There are still things I'm waiting for smartphones to do, so it's not like the companies can just stop releasing new smartphones and keep everyone happily on the Galaxy S4 for 8 years at a stretch. Until they've solved the things I want out of my smartphone, there's still work to do. Let me see...
Foremost has to be battery life. I'd like to see smartphones with much better (week-long at least) battery life, even while actually using it (most high figures for battery life are if you keep it in your pocket).
I'd like to see the return of front-facing fingerprint readers, or at least put them somewhere more accessible like on the power switch. I'm not cool with the "reach-around" maneuver on modern Samsung phones, and I end up having to turn the phone over and unlock it with two hands. Unacceptable. Face ID and the like aren't reliable for me because I wear glasses and I'd rather not have to take them off each time I want to unlock my phone -- that's even less convenient than tapping in the PIN.
"But Apple has that with the iPhone 8 and 8 Plus", you may say. But Apple doesn't have something I really, really want from the Note line: the S Pen. Give me a front-facing fingerprint reader ANNNND an S Pen on the same unit, please, without it being hopelessly outdated like the Note 4. (The Note 4 has many other problems, too, like an extremely inefficient and power-hungry CPU, pathetically slow NAND by current standards, and a very inaccurate and finnicky fingerprint reader.)
Also in my list of things Apple doesn't have that Samsung does -- that I want -- is the ability to containerize apps. Putting an app in a space where it only has access to the data and apps within that container is truly an amazing feature for people who have separate work and home lives, and even multiple home lives. Multiple Steam accounts, Discord accounts, Skype accounts? No problem; Samsung's container feature can do that. Apple is behind on this.
But Samsung isn't ahead in every category. Apple's TV app is far better and easier to use than anything Samsung has to offer, because you don't have to keep logging into 9 bajillion content-serving apps to watch videos. You just search for what you want and play it, oftentimes without knowing or caring which app is serving you the content.
Still more things that Apple is legitimately better at? Oh, yes. The Apple Watch is possibly my favorite device of all the smart devices I've ever owned. I've tried the competition; the Gear S3 Frontier doesn't hold a candle to it. In terms of ease of use and features, the Apple Watch is far more convenient than anything else (I have the Series 3 with LTE). Why, you ask? Well, being able to send and receive calls and SMS from your wrist is amazing. With prominent notifications, you can get a vibration strong enough to wake you from sleep when your alarm goes off, or if you receive an email from someone important, or even a text message. You can of course customize the notifications so you don't get startled awake for something trivial like someone uploading a Youtube video, but it is fantastic for people who need to be "on-call" on their job, or for family members who may need to reach you.
Apple Watch, continued: I can also tell someone what the weather is like (and what it'll be in hourly increments for the rest of the day) within about 3 seconds, even if my phone is in my pocket, without reaching for it, unlocking it and doing a bunch of tapping to get to the weather app. Just ask me anytime what the weather's like, what's the temperature, etc. and I'll tell you without so much as a pause in the conversation.
More Apple Watch: It also continuously monitors my heart rate and rhythm. I subscribe to the Stanford heart study, which is an app that will actually notify me if it detects an arrhythmia. The device may not be a piece of "certified health equipment", nor is it a pacemaker, but it's better than nothing for people who are on that line where they might have a family history of heart conditions, but haven't been
If the telecoms, ISPs and cable companies would build infrastructure for landline Internet and TV that was half as rugged as even something as mundane as city water or electricity service -- which, by the way, are far from infalliable and have plenty of issues -- people would be a lot more satisfied with their Internet. In reality, I'd say the majority of the people I know have line quality issues with their landline home Internet, and intermittently or constantly experience some level of packet loss varying between "makes gaming difficult/annoying" to "pretty much nothing is usable most of the time". Customer support only matters if there's a problem.
What the telcos and ISPs are doing, though, is they use the absolute cheapest shit they can find. They install indoor-grade Cat5 on the outside of the house. They use antiquated coax cables that are notorious for having signal issues unless you get them connected just right, and even then, they waste so much energy that the connector is warm when in use. Imagine if your USB or HDMI cables were warm, what that would say about those cables!
They also do a lot of installation on shoddy telephone poles. This needs to stop. They should bury the cable, and when they do, it should be jacketed in a VERY robust sleeve that will resist pinhole penetration and just about anything else for at least 50 years. The cables should have state of the art EMI interference and should not interfere with one another or any other installation in the same right-of-way.
When they install customer premise equipment, it needs to be enterprise-grade. Don't cheap out and buy a router with the absolute minimum SoC that will kind of do the job for the average user. Build something that's just going to goddamn work, even if the user decides to throw 20 devices at it with several thousand TCP connections. Don't give me this cheap shit that only has enough RAM to track 50 or 100 TCP connections.
This should all be self-explanatory, obvious, and universal. Even if you don't give a crap about your customer and just want to do the absolute minimum, you are saving $500 now to spend $10,000 later on support and service visits. Gas, time and labor, materials, etc. while your field techs and call center employees have to deal with the crappy setup you gave your customers. If you'd have done it right and done it robustly the first time, you would be able to vastly reduce your support staff, and your field techs would only have to come out to visit peoples' homes when the customer does something stupid, or the box gets struck by lightning and has to be replaced or something.
It seems that most residential-focused ISPs -- and honestly, even "business" focused ISPs except for those that are actually building out a proper, dedicated, datacenter -- have the mindset that their customers are just going to casually use the service lightly once in a while, and if it's down, they just go "oh well, the Internet is down" and go on with their lives. But we're no longer living in a world where occasionally reading the news is the limit of what people do with the Internet, just like we're no longer living in a world where all people do with electricity is turn on a reading lamp at night.
At least where I live, the power grid is pretty resilient against physical damage, even in the face of severe thunderstorms, tornadoes, lightning, and that kind of thing. If something in the "middle" of the grid goes down, like a substation, there are usually alternative routes to get the power there and people only see a brief disruption in power. If something local goes down, it's because there's been a local catastrophe, or perhaps -- once in a blue moon, but certainly not with any regularity -- there's a genuine problem with the equipment, like a bad transformer coil.
By comparison, Internet service of any description delivered via landline is a joke. Datacenters like to measure the number of "9s" they get after the 99 in their 99.9999...% uptime; consumer and residential ISPs seem to think
Maybe it's your software (or the COTS you bought) that's buggy? Zing is completely compliant with the TCK, which is a stupidly extensive test suite that ensures the JVM complies with all guarantees the Java standards make.
In other words, if you run your software against two JVMs that are in some way significantly different from one another but they both comply with the TCK, and your software doesn't work right, then your software is most likely making assumptions about the implementation that aren't guaranteed by the standard. This is *never* a good idea on any platform, especially for a language that has such an extensive, well-documented standard as Java.
Folks have similar issues with non-standard C++ code, which either won't compile or won't run properly when you move from one to another standards-compliant compiler. But this is just another reason to put in the effort to develop standards-compliant code. Those who don't, are technically at fault whenever anything breaks. If your COTS is non-standard Java that only works on the Oracle Hotspot JVM or other products based on it, I feel sorry for you, but you should complain to the vendor or stop using that product.
Actually, due to the way copyright works, his opinion still doesn't matter for new drivers. Any new driver or component added to the Linux kernel source would be considered (necessarily) a derivative work of the Linux kernel, because it would directly reference and depend upon both the specification and the implementation of probably hundreds of internal Linux functions -- for memory management, data structures, algorithms, informing the kernel about the behavior of the driver, logging, etc. I've read a few actual copyright lawyers' publicly-released opinion about this stuff, and they all agree that because the GPL enforces its terms on anything else that's considered a derivative work of a GPL work, that code would necessarily also be under GPLv2.
So, Linus could technically speaking accept a file or driver into the Linux source tree under some other license, but then Linus would be violating the copyright license of an untold number of other Linux copyright holders, whose code he is NOT entitled to re-license or to violate the GPLv2 license of those others' code by mixing it with non-GPLv2 code in violation of the "derivative works must stay GPLv2" clause of the GPLv2. Linus would potentially be on the hook for civil damages in the US at least, and he's living and working in California for a long time (probably long enough to be a naturalized citizen), so he'd most definitely be on the hook if he distributed that code to others.
Someone who contributed one or two patches to the kernel would have a tough time convincing a judge that there's any actual copyright infringement, but someone like Greg-KH, Ingo Molnar, or pretty much any Intel, Google, Red Hat or Microsoft (yes, Microsoft) employee who contributes to the codebase, would have probably touched enough internal APIs by now that certainly they'd have a claim on something that a non-trivial module would need to use. And there are a few contributors (past and current) who very vigorously believe in a "strict" interpretation of the GPLv2 and would actually be militant about enforcing their own copyright in the kernel code.
Regarding Linux, Linus Torvalds' own opinion is completely irrelevant, unless you wanted to use an extremely antiquated, practically useless version of Linux from the 90s.
The actual Linux kernel code is an incredible mish-mash of thousands of contributors' code, both companies and individuals. Each passing day brings more and more copyright owners into the Linux kernel, because each person retains copyright over their individual contributions, and each day brings at least one new kernel contributor.
In order to change the copyright license of the current Linux source code, ALL of the contributors of the current version of Linux -- or their estate in case of deceased people, or the liquidator company in case of bankrupt companies -- would need to agree to the license change. Even assuming that every last contributor could be convinced to agree to change the license, there would still be logistical problems in actually contacting all of them. Deceased contributors probably have invalid email addresses on file, and likewise for contributors who sent in patches from a corporate email address and have since moved on. Just the task of contacting the legal copyright holder of every line of Linux source becomes a nearly futile task, even if you go back to a hideously old and putrid version like kernel 2.0 or 2.2.
You might think that you could discombobulate some of the "long tail" of contributors by just removing any source lines (using a Git script) that were contributed by someone not in the top 100 contributors. If you did that, you might wonder, wouldn't that leave an almost-working kernel that the top 100 contributors could then carry forward with, fixing the little one-liners that drive-by patchers gave them here and there? And then you could surely hunt down a mere 100 contributors (or their estates) and change the license -- right? Well, no.
Per this -- http://www.remword.com/kps_res... : there are 19,817 kernel contributors just since 2005 (and quite a few more before then, as Linux was indeed quite popular and noteworthy among engineers in, say, 2004; people even made a halfway decent desktop OS based on it by this time). Half the people -- 9912 to be exact -- contributed 2 or fewer patch sets, which only amounts to 1% of the total patch sets. If we assume that, on average, over a large dataset, any two patchsets are equally likely to be any given size regardless of contributor (which may or may not be a safe assumption), that still means that 1 out of every 100 lines of code in the Linux kernel would have to be independently re-written by someone else if we removed the long tail of copyright owners.
And that would only reduce your total number of contributors by _half_; you'd still have 9000-some contributors who have 3 or more patchsets to their name. I wish that site would tell us what percentage of the codebase is contributed by, say, the last 15,000 people in the list of contributors sorted by number of patchsets. It would probably be something ridiculous like 20 or 30%, meaning that if you wanted to whittle down the number of copyright holders in the kernel to about 4000, you would have to re-write (independently, without peeking) about 1 in 5 lines of code on average. Eww.
Not to mention that randomly removing lines of code (or small but important fixes) all over the codebase would create a mess that probably wouldn't compile, and once it did compile, it'd need to be heavily tested, debugged and fixed just to get it nominally working on modern hardware.
So, yes, peoples' opinions about what the license of the Linux kernel "should be" are completely and totally irrelevant. We can't remove the code from a significant number of contributors to the kernel to whittle down the list, because doing so would spark a many-year project to get the kernel back to some semblance of what it is today. And we can't contact everyone and try to get everyone to agree to a license change, because you're probably going to be sim
Azul offers OpenJDK builds for a lot of platforms using a product called Zulu, which is free of charge if you don't want any support. IMO they're better than Oracle's OpenJDK builds because you get more platforms. I think Zulu's might also continue to get security updates for longer than Oracle is willing to provide them for old versions of Java, so if you're stuck on Java 7 or 8, this is a great alternative. Of course, updating your code so you can jump to OpenJDK 10 is better, but sometimes that can take a long time for projects hitting worst-case issues with backwards compat.
Looking at their site, they seem to offer another product that claims better latency consistency, called Zing, that is non-free. So that tells me that Zulu is mostly unmodified OpenJDK builds (although they could be marginally faster if they are compiled with different option flags or a better compiler than Oracle uses). Zing is something else entirely.
P.S. - I am not a shill for Azul. I've never done business with them, worked for them, or bought their products. But I have downloaded their free OpenJDK builds and find them much more convenient to download (with fewer nags) than Oracle Java or Oracle OpenJDK.
Android is an example of an OS that uses close to nothing from the GNU project while relying on the Linux kernel (and even then, its fork of the Linux kernel that ships on real devices is pretty far removed from the vanilla Linux kernel, with custom security modules, proprietary drivers, firmwares, etc.)
Over the years Android has replaced GPL or LGPL licensed GNU project tools with BSD or MIT licensed ones, either out of spite or because they just didn't like the license. There was never any functionality problem inherent with these tools that caused them to make the switch, and their replacements aren't significantly better.
I certainly wouldn't call modern Android "GNU/Linux". There's no GNU in it!
It's actually a FREE SOFTWARE project, not an OPEN SOURCE project, as he would tell you if he read your comment.;) But
Contributions made to upstream glibc have their copyright attributed to the FSF once they are turned over to the FSF via a copyright assignment form. See: https://www.gnu.org/licenses/w...
Since Stallman is the President of the FSF, he basically has direct authority to modify the version of glibc on the FSF's servers, because he is the highest authority of the non-profit organization that owns that copyright.
What Stallman *can't* do, due to the license that he wrote and gave to glibc, is stop you or anyone else from modifying any aspect of that code once you download it. And he isn't trying to do that here. So indeed, a fork of glibc (which has happened before and was successful; see eglibc, whose changes were eventually merged upstream) would be the only way to have a publicly-recognized copy with this modification. But it wouldn't be *the* *upstream* glibc anymore, even if you called your repo "glibc".
Stallman also technically has the authority to change the copyright license on FSF's version of glibc to any other license, which he did when they made it LGPL v2.1+. Effectively, instead of glibc just being only available under the "Library General Public License v2" (an older version of the license), it can now be released under LGPLv2.1 *or* LGPLv3 by downstream distributors. But that license change only occurred because Stallman assented to it. He could change it to plain old GPLv3 if he wanted, or he could declare the FSF's copy to be proprietary software if he wanted - he's allowed to do that under copyright law. But that still wouldn't modify the license of any copies that were previously distributed.
Jan Koum is a broad? Or some other broad clashed with their parent which is causing Jan Koum to leave? Also, why are we telling Facebook about this in a Slashdot story headline?!
What is Poetteringware, why should I be afraid of it, and how do I get it? I've never run into a single problem where I had to go out of my way to change something on RHEL7 just because it's using systemd.
Sounds to me like you're just too complacent to adapt to change.
The Windows Subsystem for Linux (WSL), while a neat project, is basically scoped to only be able to run a small subset of all software that you can run on a proper Linux kernel. So "Embrace, Extend, Extinguish" (EEE) would only be possible for a small number of use cases.
Things that don't work now and probably won't work any time soon:
- Raw sockets (i.e., IPv4/IPv6 sockets that aren't explicitly using a specific session layer like TCP or UDP)
- Native support for X11 apps with OpenGL hardware acceleration, without having to pipe the whole thing over the IP stack to a third-party X server that comes with a huge performance overhead because you're not using Direct Rendering
- Deep and comprehensive integration of desktop environments, window decorations and widgets from Linux apps (both KDE and GNOME/MATE/Cinnamon/etc., plus a few less-common ones like LXDE and XFCE) so that a Linux environment can co-exist comfortably on your screen with Windows apps without duplicating things (for example, adding all X11 windows to the Windows Taskbar's window list instead of using a KDE or Cinnamon or GNOME window list)
- Support for Linux Containers
- Deep enough support for Docker that any random Docker image you grab off the Internet and try to run in WSL will "just work" without a hitch (discounting the "Docker for Windows" project that basically starts a Linux kernel in a VM and runs Docker in there, but that's not the same thing at all and comes at a significant performance and overhead cost)
Things that are *basically* impossible per the design of LXSS, and can never happen:
- Support for custom kernel modules (the functionality of specific kernel modules like tun/tap might be supported eventually, but not *any* generic Linux kernel module)
- Support for all the various extended attributes and ACLs of any possible Linux filesystem (basically implementing the Linux VFS comprehensively, not just doing path translation like they do now)
WSL is neat, but a lot of things don't work and probably never will because of the many differences between the design of Windows and Linux and the impracticality of tying them this closely together.
Althoooough, if they implemented the userland aspects of the Windows desktop on top of a Linux kernel, *that* would be sort of an EEE. You *could* then natively run anything that'd run on Linux because your kernel would be *Linux*, and the only remaining concerns would be about drivers that exist for Windows but not for Linux (but if they went this route, I'm sure vendors that continue to support their products would grumble and reluctantly provide Linux drivers to Microsoft, in droves).
Good luck proving in court that "because his packets were this size, he *must* have been viewing porn". If that's so accurate, VPNs would be worthless in repressive states like China and Iran (and apparently Rhode Island now, too, lol) because they could just *tell* that you're visiting a site they don't like, and shut you down.
But VPNs are still very effective in many countries if your endpoint isn't a "known" VPN operator. The other option is for the ISP operator to maintain a strict whitelist of which IPs/websites are reachable from their network; a few countries have begun to think that this is the only true way forward, because if you default to routing any traffic, it's still trivially easy to bypass just about any filtering or deep scanning attempts using standard crypto like TLS.
Sure, you can probably inspect the timing and bandwidth utilization of an encrypted connection to distinguish between streaming video, working in an online office suite, uploading a video to a streaming service, or viewing a restaurant's menu, with high accuracy of being able to at least rule out one or more of those categories. But being able to tell which actual *website* is being visited, or what content is being consumed? That seems very unrealistic to me. The degree of confidence you'd have in your assertions would be, at best, around 50% or so, and almost always much lower than that. This sort of "suspicion-based reasoning" wouldn't fly in most courts in countries that uphold basic human rights.
Oh, and any timing based traffic analysis deductions can be easily defeated client-side by inserting random, non-deterministic jitter into all outbound packets. Since the server endpoint's send rate is also dependent on your client's responses (TCP ACKs), you can effectively control the delay in your server endpoint's responses by introducing small amounts of random latency into your own client's ACKs. Then you can further muddy the waters by having the endpoint pollute the encrypted tunnel with nonsense data. The most accurate conclusion that could be claimed with a high degree of certainty thereafter would be "They seem to be using a lot of throughput for some reason".
Indeed, any justice system that would allow such leaps in logic based on packet size and timing analysis (while having no idea of what the actual contents of the datastream contained) is not a justice system I'd want to be subjected to. That's getting dangerously close to guilty until proven innocent.
For most of my life, I've been a PC user (Linux when possible, but always a Windows desktop hanging around for gaming and certain cross-platform development). My use cases span all sorts of different application areas, from gaming, to productivity, video and image editing (Adobe CC products), software development, and even dabbling a bit in scientific computing.
Since about 2015 I've gradually started buying into the Apple ecosystem, and while it can never displace my Windows desktop, their devices do have distinct advantages in certain spaces. The reason I buy them is for these specific advantages, and I do so while acknowledging their drawbacks. I make informed product purchases that are not really based on emotion, but on specific qualities or features that make them better than competitors.
For example:
(1) iPhone: My first Apple product was an iPhone 6S Plus. I'm actually glad I avoided the 6, because that device is fatally flawed by design because the aluminum is too easy to bend, causing Touch IC disease. I originally bought the iPhone because I was sick and tired of security patches taking months to land on Android devices, and the sometimes extremely arbitrary and short support cycles for Android devices. Also, I've felt that overall Android devices and applications are buggier than iOS and most big iOS apps. This is with my experience on Android mostly ending around the Note 4 era. My suspicions about Apple taking smartphone security more seriously than anyone else were confirmed when their chief of security gave a talk on iOS Security at Blackhat, providing a deeper insight into the tremendous security architecture of their platform. Oh, and they now have a bug bounty program like every other good technology company (as of 2016).
(2) Macbook Pro: My first MBP was a 13" of the last gen before they introduced the Touch Bar. When the Touch Bar came out, I handed down my non-Touch Bar MBP to a family member, and got myself the Touch Bar 13". The killer features of the MBP hardware, for me, are TouchID making it both fast and secure to unlock my Mac, and the extremely good touchpad. Palm detection is flawless, it's perfectly sensitive, and clicking is faster and easier than any other touchpad I've ever used. People who think their own Windows laptop has a good touchpad haven't tried a 2016-or-newer MBP touchpad. Seriously, it's so good that I prefer it over a mouse for all use cases except FPS gaming.
I don't think the MacOS platform itself is anything incredible; it's approximately a sidegrade from Windows 10 in terms of what it can do. In some ways Windows 10 is better; Nvidia, Intel and AMD generally provide better support for OpenGL extensions on Windows than they do on Mac, so if you're using OpenGL (or developing with it), you probably will be frustrated by the level of GL extension support on MacOS. This is entirely due to Apple's inattention to that and focusing on Metal. That said, I subjectively think MacOS's disk encryption, tiered storage (Fusion Drive), overall filesystem design (APFS), and application installation management (.dmg/.pkg) is better than the scattered mess of Windows.
The main killer feature of MacOS, for me, actually might end up being supported on Windows just as well soon due to Ubuntu on Bash on Windows; anyway, that feature is the excellent UNIX compatibility. Homebrew is a lifesaver. Being able to run my favorite FOSS with basically an "apt-get" (okay, "brew install", same difference) is awesome. Windows isn't quite there yet, because they have a lot of missing features; just installing a random package, there's a fairly high chance that some functionality in the translation layer isn't there, and you won't be able to use the program. Homebrew actually compiles all the FOSS UNIX software I need *natively* for the Mac platform, since the POSIX, etc. standards are baked into the core of MacOS (based on BSD), but are just now being tacked onto Windows in a new (and largely incomplete) subsystem.
Except that the people who would take your image to post as a one-off image macro on imgur, or to use in their homework for a PowerPoint presentation, aren't likely to be able to afford the exorbitant fees you'd charge to use your image anyway. If you want to get paid for your work from the little guy, charge prices little guys can afford, and make it very easy and quick to do so. Otherwise, people will always (and I mean always) find ways to get around your smug attempts at preventing them from using their technology to accomplish their goal. Hell, they'll print out the page, scan it and crop the scanned image if they have to.
Or, you know, if you don't want people using your image, don't post it on the public Internet to be viewed by anyone. Give a thumbnail or a heavily edited / blurred version and make someone go through a paywall to see the full image.
Welcome to the 21st century, where just because you think you get to have absolute control over every possible usage of your copyrighted works, doesn't mean that you get to actually enforce said control, nor does it mean that enforcement is at all practical for you. If you gave consumers no other choice but to buy your content, they MIGHT acquiesce if you made it extremely affordable, but if you tried to charge the exorbitant fees I see floated around (hundreds of dollars per photo, if not thousands), people will either not use your content (and use the content of your competitors, who are cheaper and easier to access), or rebel against the platform and switch to something like desktop Linux to escape the jail.
The only reason iOS isn't dead is that it still provides some pretty solid alternatives to the DRM, vendor-locked ecosystem. You can still install VLC. You can still take screenshots. You can still save images from your browser and upload them to Google Drive.
The iron-fisted rule of law and the rule of imperious content creators who want to own each and every use of their works is the outdated model, not the model of free software and open content. When content hosts, search engines, etc. put up barriers to people using content the way they want to, the network naturally just routes around the damage. Won't be long before there's a Chrome and Firefox extension putting the "View Image" button right back into the Google Image Search.
It doesn't matter what you think; it's happening whether you want it or not, and too many people do it to hope to stop it, unless you imprison every man, woman and child with access to a computer or phone.
Sure, you might be able to elicit a settlement out of Google, but you're not exactly solving the problem. People are still going to use Getty images without permission. Why don't you learn from the way the Internet works, and do like everyone else does, and just route around the damage? Figure out a new way to monetize your content, instead of expecting the world to continue to pay you the same way they always have. The world has moved on. Time for you to do the same, or you'll have to find a new career.
P.S. -- What happened to democracy? When the vast majority of the people unthinkingly do some thing, it seems really backwards to have that thing be illegal. I'm not talking about 20 or 30%; I'm talking about basically everyone (except, perhaps, for people like you, who conscientiously object to the new way of doing things.) I guarantee you that over 95% of people with frequent access to a computer have committed technical "copyright violation" (according to the letter of the law) not once, not twice, but at least a dozen or so times in their lives, and that's even accounting for the many such uses that would be considered Fair Use.
Some things, like murder or car hijacking, won't ever become that popular because the crime is self-limiting; you have to physically take something away from one person to give to another. Nicking someone else's photo or music or video and using it in another context for personal use is literally harmless to the person who produced the ph
Uninterested unless they have the bravery to tell us what actual GPU chip is inside each model. CPU has not been a bottleneck for my workloads for many years, but GPU definitely is. If this is still Maxwell, I'm definitely not interested (and will be VERY annoyed).
It seems to happen approximately daily when consuming open source projects. I'll get so far down the path of trying to build out my project, then hit a broken feature that's actually (to me) a killer feature of the program/framework/library/IDE/etc., but the developers couldn't be arsed to implement it properly / keep it up-to-date / fix severe bugs with it.
Just in general. Because it happens so often that I can't list all the cases where this happens to me..NET CoreRT would be awesome if it could compile native DLLs that can be called from C, but nope. Lead devs have no interest in implementing it.
Go would be awesome if it could compile native DLLs that can be called from C, but nope. There are branches with working code out there, but nobody's managed to package it into a release of Go. Which kind of sucks, because compiling Go from source is horrible on non-Linux platforms.
Eclipse Dali's SQL to Entities reverse engineering code has been broken since 2011; it doesn't pick up most relationships and constraints between tables. So I have to use NetBeans for my SQL to JPA entities generation. Bah.
I guess I'm the "never lucky" developer -- anything I try to do, the platform I'm using is fatally flawed in exactly the area I'm exploring. But the marketing material makes it sound so great and perfect...
The "Parable of the broken window" was written in 1850. Let me clue you into something: NO ONE in the 19th century, and next to no one in the 20th century even, could have anticipated the sea change that is pervasive automation that we're seeing today. It's nowhere near the same thing as what we experienced in the industrial revolution with tractors displacing farmhands. The changes that are happening are different in kind and require different economic theories to reason about them, because the parameters involved are different.
The jobs being displaced are of entirely different categories than mindless or repetitive jobs. Kiosks are only the first step. Software developers and architects, financial analysts, dozens of types of first-level and middle managers, and even CEOs are potentially on the chopping block in the next 50 years, as machines can do their jobs cheaper and more efficiently.
Just because the economy has more money doesn't mean that money is being spent on paying out wages to people. It could be spent on paying other billionaires for use of their automated systems. As the cost of maintaining these systems decreases with reliability and durability improvements, there just isn't much need for human labor. And despite the population increasing, what I think you'll see is a much slower growth in total wages paid out to the lower 99%, or even stagnation, and sharp decreases during any hint of economic downturn of any magnitude. Meanwhile the top 1% will see exponential growth in wealth, as the 1% always retains the means to keep getting richer, and now we have governments around the world who are willing to meddle in the markets to ensure that the top 1% continue to gain in wealth even as the economy overall shrinks.
As long as our economy operates on the theory that you have to work to earn money to buy things like food, clothing and shelter that you need to live, we need productive jobs to put people in that they can perform to earn that money; otherwise they'll be homeless beggars or looking for government entitlement programs to keep them from starving to death.
If your solution is to let them starve, you're a heartless bastard. You try starving and see how much you like it.
If your solution is to pay them a universal basic income, you're on the right track.
If your solution is to wave your magic wand of 200 year-old economic theory and say "don't worry, as long as there's money people will have jobs", you're burying your head in the sand like an ostrich.
Similar kit here, except replace the SSD with a 5400 RPM HDD and Windows 8.1 with Windows 10. And McAfee. Yeah. YEAH.
These are Haswell i7 boxes with otherwise extremely high-end hardware (for a business PC), running Windows 10 v1607 x64, basically the fastest version of modern Windows (compressed RAM is awesome). But they were too cheap to spend on SSDs, so we wait. And wait. And wait.
Outlook startup, cold cache: 68 seconds.
Firefox launch, cold cache: 35 seconds.
Eclipse launch, cold cache: 3 to 5 minutes. Same Eclipse profile loads in 5 seconds on my (personal) Macbook Pro with an SSD.
`mvn verify` (run some integration tests), cache irrelevant: 25 to 35 minutes (compared to about 6 seconds on our CI server with 128 GB of RAM and an array of SSDs).
Git clone of a ~3 MB repo: 0.5 seconds networking, 5 minutes disk I/O (compared to about 0.5 seconds of disk I/O on our CI server.)
I'm gonna go out on a limb here and say that your kit isn't part of the problem. The fact that you have SSDs is the biggest factor in ensuring that your kit *isn't* the problem. But if you had 5400 RPM HDDs like me (and I'm sure zillions of other IT shops), you'd be in a world of hurt, like we are. Doubly so if you use(d) McAfee.
And no, our network isn't the bottleneck. Our network is much faster than anyone needs it to be. We have enough bandwidth that every employee can get symmetrical gigabit and fully saturate the line speed of their gigabit NIC, simultaneously. And a comfortable bit of extra beyond that. We sit on a fiber trunk.
I know every budget has to be criticized by 'the opposing party' with a list of all the wonderful things that are going to be cut, but you all DO realize that the US government is nearly $19 TRILLION in debt - or more than $50k per person in the country?
Every single program that we're paying for, essentially we're living off credit cards. We are the wealthiest nation in human history, and we still cannot afford all the crap we want.
At some point, someone has to be the grownup in the room and say "you know, that would be really nice, but we simply can't afford it".
No one disagrees with that. That's fairly obvious. But what *parts* of the budget are you going to cut?
Any cut you make hurts someone. A large percentage of all government spending is on pay for employees or contractors. Cuts cause those employees and contractors to go looking for work, and the money they used to bring in will stop being poured into the local and global economy as they scrimp and save trying to survive while looking for work.Losing your job hurts a *lot*.
On the other hand, a small tax increase that affects everyone might cost each individual $500 per year or so (in the case of a rather extreme tax increase), but (1) that level of burden isn't going to push anyone over the edge, causing them to go from "making it" to having to sell their home; and (2) the ~$125 Billion per year that you raise from it will go to continuing to fund these programs, significantly slowing the rate of incurred national debt as long as we don't spend *more* than we already have.
The problem is that the Trump administration and congress don't want to raise taxes, but they're happy to cut programs that they are ideologically opposed to. But it turns out that many of their ideologies are just plain *wrong*, like "not believing" in climate change, pro-choice, or even the general welfare of the people. Their answer to "how do you decide what to cut?" is "things we don't like".
I'm mystified as to why these companies running mission-critical apps with $$$ on the line aren't using multi-region redundancy or at least failover. Imagine if some terrorist dug up the fiber lines leading to the Ashburn primary datacenter, causing US-EAST-1 to be offline for days.
This is why you spread your resources around and have redundancies across different geographical regions. That way, the worst that could happen is users might experience a momentary lag, or maybe a couple TCP connections might get reset, but as soon as they try again it'll be up and running like normal, except that they'll be talking to a server in California or London instead of Ashburn, VA.
Surprised that so many companies don't have redundancy that this ended up costing $150M.
I haven't used wired headphones in years -- been happily on the Bluetooth bandwagon since BT 3.0 was the latest spec -- and I already have plenty of USB-C cables, including C to C and C to A.
Also, to the extent that *some of* Apple's hardware is "So Last Year" compared to the Android flagships, some of it is genuinely ahead of Android by 6 to 12 months. For instance, A10 Fusion is still today the fastest production system on chip that you can buy in a fully-functional, consumer-oriented device being produced at high volume. Both its CPU and GPU are quite a bit faster than the nearest competitor from Qualcomm and the like. Samsung's Exynos is even further behind, because their serial performance sucks; they seem to be stuck in the same "MOAR CORES" rut as AMD was for a number of years while they were irrelevant and a non-competitor to Intel. Single thread perf is king for consumer workloads (which is most HTML rendering, JavaScript execution and game engines), period, end of story.
Sure, Qualcomm will soon eclipse the A10 Fusion with something faster, but by the time an Android manufacturer puts it to market, Apple's next gen SoC (A11?) will be out in an iPhone you can go buy at your corner Best Buy or AT&T store. And it'll be the fastest again.
On the RAM front, iPhones also lag behind by about 2 years (the 7 Plus has the same RAM as the Note 4 from Q4 2014). But this brings me to my next point: Software.
I used Android phones from three different manufacturers (HTC, Samsung and Motorola) from the Android 2.0 era up til KitKat. And I swear that with every single phone, with every single stock firmware, I encountered the following, reliably: horrible, annoying bugs in my daily use case paths; awful, choppy Bluetooth with anemic range; very poor battery endurance (never more than 2 years); and some degree of bloatware that was impossible to remove/disable that had very significant storage and/or RAM footprint. I also spent most of my time with Android running a device with a widely publicly-known root exploit vulnerability (several of them exploitable remotely over the network or through apps I used), usually with some open-source exploit toolkit out there that could exploit the vulnerabilities my phone had. This is because as soon as an OTA came out fixing an exploit, another one would surface, rinse and repeat.
Lastly, despite all Google's efforts, Android devices are still extremely laggy even at the ultra high-end and perennially have performance problems, even with custom ROMs. There are so many little lags and hitches here and there that it just feels like it's running interpreted VBScript.
In my two years of using iPhones, I've never encountered a single bug of any sort. Yes, the software is not perfect and has bugs, but they're so minor or such edge cases that I've never encountered them in my daily workflow.
The performance is consistently, almost perfectly fluid, with exceedingly rare performance problems and hitches. For every single noticeable lag I've observed on an iPhone, I would've observed about a thousand of them on my Android device by using it for the same duration with similar use cases.
And Apple actually fixes their security problems quite expediently. Many OTAs have been rolled out whose entire content is just security vulnerability fixes. And often those vulnerabilities had been disclosed within the past week, or less. So the exposure time once a bug goes public is usually quite narrow or nonexistent.
Also, Apple is assertive enough to give the giant middle finger to parasites like Verizon and AT&T when they ask about the prospects of installing bloatware onto iPhones. They simply say "No". You might say that they build this into the device purchase price, but I can always purchase an unlocked iPhone and Verizon won't make a penny off of that, so there really is no financial incentive available to a carrier selling iPhones, except that *users want them* and want to use them on their favorite carrier's network.
Sound like high minded excuse to start use the platform for political purposes. All these words "bullying", "fake news", etc. are code words involved in liberal virtue signalling. "Fake news" is something that those evil right wingers do (especially it does not apply to New York Times, et al. or any garbage coming from BLM or other such outlets).
Deception, coercion, half-truths and complete fabrication are not, and have never been, tools used exclusively by people with one particular political leaning or another. They're used by leftists, rightists, centrists, libertarians, conservatives, liberals, democrats, republicans, greens, independents, tea partiers, anarcho-communists, fascists, feminists, masculinists, and everyone in between or beyond.
It might be the case that a certain number of news outlets could be liberally biased enough to use these tactics to undermine right-wing political viewpoints, but this in no way prevents or exonerates those outlets which are right-wing, from using the same tactics.
If your complaint then becomes that there are too many liberal news sources and not enough mainstream conservative news sources, then you're basically saying that you want the news to present you with lies that agree with your personal political dogma, rather than lies that attack or offend your personal political dogma.
If you think that a change of color or movement along a right/left spectrum will in any way affect the frequency and severity of lies, deception and coercion used by the mainstream media, you would be plain wrong in that belief. ANY politically motivated organization, regardless of what agenda they're pushing, is going to distribute deceptive and patently false information, also known as propaganda, that supports the agenda they are being paid to push.
The only way to return news media to reporting on objective truths observable by scientifically rigorous methods, and away from speculation, hearsay, the passing of rumors and fabrications, and opinion-slinging (all of which are inherently biased toward some particular set of beliefs, and in the context of politics, toward some particular set of political beliefs), is to forcibly separate media from financial incentive. Capitalist media is always going to be propaganda for someone.
Also, to make your argument look even more silly, Java was already ridiculously popular before Sun open sourced the code. Before that it was (mostly) freeware, but companies of all sizes were also buying support licenses for proprietary Java back in the early days. Open sourcing Java just accelerated its popularity, because, in the early days of.NET, its competition was much more platform-constrained (Windows-only, before Mono) and pricey (required a Visual Studio license to unlock some features or, in the very early days, to even get a compiler).
In fact, if you appreciate the extremely open nature of the.NET ecosystem today, then you owe thanks to Java, because Microsoft only open sourced.NET and allowed/encouraged/fostered the Mono project and cross-platform packages in Nuget because of Java/Maven's advantages that were pulling developers off of the.NET ecosystem.
Today, in 2017, there's not much reason to use Java compared to.NET if the libraries you need are supported on.NET (and even if not, you can use IKVM if you're absolutely intent on not running the Oracle HotSpot JVM, but you'll still be shipping Java bytecode). But Java led the way in platform openness and Microsoft was helpless but to follow or watch their platform crumble into obsolescence like VBScript and Cobol.
There are a precious few languages I can think of today that require you to pay money to get access to the platform and a compiler/IDE/interpreter.
All the major implementations of JavaScript are free. C#, VB.NET and all the.NET languages are free. C, C++, and Objective C are free (many free implementations exist). Ada, Go, Haskell, Python, Ruby, Perl, PHP, Lisp, and virtually all of the hundreds of "esoteric" programming languages are free.
Your argument is based around a false premise, that if someone is willing to pay money to access a programming language, it is inherently more valuable than languages that can be developed in for free.
About the only languages that are popular and non-free are ones such as VBScript and VBA (they require a Windows OS, but free as in free beer interpreters exist) and a few really unpopular and hated languages like Cobol, Progress and MUMPS.
Are you trying to say that all the free programming languages mentioned above are worse than VBScript, VBA, Cobol, Progress and MUMPS? If so, you need to have your head examined. Spend some time actually _developing with_ one of these crippled, outdated, outmoded, feature-deprived and slow as balls "non-free" programming languages, and you'll be running back as fast as you can into the arms of the nearest free language. Maybe even Java.
What's the difference? If your only other option is unreliable 3 Mbit ADSL that drops out when it rains, you're damned right people are going to use this as a home Internet connection.
If Verizon thinks they have a problem with unlimited data users on the cellular network, they can easily fix it by bringing FiOS to the 97% of the customers in their monopoly "turf" who get no service at all from them, or only ADSL. This is a problem they themselves created.
Slashdot Mirror
There are still things I'm waiting for smartphones to do, so it's not like the companies can just stop releasing new smartphones and keep everyone happily on the Galaxy S4 for 8 years at a stretch. Until they've solved the things I want out of my smartphone, there's still work to do. Let me see...
Foremost has to be battery life. I'd like to see smartphones with much better (week-long at least) battery life, even while actually using it (most high figures for battery life are if you keep it in your pocket).
I'd like to see the return of front-facing fingerprint readers, or at least put them somewhere more accessible like on the power switch. I'm not cool with the "reach-around" maneuver on modern Samsung phones, and I end up having to turn the phone over and unlock it with two hands. Unacceptable. Face ID and the like aren't reliable for me because I wear glasses and I'd rather not have to take them off each time I want to unlock my phone -- that's even less convenient than tapping in the PIN.
"But Apple has that with the iPhone 8 and 8 Plus", you may say. But Apple doesn't have something I really, really want from the Note line: the S Pen. Give me a front-facing fingerprint reader ANNNND an S Pen on the same unit, please, without it being hopelessly outdated like the Note 4. (The Note 4 has many other problems, too, like an extremely inefficient and power-hungry CPU, pathetically slow NAND by current standards, and a very inaccurate and finnicky fingerprint reader.)
Also in my list of things Apple doesn't have that Samsung does -- that I want -- is the ability to containerize apps. Putting an app in a space where it only has access to the data and apps within that container is truly an amazing feature for people who have separate work and home lives, and even multiple home lives. Multiple Steam accounts, Discord accounts, Skype accounts? No problem; Samsung's container feature can do that. Apple is behind on this.
But Samsung isn't ahead in every category. Apple's TV app is far better and easier to use than anything Samsung has to offer, because you don't have to keep logging into 9 bajillion content-serving apps to watch videos. You just search for what you want and play it, oftentimes without knowing or caring which app is serving you the content.
Still more things that Apple is legitimately better at? Oh, yes. The Apple Watch is possibly my favorite device of all the smart devices I've ever owned. I've tried the competition; the Gear S3 Frontier doesn't hold a candle to it. In terms of ease of use and features, the Apple Watch is far more convenient than anything else (I have the Series 3 with LTE). Why, you ask? Well, being able to send and receive calls and SMS from your wrist is amazing. With prominent notifications, you can get a vibration strong enough to wake you from sleep when your alarm goes off, or if you receive an email from someone important, or even a text message. You can of course customize the notifications so you don't get startled awake for something trivial like someone uploading a Youtube video, but it is fantastic for people who need to be "on-call" on their job, or for family members who may need to reach you.
Apple Watch, continued: I can also tell someone what the weather is like (and what it'll be in hourly increments for the rest of the day) within about 3 seconds, even if my phone is in my pocket, without reaching for it, unlocking it and doing a bunch of tapping to get to the weather app. Just ask me anytime what the weather's like, what's the temperature, etc. and I'll tell you without so much as a pause in the conversation.
More Apple Watch: It also continuously monitors my heart rate and rhythm. I subscribe to the Stanford heart study, which is an app that will actually notify me if it detects an arrhythmia. The device may not be a piece of "certified health equipment", nor is it a pacemaker, but it's better than nothing for people who are on that line where they might have a family history of heart conditions, but haven't been
If the telecoms, ISPs and cable companies would build infrastructure for landline Internet and TV that was half as rugged as even something as mundane as city water or electricity service -- which, by the way, are far from infalliable and have plenty of issues -- people would be a lot more satisfied with their Internet. In reality, I'd say the majority of the people I know have line quality issues with their landline home Internet, and intermittently or constantly experience some level of packet loss varying between "makes gaming difficult/annoying" to "pretty much nothing is usable most of the time". Customer support only matters if there's a problem.
What the telcos and ISPs are doing, though, is they use the absolute cheapest shit they can find. They install indoor-grade Cat5 on the outside of the house. They use antiquated coax cables that are notorious for having signal issues unless you get them connected just right, and even then, they waste so much energy that the connector is warm when in use. Imagine if your USB or HDMI cables were warm, what that would say about those cables!
They also do a lot of installation on shoddy telephone poles. This needs to stop. They should bury the cable, and when they do, it should be jacketed in a VERY robust sleeve that will resist pinhole penetration and just about anything else for at least 50 years. The cables should have state of the art EMI interference and should not interfere with one another or any other installation in the same right-of-way.
When they install customer premise equipment, it needs to be enterprise-grade. Don't cheap out and buy a router with the absolute minimum SoC that will kind of do the job for the average user. Build something that's just going to goddamn work, even if the user decides to throw 20 devices at it with several thousand TCP connections. Don't give me this cheap shit that only has enough RAM to track 50 or 100 TCP connections.
This should all be self-explanatory, obvious, and universal. Even if you don't give a crap about your customer and just want to do the absolute minimum, you are saving $500 now to spend $10,000 later on support and service visits. Gas, time and labor, materials, etc. while your field techs and call center employees have to deal with the crappy setup you gave your customers. If you'd have done it right and done it robustly the first time, you would be able to vastly reduce your support staff, and your field techs would only have to come out to visit peoples' homes when the customer does something stupid, or the box gets struck by lightning and has to be replaced or something.
It seems that most residential-focused ISPs -- and honestly, even "business" focused ISPs except for those that are actually building out a proper, dedicated, datacenter -- have the mindset that their customers are just going to casually use the service lightly once in a while, and if it's down, they just go "oh well, the Internet is down" and go on with their lives. But we're no longer living in a world where occasionally reading the news is the limit of what people do with the Internet, just like we're no longer living in a world where all people do with electricity is turn on a reading lamp at night.
At least where I live, the power grid is pretty resilient against physical damage, even in the face of severe thunderstorms, tornadoes, lightning, and that kind of thing. If something in the "middle" of the grid goes down, like a substation, there are usually alternative routes to get the power there and people only see a brief disruption in power. If something local goes down, it's because there's been a local catastrophe, or perhaps -- once in a blue moon, but certainly not with any regularity -- there's a genuine problem with the equipment, like a bad transformer coil.
By comparison, Internet service of any description delivered via landline is a joke. Datacenters like to measure the number of "9s" they get after the 99 in their 99.9999...% uptime; consumer and residential ISPs seem to think
Maybe it's your software (or the COTS you bought) that's buggy? Zing is completely compliant with the TCK, which is a stupidly extensive test suite that ensures the JVM complies with all guarantees the Java standards make.
In other words, if you run your software against two JVMs that are in some way significantly different from one another but they both comply with the TCK, and your software doesn't work right, then your software is most likely making assumptions about the implementation that aren't guaranteed by the standard. This is *never* a good idea on any platform, especially for a language that has such an extensive, well-documented standard as Java.
Folks have similar issues with non-standard C++ code, which either won't compile or won't run properly when you move from one to another standards-compliant compiler. But this is just another reason to put in the effort to develop standards-compliant code. Those who don't, are technically at fault whenever anything breaks. If your COTS is non-standard Java that only works on the Oracle Hotspot JVM or other products based on it, I feel sorry for you, but you should complain to the vendor or stop using that product.
Actually, due to the way copyright works, his opinion still doesn't matter for new drivers. Any new driver or component added to the Linux kernel source would be considered (necessarily) a derivative work of the Linux kernel, because it would directly reference and depend upon both the specification and the implementation of probably hundreds of internal Linux functions -- for memory management, data structures, algorithms, informing the kernel about the behavior of the driver, logging, etc. I've read a few actual copyright lawyers' publicly-released opinion about this stuff, and they all agree that because the GPL enforces its terms on anything else that's considered a derivative work of a GPL work, that code would necessarily also be under GPLv2.
So, Linus could technically speaking accept a file or driver into the Linux source tree under some other license, but then Linus would be violating the copyright license of an untold number of other Linux copyright holders, whose code he is NOT entitled to re-license or to violate the GPLv2 license of those others' code by mixing it with non-GPLv2 code in violation of the "derivative works must stay GPLv2" clause of the GPLv2. Linus would potentially be on the hook for civil damages in the US at least, and he's living and working in California for a long time (probably long enough to be a naturalized citizen), so he'd most definitely be on the hook if he distributed that code to others.
Someone who contributed one or two patches to the kernel would have a tough time convincing a judge that there's any actual copyright infringement, but someone like Greg-KH, Ingo Molnar, or pretty much any Intel, Google, Red Hat or Microsoft (yes, Microsoft) employee who contributes to the codebase, would have probably touched enough internal APIs by now that certainly they'd have a claim on something that a non-trivial module would need to use. And there are a few contributors (past and current) who very vigorously believe in a "strict" interpretation of the GPLv2 and would actually be militant about enforcing their own copyright in the kernel code.
Regarding Linux, Linus Torvalds' own opinion is completely irrelevant, unless you wanted to use an extremely antiquated, practically useless version of Linux from the 90s.
The actual Linux kernel code is an incredible mish-mash of thousands of contributors' code, both companies and individuals. Each passing day brings more and more copyright owners into the Linux kernel, because each person retains copyright over their individual contributions, and each day brings at least one new kernel contributor.
In order to change the copyright license of the current Linux source code, ALL of the contributors of the current version of Linux -- or their estate in case of deceased people, or the liquidator company in case of bankrupt companies -- would need to agree to the license change. Even assuming that every last contributor could be convinced to agree to change the license, there would still be logistical problems in actually contacting all of them. Deceased contributors probably have invalid email addresses on file, and likewise for contributors who sent in patches from a corporate email address and have since moved on. Just the task of contacting the legal copyright holder of every line of Linux source becomes a nearly futile task, even if you go back to a hideously old and putrid version like kernel 2.0 or 2.2.
You might think that you could discombobulate some of the "long tail" of contributors by just removing any source lines (using a Git script) that were contributed by someone not in the top 100 contributors. If you did that, you might wonder, wouldn't that leave an almost-working kernel that the top 100 contributors could then carry forward with, fixing the little one-liners that drive-by patchers gave them here and there? And then you could surely hunt down a mere 100 contributors (or their estates) and change the license -- right? Well, no.
Per this -- http://www.remword.com/kps_res... : there are 19,817 kernel contributors just since 2005 (and quite a few more before then, as Linux was indeed quite popular and noteworthy among engineers in, say, 2004; people even made a halfway decent desktop OS based on it by this time). Half the people -- 9912 to be exact -- contributed 2 or fewer patch sets, which only amounts to 1% of the total patch sets. If we assume that, on average, over a large dataset, any two patchsets are equally likely to be any given size regardless of contributor (which may or may not be a safe assumption), that still means that 1 out of every 100 lines of code in the Linux kernel would have to be independently re-written by someone else if we removed the long tail of copyright owners.
And that would only reduce your total number of contributors by _half_; you'd still have 9000-some contributors who have 3 or more patchsets to their name. I wish that site would tell us what percentage of the codebase is contributed by, say, the last 15,000 people in the list of contributors sorted by number of patchsets. It would probably be something ridiculous like 20 or 30%, meaning that if you wanted to whittle down the number of copyright holders in the kernel to about 4000, you would have to re-write (independently, without peeking) about 1 in 5 lines of code on average. Eww.
Not to mention that randomly removing lines of code (or small but important fixes) all over the codebase would create a mess that probably wouldn't compile, and once it did compile, it'd need to be heavily tested, debugged and fixed just to get it nominally working on modern hardware.
So, yes, peoples' opinions about what the license of the Linux kernel "should be" are completely and totally irrelevant. We can't remove the code from a significant number of contributors to the kernel to whittle down the list, because doing so would spark a many-year project to get the kernel back to some semblance of what it is today. And we can't contact everyone and try to get everyone to agree to a license change, because you're probably going to be sim
Azul offers OpenJDK builds for a lot of platforms using a product called Zulu, which is free of charge if you don't want any support. IMO they're better than Oracle's OpenJDK builds because you get more platforms. I think Zulu's might also continue to get security updates for longer than Oracle is willing to provide them for old versions of Java, so if you're stuck on Java 7 or 8, this is a great alternative. Of course, updating your code so you can jump to OpenJDK 10 is better, but sometimes that can take a long time for projects hitting worst-case issues with backwards compat.
Looking at their site, they seem to offer another product that claims better latency consistency, called Zing, that is non-free. So that tells me that Zulu is mostly unmodified OpenJDK builds (although they could be marginally faster if they are compiled with different option flags or a better compiler than Oracle uses). Zing is something else entirely.
P.S. - I am not a shill for Azul. I've never done business with them, worked for them, or bought their products. But I have downloaded their free OpenJDK builds and find them much more convenient to download (with fewer nags) than Oracle Java or Oracle OpenJDK.
Android is an example of an OS that uses close to nothing from the GNU project while relying on the Linux kernel (and even then, its fork of the Linux kernel that ships on real devices is pretty far removed from the vanilla Linux kernel, with custom security modules, proprietary drivers, firmwares, etc.)
Over the years Android has replaced GPL or LGPL licensed GNU project tools with BSD or MIT licensed ones, either out of spite or because they just didn't like the license. There was never any functionality problem inherent with these tools that caused them to make the switch, and their replacements aren't significantly better.
I certainly wouldn't call modern Android "GNU/Linux". There's no GNU in it!
It's actually a FREE SOFTWARE project, not an OPEN SOURCE project, as he would tell you if he read your comment. ;) But
Contributions made to upstream glibc have their copyright attributed to the FSF once they are turned over to the FSF via a copyright assignment form. See: https://www.gnu.org/licenses/w...
Since Stallman is the President of the FSF, he basically has direct authority to modify the version of glibc on the FSF's servers, because he is the highest authority of the non-profit organization that owns that copyright.
What Stallman *can't* do, due to the license that he wrote and gave to glibc, is stop you or anyone else from modifying any aspect of that code once you download it. And he isn't trying to do that here. So indeed, a fork of glibc (which has happened before and was successful; see eglibc, whose changes were eventually merged upstream) would be the only way to have a publicly-recognized copy with this modification. But it wouldn't be *the* *upstream* glibc anymore, even if you called your repo "glibc".
Stallman also technically has the authority to change the copyright license on FSF's version of glibc to any other license, which he did when they made it LGPL v2.1+. Effectively, instead of glibc just being only available under the "Library General Public License v2" (an older version of the license), it can now be released under LGPLv2.1 *or* LGPLv3 by downstream distributors. But that license change only occurred because Stallman assented to it. He could change it to plain old GPLv3 if he wanted, or he could declare the FSF's copy to be proprietary software if he wanted - he's allowed to do that under copyright law. But that still wouldn't modify the license of any copies that were previously distributed.
Jan Koum is a broad? Or some other broad clashed with their parent which is causing Jan Koum to leave? Also, why are we telling Facebook about this in a Slashdot story headline?!
What is Poetteringware, why should I be afraid of it, and how do I get it? I've never run into a single problem where I had to go out of my way to change something on RHEL7 just because it's using systemd.
Sounds to me like you're just too complacent to adapt to change.
The Windows Subsystem for Linux (WSL), while a neat project, is basically scoped to only be able to run a small subset of all software that you can run on a proper Linux kernel. So "Embrace, Extend, Extinguish" (EEE) would only be possible for a small number of use cases.
Things that don't work now and probably won't work any time soon:
- Raw sockets (i.e., IPv4/IPv6 sockets that aren't explicitly using a specific session layer like TCP or UDP)
- Native support for X11 apps with OpenGL hardware acceleration, without having to pipe the whole thing over the IP stack to a third-party X server that comes with a huge performance overhead because you're not using Direct Rendering
- Deep and comprehensive integration of desktop environments, window decorations and widgets from Linux apps (both KDE and GNOME/MATE/Cinnamon/etc., plus a few less-common ones like LXDE and XFCE) so that a Linux environment can co-exist comfortably on your screen with Windows apps without duplicating things (for example, adding all X11 windows to the Windows Taskbar's window list instead of using a KDE or Cinnamon or GNOME window list)
- Support for Linux Containers
- Deep enough support for Docker that any random Docker image you grab off the Internet and try to run in WSL will "just work" without a hitch (discounting the "Docker for Windows" project that basically starts a Linux kernel in a VM and runs Docker in there, but that's not the same thing at all and comes at a significant performance and overhead cost)
Things that are *basically* impossible per the design of LXSS, and can never happen:
- Support for custom kernel modules (the functionality of specific kernel modules like tun/tap might be supported eventually, but not *any* generic Linux kernel module)
- Support for all the various extended attributes and ACLs of any possible Linux filesystem (basically implementing the Linux VFS comprehensively, not just doing path translation like they do now)
WSL is neat, but a lot of things don't work and probably never will because of the many differences between the design of Windows and Linux and the impracticality of tying them this closely together.
Althoooough, if they implemented the userland aspects of the Windows desktop on top of a Linux kernel, *that* would be sort of an EEE. You *could* then natively run anything that'd run on Linux because your kernel would be *Linux*, and the only remaining concerns would be about drivers that exist for Windows but not for Linux (but if they went this route, I'm sure vendors that continue to support their products would grumble and reluctantly provide Linux drivers to Microsoft, in droves).
Good luck proving in court that "because his packets were this size, he *must* have been viewing porn". If that's so accurate, VPNs would be worthless in repressive states like China and Iran (and apparently Rhode Island now, too, lol) because they could just *tell* that you're visiting a site they don't like, and shut you down.
But VPNs are still very effective in many countries if your endpoint isn't a "known" VPN operator. The other option is for the ISP operator to maintain a strict whitelist of which IPs/websites are reachable from their network; a few countries have begun to think that this is the only true way forward, because if you default to routing any traffic, it's still trivially easy to bypass just about any filtering or deep scanning attempts using standard crypto like TLS.
Sure, you can probably inspect the timing and bandwidth utilization of an encrypted connection to distinguish between streaming video, working in an online office suite, uploading a video to a streaming service, or viewing a restaurant's menu, with high accuracy of being able to at least rule out one or more of those categories. But being able to tell which actual *website* is being visited, or what content is being consumed? That seems very unrealistic to me. The degree of confidence you'd have in your assertions would be, at best, around 50% or so, and almost always much lower than that. This sort of "suspicion-based reasoning" wouldn't fly in most courts in countries that uphold basic human rights.
Oh, and any timing based traffic analysis deductions can be easily defeated client-side by inserting random, non-deterministic jitter into all outbound packets. Since the server endpoint's send rate is also dependent on your client's responses (TCP ACKs), you can effectively control the delay in your server endpoint's responses by introducing small amounts of random latency into your own client's ACKs. Then you can further muddy the waters by having the endpoint pollute the encrypted tunnel with nonsense data. The most accurate conclusion that could be claimed with a high degree of certainty thereafter would be "They seem to be using a lot of throughput for some reason".
Indeed, any justice system that would allow such leaps in logic based on packet size and timing analysis (while having no idea of what the actual contents of the datastream contained) is not a justice system I'd want to be subjected to. That's getting dangerously close to guilty until proven innocent.
For most of my life, I've been a PC user (Linux when possible, but always a Windows desktop hanging around for gaming and certain cross-platform development). My use cases span all sorts of different application areas, from gaming, to productivity, video and image editing (Adobe CC products), software development, and even dabbling a bit in scientific computing.
Since about 2015 I've gradually started buying into the Apple ecosystem, and while it can never displace my Windows desktop, their devices do have distinct advantages in certain spaces. The reason I buy them is for these specific advantages, and I do so while acknowledging their drawbacks. I make informed product purchases that are not really based on emotion, but on specific qualities or features that make them better than competitors.
For example:
(1) iPhone: My first Apple product was an iPhone 6S Plus. I'm actually glad I avoided the 6, because that device is fatally flawed by design because the aluminum is too easy to bend, causing Touch IC disease. I originally bought the iPhone because I was sick and tired of security patches taking months to land on Android devices, and the sometimes extremely arbitrary and short support cycles for Android devices. Also, I've felt that overall Android devices and applications are buggier than iOS and most big iOS apps. This is with my experience on Android mostly ending around the Note 4 era. My suspicions about Apple taking smartphone security more seriously than anyone else were confirmed when their chief of security gave a talk on iOS Security at Blackhat, providing a deeper insight into the tremendous security architecture of their platform. Oh, and they now have a bug bounty program like every other good technology company (as of 2016).
(2) Macbook Pro: My first MBP was a 13" of the last gen before they introduced the Touch Bar. When the Touch Bar came out, I handed down my non-Touch Bar MBP to a family member, and got myself the Touch Bar 13". The killer features of the MBP hardware, for me, are TouchID making it both fast and secure to unlock my Mac, and the extremely good touchpad. Palm detection is flawless, it's perfectly sensitive, and clicking is faster and easier than any other touchpad I've ever used. People who think their own Windows laptop has a good touchpad haven't tried a 2016-or-newer MBP touchpad. Seriously, it's so good that I prefer it over a mouse for all use cases except FPS gaming.
I don't think the MacOS platform itself is anything incredible; it's approximately a sidegrade from Windows 10 in terms of what it can do. In some ways Windows 10 is better; Nvidia, Intel and AMD generally provide better support for OpenGL extensions on Windows than they do on Mac, so if you're using OpenGL (or developing with it), you probably will be frustrated by the level of GL extension support on MacOS. This is entirely due to Apple's inattention to that and focusing on Metal. That said, I subjectively think MacOS's disk encryption, tiered storage (Fusion Drive), overall filesystem design (APFS), and application installation management (.dmg/.pkg) is better than the scattered mess of Windows.
The main killer feature of MacOS, for me, actually might end up being supported on Windows just as well soon due to Ubuntu on Bash on Windows; anyway, that feature is the excellent UNIX compatibility. Homebrew is a lifesaver. Being able to run my favorite FOSS with basically an "apt-get" (okay, "brew install", same difference) is awesome. Windows isn't quite there yet, because they have a lot of missing features; just installing a random package, there's a fairly high chance that some functionality in the translation layer isn't there, and you won't be able to use the program. Homebrew actually compiles all the FOSS UNIX software I need *natively* for the Mac platform, since the POSIX, etc. standards are baked into the core of MacOS (based on BSD), but are just now being tacked onto Windows in a new (and largely incomplete) subsystem.
Because of the great package support for FOSS
Except that the people who would take your image to post as a one-off image macro on imgur, or to use in their homework for a PowerPoint presentation, aren't likely to be able to afford the exorbitant fees you'd charge to use your image anyway. If you want to get paid for your work from the little guy, charge prices little guys can afford, and make it very easy and quick to do so. Otherwise, people will always (and I mean always) find ways to get around your smug attempts at preventing them from using their technology to accomplish their goal. Hell, they'll print out the page, scan it and crop the scanned image if they have to.
Or, you know, if you don't want people using your image, don't post it on the public Internet to be viewed by anyone. Give a thumbnail or a heavily edited / blurred version and make someone go through a paywall to see the full image.
Welcome to the 21st century, where just because you think you get to have absolute control over every possible usage of your copyrighted works, doesn't mean that you get to actually enforce said control, nor does it mean that enforcement is at all practical for you. If you gave consumers no other choice but to buy your content, they MIGHT acquiesce if you made it extremely affordable, but if you tried to charge the exorbitant fees I see floated around (hundreds of dollars per photo, if not thousands), people will either not use your content (and use the content of your competitors, who are cheaper and easier to access), or rebel against the platform and switch to something like desktop Linux to escape the jail.
The only reason iOS isn't dead is that it still provides some pretty solid alternatives to the DRM, vendor-locked ecosystem. You can still install VLC. You can still take screenshots. You can still save images from your browser and upload them to Google Drive.
The iron-fisted rule of law and the rule of imperious content creators who want to own each and every use of their works is the outdated model, not the model of free software and open content. When content hosts, search engines, etc. put up barriers to people using content the way they want to, the network naturally just routes around the damage. Won't be long before there's a Chrome and Firefox extension putting the "View Image" button right back into the Google Image Search.
It doesn't matter what you think; it's happening whether you want it or not, and too many people do it to hope to stop it, unless you imprison every man, woman and child with access to a computer or phone.
Sure, you might be able to elicit a settlement out of Google, but you're not exactly solving the problem. People are still going to use Getty images without permission. Why don't you learn from the way the Internet works, and do like everyone else does, and just route around the damage? Figure out a new way to monetize your content, instead of expecting the world to continue to pay you the same way they always have. The world has moved on. Time for you to do the same, or you'll have to find a new career.
P.S. -- What happened to democracy? When the vast majority of the people unthinkingly do some thing, it seems really backwards to have that thing be illegal. I'm not talking about 20 or 30%; I'm talking about basically everyone (except, perhaps, for people like you, who conscientiously object to the new way of doing things.) I guarantee you that over 95% of people with frequent access to a computer have committed technical "copyright violation" (according to the letter of the law) not once, not twice, but at least a dozen or so times in their lives, and that's even accounting for the many such uses that would be considered Fair Use.
Some things, like murder or car hijacking, won't ever become that popular because the crime is self-limiting; you have to physically take something away from one person to give to another. Nicking someone else's photo or music or video and using it in another context for personal use is literally harmless to the person who produced the ph
Uninterested unless they have the bravery to tell us what actual GPU chip is inside each model. CPU has not been a bottleneck for my workloads for many years, but GPU definitely is. If this is still Maxwell, I'm definitely not interested (and will be VERY annoyed).
It seems to happen approximately daily when consuming open source projects. I'll get so far down the path of trying to build out my project, then hit a broken feature that's actually (to me) a killer feature of the program/framework/library/IDE/etc., but the developers couldn't be arsed to implement it properly / keep it up-to-date / fix severe bugs with it.
Just in general. Because it happens so often that I can't list all the cases where this happens to me. .NET CoreRT would be awesome if it could compile native DLLs that can be called from C, but nope. Lead devs have no interest in implementing it.
Go would be awesome if it could compile native DLLs that can be called from C, but nope. There are branches with working code out there, but nobody's managed to package it into a release of Go. Which kind of sucks, because compiling Go from source is horrible on non-Linux platforms.
Eclipse Dali's SQL to Entities reverse engineering code has been broken since 2011; it doesn't pick up most relationships and constraints between tables. So I have to use NetBeans for my SQL to JPA entities generation. Bah.
I guess I'm the "never lucky" developer -- anything I try to do, the platform I'm using is fatally flawed in exactly the area I'm exploring. But the marketing material makes it sound so great and perfect...
The "Parable of the broken window" was written in 1850. Let me clue you into something: NO ONE in the 19th century, and next to no one in the 20th century even, could have anticipated the sea change that is pervasive automation that we're seeing today. It's nowhere near the same thing as what we experienced in the industrial revolution with tractors displacing farmhands. The changes that are happening are different in kind and require different economic theories to reason about them, because the parameters involved are different.
The jobs being displaced are of entirely different categories than mindless or repetitive jobs. Kiosks are only the first step. Software developers and architects, financial analysts, dozens of types of first-level and middle managers, and even CEOs are potentially on the chopping block in the next 50 years, as machines can do their jobs cheaper and more efficiently.
Just because the economy has more money doesn't mean that money is being spent on paying out wages to people. It could be spent on paying other billionaires for use of their automated systems. As the cost of maintaining these systems decreases with reliability and durability improvements, there just isn't much need for human labor. And despite the population increasing, what I think you'll see is a much slower growth in total wages paid out to the lower 99%, or even stagnation, and sharp decreases during any hint of economic downturn of any magnitude. Meanwhile the top 1% will see exponential growth in wealth, as the 1% always retains the means to keep getting richer, and now we have governments around the world who are willing to meddle in the markets to ensure that the top 1% continue to gain in wealth even as the economy overall shrinks.
As long as our economy operates on the theory that you have to work to earn money to buy things like food, clothing and shelter that you need to live, we need productive jobs to put people in that they can perform to earn that money; otherwise they'll be homeless beggars or looking for government entitlement programs to keep them from starving to death.
If your solution is to let them starve, you're a heartless bastard. You try starving and see how much you like it.
If your solution is to pay them a universal basic income, you're on the right track.
If your solution is to wave your magic wand of 200 year-old economic theory and say "don't worry, as long as there's money people will have jobs", you're burying your head in the sand like an ostrich.
While we're quoting diametrically-opposed theories at one another, take a look at these for an opposing viewpoint:
https://www.youtube.com/watch?...
http://marshallbrain.com/manna...
Similar kit here, except replace the SSD with a 5400 RPM HDD and Windows 8.1 with Windows 10. And McAfee. Yeah. YEAH.
These are Haswell i7 boxes with otherwise extremely high-end hardware (for a business PC), running Windows 10 v1607 x64, basically the fastest version of modern Windows (compressed RAM is awesome). But they were too cheap to spend on SSDs, so we wait. And wait. And wait.
Outlook startup, cold cache: 68 seconds.
Firefox launch, cold cache: 35 seconds.
Eclipse launch, cold cache: 3 to 5 minutes. Same Eclipse profile loads in 5 seconds on my (personal) Macbook Pro with an SSD.
`mvn verify` (run some integration tests), cache irrelevant: 25 to 35 minutes (compared to about 6 seconds on our CI server with 128 GB of RAM and an array of SSDs).
Git clone of a ~3 MB repo: 0.5 seconds networking, 5 minutes disk I/O (compared to about 0.5 seconds of disk I/O on our CI server.)
I'm gonna go out on a limb here and say that your kit isn't part of the problem. The fact that you have SSDs is the biggest factor in ensuring that your kit *isn't* the problem. But if you had 5400 RPM HDDs like me (and I'm sure zillions of other IT shops), you'd be in a world of hurt, like we are. Doubly so if you use(d) McAfee.
And no, our network isn't the bottleneck. Our network is much faster than anyone needs it to be. We have enough bandwidth that every employee can get symmetrical gigabit and fully saturate the line speed of their gigabit NIC, simultaneously. And a comfortable bit of extra beyond that. We sit on a fiber trunk.
BTW, fuck McAfee a.k.a. "Intel Security".
I know every budget has to be criticized by 'the opposing party' with a list of all the wonderful things that are going to be cut, but you all DO realize that the US government is nearly $19 TRILLION in debt - or more than $50k per person in the country?
Every single program that we're paying for, essentially we're living off credit cards. We are the wealthiest nation in human history, and we still cannot afford all the crap we want.
At some point, someone has to be the grownup in the room and say "you know, that would be really nice, but we simply can't afford it".
No one disagrees with that. That's fairly obvious. But what *parts* of the budget are you going to cut?
Any cut you make hurts someone. A large percentage of all government spending is on pay for employees or contractors. Cuts cause those employees and contractors to go looking for work, and the money they used to bring in will stop being poured into the local and global economy as they scrimp and save trying to survive while looking for work.Losing your job hurts a *lot*.
On the other hand, a small tax increase that affects everyone might cost each individual $500 per year or so (in the case of a rather extreme tax increase), but (1) that level of burden isn't going to push anyone over the edge, causing them to go from "making it" to having to sell their home; and (2) the ~$125 Billion per year that you raise from it will go to continuing to fund these programs, significantly slowing the rate of incurred national debt as long as we don't spend *more* than we already have.
The problem is that the Trump administration and congress don't want to raise taxes, but they're happy to cut programs that they are ideologically opposed to. But it turns out that many of their ideologies are just plain *wrong*, like "not believing" in climate change, pro-choice, or even the general welfare of the people. Their answer to "how do you decide what to cut?" is "things we don't like".
I'm mystified as to why these companies running mission-critical apps with $$$ on the line aren't using multi-region redundancy or at least failover. Imagine if some terrorist dug up the fiber lines leading to the Ashburn primary datacenter, causing US-EAST-1 to be offline for days.
This is why you spread your resources around and have redundancies across different geographical regions. That way, the worst that could happen is users might experience a momentary lag, or maybe a couple TCP connections might get reset, but as soon as they try again it'll be up and running like normal, except that they'll be talking to a server in California or London instead of Ashburn, VA.
Surprised that so many companies don't have redundancy that this ended up costing $150M.
I haven't used wired headphones in years -- been happily on the Bluetooth bandwagon since BT 3.0 was the latest spec -- and I already have plenty of USB-C cables, including C to C and C to A.
Also, to the extent that *some of* Apple's hardware is "So Last Year" compared to the Android flagships, some of it is genuinely ahead of Android by 6 to 12 months. For instance, A10 Fusion is still today the fastest production system on chip that you can buy in a fully-functional, consumer-oriented device being produced at high volume. Both its CPU and GPU are quite a bit faster than the nearest competitor from Qualcomm and the like. Samsung's Exynos is even further behind, because their serial performance sucks; they seem to be stuck in the same "MOAR CORES" rut as AMD was for a number of years while they were irrelevant and a non-competitor to Intel. Single thread perf is king for consumer workloads (which is most HTML rendering, JavaScript execution and game engines), period, end of story.
Sure, Qualcomm will soon eclipse the A10 Fusion with something faster, but by the time an Android manufacturer puts it to market, Apple's next gen SoC (A11?) will be out in an iPhone you can go buy at your corner Best Buy or AT&T store. And it'll be the fastest again.
On the RAM front, iPhones also lag behind by about 2 years (the 7 Plus has the same RAM as the Note 4 from Q4 2014). But this brings me to my next point: Software.
I used Android phones from three different manufacturers (HTC, Samsung and Motorola) from the Android 2.0 era up til KitKat. And I swear that with every single phone, with every single stock firmware, I encountered the following, reliably: horrible, annoying bugs in my daily use case paths; awful, choppy Bluetooth with anemic range; very poor battery endurance (never more than 2 years); and some degree of bloatware that was impossible to remove/disable that had very significant storage and/or RAM footprint. I also spent most of my time with Android running a device with a widely publicly-known root exploit vulnerability (several of them exploitable remotely over the network or through apps I used), usually with some open-source exploit toolkit out there that could exploit the vulnerabilities my phone had. This is because as soon as an OTA came out fixing an exploit, another one would surface, rinse and repeat.
Lastly, despite all Google's efforts, Android devices are still extremely laggy even at the ultra high-end and perennially have performance problems, even with custom ROMs. There are so many little lags and hitches here and there that it just feels like it's running interpreted VBScript.
In my two years of using iPhones, I've never encountered a single bug of any sort. Yes, the software is not perfect and has bugs, but they're so minor or such edge cases that I've never encountered them in my daily workflow.
The performance is consistently, almost perfectly fluid, with exceedingly rare performance problems and hitches. For every single noticeable lag I've observed on an iPhone, I would've observed about a thousand of them on my Android device by using it for the same duration with similar use cases.
And Apple actually fixes their security problems quite expediently. Many OTAs have been rolled out whose entire content is just security vulnerability fixes. And often those vulnerabilities had been disclosed within the past week, or less. So the exposure time once a bug goes public is usually quite narrow or nonexistent.
Also, Apple is assertive enough to give the giant middle finger to parasites like Verizon and AT&T when they ask about the prospects of installing bloatware onto iPhones. They simply say "No". You might say that they build this into the device purchase price, but I can always purchase an unlocked iPhone and Verizon won't make a penny off of that, so there really is no financial incentive available to a carrier selling iPhones, except that *users want them* and want to use them on their favorite carrier's network.
I'm frighteningly
Sound like high minded excuse to start use the platform for political purposes. All these words "bullying", "fake news", etc. are code words involved in liberal virtue signalling. "Fake news" is something that those evil right wingers do (especially it does not apply to New York Times, et al. or any garbage coming from BLM or other such outlets).
Deception, coercion, half-truths and complete fabrication are not, and have never been, tools used exclusively by people with one particular political leaning or another. They're used by leftists, rightists, centrists, libertarians, conservatives, liberals, democrats, republicans, greens, independents, tea partiers, anarcho-communists, fascists, feminists, masculinists, and everyone in between or beyond.
It might be the case that a certain number of news outlets could be liberally biased enough to use these tactics to undermine right-wing political viewpoints, but this in no way prevents or exonerates those outlets which are right-wing, from using the same tactics.
If your complaint then becomes that there are too many liberal news sources and not enough mainstream conservative news sources, then you're basically saying that you want the news to present you with lies that agree with your personal political dogma, rather than lies that attack or offend your personal political dogma.
If you think that a change of color or movement along a right/left spectrum will in any way affect the frequency and severity of lies, deception and coercion used by the mainstream media, you would be plain wrong in that belief. ANY politically motivated organization, regardless of what agenda they're pushing, is going to distribute deceptive and patently false information, also known as propaganda, that supports the agenda they are being paid to push.
The only way to return news media to reporting on objective truths observable by scientifically rigorous methods, and away from speculation, hearsay, the passing of rumors and fabrications, and opinion-slinging (all of which are inherently biased toward some particular set of beliefs, and in the context of politics, toward some particular set of political beliefs), is to forcibly separate media from financial incentive. Capitalist media is always going to be propaganda for someone.
Also, to make your argument look even more silly, Java was already ridiculously popular before Sun open sourced the code. Before that it was (mostly) freeware, but companies of all sizes were also buying support licenses for proprietary Java back in the early days. Open sourcing Java just accelerated its popularity, because, in the early days of .NET, its competition was much more platform-constrained (Windows-only, before Mono) and pricey (required a Visual Studio license to unlock some features or, in the very early days, to even get a compiler).
In fact, if you appreciate the extremely open nature of the .NET ecosystem today, then you owe thanks to Java, because Microsoft only open sourced .NET and allowed/encouraged/fostered the Mono project and cross-platform packages in Nuget because of Java/Maven's advantages that were pulling developers off of the .NET ecosystem.
Today, in 2017, there's not much reason to use Java compared to .NET if the libraries you need are supported on .NET (and even if not, you can use IKVM if you're absolutely intent on not running the Oracle HotSpot JVM, but you'll still be shipping Java bytecode). But Java led the way in platform openness and Microsoft was helpless but to follow or watch their platform crumble into obsolescence like VBScript and Cobol.
There are a precious few languages I can think of today that require you to pay money to get access to the platform and a compiler/IDE/interpreter.
All the major implementations of JavaScript are free. .NET languages are free.
C#, VB.NET and all the
C, C++, and Objective C are free (many free implementations exist).
Ada, Go, Haskell, Python, Ruby, Perl, PHP, Lisp, and virtually all of the hundreds of "esoteric" programming languages are free.
Your argument is based around a false premise, that if someone is willing to pay money to access a programming language, it is inherently more valuable than languages that can be developed in for free.
About the only languages that are popular and non-free are ones such as VBScript and VBA (they require a Windows OS, but free as in free beer interpreters exist) and a few really unpopular and hated languages like Cobol, Progress and MUMPS.
Are you trying to say that all the free programming languages mentioned above are worse than VBScript, VBA, Cobol, Progress and MUMPS? If so, you need to have your head examined. Spend some time actually _developing with_ one of these crippled, outdated, outmoded, feature-deprived and slow as balls "non-free" programming languages, and you'll be running back as fast as you can into the arms of the nearest free language. Maybe even Java.
What's the difference? If your only other option is unreliable 3 Mbit ADSL that drops out when it rains, you're damned right people are going to use this as a home Internet connection.
If Verizon thinks they have a problem with unlimited data users on the cellular network, they can easily fix it by bringing FiOS to the 97% of the customers in their monopoly "turf" who get no service at all from them, or only ADSL. This is a problem they themselves created.