Agreed. Amazing that I read all the comments =>2 and saw only two mentions of the circles, which seemed brilliant to me. The ability to direct a post toward my family or my out of town friends or friends interested in linux or whatever seems absolutely necessary to making the thing usable.
Circles took a tiny bit of effort to set up and maintain, but the payoff seemed huge. I was drifting toward biting the bullet and signing up for a FB account right when G+ was launched. The obvious superiority of the interface and capabilities compared to FB (as it seemed to me) convinced me just how terrible FB really is.
In the end, the network effects killed it, I think. Many of my friends had G+ accounts, but they all also used their FB accounts to interact with family and other groups of people. While they may or may not have felt that G+ had technical advantages, it was not worth the effort to be simultaneously on FB and G+.
The FA may not talk about it, but most people are riding them on the sidewalk in San Diego. I would say about 3/4, but it probably varies by neighborhood. Some scooter riders use bike lanes and follow normal bike safety protocols around traffic, but less than 1 in 10 wears a helmet.
The "language" consists of the compiler (the contents of src/), most of the standard library (base/), and some utilities (most of the rest of the files in this repository).
Amazing how easy it is to throw in 'section 108 "h" of the copyright act gives libraries the power' and appear to be an authority. A quick google search (for an SF author's name and openlibrary) shows that they are lending books written in the 2000s that are very popular, still in print, and have authorized electronic versions, so none of the criteria are satisfied.
Success: I have been running Amazon's Kindle for PC under Wine for years. This (plus Calibre) allows me to put Kindle ebooks on my non-Kindle e-ink device. If I had an installable licensed copy of MSWindows and a PC that would run it under a VM, I would do that, but Kindle runs under Wine on my 10-year-old Thinkpad.
Failure: I have never trusted Wine enough to run Turbotax on it.
Since people are making snide comments below, and I am typing this on a 4:3 laptop (old thinkpad from ebay!) I will explain. Many tasks benefit from a minimum amount of vertical screen real estate: some UIs have too many menus and notification bars, sometimes you want a vertical stack of windows because of how things line up, sometimes you need to read large blocks of text in pdfs. A screen vertical of 8.4" with a 4:3 ratio makes a compact, easy-to-carry laptop without miniaturized keys. A screen vertical of 8.4" on a 16:9 makes for a gigantic laptop which is more of a portable desktop.
As long as a run Linux and stay away from Gnome, everything runs effectively instantly or is limited by the network even on a core 2 duo.
It is worth noting that the same issue applies to the output window of the laser truck, though to a lesser degree. Very high transmission substances are available for windows, and they are much more robust than reflective coatings. A scratch or coating of dust could probably still destroy the system.
Maybe a better way to put it is to think of three ranges: At low enough power, a coating isn't needed. At high enough power, any practical coating will be burned through. The in-between range where a reflective coating can make a difference is surprising narrow, not much more than a factor of 10 in power, because really good wide-spectrum reflectivities will be less than 99%.
The best reflectivity is fragile. A 10 W laser can burn a crater in a beautiful lab-grade mirror. (Flaw in the coating? minuscule deterioration? speck of dust?)
This can be translated into time instead. So if the laser damages the target in a microsecond, no coating will help. But if the beam has to be held on target for tens of seconds, some reflectivity will turn this into minutes and may make a difference.
he answer seems to be Javascript. When I have NoScript blocking everything, then browser load is minimal and stays minimal indefinitely regardless of the number of tabs. Certain sites that require Javascript must periodically have their tabs killed and then reloaded to keep the CPU usage reasonable.
Maybe what we need is Javascript sandboxing that can pause scripts in tabs without focus, limit CPU usage, autokill pages, and so on. I have no idea whether the engine is buggy or the site code is buggy or the frameworks are broken or whatever, but if it hasn't been fixed yet, then we need a drastic solution.
In fairness, on the Fedora page of pending tasks to compete, remote display is listed first. I think we can assume this indicates that the importance of the issue is recognized.
And if X on Wayland works as well as XQuartz, it is just barely better than useless. When I run remote programs on XQuartz, it crashes a lot. Some programs can't reasonably be used at all, because XQuartz crashes so often (accidently hitting the mouse scroll wheel while in emacs seems to cause a crash every time). Real X on Linux has always worked beautifully in comparison. If X on Wayland is only as good as XQuartz, it is worth having in an emergency but no good for daily work. YMMV.
Well, yes, all of these to various extents. Plus the need at times to be precise, when the equivalent in general language is vague. Plus when writing for your peers, it is a lot easier to use your shared language.
Any particular impenetrable paper may result from one of these causes or any combination.
It is fair to say that if an informed layperson (someone with an ongoing interest in the field, not a specialized degree) cannot get the gist of the argument, then the paper is poorly written and shouldn't have been published.
... then just use a simple text file. Come up with your own scheme for title, date and time and it will work.
Either as single file with search, or with multiple files and grep(1).
And please make things easier on yourself by setting up commands to automate your own chosen format. I have simple commands for opening a file with today's date in the name, inserting date and time into the text of the file, generating numbered lines, and so on.
Autogenerating dates and times is particularly valuable in avoiding extremely costly mistakes.
(For me, this is aliasing "mylog" to something like emacs ~/Notes/`date +"%Y%m%d"`.txt)
How can you point out that it is only MJG's work that lets Linux be more popular on new hardware than, oh, BeOS? That goes completely against the agreed-on narrative so ably established by previous posters, that Garrett is a useless whiner and who needs him. This is why it is so important to read everything before adding your own 2 cents.
When they were called on their fraud, they checked their research and confirmed they did find an thermal anomaly as predicted by Pons and Fleischmann.
Oh, you've been listening to Eugene Mallove. (Just checked Wikipedia and heard for the first time of his terrible murder, so I will limit what I say.)
Scientific laboratories are complicated places. If the people who were there say, "There is a blip in the data when we changed the settings on the something-or-other" and someone who wasn't there says "This published plot PROVES cold fusion occurred" the default should not be to believe the person who wasn't there.
The experiment is not proof that there is no cold fusion. It is proof that cold fusion is not a robust, easily obtainable effect that could be reproduced based on the information publicly available at the time.
For example, it was shown almost TWENTY YEARS AGO in Jet that the capabilities of modern experiments outstripped their ability to deal with the effects of the resulting radiation, and no one has run a large experiment using a reactor-like fuel mix since.
Since then, there has been quite a bit of evidence collected that trying to design and construct a new machine with reactor-grade shielding on a shoestring budget is not something that can be accomplished quickly.
No, the goal of every "reactor" ever built so far is to study plasma physics. To make the site ready for megawatts of flux of neutrons and gammas would cost five times as much and take 5 times as long to build. It's like testing water pumps and cooling towers for a fission plant without any uranium around--it's faster, cheaper, and safer.
One device is under construction that WILL generate more power than it takes in, and the costs and building times are much higher in consequence.
The story says the bug affected all drives equally, but a linked-to article says that the bug was isolated to TRIM commands by a group that found their 5 models of Samsung drives became regularly corrupted but their 3 models of Intel drives did not. That is not confirmation bias.
Actually people are breaking in using this feature. It involves using a box with an antenna and broadcaster. When you are 100 feet away (possibly in a building), they use the box to boost the signal and make the car believe you are 2 feet away.
It's hard to disagree with this statement, although we do not know all the details. Compare to computers—we have insecure OS's running insecure servers on insecure networks, and then people realized that maybe all this should be retrofitted with security without losing a single capability or backwards compatibility. That's definitely hard. On the other hand, we have had cars secure from RF-based attacks for a hundred years, and now suddenly they become insecure without the addition of a single meaningful capability.
I could duck tape an ipad to the dash of my 10 year old car and have something more capable and more secure than a 1 year old car.
Slashdot Mirror
Circles took a tiny bit of effort to set up and maintain, but the payoff seemed huge. I was drifting toward biting the bullet and signing up for a FB account right when G+ was launched. The obvious superiority of the interface and capabilities compared to FB (as it seemed to me) convinced me just how terrible FB really is.
In the end, the network effects killed it, I think. Many of my friends had G+ accounts, but they all also used their FB accounts to interact with family and other groups of people. While they may or may not have felt that G+ had technical advantages, it was not worth the effort to be simultaneously on FB and G+.
The FA may not talk about it, but most people are riding them on the sidewalk in San Diego. I would say about 3/4, but it probably varies by neighborhood. Some scooter riders use bike lanes and follow normal bike safety protocols around traffic, but less than 1 in 10 wears a helmet.
The "language" consists of the compiler (the contents of src/), most of the standard library (base/), and some utilities (most of the rest of the files in this repository).
https://github.com/JuliaLang/j...
Probably someone watching Oracle asserting ownership of the Java base classes. Use was allowed, but not alterations or compatible implementations.
No, they buy a license that allows lending electronic copies.
https://www.boston.com/news/te...
The pricing structure and attached permissions are completely different.
Amazing how easy it is to throw in 'section 108 "h" of the copyright act gives libraries the power' and appear to be an authority. A quick google search (for an SF author's name and openlibrary) shows that they are lending books written in the 2000s that are very popular, still in print, and have authorized electronic versions, so none of the criteria are satisfied.
Failure: I have never trusted Wine enough to run Turbotax on it.
Since people are making snide comments below, and I am typing this on a 4:3 laptop (old thinkpad from ebay!) I will explain. Many tasks benefit from a minimum amount of vertical screen real estate: some UIs have too many menus and notification bars, sometimes you want a vertical stack of windows because of how things line up, sometimes you need to read large blocks of text in pdfs. A screen vertical of 8.4" with a 4:3 ratio makes a compact, easy-to-carry laptop without miniaturized keys. A screen vertical of 8.4" on a 16:9 makes for a gigantic laptop which is more of a portable desktop. As long as a run Linux and stay away from Gnome, everything runs effectively instantly or is limited by the network even on a core 2 duo.
It is worth noting that the same issue applies to the output window of the laser truck, though to a lesser degree. Very high transmission substances are available for windows, and they are much more robust than reflective coatings. A scratch or coating of dust could probably still destroy the system.
The best reflectivity is fragile. A 10 W laser can burn a crater in a beautiful lab-grade mirror. (Flaw in the coating? minuscule deterioration? speck of dust?)
This can be translated into time instead. So if the laser damages the target in a microsecond, no coating will help. But if the beam has to be held on target for tens of seconds, some reflectivity will turn this into minutes and may make a difference.
Maybe what we need is Javascript sandboxing that can pause scripts in tabs without focus, limit CPU usage, autokill pages, and so on. I have no idea whether the engine is buggy or the site code is buggy or the frameworks are broken or whatever, but if it hasn't been fixed yet, then we need a drastic solution.
(I like the idea of moderating /. with a katana.)
In fairness, on the Fedora page of pending tasks to compete, remote display is listed first. I think we can assume this indicates that the importance of the issue is recognized.
And if X on Wayland works as well as XQuartz, it is just barely better than useless. When I run remote programs on XQuartz, it crashes a lot. Some programs can't reasonably be used at all, because XQuartz crashes so often (accidently hitting the mouse scroll wheel while in emacs seems to cause a crash every time). Real X on Linux has always worked beautifully in comparison. If X on Wayland is only as good as XQuartz, it is worth having in an emergency but no good for daily work. YMMV.
May I suggest that you remember seeing Empire Strikes Back and being puzzled by "Episide V"? That is my experience.
#blackpaintspatters
Any particular impenetrable paper may result from one of these causes or any combination.
It is fair to say that if an informed layperson (someone with an ongoing interest in the field, not a specialized degree) cannot get the gist of the argument, then the paper is poorly written and shouldn't have been published.
... then just use a simple text file. Come up with your own scheme for title, date and time and it will work. Either as single file with search, or with multiple files and grep(1).
And please make things easier on yourself by setting up commands to automate your own chosen format. I have simple commands for opening a file with today's date in the name, inserting date and time into the text of the file, generating numbered lines, and so on.
Autogenerating dates and times is particularly valuable in avoiding extremely costly mistakes.
(For me, this is aliasing "mylog" to something like emacs ~/Notes/`date +"%Y%m%d"`.txt)
How can you point out that it is only MJG's work that lets Linux be more popular on new hardware than, oh, BeOS? That goes completely against the agreed-on narrative so ably established by previous posters, that Garrett is a useless whiner and who needs him. This is why it is so important to read everything before adding your own 2 cents.
When they were called on their fraud, they checked their research and confirmed they did find an thermal anomaly as predicted by Pons and Fleischmann.
Oh, you've been listening to Eugene Mallove. (Just checked Wikipedia and heard for the first time of his terrible murder, so I will limit what I say.)
Scientific laboratories are complicated places. If the people who were there say, "There is a blip in the data when we changed the settings on the something-or-other" and someone who wasn't there says "This published plot PROVES cold fusion occurred" the default should not be to believe the person who wasn't there.
The experiment is not proof that there is no cold fusion. It is proof that cold fusion is not a robust, easily obtainable effect that could be reproduced based on the information publicly available at the time.
Since then, there has been quite a bit of evidence collected that trying to design and construct a new machine with reactor-grade shielding on a shoestring budget is not something that can be accomplished quickly.
URL bar input seems like an obvious place to do fuzz testing. Just throw random stuff at it as fast as you can, and wait for a crash.
No, the goal of every "reactor" ever built so far is to study plasma physics. To make the site ready for megawatts of flux of neutrons and gammas would cost five times as much and take 5 times as long to build. It's like testing water pumps and cooling towers for a fission plant without any uranium around--it's faster, cheaper, and safer. One device is under construction that WILL generate more power than it takes in, and the costs and building times are much higher in consequence.
The story says the bug affected all drives equally, but a linked-to article says that the bug was isolated to TRIM commands by a group that found their 5 models of Samsung drives became regularly corrupted but their 3 models of Intel drives did not. That is not confirmation bias.
Actually people are breaking in using this feature. It involves using a box with an antenna and broadcaster. When you are 100 feet away (possibly in a building), they use the box to boost the signal and make the car believe you are 2 feet away.
It's hard to disagree with this statement, although we do not know all the details. Compare to computers—we have insecure OS's running insecure servers on insecure networks, and then people realized that maybe all this should be retrofitted with security without losing a single capability or backwards compatibility. That's definitely hard. On the other hand, we have had cars secure from RF-based attacks for a hundred years, and now suddenly they become insecure without the addition of a single meaningful capability.
I could duck tape an ipad to the dash of my 10 year old car and have something more capable and more secure than a 1 year old car.