Admittedly this is a problem I did not consider because I'm from Germany and our plugs are reversible. Likewise, plugspreading is less of a problem because most power strips have the sockets at a 45-degree angle. Purely Europlug power strips are affected but few people use them because they're incompatibe with Schuko plugs.
Besides, plugspreading becomes less of a problem if you buy a multi-port USB power brick or two. I just spent fifteen bucks on a 30W three-port brick with one Quick Charge 3 port. It's not much bigger than your typical smartphone charger, has its USB ports at the end and replaces three separate chargers. That cleared up a lot of space on the power strip.
Or you can go straight for a power strip with built-in USB ports.
Firstly, this would make smartphone chargers unneccessarily big. Those things fit in a pocket and it's good that way. (The oh-so-terrible waste of space by a smartphone charger shown in TFA could have been solved by moving the charger one outlet to the right. The horror.)
Secondly, this would mean adding cables rated for 240V AC to devices that are supposed to supply 5V or maybe 12V DC. That's a waste of copper.
You are aware that Slashdot is a social media site? It's not the share-your-life kind of social media but its heavy emphasis on discussion does make it social media.
Of course the rest of your point still stands. Non-pseudonymous social media are inherently dangerous and Facebook is about as trustworthy as a guy waving around a burning torch inside a fireworks factory.
Single button press, yes. But then the till takes a few seconds each to enable the card reader and deal with the completion of the transaction because the POS system is, well, a POS. Like many POS systems are. Of course some stores are faster than that but, well, our cashiers are super fast already so even a substantial improvement over cash doesn't save a significant amount of time.
I'm from Germany, actually. Our cashiers are damn fast so there's not much of a speed bonus to going cashless. Maybe if you use your phone and have it auto-approve every transaction but we're still talking a few seconds here. I'll stick with cash.
Cash transactions are also slower than just waving your phone, which raises costs.
Not in my experience. (Note that self-service kiosks aren't really a thing outside of vending machines where I live so my experience always involves cashiers.) For cashless you have to tell the cashier to use a card, wait for them to turn on the card/NFC terminal, wait for the terminal to recognize the card/device, wait for the server to issue a challenge, enter your pin and then wait for your response to be verified. Takes twenty seconds if everything goes well and up to a minute if something goes wrong and the cashier has to restart the transaction.
For cash you give your money to someone who spends half of their workday efficiently sorting currency into and out of an organized box, then take back the change. Ten seconds if you're lucky, thirty if you're really not.
So it works really well for a very narrow definition of "Pro", which basically turns out to be Apple software engineers and third-party Mac and iOS developers.
And a whole bunch of those don't need a Mac Pro in the first place. For instance, the company I work for offers a Xamarin-based app. Our iOS build server is an older Mac mini, which is still plenty fast for our purposes - despite the fact that building a Xamarin app for iOS is extremely slow due to architectural decisions by Apple. Unless you regularly need to build something really big you're probably going to get away with something less expensive than a Mac Pro.
The ashtray is really a completely different beast than the cheese grater. It's more akin to a souped up Mac mini. Cool if you need a silent and beefy desktop computer; less cool if you need a crazy powerful workstation.
And I agree - the cheese grater gave you space for expansion, relied on standardized internal interfaces for good interoperability and had tons of ports. Those are all things Apple don't like in their hardware these days.
I fail to see how jet engines employ frequency-division multiplexing, which you insist is a necessary requirement for using the term "broadband". Of course if you use the term "frequency-division multiplexing" vaguely enough it describes any medium in which signals of differing frequencies can occur at the same time – but in that case the sun becomes an implementation of FDM since its light is not monochromatic. Either FDM describes an implementation that specifically takes advantage of the medium's capacity for multiple signals to convey multiple streams of data (which jet engines don't) or the term itself is pointless because it is synonymous with "any analog signal".
Also, again, it's plainly obvious that the usage of "broadband" for fast internet speeds was coined by means of analogy, just like we use terms like "stream" or "data flow" despite no liquids being involved. The observed thing behaves similarly enough to something we already have a name for to suggest the adaptation of the existing name.
Again, "broadband" has been coined independently by several different groups of people. The fact that some used it to describe a specific technology does not change the fact that different people have used it to describe different things. For instance, here is a patent from the 50s for an antenna that uses the term without having anything to do with FDM beyond the fact that radio signals have a finite bandwidth. Here is a paper from 1988 that refers to "broadband" acoustic signals.
The common ground for these definitions is "has a relatively large bandwidth", which would make "broadband" a natural fit. Using the term for internet speeds was a small step since in the domain of radio transmissions an increase in bandwidth generally corellates to an increase in data transmission capacity.
"Boadband" has never been particularly well defined. It can be used to mean "wideband", it can be used to mean "every signal that isn't passband" and it can be used to mean "every signal that occupies multiple non-masking channels". All of those definitions are correct. The meaning of "transmission speeds generally considered fast" has been in general use for well over two decades now, making it about as uncontroversial as the use of "to hack" in a context that doesn't involve an axe.
Broadband has never had anything to do with FDM specifically. Or rather, there have always been definitions of the term that didn't have anything to do with FDM.
You can replace a character with a different version and have it work. DC did so repeatedly with Green Lantern and despite the fact that they ended up with multiple Green Lanterns at the same time it seems to work well. The various iterations of Blue Beetle are another example (although BB is a lower-tier character than GL, of course). You do need to be very careful about doing this with A-listers but it can work if done right and in good measure.
The main problem is that Marvel hasn't been careful at all; it seems that they replaced their entire A-list roster in a short period of time. (Spider-Man just got more alternate versions of various characters, as is tradition.)
Kinect Star Wars. The game has a terrible reputation; it's often seen as a total joke. However, my experience with it is a bit different.
A friend who has an X360 wanted to host a barbecue night and wanted to get a multiplayer-friendly game for that. I talked him into getting Kinect Star Wars as a joke before we'd get to the real games. So we put it in, got everything set up and launched straight into the dance minigame.
Six hours later we had finally obtained top ranking on all songs, burned off almost as many calories as we had consumed and were convinced that Kinect Star Wars is the best game ever released for the Xbox 360. It's silly, yes, but there is something to be said in favor of a game where a gothic lolita Death Star gunner meets Boba Fett for a dance-off in an Imperial disco while a recruitment song set to the tune of YMCA plays. (Just search YouTube for "Empire Today extended", although the player there scandalously does not use Odella as his character.)
Reliable? HA! The only time I've ever had a computer completely fail, to the point of not being bootable anymore, it was an Apple MacBook Pro. The old Core Duo model from 2005 or so. That was the last Mac I bought. I still have a beige G3 that runs fine. It's useless, but it runs. Apple's build quality declined sharply in the mid-00's. Nothing about their product is "reliable" anymore. So "ancient" technology is just outdated.
I'm not pretending that Apple hardware specs aren't shitty these days but I've gotta stick up for their build quality. I have a 2010 13" MBP and a 2012 15" one. Both are still rock solid. Literally, in fact; I once dropped the 15" one and made a nice little hole in the carpet. Still runs like a champ. Unibody MBPs are essentially indestructible.
I have had an MBP fail spectacularly on me; it was the 2007 model. It died of the Geforce 8600M GT bug - essentially, Nvidia had fucked up their chip design, causing all 8600M GT chips in the world to slowly self-destruct. The 2007 MBP used the 8600M GT. Apple tried doing a free repair program but the replacement logic boards had the same GPU so they weren't long for this world either.
I'm actually kinda sad that everything Apple has released after 2002 has shit specs and ridiculously expensive purchase-time-only upgrades. I really like the build quality but with the crap Apple keeps putting out my next laptop will come from someone else.
OTOH, trying to compete with USB would've made the new standard the next FireWire: Technically better in every regard but too expensive and not ubiquitous enough. It would've found niche use and would've died at some point because everyone would've stuck with USB.
I mean, what would NewStandard-C have offered? Small plugs? Micro-USB. Faster charging? Qualcomm Quick Charge and similar standards. Support for displays? HDMI and (mini-)DisplayPort already fill the need perfectly. Faster speeds? Not interesting enough for the majority of users to warrant switching until USB comes out with the next revision that offers roughly what NewStandard-C offers, just five years later and with more overhead.
Its virtually impossible to compete with USB because it's ubiquitous and Good Enough(TM).
Exactly. It's not that the article on electroweak interaction is too hard; it's that the article is incomplete. The graphene article is chock-full of highly technical information that only concerns experts in the field but it also has plenty of information digestible by and interesting to the general public. The electroweak interaction article is an introduction and then just bare formulas.
Imagine if the Wikipedia article on object-oriented programming consisted entirely of the overview and a few code examples. No discussion of what OOP tries to achieve and how it attempts to do so. Just the bare code. It wouldn't be a terribly useful article to most people.
Speaking as an outsider, these days it feels as if all discourse in the USA exists only within the narrative of Red Team vs. Blue Team –and it doesn't actually matter what those teams represent or what they do. Even things like the well-being of the country are secondary to the question of which team "wins". That is deeply troubling –especially since the USA have nuclear weapons and a propensity to project their power wherever they want. An unstable USA is in nobody's interest.
Nobody uses Reddit's code because Reddit's code is terrible. It's a jumble of Python that relies on some very specific (and long-deprecated) Python packages, which means that you can only run it on one specific version of Python on Ubuntu 14.04. Trying to run it anywhere else (even Ubuntu 14.10) will typically end in failure.
We're in full agreement about them dropping open source sucking, though.
Unfortunately, some features will simply die. Tab groups are one such feature. I like having many tabs open, grouped by areas of current interest to me. If, for example, I'm setting up a new home server I open a "home server" group once I start planning and it stays open until the server runs smoothly. It's a killer feature for me, first provided by the TabGroups Manager addon, then by Firefox's own Panorama feature... then by TabGroups Manager again as Mozilla decided that Panorama was not important enough for them to maintain. With Firefox 57 tab groups will be completely dead.
Note that technically WebExtensions should provide the necessary APIs to support tab groups as having a third-party addon that provides tree-style tabs in a sidebar is a goal of Fx 57. However, there are a few unresolved bugs, including some marked as "some day, something might happen here to allow this to happen" and my preferred method of having a tab group bar above the tabs won't work unless they give WebExtensions the ability to add toolbars. Plus, there's the fact that a few developers of such addons have since abandoned their addons out of frustration.
I get the feeling that Mozilla should've started the transition earlier (or, alternatively, should switch over later). Right now they're still trying to figure out how the WebExtension API should look and how powerful it should be. It's probably going to take another year before they have all the features the developers actually need in there. Until then whole classes of addons will die out and many won't come back as the developers will have moved on.
Depends on the manufacturer. I have a BQ and I have no issues with a lack of updates. (The Zuk Z1 before that was a different story altogether, admittedly. To be fair, though, that was because Zuk made the mistake of partnering with Cyanogen Inc.)
Unfortunate as it is, the presence or lack of a solid update scheme is a distinguishing feature between manufacturers - one you won't find in a feature matrix. Looking into it can help you avoid making mistakes.
Time. Chances are that Intel will have on-CPU TB3 out at least one generation ahead of AMD because they probably had a head start on the integration work. This might be part of Intel's response to Zen turning out much better than AMD's last few architectures. Zen has USB 3.1 so Core having TB3 might be a reasonable answer to that.
Some people distinguish betwen "soft-bricked" (the device stops working but can still be revived with user-available measures going beyond normal configuration*), "hard-bricked" (the device stops working and can only be revived with tools unavailable to an ordinary user**) and "broken" (the device is dead and can only be replaced***). In this case the routers appear to have been hard-bricked as they stopped working and had to be physically accessed by the vendor in order to restore functionality.
* E.g. using Fastboot to flash a new firmware to an Android phone.
** E.g. using JTAG to flash a new bootloader as the device can't even go into Fastboot mode anymore.
*** E.g. my Zuk Z1.
Slashdot Mirror
Admittedly this is a problem I did not consider because I'm from Germany and our plugs are reversible. Likewise, plugspreading is less of a problem because most power strips have the sockets at a 45-degree angle. Purely Europlug power strips are affected but few people use them because they're incompatibe with Schuko plugs.
Besides, plugspreading becomes less of a problem if you buy a multi-port USB power brick or two. I just spent fifteen bucks on a 30W three-port brick with one Quick Charge 3 port. It's not much bigger than your typical smartphone charger, has its USB ports at the end and replaces three separate chargers. That cleared up a lot of space on the power strip.
Or you can go straight for a power strip with built-in USB ports.
(As for the Apple: You do realize that it only supplies 5W? Their 12W device has a less convenient form factor - in fact, the 18W charger that came with my smartphone has a smaller footprint than Apple's 12W one.)
That has its own downsides:
Firstly, this would make smartphone chargers unneccessarily big. Those things fit in a pocket and it's good that way. (The oh-so-terrible waste of space by a smartphone charger shown in TFA could have been solved by moving the charger one outlet to the right. The horror.)
Secondly, this would mean adding cables rated for 240V AC to devices that are supposed to supply 5V or maybe 12V DC. That's a waste of copper.
You are aware that Slashdot is a social media site? It's not the share-your-life kind of social media but its heavy emphasis on discussion does make it social media.
Of course the rest of your point still stands. Non-pseudonymous social media are inherently dangerous and Facebook is about as trustworthy as a guy waving around a burning torch inside a fireworks factory.
Single button press, yes. But then the till takes a few seconds each to enable the card reader and deal with the completion of the transaction because the POS system is, well, a POS. Like many POS systems are. Of course some stores are faster than that but, well, our cashiers are super fast already so even a substantial improvement over cash doesn't save a significant amount of time.
I'm from Germany, actually. Our cashiers are damn fast so there's not much of a speed bonus to going cashless. Maybe if you use your phone and have it auto-approve every transaction but we're still talking a few seconds here. I'll stick with cash.
Cash transactions are also slower than just waving your phone, which raises costs.
Not in my experience. (Note that self-service kiosks aren't really a thing outside of vending machines where I live so my experience always involves cashiers.) For cashless you have to tell the cashier to use a card, wait for them to turn on the card/NFC terminal, wait for the terminal to recognize the card/device, wait for the server to issue a challenge, enter your pin and then wait for your response to be verified. Takes twenty seconds if everything goes well and up to a minute if something goes wrong and the cashier has to restart the transaction.
For cash you give your money to someone who spends half of their workday efficiently sorting currency into and out of an organized box, then take back the change. Ten seconds if you're lucky, thirty if you're really not.
There's an app called Duet Display that turns an iOS device into a second screen with a touch bar. A friend of mine uses it and seems to like it.
So it works really well for a very narrow definition of "Pro", which basically turns out to be Apple software engineers and third-party Mac and iOS developers.
And a whole bunch of those don't need a Mac Pro in the first place. For instance, the company I work for offers a Xamarin-based app. Our iOS build server is an older Mac mini, which is still plenty fast for our purposes - despite the fact that building a Xamarin app for iOS is extremely slow due to architectural decisions by Apple. Unless you regularly need to build something really big you're probably going to get away with something less expensive than a Mac Pro.
The ashtray is really a completely different beast than the cheese grater. It's more akin to a souped up Mac mini. Cool if you need a silent and beefy desktop computer; less cool if you need a crazy powerful workstation.
And I agree - the cheese grater gave you space for expansion, relied on standardized internal interfaces for good interoperability and had tons of ports. Those are all things Apple don't like in their hardware these days.
I fail to see how jet engines employ frequency-division multiplexing, which you insist is a necessary requirement for using the term "broadband". Of course if you use the term "frequency-division multiplexing" vaguely enough it describes any medium in which signals of differing frequencies can occur at the same time – but in that case the sun becomes an implementation of FDM since its light is not monochromatic. Either FDM describes an implementation that specifically takes advantage of the medium's capacity for multiple signals to convey multiple streams of data (which jet engines don't) or the term itself is pointless because it is synonymous with "any analog signal".
Also, again, it's plainly obvious that the usage of "broadband" for fast internet speeds was coined by means of analogy, just like we use terms like "stream" or "data flow" despite no liquids being involved. The observed thing behaves similarly enough to something we already have a name for to suggest the adaptation of the existing name.
Again, "broadband" has been coined independently by several different groups of people. The fact that some used it to describe a specific technology does not change the fact that different people have used it to describe different things. For instance, here is a patent from the 50s for an antenna that uses the term without having anything to do with FDM beyond the fact that radio signals have a finite bandwidth. Here is a paper from 1988 that refers to "broadband" acoustic signals.
The common ground for these definitions is "has a relatively large bandwidth", which would make "broadband" a natural fit. Using the term for internet speeds was a small step since in the domain of radio transmissions an increase in bandwidth generally corellates to an increase in data transmission capacity.
"Boadband" has never been particularly well defined. It can be used to mean "wideband", it can be used to mean "every signal that isn't passband" and it can be used to mean "every signal that occupies multiple non-masking channels". All of those definitions are correct. The meaning of "transmission speeds generally considered fast" has been in general use for well over two decades now, making it about as uncontroversial as the use of "to hack" in a context that doesn't involve an axe.
Broadband has never had anything to do with FDM specifically. Or rather, there have always been definitions of the term that didn't have anything to do with FDM.
You can replace a character with a different version and have it work. DC did so repeatedly with Green Lantern and despite the fact that they ended up with multiple Green Lanterns at the same time it seems to work well. The various iterations of Blue Beetle are another example (although BB is a lower-tier character than GL, of course). You do need to be very careful about doing this with A-listers but it can work if done right and in good measure.
The main problem is that Marvel hasn't been careful at all; it seems that they replaced their entire A-list roster in a short period of time. (Spider-Man just got more alternate versions of various characters, as is tradition.)
Do watch the new one, though. It's really good.
We were actually sober. That game really is a pretty damn good party game.
Kinect Star Wars. The game has a terrible reputation; it's often seen as a total joke. However, my experience with it is a bit different.
A friend who has an X360 wanted to host a barbecue night and wanted to get a multiplayer-friendly game for that. I talked him into getting Kinect Star Wars as a joke before we'd get to the real games. So we put it in, got everything set up and launched straight into the dance minigame.
Six hours later we had finally obtained top ranking on all songs, burned off almost as many calories as we had consumed and were convinced that Kinect Star Wars is the best game ever released for the Xbox 360. It's silly, yes, but there is something to be said in favor of a game where a gothic lolita Death Star gunner meets Boba Fett for a dance-off in an Imperial disco while a recruitment song set to the tune of YMCA plays. (Just search YouTube for "Empire Today extended", although the player there scandalously does not use Odella as his character.)
Reliable? HA! The only time I've ever had a computer completely fail, to the point of not being bootable anymore, it was an Apple MacBook Pro. The old Core Duo model from 2005 or so. That was the last Mac I bought. I still have a beige G3 that runs fine. It's useless, but it runs. Apple's build quality declined sharply in the mid-00's. Nothing about their product is "reliable" anymore. So "ancient" technology is just outdated.
I'm not pretending that Apple hardware specs aren't shitty these days but I've gotta stick up for their build quality. I have a 2010 13" MBP and a 2012 15" one. Both are still rock solid. Literally, in fact; I once dropped the 15" one and made a nice little hole in the carpet. Still runs like a champ. Unibody MBPs are essentially indestructible.
I have had an MBP fail spectacularly on me; it was the 2007 model. It died of the Geforce 8600M GT bug - essentially, Nvidia had fucked up their chip design, causing all 8600M GT chips in the world to slowly self-destruct. The 2007 MBP used the 8600M GT. Apple tried doing a free repair program but the replacement logic boards had the same GPU so they weren't long for this world either.
I'm actually kinda sad that everything Apple has released after 2002 has shit specs and ridiculously expensive purchase-time-only upgrades. I really like the build quality but with the crap Apple keeps putting out my next laptop will come from someone else.
OTOH, trying to compete with USB would've made the new standard the next FireWire: Technically better in every regard but too expensive and not ubiquitous enough. It would've found niche use and would've died at some point because everyone would've stuck with USB.
I mean, what would NewStandard-C have offered? Small plugs? Micro-USB. Faster charging? Qualcomm Quick Charge and similar standards. Support for displays? HDMI and (mini-)DisplayPort already fill the need perfectly. Faster speeds? Not interesting enough for the majority of users to warrant switching until USB comes out with the next revision that offers roughly what NewStandard-C offers, just five years later and with more overhead.
Its virtually impossible to compete with USB because it's ubiquitous and Good Enough(TM).
Exactly. It's not that the article on electroweak interaction is too hard; it's that the article is incomplete. The graphene article is chock-full of highly technical information that only concerns experts in the field but it also has plenty of information digestible by and interesting to the general public. The electroweak interaction article is an introduction and then just bare formulas.
Imagine if the Wikipedia article on object-oriented programming consisted entirely of the overview and a few code examples. No discussion of what OOP tries to achieve and how it attempts to do so. Just the bare code. It wouldn't be a terribly useful article to most people.
Speaking as an outsider, these days it feels as if all discourse in the USA exists only within the narrative of Red Team vs. Blue Team –and it doesn't actually matter what those teams represent or what they do. Even things like the well-being of the country are secondary to the question of which team "wins". That is deeply troubling –especially since the USA have nuclear weapons and a propensity to project their power wherever they want. An unstable USA is in nobody's interest.
Eh, it'll never catch on.
Nobody uses Reddit's code because Reddit's code is terrible. It's a jumble of Python that relies on some very specific (and long-deprecated) Python packages, which means that you can only run it on one specific version of Python on Ubuntu 14.04. Trying to run it anywhere else (even Ubuntu 14.10) will typically end in failure.
We're in full agreement about them dropping open source sucking, though.
Unfortunately, some features will simply die. Tab groups are one such feature. I like having many tabs open, grouped by areas of current interest to me. If, for example, I'm setting up a new home server I open a "home server" group once I start planning and it stays open until the server runs smoothly. It's a killer feature for me, first provided by the TabGroups Manager addon, then by Firefox's own Panorama feature... then by TabGroups Manager again as Mozilla decided that Panorama was not important enough for them to maintain. With Firefox 57 tab groups will be completely dead.
Note that technically WebExtensions should provide the necessary APIs to support tab groups as having a third-party addon that provides tree-style tabs in a sidebar is a goal of Fx 57. However, there are a few unresolved bugs, including some marked as "some day, something might happen here to allow this to happen" and my preferred method of having a tab group bar above the tabs won't work unless they give WebExtensions the ability to add toolbars. Plus, there's the fact that a few developers of such addons have since abandoned their addons out of frustration.
I get the feeling that Mozilla should've started the transition earlier (or, alternatively, should switch over later). Right now they're still trying to figure out how the WebExtension API should look and how powerful it should be. It's probably going to take another year before they have all the features the developers actually need in there. Until then whole classes of addons will die out and many won't come back as the developers will have moved on.
Depends on the manufacturer. I have a BQ and I have no issues with a lack of updates. (The Zuk Z1 before that was a different story altogether, admittedly. To be fair, though, that was because Zuk made the mistake of partnering with Cyanogen Inc.)
Unfortunate as it is, the presence or lack of a solid update scheme is a distinguishing feature between manufacturers - one you won't find in a feature matrix. Looking into it can help you avoid making mistakes.
Time. Chances are that Intel will have on-CPU TB3 out at least one generation ahead of AMD because they probably had a head start on the integration work. This might be part of Intel's response to Zen turning out much better than AMD's last few architectures. Zen has USB 3.1 so Core having TB3 might be a reasonable answer to that.
Some people distinguish betwen "soft-bricked" (the device stops working but can still be revived with user-available measures going beyond normal configuration*), "hard-bricked" (the device stops working and can only be revived with tools unavailable to an ordinary user**) and "broken" (the device is dead and can only be replaced***). In this case the routers appear to have been hard-bricked as they stopped working and had to be physically accessed by the vendor in order to restore functionality.
* E.g. using Fastboot to flash a new firmware to an Android phone.
** E.g. using JTAG to flash a new bootloader as the device can't even go into Fastboot mode anymore.
*** E.g. my Zuk Z1.