Bad marketing and poor vision. Had they put focus behind certain projects (OpenSolaris, VirtualBox) they could easily be competing with both the bigger SAN providers and Citrix/VMware on the virtualization fronts.
Apple prices would've fit well into their portfolio, though.
Who's to say that Sun wouldn't have come out with something functionally similar but less-Apple? That's not a bad thing: Sun technology has always been awesome and useful.
I feel assured by Sun's awesomeness at the time that, if they'd bought Apple, they'd have taken the Newton concept and turned it into something incredible and usable. Sun was/is great at hardware design, hardware utilization, and simplified user interfaces. The Palm hegemony of the time wasn't really so awesome that Sun couldn't have taken them on, I don't think.
I've yet to have an issue doing a simple: aptitude update && aptitude upgrade && aptitude dist-upgrade (or the various equivilent across the years). Worst case, I've had issues with desktop applications or esoteric/old/consumer hardware on the newer releases.
What I was referring to was ports (where anything usable comes from), not the FreeBSD/jails/build system/bind/sendmail/whatever blob that comes with a FreeBSD minimal install. You're kidding yourself if you don't call an apple an apple, as the case may be with comparing FreeBSD's base system to a Linux distribution, then considering ports not part of the equation.
(That said, I do prefer the -model- FreeBSD uses for "base" and "packages" - I just think they do a poor job.)
On the contrary: rebuilding ports is usually perpetually necessary due to package security problems. Seems damn near half the time I update a ports tree, some prerequisite library port gets 'updated' to something that breaks a significant portion of the system, silently - something not noticeable until after you start (or even complete) your ports upgrades.
Access to ports is not a badge of pride. Someone running ports on a server, without an extensive in-house vetting and change control system for ports itself, is negligent.
No more so than any other test - as a synthetic performance test.
"Hard" petaflops/etc. numbers are pretty useless unless you're doing scientific computations. Everywhere else, things like bus throughput, subsystem IOPS and throughput, and so on are much more significant.
As for Iran, I suspect the biggest use for these computers would be in a weapons/targeting system or similar. Most of Iran's proficient scientists (who would be able to effectively utilize these systems) have long left for less oppressive regions of the globe.
I don't suppose I should mention that I've got a half dozen latest-model IBM Bladecenters in my office. I haven't much use for them, but they're there - it means we're on the up-and-up, and have some serious computing power.
(Note: these BladeCenters are empty plastic shells, quite similar to 'demo units' sales guys parade around or you might find at a typical Iranian photo shoot.)
How about Chromium? I know about a year ago I was running Chrome/Chromium quite acceptably on a 600Mhz Windows 2000 laptop with 386Mb RAM. It adapts much better than firefox, by a long shot.
As for windows 98 itself... it was rare to have a video card at that time that could support most of the resolutions necessary (horizontal) to surf the Internet these days without a degree of suffering. If you've got something that old, throw a minimalist Linux distro (eg. puppy) on it and call it done.
I'm not sure that this vehicle would be intended for in-theatre utility. Some of the design criteria seem to conflict with the rigidity and design of the frame - conflicts which would greatly limit the vehicle's capabilities:
* solid rear axle * rear-wheel drive * able to get itself un-stuck * gas engine
Honestly, without looking at the frame design, it looks like what they want is (basically) an upgraded version of the Chevy Blazer/M1008 and similar variants made for the military (on account of there being no current production vehicles that are similar). Difference: the CUCV trucks had 4WD and significantly more stowage capability, as well as an (obviously, due to low gearing and a diesel motor vs. a 6.2 liter gas engine) slower top speed.
But, looking at the frame design, I think "high-speed interceptor vehicle for a squad". I think I had a couple GI Joe vehicles that had very similar designs, as a child. Not a bad idea, but: no 4WD? Maybe they're looking for another Willy's Jeep.
I'm not really sure, personally, what kind of utility a vehicle like this would have without 4WD. Add 4WD or a fully independent/independent front drive system and it gets a lot more interesting and capable.
Like standard issue M-4s using 5.56 ammunition, with an effective range of roughly 300 yards being used in Afghanistan, where average engagements take place at ranges of 400 yards and up(And the documented reluctance of DoD to go to much more capable calibers such as 6.5mm, and the massive amount iof time it took for SCARs and ACRs to even get into the hands of troops)?
1) Short of going to something like a.308/7.62 NATO rifle again, very little has an effective range of 300+ yards. The 300+ yard engagements are against fixed emplacements - RPKs and other LMGs which have little tenacity at such ranges, or 7.62x54R rifles of similar design. Not moving to a larger caliber in a theatre which is difficult to resupply to and where cover fire is important, it makes no sense to move to a cartridge that's heavier and therefore impossible to carry a lot of. 2) Very few people are able to hit a man-size target at 300 yards under the best of conditions, nevermind hitting their head, in bad weather, with approximate but unknown ranging, elevation, and windage to account for. This applies to even the better (not best) shooters in any military: the average grunt is not going to have much luck. 3) Squads have had marksmen since WWII - men with better 'long range, accurate weapons' who are more able to hit those longer targets. They're quite effective, but they need cover-fire to allow them the liberty to make their shots. Thus, those 5.56mm cartridges serve a vital role. 4) The common bullet weight for the M16 has gone from 55 grains to 75 grains in the past 10 years. These bullets are significantly more effective and have similar ballistics characteristics to the 6.5mm cartridged bullets (in terms of arc length and trajectory) - but have the added benefit of truly tumbling on impact. 5) The Afghanistan war is not against armored units, or even those with significant body armor - the areas in which a larger cartridge is generally needed. 6) The military is not restricted to just using M16s. There are a myriad of other weapons in deployment, usually used when the tool is appropriate for the task at hand. Shotguns, the venerable M16, new automatic, recoilless shotguns, the Barrett.50 and.416 cal rifles, and computer-equiped long-range RPGs like the XM-25. 7) Compared to the modern M16 derived rifles, the SCAR and ACR are inferior in many ways. They have not undergone the marked R&D over the past 50? years that Stoner's rifle has. Not insignificantly, they aren't in the supply chain and would require significant expense to replace existing stocks with. (This is a cost-saving measure: the SCAR and ACR do not offer a significant enough improvement to justify moving to them.) 8) There are billions of rounds of 5.56 brass, powder, and bullets in the supply chain and at armories, with years of contracts to supply said rounds which need to be fulfilled. You can't just jump to another caliber "like that". (It took years and years for procurement of the M16 and ammo for it, initially. Fifteen years or so? Same for the M1 and later the M14, though with those the cartridges weren't changed at the same time. The 1903 Springfield was the result of over 30 years of frequent changes in arms and ammunition design - largely starting when Roosevelt saw the poor results from American arms in Cuba during the Spanish-American War. These drastic changes take a lot of time; rushing the matter is not so simple.)
The Navy and Air Force are other matters.:P Due to the scope of production, a lot of the shortcomings have to do with buying too few, not spending too much or designing things poorly. This has been an ongoing trend since the Carter administration. When you're talking about production numbers under a couple hundred for specialized, multi-million dollar items, the per-unit cost goes up with the fewer units you
A classic example of this is the M16 rifle. It's the very definition of 'design and deploy by committee'. There is ample information out there about it: it "shits where it eats" (gas ejection port feeds into the chamber - a design feature to reduce recoil which is also a flaw with inferior ammo and infrequent cleaning), has an undersized cartridge (the 'original' was the AR10, a.308), has weak magazine design, too tight of tolerances to accomodate excessive dirt, the buffer tube and buttstock are not strong like its predecessors - and so on. These matters were further complicated in its earlier years by "deployment by comittee" - with underpowered, inconsistent ammunition, resulting in the "it jams all the time" - and cheaping out on the weapon procurement (see: Mattel and rifles that melted).
I hear that the Stryker vehicle has suffered many of the similar bueaurocratic problems leading up to its deployment. It had to have every single little safety feature, every little electronic tool, and so on - until it was no longer suitable for its original role.
Honestly, most weapon system problems are due to the government changing the specifications - just like with any engineering project. Most of slashdot should be quite familiar with this.
Fortunately for our military, it (the AR15/M16) has (finally) become a fairly potent tool (in no small part due to large civilian ownership and therefore, continued R&D money to the companies that make them).
That said, some of your original statements are wrong:
I'm not sure what the fuck "Bad Linux" is, but here's my 0.02 USD:
I started using Ubuntu on the 8.04 release. I liked it. I installed it on my Thinkpad X30, where I'd previously run Debian Stable/Testing/Sid to try to reach a good blend of stability, performance, usability, recent packages, and the lovely dpkg/apt/aptitude combination.
I should note that I'm a 'minimalist' user - I disdain the larger desktop environments due to their bloat (though KDE 3.x I deem 'good'). The packages I use are 'minimal' and trend toward what you'd expect a sysadmin to use: disk, network, file, and server packages - as well as the awesome window manager, chromium, thunderbird, clusterssh, and a handful of others.
I left debian for desktop use for a couple reasons: I had repeated issues over a number of upgrades which I would consider "deal breakers". My CF card reader refused to work, the i810 graphics support was highly fluctuative (ie suspend worked, then it didn't, then it did...), and so on. Ubuntu (mostly) fixed these issues, and I had no issue with 8.04 whatsoever (even though its release of most of the packages was very similar to the Debian release I'd left behind).
Then I upgraded to 9.04. I had a hell of a lot of problems with misc. things not working and a little instability. A lot of the same problems I'd had in Debian were back, with some others to boot. I also had some severe IO performance degradation (seemingly due to CFQ and/or ext, but I really don't know for certain) resulting in any disk IO being crippling to the point of uselessness ("I think I'll make some coffee"). I don't overtly blame Ubuntu for it, but OpenSuSE of a similar vintage and kernel version ran just fine on that hardware.
I think that, between 9.04 (supposedly a stable release - not my experience) and 9.10 (a horrible release), Ubuntu has gained a bad name for itself as "too cutting edge" - though honestly, I don't think this is the case. Other distros have similar level of 'edge cutting'. Some of the bigger problems, I think, is that GNOME sucks, and that the Ubuntu KDE packages are just wretched. There are quite a few 'impact a lot of users' bugs out there which don't get fixed for several releases, as well - something that isn't really acceptable on a "long term support" release. And then there are the kernel stopping bugs that ship... (Part of me wonders how much of this has to do with the kernel.org boys/girls and their "just ship it, it has new features" approach to development for the last number of years...)
10.04 and 10.10 are another story. The adoption of upstart has soured me somewhat (and I hear Debian will be doing the same, too), which is particularly unsavory on the (otherwise quite appealing) Ubuntu Server. It's great for a desktop (and a fast start-up) but it greatly complicates the debugging process on a server (and isn't all that necessary to begin with, IMO). The performance is really quite good due to the later kernel versions, but the stability leaves something to be desired.
Personally, I pine for the days of "minimalist installs", when that actually meant something. Sure, I can install a full desktop in around 2Gb now, but there's a lot of cruft that will be running in the background, at all times. Getting rid of that crap takes a while on a fresh install.
I'd still say ubuntu is one of the better distros, if only because it's "debian based", and that the majority of the people who used Ubuntu "previously" went to something like an iPad.
I started using a minimalized Bing search recently because I got tired of the seeded and mostly useless Google results. It seems to be doing better for technical esoterica than Google has in the past year, though I'm not sure how much of that is due to Google breaking their search and how much of it is due to the ad placements.
Bing Maps, likewise, blows the ever loving shit out of Google Maps at this point - if you want to print it or have a bird's eye/overhead view of the target. Compared to the bird's eye on Bing, the Google sat view is a useless gimmick.
* You duck down in your chair to grab a pencil you drop * You lean over to open a desk drawer * You lean back to take a moment of reflection * You step to the side (if standing) to grab something * You're skinny and the sensors can't see you * You (potentially) don't move enough while watching something on the screen * You do a lot of back-and-forth in a small area (eg. a pharmacy, where you've got to fetch medications after looking them up, then come back to the next person) and leave the keyboard frequently without leaving the system uncontrolled (ie it's always in view).... and then you've got to take the time to log back in and for everything to load back up again. Hopefully you weren't working on something and the data got lost...
A more sensible meme would be to lock the machine when the user steps away instead of logging them out, to be sure. Hopefully the sensors are accurate. Even then, there are many cases (within the designed use case) where this probably isn't appropriate or useful. Biometric logins/unlocking would likely be a bare minimum additional component, IMO.
Right, I agree completely - because governments and individuals never pollute the environment, or otherwise generally fuck up....
Hey, people fuck up. Believe it or not, it's usually not out of malice or greed. Most people in positions of responsibility and power are fairly endeverous of putting on a good show and leaving a legacy, not just cashing out and living in some remote location while the rest of the world burns.
I'll second this - 20 years seems about ideal to me.
There are so many older books which are still useful and full of difficult-to-find, well-organized information (lots of naturalist/outdoor/history books). Think of books made by small publishing companies which have since gone out of business or been sold to larger companies. They're hard to find. Authors will even sometimes loose access to their own work as a result of things like this.
With this kind of regime, the only 'threshold' for getting into printing books would be the printing equipment. Using modern technology, many, many useful books could be one-off or limited-release again.
Yeah, seems like a bit of an excuse, or possibly an overly cautious response.
As far as I know, copyright does not prohibit derived works. This is obviously a derived work, and not trying to pass itself off as the same thing. (Sorta like if I were to write a book about a handful of devious thieves - Frank and Joe Hardy, and their sidekick, Chet - I'd simply be using the same names as the child detectives; it's a completely different story.)
Now, there may be patents or something like that on the characters - I don't know. From my eye, the characters are all different: they just use some of the same names, and the story is similar. It should fly under copyright.
Because it's not like Shakespeare or Mozart profited (almost exclusively) from their works during their lifetimes, is it? There were immitators, sure - but their works were completely theirs under the protection of laws at the time.
(Shakespeare was the writer of his day's "Friends" and "Three's Company". Sure, it was witty, and I'm glad people can "get it" after learning all the cultural colloquialisms. It's still just shlock.)
No its not. Just buy a wireless keyboard. The fact of the matter is, the only things that a consumer can't replace in a laptop is the screen, CPU and mainboard.
You forgot RAM. RAM is usually at or near the max of what the machine is capable of holding if you buy the machine in a 'capable' configuration. The "default"/minimum for the machine is usually half that. It gives you very few upgrade options. Considering laptops are usually roughly 2/3rds as capable (or less) of a comparably priced desktop (disk speed, CPU, RAM/RAM upgrade options), there's not a lot that can be done to make it go another mile when it's time to replace it.
Also, consider that laptop displays do tend to fail significantly faster than desktop varieties due to the jostling the CCFLs get, and the likelihood that they won't be getting power that's as clean as a desktop (most of those AC adapters suck), causing the LCD controller to fail.
Though given that the mainboard often has the power charging components, and laptops seem to be designed so poorly as to not run/restart/etc. without a good battery, it's almost a moot distinction. Most laptops (and other portable devices) I've had have ended their life due to the charging components, battery, etc. failing. Usually it's 3-4 years after purchase, but I've seen it happen within warranty a number of times as well. Ironically, this is at about the time when other components start to fail (hard drives, keyboard keys, etc.).
Most laptops have 1-2 years of useful lifespan before they become cost prohibitive to repair. A new battery and power supply for a laptop is often in the $200 range, making a new laptop seem mighty appealing to most users. And if the power plug socket, or something on the charging component is broken/damaged/diminished, well...
Personally, I think the biggest thing to improve laptop longevity, despite all the above, would be to modularize the power component and make a standard connector (maybe). It's already fairly isolated/modular from the rest of the system, but having the "small plug" on the back/side is a recipe for failure (board, battery, power supply). Instead, the battery should be inline from the PSU, directly. The battery could attach to the rear (under the display, as most/many do), with the PSU attaching to that in the same fashion as the battery does to the laptop. These interfaces would be identical, so you could run the PSU with the battery not inline, if you so desire. The battery would then have a pass-through circuit (akin to how a UPS does), as they're already essentially doing this internally to the laptop. Since the charge/pass through is as likely to fail as the battery on current designs (and often at the same time for similar reasons) it makes sense to combine them into a single device.
This would be suitable for most users. The only case in which it would probably not work is if someone wants to charge a number of batteries at a time, in which case they'd have to power off the machine. (This would, however, make it trivial to get another power charger for your extra batteries.) This could probably be worked around by using a single, standard AA (or similar) disposable battery cell in the chassis to provide a minute or two of battery power while you switch batteries.
How is this anything but an eventual response to an internal snafu which could've resulted in (much more expensive) litigious actions?
The GPL violations, and the resulting denial of compliance (for years, wasn't it?) was nothing but bad press. In contrast, had they admitted to the snafu right off the bat and addressed it promptly (a couple months? 6 on the outside?) it'd have been another thing entirely - the press would've been positive. GNU 'compliance' types just want the (free to the violators) adherence to the GPL - they don't care about the money or licensing aspect of it.
In my mind, this almost says less than doing nothing about it at all - "oh, we've got another release, let's include those license files and source code this time".
(What do you want to bet the code is significantly aged and with numerous vulnerabilities known for a long time?)
Er, actually, no. I want my police officers to sit around doing nothing all day.
I don't want tickets for petty infractions. I don't want them to have to memorize dozens of code books, statures, and laws.
The laws should be simple so anyone can understand - and adhere - to them.
Police should basically serve the function of fire fighters and street lights - but for the offensive acts of others and the disorderly behavior of miscreants (eg. drunk rabble rousers), not fires and darkness.
If police work is overtly difficult, it means one (or more) of three things:
1) There are too many laws and regulations on which people can be penalized 2) There are entirely too many assholes out there who do inconsiderate things that cause non-trivial hardship for others (think: someone who parks their truck in an intersection to make a delivery) 3) Society as we know it is collapsing on account of violence
I see no reason why your average police officer should be burdened of paperwork that a clerk is. Yet, they are.
Slashdot Mirror
Bad marketing and poor vision. Had they put focus behind certain projects (OpenSolaris, VirtualBox) they could easily be competing with both the bigger SAN providers and Citrix/VMware on the virtualization fronts.
Apple prices would've fit well into their portfolio, though.
Who's to say that Sun wouldn't have come out with something functionally similar but less-Apple? That's not a bad thing: Sun technology has always been awesome and useful.
I feel assured by Sun's awesomeness at the time that, if they'd bought Apple, they'd have taken the Newton concept and turned it into something incredible and usable. Sun was/is great at hardware design, hardware utilization, and simplified user interfaces. The Palm hegemony of the time wasn't really so awesome that Sun couldn't have taken them on, I don't think.
Were you replying to someone else?
I've yet to have an issue doing a simple: aptitude update && aptitude upgrade && aptitude dist-upgrade (or the various equivilent across the years). Worst case, I've had issues with desktop applications or esoteric/old/consumer hardware on the newer releases.
What I was referring to was ports (where anything usable comes from), not the FreeBSD/jails/build system/bind/sendmail/whatever blob that comes with a FreeBSD minimal install. You're kidding yourself if you don't call an apple an apple, as the case may be with comparing FreeBSD's base system to a Linux distribution, then considering ports not part of the equation.
(That said, I do prefer the -model- FreeBSD uses for "base" and "packages" - I just think they do a poor job.)
On the contrary: rebuilding ports is usually perpetually necessary due to package security problems. Seems damn near half the time I update a ports tree, some prerequisite library port gets 'updated' to something that breaks a significant portion of the system, silently - something not noticeable until after you start (or even complete) your ports upgrades.
Access to ports is not a badge of pride. Someone running ports on a server, without an extensive in-house vetting and change control system for ports itself, is negligent.
Quantum Bigfoot: the name itself defies parody. It was that bad.
Yes, but it's experimental headcrab support. That means it's full featured, you've just got to read the source to figure out how to use it.
Looks like it's another 2-3 hours of compiling packages and the odd several hours of library/package build error resolution for you!
No more so than any other test - as a synthetic performance test.
"Hard" petaflops/etc. numbers are pretty useless unless you're doing scientific computations. Everywhere else, things like bus throughput, subsystem IOPS and throughput, and so on are much more significant.
As for Iran, I suspect the biggest use for these computers would be in a weapons/targeting system or similar. Most of Iran's proficient scientists (who would be able to effectively utilize these systems) have long left for less oppressive regions of the globe.
So, photos are conclusive proof, are they?
I don't suppose I should mention that I've got a half dozen latest-model IBM Bladecenters in my office. I haven't much use for them, but they're there - it means we're on the up-and-up, and have some serious computing power.
(Note: these BladeCenters are empty plastic shells, quite similar to 'demo units' sales guys parade around or you might find at a typical Iranian photo shoot.)
How about Chromium? I know about a year ago I was running Chrome/Chromium quite acceptably on a 600Mhz Windows 2000 laptop with 386Mb RAM. It adapts much better than firefox, by a long shot.
As for windows 98 itself... it was rare to have a video card at that time that could support most of the resolutions necessary (horizontal) to surf the Internet these days without a degree of suffering. If you've got something that old, throw a minimalist Linux distro (eg. puppy) on it and call it done.
I'm not sure that this vehicle would be intended for in-theatre utility. Some of the design criteria seem to conflict with the rigidity and design of the frame - conflicts which would greatly limit the vehicle's capabilities:
* solid rear axle
* rear-wheel drive
* able to get itself un-stuck
* gas engine
Honestly, without looking at the frame design, it looks like what they want is (basically) an upgraded version of the Chevy Blazer/M1008 and similar variants made for the military (on account of there being no current production vehicles that are similar). Difference: the CUCV trucks had 4WD and significantly more stowage capability, as well as an (obviously, due to low gearing and a diesel motor vs. a 6.2 liter gas engine) slower top speed.
But, looking at the frame design, I think "high-speed interceptor vehicle for a squad". I think I had a couple GI Joe vehicles that had very similar designs, as a child. Not a bad idea, but: no 4WD? Maybe they're looking for another Willy's Jeep.
I'm not really sure, personally, what kind of utility a vehicle like this would have without 4WD. Add 4WD or a fully independent/independent front drive system and it gets a lot more interesting and capable.
Like standard issue M-4s using 5.56 ammunition, with an effective range of roughly 300 yards being used in Afghanistan, where average engagements take place at ranges of 400 yards and up(And the documented reluctance of DoD to go to much more capable calibers such as 6.5mm, and the massive amount iof time it took for SCARs and ACRs to even get into the hands of troops)?
1) Short of going to something like a .308/7.62 NATO rifle again, very little has an effective range of 300+ yards. The 300+ yard engagements are against fixed emplacements - RPKs and other LMGs which have little tenacity at such ranges, or 7.62x54R rifles of similar design. Not moving to a larger caliber in a theatre which is difficult to resupply to and where cover fire is important, it makes no sense to move to a cartridge that's heavier and therefore impossible to carry a lot of. .50 and .416 cal rifles, and computer-equiped long-range RPGs like the XM-25.
2) Very few people are able to hit a man-size target at 300 yards under the best of conditions, nevermind hitting their head, in bad weather, with approximate but unknown ranging, elevation, and windage to account for. This applies to even the better (not best) shooters in any military: the average grunt is not going to have much luck.
3) Squads have had marksmen since WWII - men with better 'long range, accurate weapons' who are more able to hit those longer targets. They're quite effective, but they need cover-fire to allow them the liberty to make their shots. Thus, those 5.56mm cartridges serve a vital role.
4) The common bullet weight for the M16 has gone from 55 grains to 75 grains in the past 10 years. These bullets are significantly more effective and have similar ballistics characteristics to the 6.5mm cartridged bullets (in terms of arc length and trajectory) - but have the added benefit of truly tumbling on impact.
5) The Afghanistan war is not against armored units, or even those with significant body armor - the areas in which a larger cartridge is generally needed.
6) The military is not restricted to just using M16s. There are a myriad of other weapons in deployment, usually used when the tool is appropriate for the task at hand. Shotguns, the venerable M16, new automatic, recoilless shotguns, the Barrett
7) Compared to the modern M16 derived rifles, the SCAR and ACR are inferior in many ways. They have not undergone the marked R&D over the past 50? years that Stoner's rifle has. Not insignificantly, they aren't in the supply chain and would require significant expense to replace existing stocks with. (This is a cost-saving measure: the SCAR and ACR do not offer a significant enough improvement to justify moving to them.)
8) There are billions of rounds of 5.56 brass, powder, and bullets in the supply chain and at armories, with years of contracts to supply said rounds which need to be fulfilled. You can't just jump to another caliber "like that". (It took years and years for procurement of the M16 and ammo for it, initially. Fifteen years or so? Same for the M1 and later the M14, though with those the cartridges weren't changed at the same time. The 1903 Springfield was the result of over 30 years of frequent changes in arms and ammunition design - largely starting when Roosevelt saw the poor results from American arms in Cuba during the Spanish-American War. These drastic changes take a lot of time; rushing the matter is not so simple.)
The Navy and Air Force are other matters. :P Due to the scope of production, a lot of the shortcomings have to do with buying too few, not spending too much or designing things poorly. This has been an ongoing trend since the Carter administration. When you're talking about production numbers under a couple hundred for specialized, multi-million dollar items, the per-unit cost goes up with the fewer units you
A classic example of this is the M16 rifle. It's the very definition of 'design and deploy by committee'. There is ample information out there about it: it "shits where it eats" (gas ejection port feeds into the chamber - a design feature to reduce recoil which is also a flaw with inferior ammo and infrequent cleaning), has an undersized cartridge (the 'original' was the AR10, a .308), has weak magazine design, too tight of tolerances to accomodate excessive dirt, the buffer tube and buttstock are not strong like its predecessors - and so on. These matters were further complicated in its earlier years by "deployment by comittee" - with underpowered, inconsistent ammunition, resulting in the "it jams all the time" - and cheaping out on the weapon procurement (see: Mattel and rifles that melted).
I hear that the Stryker vehicle has suffered many of the similar bueaurocratic problems leading up to its deployment. It had to have every single little safety feature, every little electronic tool, and so on - until it was no longer suitable for its original role.
Honestly, most weapon system problems are due to the government changing the specifications - just like with any engineering project. Most of slashdot should be quite familiar with this.
Fortunately for our military, it (the AR15/M16) has (finally) become a fairly potent tool (in no small part due to large civilian ownership and therefore, continued R&D money to the companies that make them).
That said, some of your original statements are wrong:
I wonder how long it'll be until FreeBSD rolls a security update out for this.
I'm not sure what the fuck "Bad Linux" is, but here's my 0.02 USD:
I started using Ubuntu on the 8.04 release. I liked it. I installed it on my Thinkpad X30, where I'd previously run Debian Stable/Testing/Sid to try to reach a good blend of stability, performance, usability, recent packages, and the lovely dpkg/apt/aptitude combination.
I should note that I'm a 'minimalist' user - I disdain the larger desktop environments due to their bloat (though KDE 3.x I deem 'good'). The packages I use are 'minimal' and trend toward what you'd expect a sysadmin to use: disk, network, file, and server packages - as well as the awesome window manager, chromium, thunderbird, clusterssh, and a handful of others.
I left debian for desktop use for a couple reasons: I had repeated issues over a number of upgrades which I would consider "deal breakers". My CF card reader refused to work, the i810 graphics support was highly fluctuative (ie suspend worked, then it didn't, then it did...), and so on. Ubuntu (mostly) fixed these issues, and I had no issue with 8.04 whatsoever (even though its release of most of the packages was very similar to the Debian release I'd left behind).
Then I upgraded to 9.04. I had a hell of a lot of problems with misc. things not working and a little instability. A lot of the same problems I'd had in Debian were back, with some others to boot. I also had some severe IO performance degradation (seemingly due to CFQ and/or ext, but I really don't know for certain) resulting in any disk IO being crippling to the point of uselessness ("I think I'll make some coffee"). I don't overtly blame Ubuntu for it, but OpenSuSE of a similar vintage and kernel version ran just fine on that hardware.
I think that, between 9.04 (supposedly a stable release - not my experience) and 9.10 (a horrible release), Ubuntu has gained a bad name for itself as "too cutting edge" - though honestly, I don't think this is the case. Other distros have similar level of 'edge cutting'. Some of the bigger problems, I think, is that GNOME sucks, and that the Ubuntu KDE packages are just wretched. There are quite a few 'impact a lot of users' bugs out there which don't get fixed for several releases, as well - something that isn't really acceptable on a "long term support" release. And then there are the kernel stopping bugs that ship... (Part of me wonders how much of this has to do with the kernel.org boys/girls and their "just ship it, it has new features" approach to development for the last number of years...)
10.04 and 10.10 are another story. The adoption of upstart has soured me somewhat (and I hear Debian will be doing the same, too), which is particularly unsavory on the (otherwise quite appealing) Ubuntu Server. It's great for a desktop (and a fast start-up) but it greatly complicates the debugging process on a server (and isn't all that necessary to begin with, IMO). The performance is really quite good due to the later kernel versions, but the stability leaves something to be desired.
Personally, I pine for the days of "minimalist installs", when that actually meant something. Sure, I can install a full desktop in around 2Gb now, but there's a lot of cruft that will be running in the background, at all times. Getting rid of that crap takes a while on a fresh install.
I'd still say ubuntu is one of the better distros, if only because it's "debian based", and that the majority of the people who used Ubuntu "previously" went to something like an iPad.
I started using a minimalized Bing search recently because I got tired of the seeded and mostly useless Google results. It seems to be doing better for technical esoterica than Google has in the past year, though I'm not sure how much of that is due to Google breaking their search and how much of it is due to the ad placements.
Bing Maps, likewise, blows the ever loving shit out of Google Maps at this point - if you want to print it or have a bird's eye/overhead view of the target. Compared to the bird's eye on Bing, the Google sat view is a useless gimmick.
Wow, that is potentially (highly) irritating.
Imagine:
* You duck down in your chair to grab a pencil you drop ... and then you've got to take the time to log back in and for everything to load back up again. Hopefully you weren't working on something and the data got lost...
* You lean over to open a desk drawer
* You lean back to take a moment of reflection
* You step to the side (if standing) to grab something
* You're skinny and the sensors can't see you
* You (potentially) don't move enough while watching something on the screen
* You do a lot of back-and-forth in a small area (eg. a pharmacy, where you've got to fetch medications after looking them up, then come back to the next person) and leave the keyboard frequently without leaving the system uncontrolled (ie it's always in view).
A more sensible meme would be to lock the machine when the user steps away instead of logging them out, to be sure. Hopefully the sensors are accurate. Even then, there are many cases (within the designed use case) where this probably isn't appropriate or useful. Biometric logins/unlocking would likely be a bare minimum additional component, IMO.
Right, I agree completely - because governments and individuals never pollute the environment, or otherwise generally fuck up. ...
Hey, people fuck up. Believe it or not, it's usually not out of malice or greed. Most people in positions of responsibility and power are fairly endeverous of putting on a good show and leaving a legacy, not just cashing out and living in some remote location while the rest of the world burns.
That would be awesome. It would totally give me the excuse to push the red button.
I'll second this - 20 years seems about ideal to me.
There are so many older books which are still useful and full of difficult-to-find, well-organized information (lots of naturalist/outdoor/history books). Think of books made by small publishing companies which have since gone out of business or been sold to larger companies. They're hard to find. Authors will even sometimes loose access to their own work as a result of things like this.
With this kind of regime, the only 'threshold' for getting into printing books would be the printing equipment. Using modern technology, many, many useful books could be one-off or limited-release again.
Yeah, seems like a bit of an excuse, or possibly an overly cautious response.
As far as I know, copyright does not prohibit derived works. This is obviously a derived work, and not trying to pass itself off as the same thing. (Sorta like if I were to write a book about a handful of devious thieves - Frank and Joe Hardy, and their sidekick, Chet - I'd simply be using the same names as the child detectives; it's a completely different story.)
Now, there may be patents or something like that on the characters - I don't know. From my eye, the characters are all different: they just use some of the same names, and the story is similar. It should fly under copyright.
Because it's not like Shakespeare or Mozart profited (almost exclusively) from their works during their lifetimes, is it? There were immitators, sure - but their works were completely theirs under the protection of laws at the time.
(Shakespeare was the writer of his day's "Friends" and "Three's Company". Sure, it was witty, and I'm glad people can "get it" after learning all the cultural colloquialisms. It's still just shlock.)
No its not. Just buy a wireless keyboard. The fact of the matter is, the only things that a consumer can't replace in a laptop is the screen, CPU and mainboard.
You forgot RAM. RAM is usually at or near the max of what the machine is capable of holding if you buy the machine in a 'capable' configuration. The "default"/minimum for the machine is usually half that. It gives you very few upgrade options. Considering laptops are usually roughly 2/3rds as capable (or less) of a comparably priced desktop (disk speed, CPU, RAM/RAM upgrade options), there's not a lot that can be done to make it go another mile when it's time to replace it.
Also, consider that laptop displays do tend to fail significantly faster than desktop varieties due to the jostling the CCFLs get, and the likelihood that they won't be getting power that's as clean as a desktop (most of those AC adapters suck), causing the LCD controller to fail.
Though given that the mainboard often has the power charging components, and laptops seem to be designed so poorly as to not run/restart/etc. without a good battery, it's almost a moot distinction. Most laptops (and other portable devices) I've had have ended their life due to the charging components, battery, etc. failing. Usually it's 3-4 years after purchase, but I've seen it happen within warranty a number of times as well. Ironically, this is at about the time when other components start to fail (hard drives, keyboard keys, etc.).
Most laptops have 1-2 years of useful lifespan before they become cost prohibitive to repair. A new battery and power supply for a laptop is often in the $200 range, making a new laptop seem mighty appealing to most users. And if the power plug socket, or something on the charging component is broken/damaged/diminished, well...
Personally, I think the biggest thing to improve laptop longevity, despite all the above, would be to modularize the power component and make a standard connector (maybe). It's already fairly isolated/modular from the rest of the system, but having the "small plug" on the back/side is a recipe for failure (board, battery, power supply). Instead, the battery should be inline from the PSU, directly. The battery could attach to the rear (under the display, as most/many do), with the PSU attaching to that in the same fashion as the battery does to the laptop. These interfaces would be identical, so you could run the PSU with the battery not inline, if you so desire. The battery would then have a pass-through circuit (akin to how a UPS does), as they're already essentially doing this internally to the laptop. Since the charge/pass through is as likely to fail as the battery on current designs (and often at the same time for similar reasons) it makes sense to combine them into a single device.
This would be suitable for most users. The only case in which it would probably not work is if someone wants to charge a number of batteries at a time, in which case they'd have to power off the machine. (This would, however, make it trivial to get another power charger for your extra batteries.) This could probably be worked around by using a single, standard AA (or similar) disposable battery cell in the chassis to provide a minute or two of battery power while you switch batteries.
How is this anything but an eventual response to an internal snafu which could've resulted in (much more expensive) litigious actions?
The GPL violations, and the resulting denial of compliance (for years, wasn't it?) was nothing but bad press. In contrast, had they admitted to the snafu right off the bat and addressed it promptly (a couple months? 6 on the outside?) it'd have been another thing entirely - the press would've been positive. GNU 'compliance' types just want the (free to the violators) adherence to the GPL - they don't care about the money or licensing aspect of it.
In my mind, this almost says less than doing nothing about it at all - "oh, we've got another release, let's include those license files and source code this time".
(What do you want to bet the code is significantly aged and with numerous vulnerabilities known for a long time?)
Er, actually, no. I want my police officers to sit around doing nothing all day.
I don't want tickets for petty infractions. I don't want them to have to memorize dozens of code books, statures, and laws.
The laws should be simple so anyone can understand - and adhere - to them.
Police should basically serve the function of fire fighters and street lights - but for the offensive acts of others and the disorderly behavior of miscreants (eg. drunk rabble rousers), not fires and darkness.
If police work is overtly difficult, it means one (or more) of three things:
1) There are too many laws and regulations on which people can be penalized
2) There are entirely too many assholes out there who do inconsiderate things that cause non-trivial hardship for others (think: someone who parks their truck in an intersection to make a delivery)
3) Society as we know it is collapsing on account of violence
I see no reason why your average police officer should be burdened of paperwork that a clerk is. Yet, they are.