Very true. What should have happened, in an ideal world, and I'm not sarcastic here, was that a flight attendant would react to this "revelation" by looking at the customer and telling them "Are you fucking stupid?", perhaps in a more politically correct way. And that should have been the end of it.
The problem is systems where you need the entire GUI and a big chunk of the userland applications stack working to be able to read logs.
And of course that's not the case with systemd. So, what was your point again?
P.S. Even if it were true, it's rather trivial to have a statically linked Gnome or Qt application that runs on the framebuffer and can be started from busybox that got started straight from init or systemd.
As a very part-time server admin, I do in fact care about boot times, given that when the server is down, the phones are down, the security system's video recording is down, the internet access and internal networking is down, the environmental controls for the building are down, etc. Yes, one could argue that all of those should be in their own little VMs, but if you're doing security updates, it boils down to having update everything sooner or later - might as well have it all on one machine.
The problem with some aspects of Unix philosophy is that eventually you need a better design goal than "slice and dice". Sometimes you do actually need an integrated piece of software that handles more than one job coherently. The Unix philosophy works well for certain kinds of tools, but it utterly breaks others. I consider the init system, as usually implemented, to be a sad mess that was state-of-the-art maybe 30 years ago. It still made some sense when linux 1.0 was the hot new thing, and it was considered risky to run a departmental Linux server (as opposed to, say, some BSD). We now know better about software design I think, and I in fact like systemd. It offers way more functionality without having to write custom scripts to implement said functionality. Sure, one might argue, perhaps the problem was not init itself, but the scripts that made it tick. At the end of the day, though, I don't care: the init proponents could never deliver systemd's functionality in the scripts that come with init implementations. They've lost at their own game. There's no point in grumbling about that. I need functionality, systemd delivers it, init doesn't. End of story.
I think that it all boils down to necessary and necessarily deadly experimental aspect of the sciences and engineering. Just as we had to learn fracture mechanics by killing a bunch of people in BOAC Comets, we'll have to learn crop genetic engineering by fucking things up here and there. It's the necessary price of progress, I think.
wireless access points can be had for just a few dollars these days
What? I very much doubt this SSID was broadcast by a stand-alone AP. It was, likely, due to default behavior of old versions of Windows, setting up an ad-hoc network with an SSID of the last seen access point. Someone somewhere has jokingly set their SSID to "Al-Quida...", and there was that one Windows-running laptop that someone had that picked up on that SSID and kept broadcasting it. Even if someone set such an SSID on purpose on their mobile device, it's still irrelevant and inactionable.
Delaying a flight over this shows how much technical ineptitude is there.
I think that the slightly sad part of this story is that a source-level port of IE6 to, say, pango/cairo running on a frame buffer, would be much faster than Firefox when it comes to rendering basic web. Of course its JS completely sucked both performance- and functionality-wise.
The thing is that GMO is like coding: you can "write" all sorts of things. My only concern, and a concern that I think everyone somewhate educated should share, is that miRNA goes straight from the food we eat into our cells and can serve regulatory actions there. So if, due to insufficient testing, you release a GMO that produces miRNA that is deleterious to some animals (including us, humans), you might turn a whole lot of plants into toxic plants, and have a really hard job of eradicating them (they are fucking purposefully planted, after all). The biggest issue is that such deleterious miRNA might be a slow killer. Sure, there are a lot of "what ifs", but the truth is that as of now, there's zero animal testing of GMO, and zero testing of GMO for possible negative effects on humans etc. That must change, since it can't be but expected that over time, the extent of genetic modifications in GMOs will grow. So far we've been seemingly lucky. Who knows how long the luck will last.
The weirdest thing is when you grow up in a European country that has a more reality-based food pyramid instead of a politics-based one. Then you go to the U.S., look at a cereal carton, and go WTF?!
The first reference doesn't talk about evils of GMO, but about the evils of a particular herbicide. The second one talks about miRNA and how genetic material transfers directly from the food we eat into our bodies. This is not by itself pro- or anti-GMO, it's merely a strong point that supports proper testing of GMO foods - something that, admittedly, Monsanto has long argued unnecessary. Again, this doesn't make any particular GMO dangerous, it merely prompts at what should we look at when testing such organisms for consumption by humans and livestock. The third reference shows some fallout from RoundUp-resistance genes jumping from crops to weeds. Again, this doesn't show any danger ingerent in GMOs themselves, but in a particular modification. Just as software development techniques can be used for good and bad, the genetic modifications can be used for good and bad. We need to learn how to use them for good. DUH:)
Has Monstanto been demonstrably lying through its teeth to the public, repeatedly? Sure. There's no news here.
I'm not anyone's astroturfer, I merely recognize the fact that doing the desired genomic modifications directly, in a focused manner, is much more efficient than the inbreeding needed for standard breeding. Standard breeding is simply what the old technology allowed: we had no other means of tweaking the genome. Now we do. Sure we can create organisms that can be dangerous to some ecosystems, but in reality there's so much competition between the organisms that it isn't like we will immediately wipe out all edible plants worldwide. There can be bad things happenening here and there, that's a given, but that is a given no matter what genetic modification methods you use. That's also a given if you do nothing specific at all. Of course with GMO techniques, we're way more efficient at potentitally screwing up, so screwups are to be expected more often, even screwups that can have temporarily bad economic effects (say raising price of crop x by a small integer factor). In most cases, though, we know enough to weather such fuckups in the developed world. I certainly agree that fallout from GMO "oopses" can royally fuck up agriculture in less developed countries, or in areas with less diversity in food crops.
Look, we're just using desktop machines to do gene splicing, not the plants themselves. It's way more efficient that way. A lot of this "bioreactor" gene splicing consisted of waiting for the desired mutation to appear. Let's repeat: the humans often waited for the right mutation to appear, and then tried hard to preserve and spread those mutations in face of negative selection pressures. This is really no different than getting some jellyfish gene into a food crop, just that the road to the desired effect is much shorter.
So, you truly believe that the pests won't quickly evolve resistance to whatever mechanism is spread by those modified genes? Thanks, that's all I need to know. NEXT please.
It's really simple. There are commercial products (both hardware and software) that are offered for sale only to government agencies. You literally can't buy them if you don't work for some government entity, or if you don't have a direct and explicit authorization from a government entity. They are nonattributable to any particular government agency, but everyone in the know knows what company makes the product. The fact that you know the manufacturer doesn't make it attributable.
The major problem is that no matter how far it is, eventually it will be done, and at that point you're literally dealing with artificial life, but the entire infrastructure is not designed as if you had a lab with a highly infective pathogen.
Slashdot should have an automatic "-1" score assigned to any post that contains ("car" or "vehicle") and "high rate of speed". Some undereducated journalist somewhere has used it once, and it's spreading like a disease. Do words have no meaning at all to you?
For some reason I thought that Haier was the typical race-to-the-bottom appliance manufacturer that peddles short-lived, underengineered crap. I expect that the UX of this "innovation" will be a trainwreck. Why is this news?
I hate to sort of point out the obvious, but the muscles useful for push-ups are not necessarily useful for much else. I do all of my own house renovation, going up and down scaffoldings, carrying materials (including 90lb buckets of concrete), sanding drywall (best workout ever), I fix my cars including wrangling engines and transmissions around, etc. I can't do push ups, it wasn't something I ever had a need to do. It seems like a skill for its own sake.
Heck, the whole process to build Qt and its browser-du-jour (webkit then, chromium now) amounts to configure followed by make, and that's whether you're on Windows or Unix.
IBM's mainframe processing power is a bit different than your typical desktop processing power. Mainframes have, historically, never been used for much numerical computation, but are heavy data/string pushers. All the while a desktop CPU has several subsystems optimized to crunch numbers in a way that is not useful at all in a mainframe that pushes, say, product data around.
Amazon Prime isn't a handout, let there be no doubts about that. Even if you anecdotally have used it as a handout (I did, too), averaged across the customer base it's not a handout.
Slashdot Mirror
Very true. What should have happened, in an ideal world, and I'm not sarcastic here, was that a flight attendant would react to this "revelation" by looking at the customer and telling them "Are you fucking stupid?", perhaps in a more politically correct way. And that should have been the end of it.
The problem is systems where you need the entire GUI and a big chunk of the userland applications stack working to be able to read logs.
And of course that's not the case with systemd. So, what was your point again?
P.S. Even if it were true, it's rather trivial to have a statically linked Gnome or Qt application that runs on the framebuffer and can be started from busybox that got started straight from init or systemd.
As a very part-time server admin, I do in fact care about boot times, given that when the server is down, the phones are down, the security system's video recording is down, the internet access and internal networking is down, the environmental controls for the building are down, etc. Yes, one could argue that all of those should be in their own little VMs, but if you're doing security updates, it boils down to having update everything sooner or later - might as well have it all on one machine.
The problem with some aspects of Unix philosophy is that eventually you need a better design goal than "slice and dice". Sometimes you do actually need an integrated piece of software that handles more than one job coherently. The Unix philosophy works well for certain kinds of tools, but it utterly breaks others. I consider the init system, as usually implemented, to be a sad mess that was state-of-the-art maybe 30 years ago. It still made some sense when linux 1.0 was the hot new thing, and it was considered risky to run a departmental Linux server (as opposed to, say, some BSD). We now know better about software design I think, and I in fact like systemd. It offers way more functionality without having to write custom scripts to implement said functionality. Sure, one might argue, perhaps the problem was not init itself, but the scripts that made it tick. At the end of the day, though, I don't care: the init proponents could never deliver systemd's functionality in the scripts that come with init implementations. They've lost at their own game. There's no point in grumbling about that. I need functionality, systemd delivers it, init doesn't. End of story.
I think that it all boils down to necessary and necessarily deadly experimental aspect of the sciences and engineering. Just as we had to learn fracture mechanics by killing a bunch of people in BOAC Comets, we'll have to learn crop genetic engineering by fucking things up here and there. It's the necessary price of progress, I think.
wireless access points can be had for just a few dollars these days
What? I very much doubt this SSID was broadcast by a stand-alone AP. It was, likely, due to default behavior of old versions of Windows, setting up an ad-hoc network with an SSID of the last seen access point. Someone somewhere has jokingly set their SSID to "Al-Quida ...", and there was that one Windows-running laptop that someone had that picked up on that SSID and kept broadcasting it. Even if someone set such an SSID on purpose on their mobile device, it's still irrelevant and inactionable.
Delaying a flight over this shows how much technical ineptitude is there.
I think that the slightly sad part of this story is that a source-level port of IE6 to, say, pango/cairo running on a frame buffer, would be much faster than Firefox when it comes to rendering basic web. Of course its JS completely sucked both performance- and functionality-wise.
The thing is that GMO is like coding: you can "write" all sorts of things. My only concern, and a concern that I think everyone somewhate educated should share, is that miRNA goes straight from the food we eat into our cells and can serve regulatory actions there. So if, due to insufficient testing, you release a GMO that produces miRNA that is deleterious to some animals (including us, humans), you might turn a whole lot of plants into toxic plants, and have a really hard job of eradicating them (they are fucking purposefully planted, after all). The biggest issue is that such deleterious miRNA might be a slow killer. Sure, there are a lot of "what ifs", but the truth is that as of now, there's zero animal testing of GMO, and zero testing of GMO for possible negative effects on humans etc. That must change, since it can't be but expected that over time, the extent of genetic modifications in GMOs will grow. So far we've been seemingly lucky. Who knows how long the luck will last.
The weirdest thing is when you grow up in a European country that has a more reality-based food pyramid instead of a politics-based one. Then you go to the U.S., look at a cereal carton, and go WTF?!
The first reference doesn't talk about evils of GMO, but about the evils of a particular herbicide. The second one talks about miRNA and how genetic material transfers directly from the food we eat into our bodies. This is not by itself pro- or anti-GMO, it's merely a strong point that supports proper testing of GMO foods - something that, admittedly, Monsanto has long argued unnecessary. Again, this doesn't make any particular GMO dangerous, it merely prompts at what should we look at when testing such organisms for consumption by humans and livestock. The third reference shows some fallout from RoundUp-resistance genes jumping from crops to weeds. Again, this doesn't show any danger ingerent in GMOs themselves, but in a particular modification. Just as software development techniques can be used for good and bad, the genetic modifications can be used for good and bad. We need to learn how to use them for good. DUH :)
Has Monstanto been demonstrably lying through its teeth to the public, repeatedly? Sure. There's no news here.
I'm not anyone's astroturfer, I merely recognize the fact that doing the desired genomic modifications directly, in a focused manner, is much more efficient than the inbreeding needed for standard breeding. Standard breeding is simply what the old technology allowed: we had no other means of tweaking the genome. Now we do. Sure we can create organisms that can be dangerous to some ecosystems, but in reality there's so much competition between the organisms that it isn't like we will immediately wipe out all edible plants worldwide. There can be bad things happenening here and there, that's a given, but that is a given no matter what genetic modification methods you use. That's also a given if you do nothing specific at all. Of course with GMO techniques, we're way more efficient at potentitally screwing up, so screwups are to be expected more often, even screwups that can have temporarily bad economic effects (say raising price of crop x by a small integer factor). In most cases, though, we know enough to weather such fuckups in the developed world. I certainly agree that fallout from GMO "oopses" can royally fuck up agriculture in less developed countries, or in areas with less diversity in food crops.
Look, we're just using desktop machines to do gene splicing, not the plants themselves. It's way more efficient that way. A lot of this "bioreactor" gene splicing consisted of waiting for the desired mutation to appear. Let's repeat: the humans often waited for the right mutation to appear, and then tried hard to preserve and spread those mutations in face of negative selection pressures. This is really no different than getting some jellyfish gene into a food crop, just that the road to the desired effect is much shorter.
So, you truly believe that the pests won't quickly evolve resistance to whatever mechanism is spread by those modified genes? Thanks, that's all I need to know. NEXT please.
Think of what it takes to drive someone nuts. It takes precisely that: too nutty to be believeable, yet technically rather trivial.
Hypocrites can still be right, you know.
It's really simple. There are commercial products (both hardware and software) that are offered for sale only to government agencies. You literally can't buy them if you don't work for some government entity, or if you don't have a direct and explicit authorization from a government entity. They are nonattributable to any particular government agency, but everyone in the know knows what company makes the product. The fact that you know the manufacturer doesn't make it attributable.
That might well be the case :)
The major problem is that no matter how far it is, eventually it will be done, and at that point you're literally dealing with artificial life, but the entire infrastructure is not designed as if you had a lab with a highly infective pathogen.
Slashdot should have an automatic "-1" score assigned to any post that contains ("car" or "vehicle") and "high rate of speed". Some undereducated journalist somewhere has used it once, and it's spreading like a disease. Do words have no meaning at all to you?
For some reason I thought that Haier was the typical race-to-the-bottom appliance manufacturer that peddles short-lived, underengineered crap. I expect that the UX of this "innovation" will be a trainwreck. Why is this news?
I'm sure I could pull off a brisk walk with plenty of gear. What's that got to do with fucking push-ups that everyone is so fixated with?
I hate to sort of point out the obvious, but the muscles useful for push-ups are not necessarily useful for much else. I do all of my own house renovation, going up and down scaffoldings, carrying materials (including 90lb buckets of concrete), sanding drywall (best workout ever), I fix my cars including wrangling engines and transmissions around, etc. I can't do push ups, it wasn't something I ever had a need to do. It seems like a skill for its own sake.
Heck, the whole process to build Qt and its browser-du-jour (webkit then, chromium now) amounts to configure followed by make, and that's whether you're on Windows or Unix.
IBM's mainframe processing power is a bit different than your typical desktop processing power. Mainframes have, historically, never been used for much numerical computation, but are heavy data/string pushers. All the while a desktop CPU has several subsystems optimized to crunch numbers in a way that is not useful at all in a mainframe that pushes, say, product data around.
Amazon Prime isn't a handout, let there be no doubts about that. Even if you anecdotally have used it as a handout (I did, too), averaged across the customer base it's not a handout.