"Desktop" is a fantastic way to distinguish between a non-portable computer and a portable "laptop" one, but to draw a line between a "desktop" and a "workstation" is asinine.
I build my own desktop computers, and the parts list is basically the same no matter the role: a case, mobo, CPU, some RAM, a storage device (HDD, SSD, whatever), and, optionally, a GPU. Maybe some other specialty stuff (I consider removable storage to be a specialty now), but those are the basics. Whether that mobo/CPU/GPU are optimized for one workload over another is a matter for consideration when selecting the parts, but there's no "workstation sauce" that makes one box a "workstation" and another box that lacks it is a "desktop".
Granted, the definition of workstation vs. desktop is somewhat vague. From the context I've seen it used in, it might be something like "powerful computer that is built for reliability rather than flashy effects". Some features that come to mind are:
-CPU comparable in performance to a high end gaming PC, but with support for reliability features like ECC RAM. -Similar priorities for the mainboard, with generously dimensioned, high quality electronic parts instead of optical bling -The reliability features mentioned above are actually used -NO overclocking -Depending on how much a crash would mean in lost productivity, maybe a redundant power supply, a USV or both. That is something you usually don't see in a consumer-class PC.
As I understand it, with the proposed patch Intel CPUs would by default run in a faster (not fully patched) but unsafe mode:
then you have to set the IBRS bit once at boot time to *ask* it not to be broken.
In Linus' place, I'd let Linux run Intel CPUs by default in the safer mode and take the performance hit, in the interest of protecting users who are not familiar with the details and risks of Meltdown. If Intel does not like that, fuck them;-)
If the admin wants to, they can decide not to set the IBRS bit and thus sacrifice security for speed, but that should be opt-out and the admin's responsibility.
Indeed, look for good bands elsewhere. I found YouTube is surprisingly useful there, in three ways:
1) The "Up next" column contains a mix of stuff I already have seen/listened to (might have to switch off cookies now that I think about it, but I digress) and similar stuff. The "similar" category is often a good start for finding other good bands.
2) Some guys have created YouTube channels with their preferred music. In some cases with permission of the labels, or at least they claim so. As an example, the channel of "Black Metal Promotion": https://www.youtube.com/channel/UCzCWehBejA23yEz3zp7jlcg. As far from radio music as you can get...
3) Another version of 2) are forums where people post links to music on YouTube.
Interestingly, those links are not often killed off via DMCA takedown notices. Either the (small) labels cannot afford an army of lawyers or they are appreciating the free publicity. As in, losing some sales to freeloaders is still better than to languish in obscurity.
I suspect that the bigger "out of the box" applications won't be converted at all, instead the workers will be told what they use in the future. Preferably not too much new stuff at a time, so they don't have to relearn everything in short order.
The real problem might be with small, proprietary stuff. As a hypothetical example, the management software for the munipical public transport, which was written by some small software vendor 10 years ago. Of course, it is closed source, the vendor is bankrupt by now and the source code has been lost at some point.
My understanding is that Munich had to keep a fraction of machines under windows, because some of their proprietary software was not easily migrated to open source. But they still wanted to (mostly) avoid the expense of migrating to a new version of Windows, which would have required hardware updates as well as new licenses. So they went ahead and got into some technical difficulties, as well as push back from users.
Overall, I think Barcelona has the better strategy here, even if it will take them longer. Both in terms of a smooth transition on the technical side and in terms of less excuses for unwilling users.
Because if you replace the software in smaller increments, the claim the whole system sucks does not work anymore. Instead, you can require people to be more specific with their complaints. Such as Joe Shmoe saying "Libre Office does not work with my documents". Then a support guy can visit Joe and ask him to demonstrate the problem, and how to fix it will become more obvious. - If only Joe did not understand how to use that feature in Libre Office, show him. - If many employees have problems using Libre Office, your training program might suck. Improve it, maybe invest in more training time for each employee. - If it is a genuine bug, work with the Libre Office developers to fix it. Maybe actually hire some developers for that.
how convenient for Wintel.. get that fancy new CPU and another copy of Windows 10, because what you have today is OEM version, so with a new CPU you'll need a new Windows copy.
And don't forget a new motherboard, because the CPU socket changed from Socket 1150 to Socket 1151 with Skylake. Well, perhaps some people will actually throw away their computers and I can loot some memory from those. Not gonna touch the Intel CPUs though;)
I don't think this is relevant to GP's argumentation. He (she?) wrote that those who do not update the drivers don't need to agree to the new license. No rooting is necessary here. Legally, it means that the desire of Nvidia to control the use of their hardware can be avoided. At least in the short term.
Practically, the problem will resurface when the current hardware is obsolete/gets unreliable because of age and needs replacing, including drivers for the new hardware. Then the license terms won't be so easily avoided anymore.
Time to switch to AMD. Not only don't they have such clauses, they are actively putting themselves into a situation where a future management cannot easily pull a Nvidia anymore. I mean the open source driver development that gradually replaces the closed source drivers at AMD. Those licenses not be revoked for already released versions.
Unless an organization is already heavily invested in CUDA, they might go with OpenCL instead so they can use AMD consumer stuff instead of Quadros. Even where GeForce versus AMD Vega currently favors GeForce, Quadro prices will make sure that GeForce versus Vega turns that into a win for AMD in terms of investment costs.
In academia, that would also lead to the effect that new developers are more often trained on OpenCL and less on CUDA. That could lead to the sort of long-term win Microsoft Visual Studio had over the Borland development tools.
That's why there is a Windows 10 LTSB branch, not all customers are willing to take automated feature updates. I'm not sure about the handling of security hotfixes though, a quick skimming of some articles I found via Google did not yield specifics about automatic or manual security patches.
Otherwise I agree, anyone who uses a non-LTSB version of Windows 10 for critical stuff has it coming...
The difference is that in Germany, Biogas plants get to sell their electricity for a guaranteed price. Yes, it's a subsidy. But IMHO not worse than subsidies for other power plants, such as Hinkley Point in Great Britain. In that case, a new nuclear power plant is going to get a subsidy.
Arguably, the real limit for responsible drivers is the speed rating of the tires. With some ad-hoc Google research I find no tires for heavy trucks that are rated more than speed class "M", which is 81 mph or 130 km/h.
No, it's not a myth. A company I worked for had trucks that would go that fast. We had several tractors with Caterpillar 3406B engines set at 425 horsepower. One was always getting worse fuel mileage then the others. We had the local Cat dealer check out the tractor. They went into the computerized fuel system on the engine and were able to show how fast the driver had been running with it. He had been hitting 118 out on I-10 through Arizona pretty regularly. With a few keystrokes they cut his top end back to 85 MPH and the fuel mileage improved a lot. A lot of the tractors on the road have Detoit Diesel 60 series engines set at 500 horsepower and they will also easily run over a hundered if they are geared right and haven't had the top speed set down. Is it smart to run that fast? No way. It takes a lot more distance to stop as you get rolling faster. Trucks running close together to "draft" are just an accident waiting to happen.
But I guess in 2001, speed limiters were not as widespread as today...
Lets do at least a bit of math here. My starting assumptions are: 1) the statement about Poland's emissions -- at 650gCO2/kWh -- is correct 2) a typical electric car needs around 20kWh/100km (that's a number I remember from some real life tests in recent years). See https://greentransportation.info/energy-transportation/kwh-evcars-gizmos.html for example.
Then the 650gCO2/kWh translate to 130gCO2/km in terms of CO2 emission. Which is about the same a fairly economical IC car produces. Other countries than Poland may be much better if they use a lot of regenerative energies.
Call me backwards, but I'm still using SeaMonkey. It has a few weaknesses, such as the handling of formatted text in the mail client, but by and large it still offers the traditional Mozilla GUI which I like better than current Firefox. I guess you could call that the ultimate Classic Theme Extension;)
If you care more about quality than having the very latest, you can get somewhat older models much cheaper unless things have changed in the last 20 years. Back in the day, I bought a pair of JBL L90 for 2000 DM (about 1000 euro) when the recommended retail price a few years before was 5000 DM or 2500 Euro. And that was not unusual, other speakers with similar prices were similarly marked down after a few years.
Now speaker technology does not move that fast anymore, being fairly mature. So getting a somewhat older model for 40% of the original price is usually a quite attractive proposal.
I have to agree that it is a "legacy system", and will probably go the way of the dodo for you if (when?) you ditch Windows.
But as long as you keep a copy of the last decent installer, and Microsoft does not ditch the Windows Core Audio APIs, you can use it on Windows as long as you want. The owner cannot keep you from that. Just don't make the mistake and downgrade to Windows 10 S;)
Still using version 5.66 lite on Win7 and quite pleased with it for playing audio. For video I use VLC though.
About XMMS: Source is available at http://www.xmms.org/ (http://www.xmms.org/download.php). They offer XMMS 1.2.11, which arguably is V1. Binaries appear to be missing though. Looks like the site has not been maintained for a while.
Fully agree to parent. The most likely reason for using another player is that I might switch away from Windows eventually. Though I might check out the alternatives that have been suggested here.
I'm only moderately optimistic about the possible new version, however. With that kind of sequel, developers often forget about optimization and keeping the memory usage in check.
Considering the sequence of events - employee writes controversial essay and gets fired shortly after it becomes public, I think Google management would need a pretty good argumentation outside the essay to claim another reason for firing Damore.
BTW, I've read the essay and it avoids outright claims of women being inferior. What it does say is that men and women have slightly different aptitudes on average and that for coding jobs men might be more suited. Also, it is worded politely enough to give no offense on that account.
Overall, I think Damore's essay counts as fair debate. It may be legal for Google to fire him, but it is certainly a dick move.
The author of the article links to a Deviantart photo as the "work prolifically online posted". Even the most cursory examination should have detected the "Cristina Matei - Selfportrait" below the photo. No "Mia Ash" here. I agree Cristina is cute though;)
I think Slashdot should consider banning TFA (in this context "The Fucking Author") Chicksdaddy from posting articles to Slashdot. It would improve the quality of the site.
I also think the front is a styling miss (but not the rest of the car). My misgivings go the other way though: I think the front still looks too much like that of a car with a grille opening. If you don't need a grille, why cling to the traditional form instead of making it even more sleek? The front area of the General Motors EV 1 does look better IMHO. If you still need a small air inlet, it could go left and right of the license plate. Imagine something like the GM Impact, with the two central air inlets "pulled apart" and located next to the fender area. The license plate could go in the middle.
Until a new kernel is broken for your specific hardware configuration, and you don't have a working fallback. We could discuss various compromises like "keep the last three versions" but my point is that breaking your system is worse than an update not going through. And if an update fails to install because the disk is full, I hope there is an easily understood error message about it, so the user can fix it without a degree in CS. But that is a slightly different topic.
I don't entirely disagree, but please make autoremoval of old kernels opt-in. Or maybe an option you have to explicitly choose after installing a new kernel (although that would not be quite "auto" anymore;)
I'd settle for 5 years in the first step, but extend it to software, including embedded software. The scope of the software warranty could be similar to what Microsoft does during extended support: Fixes for vulnerabilities, but no new features.
I think that would already send many vendors into a panic;)
The EU already has mandatory warranties, when a product is sold to a private consumer.
The minimum is two years, but there is a catch: After six months, the burden of proof shifts from the seller to the buyer. In many cases, that means YOU have now to prove that you did not mishandle the product. Make that easier for the consumer, and perhaps extend the two years too. A lot of companies will now have to up their standard of quality.
Slashdot Mirror
"Desktop" is a fantastic way to distinguish between a non-portable computer and a portable "laptop" one, but to draw a line between a "desktop" and a "workstation" is asinine.
I build my own desktop computers, and the parts list is basically the same no matter the role: a case, mobo, CPU, some RAM, a storage device (HDD, SSD, whatever), and, optionally, a GPU. Maybe some other specialty stuff (I consider removable storage to be a specialty now), but those are the basics. Whether that mobo/CPU/GPU are optimized for one workload over another is a matter for consideration when selecting the parts, but there's no "workstation sauce" that makes one box a "workstation" and another box that lacks it is a "desktop".
Granted, the definition of workstation vs. desktop is somewhat vague. From the context I've seen it used in, it might be something like "powerful computer that is built for reliability rather than flashy effects". Some features that come to mind are:
-CPU comparable in performance to a high end gaming PC, but with support for reliability features like ECC RAM.
-Similar priorities for the mainboard, with generously dimensioned, high quality electronic parts instead of optical bling
-The reliability features mentioned above are actually used
-NO overclocking
-Depending on how much a crash would mean in lost productivity, maybe a redundant power supply, a USV or both. That is something you usually don't see in a consumer-class PC.
As I understand it, with the proposed patch Intel CPUs would by default run in a faster (not fully patched) but unsafe mode:
then you have to set the IBRS bit once at boot time to *ask* it not to be broken.
In Linus' place, I'd let Linux run Intel CPUs by default in the safer mode and take the performance hit, in the interest of protecting users who are not familiar with the details and risks of Meltdown. If Intel does not like that, fuck them ;-)
If the admin wants to, they can decide not to set the IBRS bit and thus sacrifice security for speed, but that should be opt-out and the admin's responsibility.
Don't listen to radio metal.
Indeed, look for good bands elsewhere. I found YouTube is surprisingly useful there, in three ways:
1) The "Up next" column contains a mix of stuff I already have seen/listened to (might have to switch off cookies now that I think about it, but I digress) and similar stuff. The "similar" category is often a good start for finding other good bands.
2) Some guys have created YouTube channels with their preferred music. In some cases with permission of the labels, or at least they claim so. As an example, the channel of "Black Metal Promotion": https://www.youtube.com/channel/UCzCWehBejA23yEz3zp7jlcg. As far from radio music as you can get...
3) Another version of 2) are forums where people post links to music on YouTube.
Interestingly, those links are not often killed off via DMCA takedown notices. Either the (small) labels cannot afford an army of lawyers or they are appreciating the free publicity. As in, losing some sales to freeloaders is still better than to languish in obscurity.
I suspect that the bigger "out of the box" applications won't be converted at all, instead the workers will be told what they use in the future. Preferably not too much new stuff at a time, so they don't have to relearn everything in short order.
The real problem might be with small, proprietary stuff. As a hypothetical example, the management software for the munipical public transport, which was written by some small software vendor 10 years ago. Of course, it is closed source, the vendor is bankrupt by now and the source code has been lost at some point.
My understanding is that Munich had to keep a fraction of machines under windows, because some of their proprietary software was not easily migrated to open source. But they still wanted to (mostly) avoid the expense of migrating to a new version of Windows, which would have required hardware updates as well as new licenses. So they went ahead and got into some technical difficulties, as well as push back from users.
Overall, I think Barcelona has the better strategy here, even if it will take them longer. Both in terms of a smooth transition on the technical side and in terms of less excuses for unwilling users.
Because if you replace the software in smaller increments, the claim the whole system sucks does not work anymore. Instead, you can require people to be more specific with their complaints. Such as Joe Shmoe saying "Libre Office does not work with my documents". Then a support guy can visit Joe and ask him to demonstrate the problem, and how to fix it will become more obvious.
- If only Joe did not understand how to use that feature in Libre Office, show him.
- If many employees have problems using Libre Office, your training program might suck. Improve it, maybe invest in more training time for each employee.
- If it is a genuine bug, work with the Libre Office developers to fix it. Maybe actually hire some developers for that.
how convenient for Wintel.. get that fancy new CPU and another copy of Windows 10, because what you have today is OEM version, so with a new CPU you'll need a new Windows copy.
And don't forget a new motherboard, because the CPU socket changed from Socket 1150 to Socket 1151 with Skylake. ;)
Well, perhaps some people will actually throw away their computers and I can loot some memory from those. Not gonna touch the Intel CPUs though
In this case, a terrorist who plans who-knows-what might not be deterred by a license clause.
I don't think this is relevant to GP's argumentation. He (she?) wrote that those who do not update the drivers don't need to agree to the new license. No rooting is necessary here. Legally, it means that the desire of Nvidia to control the use of their hardware can be avoided. At least in the short term.
Practically, the problem will resurface when the current hardware is obsolete/gets unreliable because of age and needs replacing, including drivers for the new hardware. Then the license terms won't be so easily avoided anymore.
Time to switch to AMD. Not only don't they have such clauses, they are actively putting themselves into a situation where a future management cannot easily pull a Nvidia anymore. I mean the open source driver development that gradually replaces the closed source drivers at AMD. Those licenses not be revoked for already released versions.
Unless an organization is already heavily invested in CUDA, they might go with OpenCL instead so they can use AMD consumer stuff instead of Quadros. Even where GeForce versus AMD Vega currently favors GeForce, Quadro prices will make sure that GeForce versus Vega turns that into a win for AMD in terms of investment costs.
In academia, that would also lead to the effect that new developers are more often trained on OpenCL and less on CUDA. That could lead to the sort of long-term win Microsoft Visual Studio had over the Borland development tools.
That's why there is a Windows 10 LTSB branch, not all customers are willing to take automated feature updates. I'm not sure about the handling of security hotfixes though, a quick skimming of some articles I found via Google did not yield specifics about automatic or manual security patches.
Otherwise I agree, anyone who uses a non-LTSB version of Windows 10 for critical stuff has it coming...
The difference is that in Germany, Biogas plants get to sell their electricity for a guaranteed price. Yes, it's a subsidy. But IMHO not worse than subsidies for other power plants, such as Hinkley Point in Great Britain. In that case, a new nuclear power plant is going to get a subsidy.
Arguably, the real limit for responsible drivers is the speed rating of the tires. With some ad-hoc Google research I find no tires for heavy trucks that are rated more than speed class "M", which is 81 mph or 130 km/h.
This said, if you look on old forum threads like http://boards.straightdope.com/sdmb/showthread.php?t=97985, you can find statements like this:
No, it's not a myth. A company I worked for had trucks that would go that fast. We had several tractors with Caterpillar 3406B engines set at 425 horsepower. One was always getting worse fuel mileage then the others. We had the local Cat dealer check out the tractor. They went into the computerized fuel system on the engine and were able to show how fast the driver had been running with it. He had been hitting 118 out on I-10 through Arizona pretty regularly. With a few keystrokes they cut his top end back to 85 MPH and the fuel mileage improved a lot. A lot of the tractors on the road have Detoit Diesel 60 series engines set at 500 horsepower and they will also easily run over a hundered if they are geared right and haven't had the top speed set down. Is it smart to run that fast? No way. It takes a lot more distance to stop as you get rolling faster. Trucks running close together to "draft" are just an accident waiting to happen.
But I guess in 2001, speed limiters were not as widespread as today...
Lets do at least a bit of math here. My starting assumptions are:
1) the statement about Poland's emissions -- at 650gCO2/kWh -- is correct
2) a typical electric car needs around 20kWh/100km (that's a number I remember from some real life tests in recent years). See https://greentransportation.info/energy-transportation/kwh-evcars-gizmos.html for example.
Then the 650gCO2/kWh translate to 130gCO2/km in terms of CO2 emission. Which is about the same a fairly economical IC car produces. Other countries than Poland may be much better if they use a lot of regenerative energies.
Call me backwards, but I'm still using SeaMonkey. It has a few weaknesses, such as the handling of formatted text in the mail client, but by and large it still offers the traditional Mozilla GUI which I like better than current Firefox. ;)
I guess you could call that the ultimate Classic Theme Extension
If you care more about quality than having the very latest, you can get somewhat older models much cheaper unless things have changed in the last 20 years. Back in the day, I bought a pair of JBL L90 for 2000 DM (about 1000 euro) when the recommended retail price a few years before was 5000 DM or 2500 Euro. And that was not unusual, other speakers with similar prices were similarly marked down after a few years.
Now speaker technology does not move that fast anymore, being fairly mature. So getting a somewhat older model for 40% of the original price is usually a quite attractive proposal.
I have to agree that it is a "legacy system", and will probably go the way of the dodo for you if (when?) you ditch Windows.
But as long as you keep a copy of the last decent installer, and Microsoft does not ditch the Windows Core Audio APIs, you can use it on Windows as long as you want. The owner cannot keep you from that. Just don't make the mistake and downgrade to Windows 10 S ;)
Still using version 5.66 lite on Win7 and quite pleased with it for playing audio. For video I use VLC though.
About XMMS:
Source is available at http://www.xmms.org/ (http://www.xmms.org/download.php). They offer XMMS 1.2.11, which arguably is V1. Binaries appear to be missing though. Looks like the site has not been maintained for a while.
Fully agree to parent. The most likely reason for using another player is that I might switch away from Windows eventually. Though I might check out the alternatives that have been suggested here.
I'm only moderately optimistic about the possible new version, however. With that kind of sequel, developers often forget about optimization and keeping the memory usage in check.
Considering the sequence of events - employee writes controversial essay and gets fired shortly after it becomes public, I think Google management would need a pretty good argumentation outside the essay to claim another reason for firing Damore.
BTW, I've read the essay and it avoids outright claims of women being inferior. What it does say is that men and women have slightly different aptitudes on average and that for coding jobs men might be more suited. Also, it is worded politely enough to give no offense on that account.
Overall, I think Damore's essay counts as fair debate. It may be legal for Google to fire him, but it is certainly a dick move.
The author of the article links to a Deviantart photo as the "work prolifically online posted". Even the most cursory examination should have detected the "Cristina Matei - Selfportrait" below the photo. No "Mia Ash" here. I agree Cristina is cute though ;)
I think Slashdot should consider banning TFA (in this context "The Fucking Author") Chicksdaddy from posting articles to Slashdot. It would improve the quality of the site.
I also think the front is a styling miss (but not the rest of the car). My misgivings go the other way though:
I think the front still looks too much like that of a car with a grille opening. If you don't need a grille, why cling to the traditional form instead of making it even more sleek?
The front area of the General Motors EV 1 does look better IMHO.
If you still need a small air inlet, it could go left and right of the license plate. Imagine something like the GM Impact, with the two central air inlets "pulled apart" and located next to the fender area. The license plate could go in the middle.
Until a new kernel is broken for your specific hardware configuration, and you don't have a working fallback.
We could discuss various compromises like "keep the last three versions" but my point is that breaking your system is worse than an update not going through.
And if an update fails to install because the disk is full, I hope there is an easily understood error message about it, so the user can fix it without a degree in CS. But that is a slightly different topic.
I don't entirely disagree, but please make autoremoval of old kernels opt-in. Or maybe an option you have to explicitly choose after installing a new kernel (although that would not be quite "auto" anymore ;)
I'd settle for 5 years in the first step, but extend it to software, including embedded software. The scope of the software warranty could be similar to what Microsoft does during extended support:
Fixes for vulnerabilities, but no new features.
I think that would already send many vendors into a panic ;)
The EU already has mandatory warranties, when a product is sold to a private consumer.
The minimum is two years, but there is a catch:
After six months, the burden of proof shifts from the seller to the buyer. In many cases, that means YOU have now to prove that you did not mishandle the product. Make that easier for the consumer, and perhaps extend the two years too. A lot of companies will now have to up their standard of quality.