So, we can either continue to support all of those modes, or we can drop the "old cruft" and maybe the people who aren't so cheap as to be unwilling to spend $20 on a video card can have something that looks a bit nicer.
Hang on... The PVR-350 isn't a $20 video card, it's a very specialized card, which I don't believe there's a direct replacement for, anywhere. On the same token, XvMC is also the only hardware acceleration supported by many video cards which support component & SVideo output, which can be pretty important on slower hardware to support playback of HDTV broadcasts... I was using it for years, until just recently, though I admit I wouldn't have had a problem using older versions of the software.
And I had a hell of a time recently, finding an NVidia PCI-Express video card for sale that supports SVideo-output, because I don't feel like entirely throwing away the big old TV. Even if I opt to spend a $1,000+ to replace my old TV, I'll probably still keep the old one, delegated to the game-room, or some such, and want to be able to pump out a picture to it.
I was lucky to find an old, refurbished NVidia 8500 PCIe card for sale (sold-out now, good-luck everyone else...), so I'm not stuck on AGP or XvMC to drive my trusty old TV, but ironically, the bracket it came with was BLOCKING the SVideo plug entirely, requiring me to drill it out to use it, and showing just how much love there is for those of us not on the bleeding-edge of the throw-away culture. I'd think with the recession our ranks would be growing, but I sure don't see it in the Linux community, where it used to be a very strong segment.
You can find some amazing deals out there, if you look for them:
Linux has always made it possible to squeeze all the life out of old, cheap hardware like the above, but it seems that thread in the community is getting thinner by the day. In the case of the above, RHEL5 drivers for the intel video work vastly better than the KMS-requiring drivers in RHEL6.x / Fedora, and no, it doesn't seem to want to work with any add-in PCI 2.2 cards, but who would?
If you want to get up an hour early in the summer, get your ass out of bed!
Why should the rest of us screw up our sleep rhythms because you don't want to reset your alarm clock?
Because most people DO STUFF. Getting up arbitrarily an hour earlier, will just mean you're sitting around doing nothing for an hour, until stores open, you start work, etc.
If you're so concerned about your sleep, why don't you START getting up a few minutes earlier each day, leading up to DST? Are you so dense and helpless that you can't plan a week or two ahead?
The thing that makes good tablets and smartphones useful just don't exist in cheap tablets. Little things like a less responsive touch screen take them from being a pleasure to use, to an exercise in frustration. The same goes for the speed and capacity of the device to a lesser extent. Software is a huge issue for the average person, too, and the cheaper devices have crap. Of course we can root it and install 3rd party Android distros that will work, but that won't appeal to the mass market so much,
Yes, sadly that IS true NOW, but it's only been in the past few years that there was such a steep decline of experts and increase of worthless noise. It certainly wasn't true during 9/11 which most definitely stands as perhaps the defining moment of/. Lots of people were going elsewhere for their news, and NOT GETTING IT./. simply had the capacity to support the traffic spike, a large enough readership contributing information, and enough smart people contributing insights, that it had the best coverage of a world-changing event right in the middle of the fog of war and flood of baseless rumors from even reputable news organizations. It is one of the most interesting times in my lifetime, and I highly recomend reading the 9/11/. threads, and listening to Howard Stern's broadcast to get a real sense of what it was like on that first terrible and confusing day.
I was reminded of that moment a short while ago when the news of the Virginia earthquake broke the mold of ancient news, and immediately showed up on the front page. It didn't mean anything, it just showed a little bit of that old spark is still here. Mostly, it just triggered the memory, and I don't believe/. really can be the same again.
No one says a website is "yahooed" or "googled" or "Drudged" if it's been hit by a enmass of visitors after being posted on slashdot or any other site.
Indeed not... They say it has been "Farked", which can best be described as the/. for morons, or the hive mind that spat out the memes before the Anonymous/b/tards.
Can't really make anything of it, though. Reddit and Dig just don't lend themselves to convertion into unique verbs.
Personally, I've seen slashdot go through a lot, but I'm not sure if it will handle the latest wave... Editors throwing out flamebait stories after trolling stories, after another, increasing the noise greatly. It also seems the number of stories has risen, so they're off the front page faster, and don't lend themselves to discussion, just the first few bits of commentary, then irrelevent... This coincided with a dramatic change in moderation, where the insightful and highly technical post get modded down or lost, and the feel-good flame that can't be bothered with pesky facts gets the +5 mod. I've seen/. go through a lot, but I'm not so sure it's going to survive this one, as a lot of the amazingly smart people have left, or at least barely ever contribute anymore. I've dramatically cut back on my contributions the same way. Will/. get better, and even if it does, tomorrow, will there be anything left of the old crew?
Sounds like old-timers syndrome, I know, but it's not. Sometimes I'll look through old forgotten bookmarks to slashdot stories, or turn something up in a web search, and be astonished at how vastly higher the concentration of real, insightful, experts was, on a number of different subjects. Now there's an awful lot of highly rated noise and misinformation from people who know just enough to be dangerous. Check the old threads... It's amazing what a change it was... it didn't even seem like a dramatic change over the course of months and years, though there were certainly vocal objections when it started.
I admit, I've seen/. come through some very bad periods, most often due to the moderation system losing it's smarts, or sometimes being intentionally gamed by the flamers and trolls. "First post" gags are emblematic of that kind of gamesmanship way back when, but this seems different than all the rest. Editors have contributed greatly to the decline, and it seems like this is the/. Equivalent of the Endless September.
I wouldn't think CmdrTaco's leaving would improve things, but there's always that chance in a shake-up, and I can only hope for some, any deus ex to come out and start to fix it. I too learned many things from/. over the years, but that capacity for such high-level intelligent discussions appears to have been gone for perhaps a few years now.
if you find a part of fedora that does not work with your particular setup then you can just replace it
The biggest part that doesn't work is THE INSTALLER. If you've got a drop-in replacement for that, POINT ME TO IT, because ANACONDA was never impressive, and it's only gotten worse (VNC mode, my ass!).
Or go back to Windows/AppleOS.
Actually, I'd rather go back to a decent Linux distro, say, Slackware... Or maybe FreeBSD, with that wonderful ZFS support (rather than waiting forever for btrfs).
even high-end enterprise hardware is not up against those limits yet.
That's utterly idiotic. Enterprise SATA drives may lag behind consumer drives, but not by much, and enterprises will ALWAYS have raid arrays (multiplying the sizes), while consumers rarely do. Enterprise SAS drives are 750GBytes now, so it only takes a small array to break the 2TB mark there.
IN AN ENTERPRISE, I had to deal with low-end servers that had over 2TB boot volumes, years ago, running on RHEL4.x (not a typo... RHEL4.x, as in BEFORE RHEL5, and way before RHEL6). It was a PITA, and I can't believe how far behind RedHat is on bringing out full and stable support of GPT installation and booting.
I can't !@##$%^& believe how Redhat and Fedora can't handle GPT right to save their !@#$@#% lives...
RHEL5 should have included the GRUB patches to handle booting from GPT partitions... They did not. When it was released, 1TB SATA drives were available. RAID-5 with 4x1TB drives? Sorry, no, can't boot from it, unless your RAID controller is smart enough to divvy it up into pieces to make up for for the idiots at RH. I've gone through all kinds of !@##$% because RHEL5 is THE enterprise Linux operating system, and yet it doesn't !@##@$% support installing to, or booting from GPT partitions, which means you're limited to 2TB volumes, max.
Okay, so they made a bad decision, but newer versions will solve all our problems, right? Wrong! Fedora goes back and forth with @$#$%%$^@ bugs around ANACONDA and GPT. Today, I can boot-up with a Fedora 15 disc, go through the menus, take a quick look at the layout, and find I've got 1/3rd unallocated on my 3TB hard drive, because it's using old msdos partitions, and there's no way for me to tell it to use GPT. @#$#$^$%! Put a GPT signature on it you say? Okay, now ANACONDA detects the disk is corrupted and asks if it should abort or wipe it out...
Partition everything manually, you say? Well I would, but GNU parted is absolutely the most god-awful tool I've ever used...
Okay here goes... mkpart 1GB 10GB WARNING: Not aligned, performance will be terrible... Ignore/Cancel?
WTF? It converts my human units into billions of sectors, and can't be bothered to round it off to the nearest multiple of 8, or friggin' ask me if I want it to do so? Who the hell made this crap? Math is what computers are so damn good at, WTF do I need to pull out a pocket calculator to partition my !@#$#$% computer in the year 2011?
Redhat drives me nuts. Imagine if the most popular luxury sports-car maker out there engineered their cars so that they couldn't be driven for more than 1hr straight, before shutting-down and needing to be restarted. That's the kind of fundamental stupidity we're talking about, here. Middle-of-the-road consumer hardware is over 2TB now, when are they going to fix this !@#$#%?
I know, I know, I don't need to use parted. Someone else had half a brain and hated parted, too, and made gdisk just for this purpose: http://www.rodsbooks.com/gdisk/ It doesn't take away from the sheer idiocy of either project...
In the past three years, solid state disks have dropped by more than a factor of twenty. Thus, SSDs are dropping in price more than four times as fast as hard drives. If we extrapolate that into the future, the current order of magnitude will go away in about seven or eight years without this added burden on hard drives.
And if we extrapolate further, within 10 years manufacturers will be PAYING YOU to take SSDs off their hands.
Your non-root example is nonsense. You can install packages under an alternate root, as a normal user, quite easily. It's up to you to set your PATH and LD_LIBRARY_PATH properly, but that's a one-time thing.
Hell, just jail/chroot the user into a folder, and build-out a full system under that tree. Your package manager will be none-the-wiser.
You're way off. Yes, the kernel of any OS is basically a hypervisor, but no, there's NO REASON for that to change. VMWare's approach was never a good one, and it was strictly useful because of brain-dead OSes out there which aren't stable, and don't allow you to basically have your own entire OS running at user-level, while any Unix system can do exactly that. Paravirtualization only takes this a step further, so you have your own entirely seperate kernel as well, but again, except in a few narrow cirumstances, this isn't needed. One kernel can run as many separate userlands as you like, with entirely different programs, libs, etc.
And Paravirtualization isn't the only thing blurring the line of virtualization. Try KVM... The kernel is the hypervisor again.
Frankly, for all Unix virtualization, we need a library that makes people THINK they're running as root within their userland on a multiuser system. Call it para-paravirtualization...
No MITM attacks, at least not easily (need to crack an entire CA, or at least steal the secret key).
If I can change a couple lines in the whois info for your DNS record, I will have a valid SSL certificate (from a reputible authority) for your domain in a matter of minutes.
DNSSEC is better because all mistakes the certificate authorities can make are completely taken out of the equation, and you ONLY have to trust the DNS authorities.
Actually, I'm with the GP. All SSL has ever told us, and all that DNSSEC will ever tell us, is that you're talking to a guy who controls the DNS record... Your registrar, or anyone along the chain could change your whois data, and get a valid SSL certificate issued to themselves, or could change your DNSSEC info and redirect you to a new server.
It's much more agile if everyone is their own registrar. You need to somehow verify the organization is legit before submitting your data to them, but SSL and DNSSEC certainly won't do this, so it's nearly worthless. Once you've decided to accept a self-signed certificate, any other organization taking control of that domain will NOT be able to pretend they are the previous people, because their self signed or purchased cert doesn't match.
A different authoritity for each domain, and we're all better off. Yes, you have the problem of verifying the key for each new site, but at least we're not pretending that DNS control proves your are who you say you are.
Blackberry, HP and Google told him to take a hike so the only credible option left was WP7.
That's not true. It's right there on the first page that Nokia told Google to take a hike, when Google refused to give them a competitive advantage over other phone makers:
He tried to negotiate a deal with Google to run Android, but Google refused to give the world's biggest phonemaker any advantages over its smaller partners
So instead they go with Windows Phone 7? What?
Elop says to the crowd. "We'd be just another company distributing Android. That's not Nokia! We need to fight!"
Instead of being just another company distributing Android, they're just another company distributing Windows Phone 7. Awesome upgrade! Get-in on the ground-floor of something no-one wants, in exchange for, still, no exclusivity to speak of.
Of course we know what they got... Over a billion in payouts from Microsoft. And people say it's hard to beat "free" (Android).
Key takeaway is that hiring open source evangelists to design a mobile OS(i.e Meego) failed
Nokia was horribly mis-managed. Who were the open source evangelists in management and director-level positions that are singularly to-blame for everything bad that happened to the company over the past half-decade?
N900 is worlds better than anything iOS/Android-laden: instead of a limited toy OS with a browser, media player and fart apps, it has a general purpose operating system in a smartphone-sized form -- effectively a very, very small laptop
SSH / VNC / RDP support on Android is pretty damn good, all with free apps. I'm disappointed I've never found an NX client for Android, but I can live without it. Since I do real work on my Android phone, and it has entirely eliminated my need for a laptop, you've failed to convince me that the N900 is awesome, and Android sucks. And yes, you can get a decent Android slider for sub-$200, with a $25/mo unlimited data plan thanks to Virgin Mobile.
We think of the WWW as being an OK thing these days, only because of years of hard-fought experience. Think of Geocities pages with wild-ass layouts. But back in the early days, imagine your bank, e-commerce sites, university sites, government sites, search engines, etc., all looked like they were put together by a child, and looked like Geocities.
HTML has no standard layout. It's completely random. Whatever the hell you want to do. Whatever technologies you want to throw in. It took two decades for everyone to figure out that a navigation column (NOT A FRAME, NOT A SITEMAP, NOT AN IMAGE MAP), and with minimal changes as you navigate, is universally the way to go. So many times it was a nightmare trying to navigate a site, and you may NEVER be able to figure out how to get back to where you were, because the navigation column keeps changing the available options with each click. It was insane, and really, still isn't great.
Still today, you have to LEARN every new website you visit, like a child. Is the nav bar something you have to click on, or will it pop-up more options when you hover over it? Are these top-level categories that will just take you to more navigation pages, or will they take you to a page on the subject with high-level information, or such?
And how many years have we been hearing about the semantic web, and more and more markup to fill-in the gaps? I've got news for you... Gopher was great. It was strict about the format and layout of the page. Each page like navigating a book, but more importantly, every page worked exactly the same way. Imagine if the entire internet looked like a nice clean Wikipedia page... Mostly body text, links which take you to a list of references at the bottom, etc. No random layout crap. No advanced parsing to figure it all out. No need to learn to navigate anew with each new site you visit.
Hell, look no further than slashdot. With each revision I just keep changing settings to give me as old an simple of a layout as I can get. Nested comments, threshold 3, no-nonsense. HTTP and HTML were an anti-feature... The freedom to add cruft, and turn a nice interactive book into an awful DVD Menu of a thing. Things are only slowly improving as web designers and users discover how awful all the flexibility is, and collectively move towards a standard layout, which unfortunately is poorly typed and implemented a hundred different ways.
If Gopher had simply been improved, and HTML had never been, we wouldn't have "simple" web pages, or "mobile" web pages, because the layout on a phone would be just as good as on a desktop. So very much was lost, and very little was gained.
For experts, they are also great, because they reward knowledge, are searchable, and save screen real estate.
Here's the problem... Reading comments here, or anywhere else for that matter, you'll see EVERYONE disagrees with you. Nobody likes the new interfaces, nobody finds them easier to use. etc.
You can spout off on how great they are all you want, but most of the people who are there using it, hate everything about it.
His skills are orthogonal to UI, and his decisions were bad in the past (KDE!! WTF!?).
I fail to see your point. I'm not a fan of KDE, but I admit that until KDE4, it was a solid environment, with a control center that allowed you to change damn near any setting you could want to, either in the DE or in the system in general. With 4.x they've sadly regressed into copying the worst aspects of Windows, much as GNOME 3 has copied the worst of OS X.
Most importantly, they are designed by specialists, with the user in mind, and actual tests, with actual users. A kernel developers opinion is not that relevant here.
Appeals to authority are worthless... Experts make stupid decisions all the time. User studies are NOTORIOUS for getting things utterly and totally wrong. People will say they love X during testing, then when it comes to actually using X, find it absolutely horrible.
In short, with Windows having such a large market share, imitating Windows EXACTLY would get great reviews, and the experts will spout about how new and innovative and shiny it is. Yet that doesn't mean the interface isn't a horrible piece of crap.
It's about the same as copying the controls in a car to a helicopter... Sure, everyone will be comfortable with it right away, and say what a great interface it is, but when it comes time to use it for real, day to day work, you'll find it's horribly crippled, and you can't actually do your job with it.
My desktop of choice? Blackbox (with plenty of config changes to make it behave like Openbox 2.x did by default...). I can launch any application in a fraction of a second; shade, iconify; maximize vertically, horizontally, or both; restore, switch to, etc., faster than I can in any other desktop environment. It takes up tiny amount of screen real-estate, while GNOME and KDE are sucking up ever more screen space with each release... It give me instant access to all the menus I need, just about anywhere my cursor happens to be, on-screen.
You want some usability testing? Grab a stopwatch and time a long-time blackbox user against a long-time GNOME user... Who do you think will get slaughtered?
Now, Blackbox has a learning curve, I admit. But Fluxbox, which comes with a menu button and iconified windows on the (tiny) Toolbar, is basically a lightweight and compact mimic of Windows 95, and any idiot can pick it up and run with it, while still having the full power available as soon as they discover it. And when it comes time to configure it, you aren't spending hours in abstracted control-panel rip-offs, which require an exhaustive search to figure out how the hell to add a couple keyboard shortcuts... Hell, you can look through every possible config option in 30 seconds. Now it's of course not a full DE, but it gets you 95% of the way there, without tying you to a cement truck and sapping all productivity. Throw in a couple supporting apps and you're gold.
Like I said... stopwatch. Completely objective measurement of the utility of an interface. Not subjective, PR, "designer" circle-jerk, bullshit flavor-of-the-week, crap. And anybody who uses it day-in and day-out cares a hell of a lot about what the stopwatch has to say. New users would benefit greatly from the stopwatch test as well.
im sure that nuclear waste can be stored safely, somewhere, some how. but the current nuclear industry is so obsessed with lying, disinformation, and corruption, that i wouldn't trust it to clean the dishes at a restaurant let alone run something like the Fukushima plant.
What does the "nuclear industry" have to do with anything? In the US, it's the DoE that makes the decisions. The DoE happen to have pretty much the world's top experts in the field, thanks to building and maintaining the top nulcear weapon stockpile, and has a few neat toys like periodically aquiring one after another of the world's fastest supercomputers...
In Japan, TEPCO was widely known to be untrustworthy before the disasterw and NISA was criticized for being in-bed with industry. I have not, however, seen such criticism of the DoE from pretty much any quarter at all. In fact the DoE seems to be about the most trusted and least controvercial of the large government agencies that I can think of.
The scavenging device could piggy-back solar energy panels so that, when the system stops generating power at sundown, the wireless energy could be used overnight to increase the battery charge or to prevent power leakage. The devices would also be useful in remote areas where an outage of a traditional power source could be flagged by sending a distress signal from an antenna-powered unit.
These are incredibly stupid ideas...
If you're using even the tiniest of solar panels for power, the extra power from this thing wont even register. Maybe if you're using it as a backup for a tiny thermocouple it'll help, but that's about the only other power source in the same ballpark...
If you've got a "power source" and want to send a distress signal when it goes out, you store up a bit of power from your main source, and can completely forego this antenna. The derission of batteries is nonsense (give me one NiMH cell...), but even if you buy the premise, one of those mentioned "ultracapacitors" charged from the grid would be vastly more suitable than using this power antenna at all...
Considering the height of radio/TV towers, the direct path of travel is mostly going to be into the ground anyway. The energy this would pick up would be wasted anyway.
This is only true if you're fairly close to the broadcasting antenna. Thanks to the curvature of the earth, what's on the top of a mountain, 50 miles away, is now at ground-level... Any yes, at that range your neighbors on the opposite side are picking up TV signals barely above ground level where you are, if not picking-up on signals that are actually reflecting off the ground. You often see high-gain antennas angled downward for these reasons, and there certainly is some opportunity to degrade signal strength for your neighbors.
What jobs would those be? Where are they? The fear that people have is justified because there simply aren't enough jobs if you take away repetitive labor.
What you're saying is the fear everyone has had for the past couple centuries. The fact that there are jobs for us humans RIGHT NOW is incontrovertible proof of the theory.
Of course the jobs aren't around RIGHT NOW. They slowly come along as workers get freed-up from other lines of work. The entire IT industry counts as an example. A century ago, we would have either been working on the farm, or stamping out metal, day-in and day-out. Now that IT has helped automate much of that, IT jobs have become viable. I don't see any reason to believe that the process which has been constantly ongoing for centuries, will suddenly cease working and come to a stop in the near future.
If you want to see socialism in action look at Sweden, if you want to see capitalism in action look at China,
China isn't capitalist by a long stretch. They've got extensive subsidies, partial state ownership of many large companies, heavy regulations about how foreign companies can and can't invest, etc. China is an example of totalitarianism mixed with some capitalism (whereas it's always been democracy elsewhere). It's most certainly NOT pure capitalism.
Slashdot Mirror
Hang on... The PVR-350 isn't a $20 video card, it's a very specialized card, which I don't believe there's a direct replacement for, anywhere. On the same token, XvMC is also the only hardware acceleration supported by many video cards which support component & SVideo output, which can be pretty important on slower hardware to support playback of HDTV broadcasts... I was using it for years, until just recently, though I admit I wouldn't have had a problem using older versions of the software.
And I had a hell of a time recently, finding an NVidia PCI-Express video card for sale that supports SVideo-output, because I don't feel like entirely throwing away the big old TV. Even if I opt to spend a $1,000+ to replace my old TV, I'll probably still keep the old one, delegated to the game-room, or some such, and want to be able to pump out a picture to it.
I was lucky to find an old, refurbished NVidia 8500 PCIe card for sale (sold-out now, good-luck everyone else...), so I'm not stuck on AGP or XvMC to drive my trusty old TV, but ironically, the bracket it came with was BLOCKING the SVideo plug entirely, requiring me to drill it out to use it, and showing just how much love there is for those of us not on the bleeding-edge of the throw-away culture. I'd think with the recession our ranks would be growing, but I sure don't see it in the Linux community, where it used to be a very strong segment.
You can find some amazing deals out there, if you look for them:
http://www.geeks.com/details.asp?invtid=SAMBA845V-24-4-R&cat=SYS
Linux has always made it possible to squeeze all the life out of old, cheap hardware like the above, but it seems that thread in the community is getting thinner by the day. In the case of the above, RHEL5 drivers for the intel video work vastly better than the KMS-requiring drivers in RHEL6.x / Fedora, and no, it doesn't seem to want to work with any add-in PCI 2.2 cards, but who would?
Because most people DO STUFF. Getting up arbitrarily an hour earlier, will just mean you're sitting around doing nothing for an hour, until stores open, you start work, etc.
If you're so concerned about your sleep, why don't you START getting up a few minutes earlier each day, leading up to DST? Are you so dense and helpless that you can't plan a week or two ahead?
I've used cheap tablets. You obviously haven't.
The thing that makes good tablets and smartphones useful just don't exist in cheap tablets. Little things like a less responsive touch screen take them from being a pleasure to use, to an exercise in frustration. The same goes for the speed and capacity of the device to a lesser extent. Software is a huge issue for the average person, too, and the cheaper devices have crap. Of course we can root it and install 3rd party Android distros that will work, but that won't appeal to the mass market so much,
Yes, sadly that IS true NOW, but it's only been in the past few years that there was such a steep decline of experts and increase of worthless noise. It certainly wasn't true during 9/11 which most definitely stands as perhaps the defining moment of /. Lots of people were going elsewhere for their news, and NOT GETTING IT. /. simply had the capacity to support the traffic spike, a large enough readership contributing information, and enough smart people contributing insights, that it had the best coverage of a world-changing event right in the middle of the fog of war and flood of baseless rumors from even reputable news organizations. It is one of the most interesting times in my lifetime, and I highly recomend reading the 9/11 /. threads, and listening to Howard Stern's broadcast to get a real sense of what it was like on that first terrible and confusing day.
I was reminded of that moment a short while ago when the news of the Virginia earthquake broke the mold of ancient news, and immediately showed up on the front page. It didn't mean anything, it just showed a little bit of that old spark is still here. Mostly, it just triggered the memory, and I don't believe /. really can be the same again.
Indeed not... They say it has been "Farked", which can best be described as the /. for morons, or the hive mind that spat out the memes before the Anonymous /b/tards.
Can't really make anything of it, though. Reddit and Dig just don't lend themselves to convertion into unique verbs.
Personally, I've seen slashdot go through a lot, but I'm not sure if it will handle the latest wave... Editors throwing out flamebait stories after trolling stories, after another, increasing the noise greatly. It also seems the number of stories has risen, so they're off the front page faster, and don't lend themselves to discussion, just the first few bits of commentary, then irrelevent... This coincided with a dramatic change in moderation, where the insightful and highly technical post get modded down or lost, and the feel-good flame that can't be bothered with pesky facts gets the +5 mod. I've seen /. go through a lot, but I'm not so sure it's going to survive this one, as a lot of the amazingly smart people have left, or at least barely ever contribute anymore. I've dramatically cut back on my contributions the same way. Will /. get better, and even if it does, tomorrow, will there be anything left of the old crew?
Sounds like old-timers syndrome, I know, but it's not. Sometimes I'll look through old forgotten bookmarks to slashdot stories, or turn something up in a web search, and be astonished at how vastly higher the concentration of real, insightful, experts was, on a number of different subjects. Now there's an awful lot of highly rated noise and misinformation from people who know just enough to be dangerous. Check the old threads... It's amazing what a change it was... it didn't even seem like a dramatic change over the course of months and years, though there were certainly vocal objections when it started.
I admit, I've seen /. come through some very bad periods, most often due to the moderation system losing it's smarts, or sometimes being intentionally gamed by the flamers and trolls. "First post" gags are emblematic of that kind of gamesmanship way back when, but this seems different than all the rest. Editors have contributed greatly to the decline, and it seems like this is the /. Equivalent of the Endless September.
I wouldn't think CmdrTaco's leaving would improve things, but there's always that chance in a shake-up, and I can only hope for some, any deus ex to come out and start to fix it. I too learned many things from /. over the years, but that capacity for such high-level intelligent discussions appears to have been gone for perhaps a few years now.
The biggest part that doesn't work is THE INSTALLER. If you've got a drop-in replacement for that, POINT ME TO IT, because ANACONDA was never impressive, and it's only gotten worse (VNC mode, my ass!).
Actually, I'd rather go back to a decent Linux distro, say, Slackware... Or maybe FreeBSD, with that wonderful ZFS support (rather than waiting forever for btrfs).
That's utterly idiotic. Enterprise SATA drives may lag behind consumer drives, but not by much, and enterprises will ALWAYS have raid arrays (multiplying the sizes), while consumers rarely do. Enterprise SAS drives are 750GBytes now, so it only takes a small array to break the 2TB mark there.
IN AN ENTERPRISE, I had to deal with low-end servers that had over 2TB boot volumes, years ago, running on RHEL4.x (not a typo... RHEL4.x, as in BEFORE RHEL5, and way before RHEL6). It was a PITA, and I can't believe how far behind RedHat is on bringing out full and stable support of GPT installation and booting.
I can't !@##$%^& believe how Redhat and Fedora can't handle GPT right to save their !@#$@#% lives...
RHEL5 should have included the GRUB patches to handle booting from GPT partitions... They did not. When it was released, 1TB SATA drives were available. RAID-5 with 4x1TB drives? Sorry, no, can't boot from it, unless your RAID controller is smart enough to divvy it up into pieces to make up for for the idiots at RH. I've gone through all kinds of !@##$% because RHEL5 is THE enterprise Linux operating system, and yet it doesn't !@##@$% support installing to, or booting from GPT partitions, which means you're limited to 2TB volumes, max.
Okay, so they made a bad decision, but newer versions will solve all our problems, right? Wrong! Fedora goes back and forth with @$#$%%$^@ bugs around ANACONDA and GPT. Today, I can boot-up with a Fedora 15 disc, go through the menus, take a quick look at the layout, and find I've got 1/3rd unallocated on my 3TB hard drive, because it's using old msdos partitions, and there's no way for me to tell it to use GPT. @#$#$^$%! Put a GPT signature on it you say? Okay, now ANACONDA detects the disk is corrupted and asks if it should abort or wipe it out...
Partition everything manually, you say? Well I would, but GNU parted is absolutely the most god-awful tool I've ever used...
Okay here goes... mkpart 1GB 10GB
WARNING: Not aligned, performance will be terrible... Ignore/Cancel?
WTF? It converts my human units into billions of sectors, and can't be bothered to round it off to the nearest multiple of 8, or friggin' ask me if I want it to do so? Who the hell made this crap? Math is what computers are so damn good at, WTF do I need to pull out a pocket calculator to partition my !@#$#$% computer in the year 2011?
Redhat drives me nuts. Imagine if the most popular luxury sports-car maker out there engineered their cars so that they couldn't be driven for more than 1hr straight, before shutting-down and needing to be restarted. That's the kind of fundamental stupidity we're talking about, here. Middle-of-the-road consumer hardware is over 2TB now, when are they going to fix this !@#$#%?
I know, I know, I don't need to use parted. Someone else had half a brain and hated parted, too, and made gdisk just for this purpose: http://www.rodsbooks.com/gdisk/ It doesn't take away from the sheer idiocy of either project...
Have you seen either in retail stores? Then they're completely unaffected. IBM spun-off its retail PC business, too, yet AIX remains alive.
And if we extrapolate further, within 10 years manufacturers will be PAYING YOU to take SSDs off their hands.
No, your conclusions aren't any less flawed.
Your non-root example is nonsense. You can install packages under an alternate root, as a normal user, quite easily. It's up to you to set your PATH and LD_LIBRARY_PATH properly, but that's a one-time thing.
Hell, just jail/chroot the user into a folder, and build-out a full system under that tree. Your package manager will be none-the-wiser.
You're way off. Yes, the kernel of any OS is basically a hypervisor, but no, there's NO REASON for that to change. VMWare's approach was never a good one, and it was strictly useful because of brain-dead OSes out there which aren't stable, and don't allow you to basically have your own entire OS running at user-level, while any Unix system can do exactly that. Paravirtualization only takes this a step further, so you have your own entirely seperate kernel as well, but again, except in a few narrow cirumstances, this isn't needed. One kernel can run as many separate userlands as you like, with entirely different programs, libs, etc.
And Paravirtualization isn't the only thing blurring the line of virtualization. Try KVM... The kernel is the hypervisor again.
Frankly, for all Unix virtualization, we need a library that makes people THINK they're running as root within their userland on a multiuser system. Call it para-paravirtualization...
If I can change a couple lines in the whois info for your DNS record, I will have a valid SSL certificate (from a reputible authority) for your domain in a matter of minutes.
DNSSEC is better because all mistakes the certificate authorities can make are completely taken out of the equation, and you ONLY have to trust the DNS authorities.
Actually, I'm with the GP. All SSL has ever told us, and all that DNSSEC will ever tell us, is that you're talking to a guy who controls the DNS record... Your registrar, or anyone along the chain could change your whois data, and get a valid SSL certificate issued to themselves, or could change your DNSSEC info and redirect you to a new server.
It's much more agile if everyone is their own registrar. You need to somehow verify the organization is legit before submitting your data to them, but SSL and DNSSEC certainly won't do this, so it's nearly worthless. Once you've decided to accept a self-signed certificate, any other organization taking control of that domain will NOT be able to pretend they are the previous people, because their self signed or purchased cert doesn't match.
A different authoritity for each domain, and we're all better off. Yes, you have the problem of verifying the key for each new site, but at least we're not pretending that DNS control proves your are who you say you are.
That's not true. It's right there on the first page that Nokia told Google to take a hike, when Google refused to give them a competitive advantage over other phone makers:
So instead they go with Windows Phone 7? What?
Instead of being just another company distributing Android, they're just another company distributing Windows Phone 7. Awesome upgrade! Get-in on the ground-floor of something no-one wants, in exchange for, still, no exclusivity to speak of.
Of course we know what they got... Over a billion in payouts from Microsoft. And people say it's hard to beat "free" (Android).
Nokia was horribly mis-managed. Who were the open source evangelists in management and director-level positions that are singularly to-blame for everything bad that happened to the company over the past half-decade?
SSH / VNC / RDP support on Android is pretty damn good, all with free apps. I'm disappointed I've never found an NX client for Android, but I can live without it. Since I do real work on my Android phone, and it has entirely eliminated my need for a laptop, you've failed to convince me that the N900 is awesome, and Android sucks. And yes, you can get a decent Android slider for sub-$200, with a $25/mo unlimited data plan thanks to Virgin Mobile.
So the N900 is special, how exactly?
We think of the WWW as being an OK thing these days, only because of years of hard-fought experience. Think of Geocities pages with wild-ass layouts. But back in the early days, imagine your bank, e-commerce sites, university sites, government sites, search engines, etc., all looked like they were put together by a child, and looked like Geocities.
HTML has no standard layout. It's completely random. Whatever the hell you want to do. Whatever technologies you want to throw in. It took two decades for everyone to figure out that a navigation column (NOT A FRAME, NOT A SITEMAP, NOT AN IMAGE MAP), and with minimal changes as you navigate, is universally the way to go. So many times it was a nightmare trying to navigate a site, and you may NEVER be able to figure out how to get back to where you were, because the navigation column keeps changing the available options with each click. It was insane, and really, still isn't great.
Still today, you have to LEARN every new website you visit, like a child. Is the nav bar something you have to click on, or will it pop-up more options when you hover over it? Are these top-level categories that will just take you to more navigation pages, or will they take you to a page on the subject with high-level information, or such?
And how many years have we been hearing about the semantic web, and more and more markup to fill-in the gaps? I've got news for you... Gopher was great. It was strict about the format and layout of the page. Each page like navigating a book, but more importantly, every page worked exactly the same way. Imagine if the entire internet looked like a nice clean Wikipedia page... Mostly body text, links which take you to a list of references at the bottom, etc. No random layout crap. No advanced parsing to figure it all out. No need to learn to navigate anew with each new site you visit.
Hell, look no further than slashdot. With each revision I just keep changing settings to give me as old an simple of a layout as I can get. Nested comments, threshold 3, no-nonsense. HTTP and HTML were an anti-feature... The freedom to add cruft, and turn a nice interactive book into an awful DVD Menu of a thing. Things are only slowly improving as web designers and users discover how awful all the flexibility is, and collectively move towards a standard layout, which unfortunately is poorly typed and implemented a hundred different ways.
If Gopher had simply been improved, and HTML had never been, we wouldn't have "simple" web pages, or "mobile" web pages, because the layout on a phone would be just as good as on a desktop. So very much was lost, and very little was gained.
Here's the problem... Reading comments here, or anywhere else for that matter, you'll see EVERYONE disagrees with you. Nobody likes the new interfaces, nobody finds them easier to use. etc.
You can spout off on how great they are all you want, but most of the people who are there using it, hate everything about it.
I fail to see your point. I'm not a fan of KDE, but I admit that until KDE4, it was a solid environment, with a control center that allowed you to change damn near any setting you could want to, either in the DE or in the system in general. With 4.x they've sadly regressed into copying the worst aspects of Windows, much as GNOME 3 has copied the worst of OS X.
Appeals to authority are worthless... Experts make stupid decisions all the time. User studies are NOTORIOUS for getting things utterly and totally wrong. People will say they love X during testing, then when it comes to actually using X, find it absolutely horrible.
In short, with Windows having such a large market share, imitating Windows EXACTLY would get great reviews, and the experts will spout about how new and innovative and shiny it is. Yet that doesn't mean the interface isn't a horrible piece of crap.
It's about the same as copying the controls in a car to a helicopter... Sure, everyone will be comfortable with it right away, and say what a great interface it is, but when it comes time to use it for real, day to day work, you'll find it's horribly crippled, and you can't actually do your job with it.
My desktop of choice? Blackbox (with plenty of config changes to make it behave like Openbox 2.x did by default...). I can launch any application in a fraction of a second; shade, iconify; maximize vertically, horizontally, or both; restore, switch to, etc., faster than I can in any other desktop environment. It takes up tiny amount of screen real-estate, while GNOME and KDE are sucking up ever more screen space with each release... It give me instant access to all the menus I need, just about anywhere my cursor happens to be, on-screen.
You want some usability testing? Grab a stopwatch and time a long-time blackbox user against a long-time GNOME user... Who do you think will get slaughtered?
Now, Blackbox has a learning curve, I admit. But Fluxbox, which comes with a menu button and iconified windows on the (tiny) Toolbar, is basically a lightweight and compact mimic of Windows 95, and any idiot can pick it up and run with it, while still having the full power available as soon as they discover it. And when it comes time to configure it, you aren't spending hours in abstracted control-panel rip-offs, which require an exhaustive search to figure out how the hell to add a couple keyboard shortcuts... Hell, you can look through every possible config option in 30 seconds. Now it's of course not a full DE, but it gets you 95% of the way there, without tying you to a cement truck and sapping all productivity. Throw in a couple supporting apps and you're gold.
Like I said... stopwatch. Completely objective measurement of the utility of an interface. Not subjective, PR, "designer" circle-jerk, bullshit flavor-of-the-week, crap. And anybody who uses it day-in and day-out cares a hell of a lot about what the stopwatch has to say. New users would benefit greatly from the stopwatch test as well.
The study was so long ago that most references to it are gone. Anything that says "yet" should be ignored.
What does the "nuclear industry" have to do with anything? In the US, it's the DoE that makes the decisions. The DoE happen to have pretty much the world's top experts in the field, thanks to building and maintaining the top nulcear weapon stockpile, and has a few neat toys like periodically aquiring one after another of the world's fastest supercomputers...
In Japan, TEPCO was widely known to be untrustworthy before the disasterw and NISA was criticized for being in-bed with industry. I have not, however, seen such criticism of the DoE from pretty much any quarter at all. In fact the DoE seems to be about the most trusted and least controvercial of the large government agencies that I can think of.
Prison rape is more a popular plot device, and cultural running-gag than an actual problem in the US. Aesop's fables are probably the closest analogy.
Studies have shown the problem is, in fact, very limited: http://www.usatoday.com/news/nation/2006-01-17-prison-rape_x.htm
These are incredibly stupid ideas...
If you're using even the tiniest of solar panels for power, the extra power from this thing wont even register. Maybe if you're using it as a backup for a tiny thermocouple it'll help, but that's about the only other power source in the same ballpark...
If you've got a "power source" and want to send a distress signal when it goes out, you store up a bit of power from your main source, and can completely forego this antenna. The derission of batteries is nonsense (give me one NiMH cell...), but even if you buy the premise, one of those mentioned "ultracapacitors" charged from the grid would be vastly more suitable than using this power antenna at all...
This is only true if you're fairly close to the broadcasting antenna. Thanks to the curvature of the earth, what's on the top of a mountain, 50 miles away, is now at ground-level... Any yes, at that range your neighbors on the opposite side are picking up TV signals barely above ground level where you are, if not picking-up on signals that are actually reflecting off the ground. You often see high-gain antennas angled downward for these reasons, and there certainly is some opportunity to degrade signal strength for your neighbors.
What you're saying is the fear everyone has had for the past couple centuries. The fact that there are jobs for us humans RIGHT NOW is incontrovertible proof of the theory.
Of course the jobs aren't around RIGHT NOW. They slowly come along as workers get freed-up from other lines of work. The entire IT industry counts as an example. A century ago, we would have either been working on the farm, or stamping out metal, day-in and day-out. Now that IT has helped automate much of that, IT jobs have become viable. I don't see any reason to believe that the process which has been constantly ongoing for centuries, will suddenly cease working and come to a stop in the near future.
China isn't capitalist by a long stretch. They've got extensive subsidies, partial state ownership of many large companies, heavy regulations about how foreign companies can and can't invest, etc. China is an example of totalitarianism mixed with some capitalism (whereas it's always been democracy elsewhere). It's most certainly NOT pure capitalism.