Either you're replying to the wrong person, or you _completely_ misinterpreted my post. Or you're just stupid and have such pent up hatred inside you that you just spew it at anyone and everyone at random. Anonymously, too, of course, which is why you feel perfectly fine throwing such vehement personal attacks at me. I bet you wouldn't say the same to my face.
Your sarcastic tone was unnecessary to get your point across. In the long run it's impossible to argue with the benefits of industrial efficiency, and robots are a clear winner over humans for efficiency.
The problem is that society (an umbrella term encompassing individuals and their attitudes; government lawmakers and executives; and corporations' leadership) collectively has few ideas (and even fewer plans to actually implement those ideas) about what to do to take care of the laborers whose jobs are being taken away by this efficiency. We continue to see global population growth; there are more people than ever, but fewer jobs are needed as automation increases.
The whole "let them eat cake" philosophy won't work. You're talking about a 21st century revolution in the way business is conducted. You can't expect the current societal structures and economic theories to continue to work when you're making such a drastic change. The change is ultimately for the better, but only if we change our society to look after the people who will be out of work.
Let's hope that industrial efficiency and automation helps us reach the high ground, instead of delving into a horrid dystopia.
Good point. The only cap policy that I think would be reasonable would be to set all the slower connections to something more generous like 600 GB and the top tier connections to 1 TB, then implement a policy to increase the caps by 15% per year (doubling time = 4.96 years). The exact year over year cap growth would have to be based on a scientific study of the per-user increase in fixed line broadband Internet yearly data consumption in the US.
Caps should be in place to prevent abuse, not to artificially punish regular users doing a reasonable amount of work / activity. I always like the public water analogy.
Now granted, public water is metered instead of a flat rate, but let's say hypothetically that it were flat rate.
If you turn on all the sinks and showers in your house and just leave them on 24/7, you're going to get a call from the water provider, no matter who it is, whether you're on a flat rate plan or metered. Even if you're paying for the water on a per-volume basis, even if that makes them a significant amount of money, they'd rather you *not* use their resource to the point of exhaustion, because it impacts other customers.
If, on the other hand, you happen to be a pool enthusiast and have a gigantic pool filling your backyard, and invite neighbors and friends over to dirty up your pool and have to frequently drain and re-fill it, you could end up using many times more water than your neighbors. But you're doing it for what is nominally a legitimate *purpose* - you aren't just running it down the drain because you can; you're doing it for entertainment/social purposes. You're also probably using water at a much slower rate than the guy who leaves all his faucets and showers on 24/7.
If caps are low enough that the pool enthusiast can face punitive fees or risk being disconnected from the water supply, that's *broken*. The only guy they should be catching in their net is overt resource *abusers* (whether intentional or accidental; maybe they have a virus that is pegging their connection as part of a DDoS botnet).
Well if we assume (naively) that Google's intent is to make it more convenient and faster for users to unlock their phones, why not just standardize on technology that mimics the iPhone's Touch ID? The same button I press to turn on my screen is simultaneously scanning my finger to determine if I'm the authorized user. That level of convenience (with a fair bit of security, short of someone forcing you to unlock your own phone) is hard to surpass.
Press button, unlock phone. No typing passwords or PINs, no trying to remember the way you walked yesterday, no finagling with voice intonation, no combing your hair so you look the same to the camera... Perfectly reliable and secure enough for most.
Probably we have to assume the reason they're doing this is *not* to benefit user convenience.
On the other hand, if the CIA (or any other Federal or local organization, whether related to law enforcement or not) wishes to come into your house, the following are typically true:
(1) You know about it. (2) It costs them a *significant* amount of money (have to pay the people to go out and knock on / bust down your door). (3) There is huge risk of negative PR for them if they don't find what they're looking for. (4) They need a warrant from a judge. (5) Because of all the above, they have to be pretty darn sure that you're involved in some kind of crime before they do it.
None of these factors will necessarily be true if we allow the government to have encryption backdoors. They can just passively monitor the population whenever they choose to (which, other factors notwithstanding, would be "always") for any signs of disobedience. And in their quest to be ever-watchful and more and more effective at fighting crime and terrorism, they will soon step up their efforts to "next-level" attack prevention, like thought police -- just typing a few characters into Google, writing an opinion piece, or expressing certain ideas could get you labeled as a deviant and thrown away in jail.
In closing, I will quote you the mission of the CIA from Wikipedia:
"The Central Intelligence Agency is a civilian foreign intelligence service of the U.S. Government, tasked with gathering, processing and analyzing national security information from around the world, primarily through the use of human intelligence."
In what world is it anywhere remotely within their jurisdiction, for an organization that is not law enforcement and whose gaze should be *outward* to other countries rather than *inward* to the US, to insert themselves into the communications of US citizens, in an automated, computerized way (instead of "human" as the mission says), for the purposes of law enforcement, which is not at all part of their mission?
No, I'm not okay with the CIA deciding they'd like to get into my online presence in any capacity beyond what I post publicly. Private means private. I'm perfectly fine with losing my data permanently if I lose the access credential, precisely because making a "oops" key is exactly as insecure as making a backdoor for the three-letter acronym agencies.
And like I said, whereas it requires a number of checks and balances accompanied with a high degree of confidence for these guys to come knocking at your house, it requires basically nothing at all -- not even the faintest hint of suspicion -- for them to decrypt, monitor and analyze your private data. Your only defense is to swallow the key and pray there's no backdoor in your crypto.
The premium paid for iPhones goes to designing an OS kernel and Bluetooth stack that doesn't drop out every 5 fucking minutes like Android's does because of a bug that Google considers to be of "small" priority (Android bugtracker #95294 and many others) (I've tested it with 5 different phones across 3 manufacturers, with 3 different Android major releases, with 5 different pair of Bluetooth headsets from 5 different manufacturers with 3 different Bluetooth standards, and it reliably drops out on Android while never dropping out, not even once, on my iPhone, after listening to music and taking calls on my iPhone for about 8 months in various noisy wireless environments.)
The premium paid for iPhones goes to getting a phone completely free of bloatware, that can run faster and smoother in 2 GiB of RAM than an Android phone can run in 4 GiB (or even 6, like the upcoming Note 6).
Also, iPhone 6S had the objectively fastest mobile GPU in the world in a smartphone form factor upon release. You're paying for first access to top-shelf hardware. Android didn't/couldn't answer until the release of the Galaxy S7 in March, months after iPhone users had access to FinFET CPUs and GPUs.
Mac Pros cost so much because they use enterprise-grade hardware, like Xeon processors, enterprise-grade SSDs, and Tesla or Quadro or FirePro GPUs. This kit legitimately costs more than consumer-grade kit if you buy it straight from the original hardware manufacturer, so of course Apple is going to pass on those prices to the customer. The mark-up on a high-end Mac Pro isn't that much compared to a similarly kitted out HP or Dell enterprise workstation.
Apple does occasionally screw users with unreasonable prices, but there are plenty of examples of legitimate technical and engineering reasons for their high prices on many of their product lines. The one thing you have to avoid with Apple hardware is buying into it near the end of the product's lifecycle or close to the anticipated launch of the next gen. They don't drop their starting prices almost all over the course of a product's lifespan, until the new one is actually out on the market. So right now is an absolutely horrendous time to buy a new MacBook Pro.
$1.29 for a single song in iTunes is still high, though.
I use Start10 which works equally well (but isn't free). However I do not dislike Classic Shell. It just seems to be a little less "pretty" than Start10:)
As a savvy owner of a Prius c hybrid, I think I have some insight into this... Basically, the brake pads *are* used quite a lot by aggressive drivers who tailgate and have to brake hard when the car in front of them slows down. People who drive with a proper following distance ahead of them will rarely have to use the disc brakes.
Hybrid vehicles (and EVs, probably) have smaller brake pads than similarly sized conventional vehicles (though the actual stopping power of the disc brakes in an emergency is just as good as regular cars). The brake pads are about half as thick on my Prius c as the brake pads on my Honda Civic. That's because the manufacturer expects you to use them less often. I'm sure there are some insane drivers out there who can burn through the brake pads on a vehicle like mine in well under 50,000 miles, but those same people would burn through the brakes on any vehicle just as quickly.
I've learned to "feel" the difference between the cut-over between regenerative braking and the disc brakes. The disc brakes slow you down WAY faster. There's not a discrete and obvious jolt when you gradually depress the brake pedal; it's incredibly smooth; but to use an analogy, as long as I'm slowing down at about the same rate as a truck can slow down when using the jake brake (engine braking - that loud "farting" sound that large trucks sometimes make when slowing down), I'm using the regenerative braking system only. If I'm slowing down much faster than that, the disc brakes are being engaged (the brakes and the regenerative braking can be active at the same time, unless you are braking at what would be considered "emergency" speeds, in which case the regenerative braking system disengages, perhaps because it can't handle that amount of torque or current).
As for the article itself: 24 percent?! That's total bullshit.
The Prius c is literally a Yaris Hybrid (it's marketed as such in some parts of the world). It's the Toyota Yaris -- a compact car -- with the Toyota Hybrid Synergy Drive in it. So, it's a Yaris, *plus* the weight of the HSD.
The curb weight of the Yaris is 2335 pounds. The curb weight of the Prius c is 2500 pounds. That's only a 7.066% increase. That's a far, far cry from the 24% the article cites.
OK, you say, let's look at *electric* vehicles specifically, not just hybrids. Because hybrids don't have to lug around 500 pounds of lithium-ion batteries. Hybrid batteries tend to weigh under 200 pounds, with the smallest hybrids' ~1 kWh batteries weighing less than 100 pounds.
Let's take the Chevy Spark. The conventional Spark weighs in around 2270 pounds. The EV? 3000 pounds. That's a 32% increase for basically the same passenger and cargo volume. Fair enough. But 3000 pounds isn't out of this world, and is in the ballpark of many upscale compact cars like the (conventional) Honda Civic.
Another example. The 2016 Nissan Leaf weighs around 3150 pounds. I did some research to try and find a conventional vehicle with similar interior measurements (headroom, cargo space, etc.) and I came up with this: The 2016 Honda Civic EX has a total (usable) interior volume of 110.1 cubic feet with a curb weight of 2799 pounds. The 2016 Nissan Leaf has an interior volume of 116 cubic feet. So for 6 more cubic feet of interior (5.4% more), the vehicle weighs 351 pounds (25.4%) more.
Based on these limited comparisons, it seems like the article's claim about the increased weight of electric vehicles is factual. However, it is absolutely not valid to make the leap to saying that plug-in hybrids or conventional hybrids are anywhere near as bad in terms of added weight.
What I'm not convinced of, however, is the severity and environmental impact of tire and brake wear, regardless of vehicle weight. EVs and hybrids also run with low rolling resistance tires, which should reduce the amount of tire "stuff" in the air, in any case. Did they take that into account?
However, switching out a gasoline engine for a TDI diesel engine adds about 300 pounds to a sedan-
The only part of your post that I agree with is the assertion that the USDA should not be running this. Agriculture and Internet service are completely orthogonal aspects of living. Just because the USDA has traditionally done things that benefit farmers (which represent a percentage of the rural population) doesn't mean that they should be levied with sticking up for every possible interest that the rural community could have in any government activity whatsoever.
The FCC should be doing this. It's pretty obvious.
So, Mr. "IT Worker", how do you enjoy waiting a minimum of 20 seconds (assuming 100% utilization, no other data flows competing for throughput, no congestion or signal degradation, and zero protocol overhead) to save a 10 MB Excel file to a SMB shared drive?
How do you enjoy waiting a minimum of 350 seconds (again, assuming no overhead, etc.) to download a 175 MB Eclipse installation?
How is it waiting a day or more to download a major update package for Windows or Office or a distribution upgrade for your GNU/Linux distro (depending on what you use)?
At 4 Mbps, I can't imagine you get very much work done, unless your job is primarily involving files on your local system with very little connectivity or data exchange with your company's servers.
The connection provided by my company at my work site (in a corporate office building) caps out around 20 Mbps, with congestion during peak hours pushing it down around 8 to 10 Mbps down, 2 to 4 Mbps up. Everyone complains that we can barely get any work done.
We have another network in the building specifically for the purpose of interacting with one of our customers; it's a direct fiber link to their headquarters' gateway. Their network has something like 100 Gbps of available bandwidth, with individual clients able to effectively max out the gigabit ethernet port, even during peak hours. We *love* working for this customer because we're so much more productive than when we're forced to use our company's internal network.
So, at this point, we have: myself, for whose job a 10 Mbps connection is grossly inadequate; and *you*, for whose job a 4 Mbps connection is apparently adequate or even good. That's a small enough sample as to call either one of our accounts anecdotal.
The problem is that you can't really look at a population of people overall and just pull a number out of your ass and say "that number is enough for anyone!". When you do that, you are effectively limiting the professions and the types of work that person can pursue from that location. I'm sure I'm not the only one whose job requires a bit more than 4 Mbps, and I'm sure you're not the only one whose job could be accomplished with a 2800 baud modem. There has to be a balance somewhere, but I think it lies a good ways north of 4 Mbps.
It's incredibly narrow-minded for one to think that:
(1) This grant program is exclusively for "working poor" (or poor people, period) to get Internet access; and (2) All a poor person would/should/could want to do with the Internet is upload resumes so they can stop being poor / living in rural areas.
This post displays a stunning lack of empathy for people who have made a choice to live in a less densely populated area. First of all, not everyone who lives in a rural area is poor. There are plenty of middle-class and upper-middle-class people who have larger land holdings or have fewer neighbors than the ISPs would ideally prefer for instant ROI. Some of these are technologists, or potential customers of technology companies. They might be gamers, media consumers, software developers, architects, or any number of other professions where access to serious Internet connectivity can be a huge boon.
Not only that, but if their employer is amenable to the arrangement, having access to a modern Internet connection might even allow these people to work from home. This saves them thousands of dollars in fuel per year vs. having to drive many miles into the office, and also prevents the vehicle emissions that would have been necessary for them to get there.
Having access to a very fast Internet pipe can change someone's life in a revolutionary way. It can enable people to embark upon careers that are not possible without a fast Internet connection. To name a few:
- Youtube/Twitch/etc. streamer (gamer, political, scientific/educational, or other content) - not possible without fast upstream because otherwise video upload times make it impractical
- Podcaster - same as video above, but requires less upstream because it's just audio
- Pro gamer - without a fast connection for rapid access to new games and patches, and low latency, one cannot reach the pinnacle of pro gaming, no matter how hard they try (unless you intend to become a pro at a turn-based game where latency is relatively unimportant)
- Remote knowledge worker / engineer - A lot of the top tech companies are hiring talent remotely without regard for where they live. It's only "old world" businesses, and those with extreme security requirements like working for contractors or government touching classified data, that have hard requirements for working on-site. Almost any job in IT, plus many jobs in management, finance and administration, can be done remotely just as well as face to face... but only with good Internet connectivity. If your work output primarily involves taking information in one form (instructions, requirements, source data, whatever) and transforming it into another form (source code, metrics, stats, equations, papers, emails, whatever), you can work remotely if you have a good computer and a fast Internet connection. The remote work movement is helping to conserve a ridiculous amount of transportation fuel, but it can get even better in the future, and fast Internet is a critical first step.
The Internet is not only used for uploading resumes. Stop thinking like someone stuck in 1996, and recognize that this program is intended to open up possibilities for the rural population (about 15% of the US population, over 42 million people per the latest census) that crosses economic class boundaries and would not be possible otherwise. As someone else said, these ISPs simply would never, ever bring service to these remote areas if not for government subsidy, because these areas don't otherwise meet the company's ROI goals. These people should not be penalized for living in rural areas. And by the way, they're paying taxes, too, so it's part of their own tax money that goes to fund these programs.
Any time your data is being stored in an unencrypted format (or encrypted with keyword indexing, whatever, same difference) on a server you don't control, you should bring with you the *expectation* that the company hosting your data, and/or potential political or corporate adversaries, can and WILL access that data.
If you're OK with that, then more power to you. I doubt NSA or a corporate competitor cares about your pictures of playing fetch with your dog. They might care a bit about a copy of a secret agreement like the TPP, though.
Actually, they probably already have a forced MITM proxy (requiring you to trust a self-signed root CA from the gateway) that decrypts all traffic going across the wire.
If they have such a MITM -- and I can't imagine that they don't, since I know this is an extremely common thing in Federal IT -- then they would actually gain enormous insight into *who* is leaking, simply by allowing people to connect to these sites like Cloakroom and observing the traffic.
I will concede, however, the argument that if you perform an off-the-record (sneakernet, etc) key exchange with a trusted individual in order to begin your transaction with something like Cloakroom, then you could still privately communicate with Cloakroom even if the hostile gateway is performing MITM. All you have to do is paste your ciphertext into a text box on an HTML form and submit.
In that case, the Fed would know *that* you sent something to Cloakroom, because they'd know the IP of the computer that submitted the request and who was logged on via their PIV card, but they wouldn't know what you said -- whether something completely innocuous like "Nancy Pelosi has weird eyebrows" or something actually politically impactful.
Most corporate Exchange servers are either behind a VPN or proxy; have some third-party authentication wrapper (or even worse, an in-house custom one); or straight up don't let you access it without being hard-wired to their network with a "blessed" computer.
Most people who use an Exchange server for work won't be able to use this...
I use Apple products only when they make sense for me. Not sucked into the reality distortion field. The iPhone is a great phone (mainly because of its software), but the Apple Watch is too flawed of a device to be desirable in any way.
Of course, but that's only 10% of the job. The real hard parts include things like:
1. "Scripting the scripts": learning how and when to use each script and in what context. Often it is difficult to express exactly what conditions may prompt a given script to be run, especially if those conditions involve human factors (soandso forgets their password; the company gets sued and needs to put a legal hold on email deletion; etc.) If people could write scripts to automate the triggering of lower level scripts under the right conditions, they'd already have done that (for example, scripts that merely need to be run periodically get set up as a cron job) because we all know that most sysadmins are lazy and will seek to automate things to make life easier for them -- unless those things turn out to be difficult-to-impossible to automate.
2. Change: except for a few niche areas like Certificate Authorities, IRC servers, mail servers, and OS package repositories, the software configuration of many environments necessarily needs to change over time. Not only to keep ahead of obsolescence curves (end of security updates / support from vendors, etc.) but to actually provide new functionality and features requested by the customers (external or internal). There are very few server services that can just remain static for decades. The more change and upheaval you have in your environment, the less valuable it is to "button everything up" into a nice, well-oiled, automated machine. And the process of that automation requires significant manpower and extremely good documentation of exactly what everything does. Even in the best case, though, you'll probably end up re-doing large parts of it every six to seven years, since that's the support period (security updates lifecycle) for RHEL, Ubuntu LTS, etc.
3. Cost-benefit: Higher-level server automation can't write itself, because you have to tell the computer when to do what, under what conditions. Writing management engines that intelligently keep track of this stuff requires significant development time, and if the conditions are complex, you'd better write it in a robust language like C/C++/Java/C#/Rust/.... instead of Bash (shell scripts are not especially good at error handling and are not especially well optimized for large bodies of code). So after you have invested all that time in whiz-banging the whole thing into a web app with a few buttons, is it really worth the savings vs. waiting for a sysadmin to type a few commands manually into the console?
4. Frequency: Many (most?) system administration tasks are not done with terrible frequency. If something's only done once or twice a year, it's almost guaranteed not to be worth automating, even if the process takes a full day or two for the admin. So we typically only focus on automating tasks that need to be done multiple times per week. The more frequent, the more automated it probably is.
5. External environment: This mostly falls under "Change" above, but major vulnerabilities like the SHA1 weakness, Heartbleed, etc. occur once in a while and they can require an arbitrary amount of major change and upheaval on the server side, sometimes with very high priority (timetable doesn't allow for figuring out how to automate it). These things can happen at any time and they require someone to always be standing-by, at the ready, possessing the general knowledge of the system architecture as well as detailed knowledge of the commands to make changes on the system, so that they can implement the solution to the problem (which can't possibly be known in advance because the problem comes from an external source, so you can't automate it in advance).
Until/unless we get True AI (which I don't believe will ever happen), the world will need sysadmins. While the general trend towards higher level automation continues, it's not such a severe problem that admins will just get laid off if they don't know how to code. I don't buy that for a second, even though I strongly believe that it's beneficial for admins
Anyone who trusts their data to ext4 is grossly misinformed and playing with fire. Almost any other filesystem with a journal is less likely to eat your data than ext4. If you're using ext4 it should be for data that you really don't care if you lose / if the system running it crashes until you have a chance to reboot. In other words, it's fine for laptops where people check mail and browse facebook, but it has absolutely no place in the datacenter.
It's worth mentioning that a subset of the Linux kernel copyright holders, including some of the most influential and prolific contributors, believe that the situation around the binary modules from Nvidia, AMD, etc. is similar to this ZFS module.
Just because it's gone unchallenged for a number of years doesn't mean that the issue is settled or buried. It's never been challenged or ruled-on in court AFAIK, and until it has, it's anyone's guess what the law will actually decide when interpreting the GPLv2 and copyright law.
It really doesn't matter if you want to label individuals as "extremist" for thinking a particular way about the law, though I'll admit that's a better term than Barbara's use of the term "freetard" -- again, namecalling just weakens one's argument. The only thing that matters is the way that a judge will ultimately interpret the relevant texts to determine if there is any violation of law, and what the penalties are if so.
Slashdot Mirror
And
(5) Isn't chock-full of functionality bugs and compatibility/interoperability problems (in Bluetooth, NFC, WiFi / WiFi Direct, wireless display (Miracast or WiDi or...)
(6) Adopts a "make it work well or don't ship it" policy - a lot of the gimmicky features in recent smartphones don't work all that well, and are really buggy or highly dependent upon individual users' use case (or biology, or...) meeting some limited design expectations. The best smartphones simply exclude gimmicks that aren't properly refined yet.
(7) Stop sacrificing battery life at the altar of thinness. Not to mention thinness also makes the phone harder to hold in the hand...
(8) Improve UI responsiveness and performance to the level of Apple. Reign in bad-behaving apps that try to stay awake all the time (or let the user control this - best option). Can't be that hard if Apple can do it.
(9) Bring back MicroSD.
(10) Bring back removable batteries.
(11) Better software support for using the phone as a get-me-over workstation with bluetooth / HDMI / Miracast as possible venues for keyboard/video/mouse
My first "new-age" smartphone (discounting those horrid old 3G Windows Mobile phones with a stylus in the mid-2000s) was a Motorola Droid 2. For a number of years, Motorola was well-known and respected among smartphone users for:
- Shipping fairly high-end kit, though perhaps not always the latest and greatest
- Very good power efficiency (for Android)
- A lack of the excessive amount of crapware you get on most phones; only the bare minimum the carrier forces on you
- A close-to-vanilla Android experience
- Great build quality and premium feel
- Reasonable prices - they were never the most expensive in the marketplace
- Generous battery capacity -- which, when combined with the power efficiency of their tuned SoCs, led to awesome battery life without any external batteries or extended batteries
- One of the less-hyped smartphone manufacturers (compared to Apple and Samsung) that still churned out well-engineered products and listened to their customers
Unfortunately these virtues seem to have fallen by the wayside to an extent, and the dominance of Samsung, (LG?), and Apple has pushed them out of the market it seems.
The only effect Lenovo could possibly have on them is to force them to cheapen their build. Everything Lenovo touches turns to cheap plastic.
Actually, I get the impression that Wheeler at some point might have been backstabbed or financially or personally harmed by one of his colleagues or associates while he was in the cable industry. Now that he's a regulator, he seems to be standing in the way of letting cable companies run amok with plans to further and further monetize their customer base.
I think, if we had no regulation, Comcast would gladly charge $10 per GB just like mobile carriers do. I'm not saying we have an awesome regulator who always goes to bat for the little guy -- don't make me laugh -- but I'm saying it could be worse. And it probably will get worse in the future. We will wistfully look back at 2016 as one of "the good years".
No, they can't tell them that.
First, it's already a done deal. Samsung Knox, TouchWiz, and on and on -- and different, incompatible versions from each IHV -- are considered by these companies' executives to be their "crown jewels". They're "distinguishing" factors that set them apart from other handset makers. You would actually break a fair number of Android apps by going back and retroactively removing these incompatible subsystems, since a number of apps actually use them.
Second, the partners have the source code. Almost all of it save for perhaps the Google-branded applications suite. The code is also liberally licensed. If Google tried to put their foot down, the manufacturers would just ignore them and fork Android. See: CyanogenMod, Amazon FireOS.
Third, Google has no direct relationship with carriers. Carriers interface with the IHVs, not with Google (except, perhaps, for Nexus devices). As long as the IHVs are okay with slow security updates and love to make money by dumping their own crapware on the phones (and letting the carriers do the same), they have no motivation to change their ways.
Naming and shaming won't do much to change this momentum. It's a very poorly engineered, overly bureaucratic process created by tight competition and a tendency (by both IHVs and Google) to want to continually tighten up the ecosystem and make things more proprietary and less open.
Apple has really cornered the market on rapid OTAs and addressing security issues quickly. Google has a chance of matching them with first-party phones and Nexus devices, but that's it. Outside Nexus, Android is the wild wild west, kinda like the old Windows 3.1 / 95 days, but even worse in some respects.
There is no happy medium that's both open and secure because the market doesn't demand it. People keep buying Android's mostly-closed, buggy, insecure crap and Apple's completely-closed, less buggy, probably-more-secure-but-we-don't-really-know walled garden crap.
I'm trying to deduce what kind of belief system might be behind such a post, and I concluded (charitably) that you are being sarcastic with the intent to imply that we *shouldn't* be displacing workers with automation.
Great, OK.
So let's just completely do away with all automated information systems. We can keep computers because human beings can manually input data into them for public benefit (e.g. Wikipedia), but oh no, we can't automate anything, because that takes *jobs*.
Want to buy a new piece of electronic equipment? Head to your local Best Buy. We don't need no stinkin' automated Amazon warehouse taking all our jerbs! Oh, and by the way, that piece of electronics you just bought is hand-soldered, hand-machined, and completely unique. It's "Artisan". It's also anyone's guess as to whether the thing actually works. Why is my coworker's laptop 20% faster and much more stable than mine? Why, because a more experienced Artisan crafted it, of course.
Want to file for retirement? Head to your local Social Security office. We don't need no stinkin' online form taking all our jerbs!
Want to safely operate an airplane? Better bring all those Flight Engineers out of retirement and re-expand cockpits with a third seat and a console full of analog dials and switches. Hey, at least if he flips the wrong switch, the computer won't be there to automatically warn him that he's a stupid idiot about to kill 220 people, right?
Want to buy a train ticket? Great. Go talk to the lady at the counter, rather than sliding your credit card into the machine. At least she'll smile at you, possibly share interesting and engaging stories about her life, and help you out with a free ticket on that one day when you forgot your wallet. (This one is sincere; I really do like ticket agents and the personal service quality they bring to the experience of public transportation.)
Here's the thing, though. There are some jobs that computers and machines (or machines controlled by computers) do better than humans. Repetitive, mindless jobs. Mechanical jobs.
The reason we're able to produce so much stuff so efficiently today is because we use all these machines and all this automation to increase yield without having to pay umpteen workers to do it.
Heck, I think it would be nigh impossible to feed the world's population today without modern agriculture. Think of how many more jobs it would take to till fields by hand with scythes and manure, compared to tractors and artificial chemicals! We have an incredible capacity for converting petroleum into food. But that's suddenly a bad thing because of the potential jobs that could be handed out if we didn't do that? (BTW, it's worth mentioning that nobody *needs* to go hungry in the world today with our current food production. Hunger is a distribution (of wealth, and food) problem, not a capacity problem. It would definitely be a capacity problem if we resorted to fully manual farming, though.)
If you acknowledge these points, then you must acknowledge that industrial efficiency is beneficial. And once you accept that, the whole shaky house of cards built around "dem machines taken our jerbs!" falls down. It's only "your" job as long as it's needed.
But, as I emphasized in my first post, I don't think we can just sit back and not care about all the displaced workers. We need to have a good solution in place to ensure that these people can lead happy, fulfilled lives and also present them with opportunities to change to careers that have not been automated (yet), or indeed, careers that might never be automatable.
Oh, and for the record, I live close to the poverty line. Above it, yes, but nowhere near sniffing 6 figs. I'm a 50 percenter at best.
Uh...... what? Let me get this straight:
I write a post expressing concern for the welfare of the people who are losing jobs to automation (the "99%" in your undirected diatribe of rage against me).
You come back then accusing me of basically being a consumer (which, by the way, would have similar attitudes and behaviors in ANY advanced industrialized country; the US just happens to be marginally better at throwing larger amounts of cheaper crap at its citizens to buy, due to favorable foreign trade laws.)
Either you're replying to the wrong person, or you _completely_ misinterpreted my post. Or you're just stupid and have such pent up hatred inside you that you just spew it at anyone and everyone at random. Anonymously, too, of course, which is why you feel perfectly fine throwing such vehement personal attacks at me. I bet you wouldn't say the same to my face.
Your sarcastic tone was unnecessary to get your point across. In the long run it's impossible to argue with the benefits of industrial efficiency, and robots are a clear winner over humans for efficiency.
The problem is that society (an umbrella term encompassing individuals and their attitudes; government lawmakers and executives; and corporations' leadership) collectively has few ideas (and even fewer plans to actually implement those ideas) about what to do to take care of the laborers whose jobs are being taken away by this efficiency. We continue to see global population growth; there are more people than ever, but fewer jobs are needed as automation increases.
The whole "let them eat cake" philosophy won't work. You're talking about a 21st century revolution in the way business is conducted. You can't expect the current societal structures and economic theories to continue to work when you're making such a drastic change. The change is ultimately for the better, but only if we change our society to look after the people who will be out of work.
Let's hope that industrial efficiency and automation helps us reach the high ground, instead of delving into a horrid dystopia.
Still relevant: http://marshallbrain.com/manna...
Good point. The only cap policy that I think would be reasonable would be to set all the slower connections to something more generous like 600 GB and the top tier connections to 1 TB, then implement a policy to increase the caps by 15% per year (doubling time = 4.96 years). The exact year over year cap growth would have to be based on a scientific study of the per-user increase in fixed line broadband Internet yearly data consumption in the US.
Caps should be in place to prevent abuse, not to artificially punish regular users doing a reasonable amount of work / activity. I always like the public water analogy.
Now granted, public water is metered instead of a flat rate, but let's say hypothetically that it were flat rate.
If you turn on all the sinks and showers in your house and just leave them on 24/7, you're going to get a call from the water provider, no matter who it is, whether you're on a flat rate plan or metered. Even if you're paying for the water on a per-volume basis, even if that makes them a significant amount of money, they'd rather you *not* use their resource to the point of exhaustion, because it impacts other customers.
If, on the other hand, you happen to be a pool enthusiast and have a gigantic pool filling your backyard, and invite neighbors and friends over to dirty up your pool and have to frequently drain and re-fill it, you could end up using many times more water than your neighbors. But you're doing it for what is nominally a legitimate *purpose* - you aren't just running it down the drain because you can; you're doing it for entertainment/social purposes. You're also probably using water at a much slower rate than the guy who leaves all his faucets and showers on 24/7.
If caps are low enough that the pool enthusiast can face punitive fees or risk being disconnected from the water supply, that's *broken*. The only guy they should be catching in their net is overt resource *abusers* (whether intentional or accidental; maybe they have a virus that is pegging their connection as part of a DDoS botnet).
Well if we assume (naively) that Google's intent is to make it more convenient and faster for users to unlock their phones, why not just standardize on technology that mimics the iPhone's Touch ID? The same button I press to turn on my screen is simultaneously scanning my finger to determine if I'm the authorized user. That level of convenience (with a fair bit of security, short of someone forcing you to unlock your own phone) is hard to surpass.
Press button, unlock phone. No typing passwords or PINs, no trying to remember the way you walked yesterday, no finagling with voice intonation, no combing your hair so you look the same to the camera... Perfectly reliable and secure enough for most.
Probably we have to assume the reason they're doing this is *not* to benefit user convenience.
*That* sounds secure. /s
On the other hand, if the CIA (or any other Federal or local organization, whether related to law enforcement or not) wishes to come into your house, the following are typically true:
(1) You know about it.
(2) It costs them a *significant* amount of money (have to pay the people to go out and knock on / bust down your door).
(3) There is huge risk of negative PR for them if they don't find what they're looking for.
(4) They need a warrant from a judge.
(5) Because of all the above, they have to be pretty darn sure that you're involved in some kind of crime before they do it.
None of these factors will necessarily be true if we allow the government to have encryption backdoors. They can just passively monitor the population whenever they choose to (which, other factors notwithstanding, would be "always") for any signs of disobedience. And in their quest to be ever-watchful and more and more effective at fighting crime and terrorism, they will soon step up their efforts to "next-level" attack prevention, like thought police -- just typing a few characters into Google, writing an opinion piece, or expressing certain ideas could get you labeled as a deviant and thrown away in jail.
In closing, I will quote you the mission of the CIA from Wikipedia:
"The Central Intelligence Agency is a civilian foreign intelligence service of the U.S. Government, tasked with gathering, processing and analyzing national security information from around the world, primarily through the use of human intelligence."
In what world is it anywhere remotely within their jurisdiction, for an organization that is not law enforcement and whose gaze should be *outward* to other countries rather than *inward* to the US, to insert themselves into the communications of US citizens, in an automated, computerized way (instead of "human" as the mission says), for the purposes of law enforcement, which is not at all part of their mission?
No, I'm not okay with the CIA deciding they'd like to get into my online presence in any capacity beyond what I post publicly. Private means private. I'm perfectly fine with losing my data permanently if I lose the access credential, precisely because making a "oops" key is exactly as insecure as making a backdoor for the three-letter acronym agencies.
And like I said, whereas it requires a number of checks and balances accompanied with a high degree of confidence for these guys to come knocking at your house, it requires basically nothing at all -- not even the faintest hint of suspicion -- for them to decrypt, monitor and analyze your private data. Your only defense is to swallow the key and pray there's no backdoor in your crypto.
I use telnet as my web browser, you insensitive clod!
The premium paid for iPhones goes to designing an OS kernel and Bluetooth stack that doesn't drop out every 5 fucking minutes like Android's does because of a bug that Google considers to be of "small" priority (Android bugtracker #95294 and many others) (I've tested it with 5 different phones across 3 manufacturers, with 3 different Android major releases, with 5 different pair of Bluetooth headsets from 5 different manufacturers with 3 different Bluetooth standards, and it reliably drops out on Android while never dropping out, not even once, on my iPhone, after listening to music and taking calls on my iPhone for about 8 months in various noisy wireless environments.)
The premium paid for iPhones goes to getting a phone completely free of bloatware, that can run faster and smoother in 2 GiB of RAM than an Android phone can run in 4 GiB (or even 6, like the upcoming Note 6).
Also, iPhone 6S had the objectively fastest mobile GPU in the world in a smartphone form factor upon release. You're paying for first access to top-shelf hardware. Android didn't/couldn't answer until the release of the Galaxy S7 in March, months after iPhone users had access to FinFET CPUs and GPUs.
Mac Pros cost so much because they use enterprise-grade hardware, like Xeon processors, enterprise-grade SSDs, and Tesla or Quadro or FirePro GPUs. This kit legitimately costs more than consumer-grade kit if you buy it straight from the original hardware manufacturer, so of course Apple is going to pass on those prices to the customer. The mark-up on a high-end Mac Pro isn't that much compared to a similarly kitted out HP or Dell enterprise workstation.
Apple does occasionally screw users with unreasonable prices, but there are plenty of examples of legitimate technical and engineering reasons for their high prices on many of their product lines. The one thing you have to avoid with Apple hardware is buying into it near the end of the product's lifecycle or close to the anticipated launch of the next gen. They don't drop their starting prices almost all over the course of a product's lifespan, until the new one is actually out on the market. So right now is an absolutely horrendous time to buy a new MacBook Pro.
$1.29 for a single song in iTunes is still high, though.
I use Start10 which works equally well (but isn't free). However I do not dislike Classic Shell. It just seems to be a little less "pretty" than Start10 :)
As a savvy owner of a Prius c hybrid, I think I have some insight into this... Basically, the brake pads *are* used quite a lot by aggressive drivers who tailgate and have to brake hard when the car in front of them slows down. People who drive with a proper following distance ahead of them will rarely have to use the disc brakes.
Hybrid vehicles (and EVs, probably) have smaller brake pads than similarly sized conventional vehicles (though the actual stopping power of the disc brakes in an emergency is just as good as regular cars). The brake pads are about half as thick on my Prius c as the brake pads on my Honda Civic. That's because the manufacturer expects you to use them less often. I'm sure there are some insane drivers out there who can burn through the brake pads on a vehicle like mine in well under 50,000 miles, but those same people would burn through the brakes on any vehicle just as quickly.
I've learned to "feel" the difference between the cut-over between regenerative braking and the disc brakes. The disc brakes slow you down WAY faster. There's not a discrete and obvious jolt when you gradually depress the brake pedal; it's incredibly smooth; but to use an analogy, as long as I'm slowing down at about the same rate as a truck can slow down when using the jake brake (engine braking - that loud "farting" sound that large trucks sometimes make when slowing down), I'm using the regenerative braking system only. If I'm slowing down much faster than that, the disc brakes are being engaged (the brakes and the regenerative braking can be active at the same time, unless you are braking at what would be considered "emergency" speeds, in which case the regenerative braking system disengages, perhaps because it can't handle that amount of torque or current).
As for the article itself: 24 percent?! That's total bullshit.
The Prius c is literally a Yaris Hybrid (it's marketed as such in some parts of the world). It's the Toyota Yaris -- a compact car -- with the Toyota Hybrid Synergy Drive in it. So, it's a Yaris, *plus* the weight of the HSD.
The curb weight of the Yaris is 2335 pounds. The curb weight of the Prius c is 2500 pounds. That's only a 7.066% increase. That's a far, far cry from the 24% the article cites.
OK, you say, let's look at *electric* vehicles specifically, not just hybrids. Because hybrids don't have to lug around 500 pounds of lithium-ion batteries. Hybrid batteries tend to weigh under 200 pounds, with the smallest hybrids' ~1 kWh batteries weighing less than 100 pounds.
Let's take the Chevy Spark. The conventional Spark weighs in around 2270 pounds. The EV? 3000 pounds. That's a 32% increase for basically the same passenger and cargo volume. Fair enough. But 3000 pounds isn't out of this world, and is in the ballpark of many upscale compact cars like the (conventional) Honda Civic.
Another example. The 2016 Nissan Leaf weighs around 3150 pounds. I did some research to try and find a conventional vehicle with similar interior measurements (headroom, cargo space, etc.) and I came up with this: The 2016 Honda Civic EX has a total (usable) interior volume of 110.1 cubic feet with a curb weight of 2799 pounds. The 2016 Nissan Leaf has an interior volume of 116 cubic feet. So for 6 more cubic feet of interior (5.4% more), the vehicle weighs 351 pounds (25.4%) more.
Based on these limited comparisons, it seems like the article's claim about the increased weight of electric vehicles is factual. However, it is absolutely not valid to make the leap to saying that plug-in hybrids or conventional hybrids are anywhere near as bad in terms of added weight.
What I'm not convinced of, however, is the severity and environmental impact of tire and brake wear, regardless of vehicle weight. EVs and hybrids also run with low rolling resistance tires, which should reduce the amount of tire "stuff" in the air, in any case. Did they take that into account?
However, switching out a gasoline engine for a TDI diesel engine adds about 300 pounds to a sedan-
The only part of your post that I agree with is the assertion that the USDA should not be running this. Agriculture and Internet service are completely orthogonal aspects of living. Just because the USDA has traditionally done things that benefit farmers (which represent a percentage of the rural population) doesn't mean that they should be levied with sticking up for every possible interest that the rural community could have in any government activity whatsoever.
The FCC should be doing this. It's pretty obvious.
So, Mr. "IT Worker", how do you enjoy waiting a minimum of 20 seconds (assuming 100% utilization, no other data flows competing for throughput, no congestion or signal degradation, and zero protocol overhead) to save a 10 MB Excel file to a SMB shared drive?
How do you enjoy waiting a minimum of 350 seconds (again, assuming no overhead, etc.) to download a 175 MB Eclipse installation?
How is it waiting a day or more to download a major update package for Windows or Office or a distribution upgrade for your GNU/Linux distro (depending on what you use)?
At 4 Mbps, I can't imagine you get very much work done, unless your job is primarily involving files on your local system with very little connectivity or data exchange with your company's servers.
The connection provided by my company at my work site (in a corporate office building) caps out around 20 Mbps, with congestion during peak hours pushing it down around 8 to 10 Mbps down, 2 to 4 Mbps up. Everyone complains that we can barely get any work done.
We have another network in the building specifically for the purpose of interacting with one of our customers; it's a direct fiber link to their headquarters' gateway. Their network has something like 100 Gbps of available bandwidth, with individual clients able to effectively max out the gigabit ethernet port, even during peak hours. We *love* working for this customer because we're so much more productive than when we're forced to use our company's internal network.
So, at this point, we have: myself, for whose job a 10 Mbps connection is grossly inadequate; and *you*, for whose job a 4 Mbps connection is apparently adequate or even good. That's a small enough sample as to call either one of our accounts anecdotal.
The problem is that you can't really look at a population of people overall and just pull a number out of your ass and say "that number is enough for anyone!". When you do that, you are effectively limiting the professions and the types of work that person can pursue from that location. I'm sure I'm not the only one whose job requires a bit more than 4 Mbps, and I'm sure you're not the only one whose job could be accomplished with a 2800 baud modem. There has to be a balance somewhere, but I think it lies a good ways north of 4 Mbps.
It's incredibly narrow-minded for one to think that:
(1) This grant program is exclusively for "working poor" (or poor people, period) to get Internet access; and
(2) All a poor person would/should/could want to do with the Internet is upload resumes so they can stop being poor / living in rural areas.
This post displays a stunning lack of empathy for people who have made a choice to live in a less densely populated area. First of all, not everyone who lives in a rural area is poor. There are plenty of middle-class and upper-middle-class people who have larger land holdings or have fewer neighbors than the ISPs would ideally prefer for instant ROI. Some of these are technologists, or potential customers of technology companies. They might be gamers, media consumers, software developers, architects, or any number of other professions where access to serious Internet connectivity can be a huge boon.
Not only that, but if their employer is amenable to the arrangement, having access to a modern Internet connection might even allow these people to work from home. This saves them thousands of dollars in fuel per year vs. having to drive many miles into the office, and also prevents the vehicle emissions that would have been necessary for them to get there.
Having access to a very fast Internet pipe can change someone's life in a revolutionary way. It can enable people to embark upon careers that are not possible without a fast Internet connection. To name a few:
- Youtube/Twitch/etc. streamer (gamer, political, scientific/educational, or other content) - not possible without fast upstream because otherwise video upload times make it impractical
- Podcaster - same as video above, but requires less upstream because it's just audio
- Pro gamer - without a fast connection for rapid access to new games and patches, and low latency, one cannot reach the pinnacle of pro gaming, no matter how hard they try (unless you intend to become a pro at a turn-based game where latency is relatively unimportant)
- Remote knowledge worker / engineer - A lot of the top tech companies are hiring talent remotely without regard for where they live. It's only "old world" businesses, and those with extreme security requirements like working for contractors or government touching classified data, that have hard requirements for working on-site. Almost any job in IT, plus many jobs in management, finance and administration, can be done remotely just as well as face to face... but only with good Internet connectivity. If your work output primarily involves taking information in one form (instructions, requirements, source data, whatever) and transforming it into another form (source code, metrics, stats, equations, papers, emails, whatever), you can work remotely if you have a good computer and a fast Internet connection. The remote work movement is helping to conserve a ridiculous amount of transportation fuel, but it can get even better in the future, and fast Internet is a critical first step.
The Internet is not only used for uploading resumes. Stop thinking like someone stuck in 1996, and recognize that this program is intended to open up possibilities for the rural population (about 15% of the US population, over 42 million people per the latest census) that crosses economic class boundaries and would not be possible otherwise. As someone else said, these ISPs simply would never, ever bring service to these remote areas if not for government subsidy, because these areas don't otherwise meet the company's ROI goals. These people should not be penalized for living in rural areas. And by the way, they're paying taxes, too, so it's part of their own tax money that goes to fund these programs.
Any time your data is being stored in an unencrypted format (or encrypted with keyword indexing, whatever, same difference) on a server you don't control, you should bring with you the *expectation* that the company hosting your data, and/or potential political or corporate adversaries, can and WILL access that data.
If you're OK with that, then more power to you. I doubt NSA or a corporate competitor cares about your pictures of playing fetch with your dog. They might care a bit about a copy of a secret agreement like the TPP, though.
Actually, they probably already have a forced MITM proxy (requiring you to trust a self-signed root CA from the gateway) that decrypts all traffic going across the wire.
If they have such a MITM -- and I can't imagine that they don't, since I know this is an extremely common thing in Federal IT -- then they would actually gain enormous insight into *who* is leaking, simply by allowing people to connect to these sites like Cloakroom and observing the traffic.
I will concede, however, the argument that if you perform an off-the-record (sneakernet, etc) key exchange with a trusted individual in order to begin your transaction with something like Cloakroom, then you could still privately communicate with Cloakroom even if the hostile gateway is performing MITM. All you have to do is paste your ciphertext into a text box on an HTML form and submit.
In that case, the Fed would know *that* you sent something to Cloakroom, because they'd know the IP of the computer that submitted the request and who was logged on via their PIV card, but they wouldn't know what you said -- whether something completely innocuous like "Nancy Pelosi has weird eyebrows" or something actually politically impactful.
Most corporate Exchange servers are either behind a VPN or proxy; have some third-party authentication wrapper (or even worse, an in-house custom one); or straight up don't let you access it without being hard-wired to their network with a "blessed" computer.
Most people who use an Exchange server for work won't be able to use this...
I use Apple products only when they make sense for me. Not sucked into the reality distortion field. The iPhone is a great phone (mainly because of its software), but the Apple Watch is too flawed of a device to be desirable in any way.
Of course, but that's only 10% of the job. The real hard parts include things like:
1. "Scripting the scripts": learning how and when to use each script and in what context. Often it is difficult to express exactly what conditions may prompt a given script to be run, especially if those conditions involve human factors (soandso forgets their password; the company gets sued and needs to put a legal hold on email deletion; etc.) If people could write scripts to automate the triggering of lower level scripts under the right conditions, they'd already have done that (for example, scripts that merely need to be run periodically get set up as a cron job) because we all know that most sysadmins are lazy and will seek to automate things to make life easier for them -- unless those things turn out to be difficult-to-impossible to automate.
2. Change: except for a few niche areas like Certificate Authorities, IRC servers, mail servers, and OS package repositories, the software configuration of many environments necessarily needs to change over time. Not only to keep ahead of obsolescence curves (end of security updates / support from vendors, etc.) but to actually provide new functionality and features requested by the customers (external or internal). There are very few server services that can just remain static for decades. The more change and upheaval you have in your environment, the less valuable it is to "button everything up" into a nice, well-oiled, automated machine. And the process of that automation requires significant manpower and extremely good documentation of exactly what everything does. Even in the best case, though, you'll probably end up re-doing large parts of it every six to seven years, since that's the support period (security updates lifecycle) for RHEL, Ubuntu LTS, etc.
3. Cost-benefit: Higher-level server automation can't write itself, because you have to tell the computer when to do what, under what conditions. Writing management engines that intelligently keep track of this stuff requires significant development time, and if the conditions are complex, you'd better write it in a robust language like C/C++/Java/C#/Rust/.... instead of Bash (shell scripts are not especially good at error handling and are not especially well optimized for large bodies of code). So after you have invested all that time in whiz-banging the whole thing into a web app with a few buttons, is it really worth the savings vs. waiting for a sysadmin to type a few commands manually into the console?
4. Frequency: Many (most?) system administration tasks are not done with terrible frequency. If something's only done once or twice a year, it's almost guaranteed not to be worth automating, even if the process takes a full day or two for the admin. So we typically only focus on automating tasks that need to be done multiple times per week. The more frequent, the more automated it probably is.
5. External environment: This mostly falls under "Change" above, but major vulnerabilities like the SHA1 weakness, Heartbleed, etc. occur once in a while and they can require an arbitrary amount of major change and upheaval on the server side, sometimes with very high priority (timetable doesn't allow for figuring out how to automate it). These things can happen at any time and they require someone to always be standing-by, at the ready, possessing the general knowledge of the system architecture as well as detailed knowledge of the commands to make changes on the system, so that they can implement the solution to the problem (which can't possibly be known in advance because the problem comes from an external source, so you can't automate it in advance).
Until/unless we get True AI (which I don't believe will ever happen), the world will need sysadmins. While the general trend towards higher level automation continues, it's not such a severe problem that admins will just get laid off if they don't know how to code. I don't buy that for a second, even though I strongly believe that it's beneficial for admins
Anyone who trusts their data to ext4 is grossly misinformed and playing with fire. Almost any other filesystem with a journal is less likely to eat your data than ext4. If you're using ext4 it should be for data that you really don't care if you lose / if the system running it crashes until you have a chance to reboot. In other words, it's fine for laptops where people check mail and browse facebook, but it has absolutely no place in the datacenter.
It's worth mentioning that a subset of the Linux kernel copyright holders, including some of the most influential and prolific contributors, believe that the situation around the binary modules from Nvidia, AMD, etc. is similar to this ZFS module.
Just because it's gone unchallenged for a number of years doesn't mean that the issue is settled or buried. It's never been challenged or ruled-on in court AFAIK, and until it has, it's anyone's guess what the law will actually decide when interpreting the GPLv2 and copyright law.
It really doesn't matter if you want to label individuals as "extremist" for thinking a particular way about the law, though I'll admit that's a better term than Barbara's use of the term "freetard" -- again, namecalling just weakens one's argument. The only thing that matters is the way that a judge will ultimately interpret the relevant texts to determine if there is any violation of law, and what the penalties are if so.