Cared to check yet which ones are even reappearing from the list for Dothan/Banias? That would make the number of 34 in so-and-so many days even more pointless.
As all the DOM in IE is exposed as COM objects to friendly BHOs (Google Toolbar and the like) and the less friendly ones (countless spyware), and the scripting on a web page is manipulating the very same objects, it's more than likely that the compatibility syntax will just be a new way to access the very same MSXML COM object.
Nope, the point is that the user will be able to turn off creation off "marked safe for scripting" ActiveX objects by ProgID. Any J(ava)Script/DOM object in IE is an ActiveX object, more precisely a IDispatch object. This has some technical consequences, as it means that reference counting, not garbage collection, is used.
With stricter regulations of the maximum standby power instead, it would make far more sense to outlaw light bulbs or even inadescents (LED only!) before making a real switch mandatory.
Maybe we should outlaw everything with a battery to, due to the losses in charging and discharging. By the way, a DC standard wiring would certainly be nice and encourage use of reasonably efficient DC-DC-conversion if they need another voltage. That would save a lot in standby (as AC-DC conversion tends to get quite inefficient when the drain is far lower than the normal drain the system is designed for), but also in use, with the simple assumption that the main transformer in the circuit would be high-quality.
Just refreshing RAM is comparatively cheap. A laptop can stay in that mode for hours and hours on the battery. However, most PC power supplies have very low efficiency at low power (while real standby power for wake-on-LAN is only a few W). I'm not sure about the Mac, though, but I would doubt they have spent the premium to make the circuits fully adaptive for the complete range.
You might save time, of course, there is no denial of that. Saving energy by the process is a kind of weird question. Will your saved time result in the machine staying in sleep mode for one minute longer, or will you do actual work for one more minute? The shift in power usage for an idle and active desktop system is not that significant, at least not when the shift won't involve heavy duty for the GPU in either case.
On the other hand, the sleep mode will also induce almost all of the material fatigue in different components that turning off would give. The HD will stop and so on.
Long live Suspend-to-disk, no matter what OS it is. Yes, it will take longer to resume than suspend-to-RAM, but it's still often quicker than a clean boot, and certainly quicker than a clean boot + resuming work where it was.
Sorry to burst your bubble, but the heat you get is very very close to the power used. Of course, if you draw it all in LEDs and let them shine out of your window or something, then you might get a difference.
The real problem with this reasoning is that the generator of the original electricity was possibly going around drive by a turbine that was driven by heat. The efficiency of that transfer is far below 50 %. Only if your house is electrically heated, without employing a phase-change heat exchange (a reversed fridge for the air leaving the building, making the outside a little cooler) it's equivalent and one can still argue about how to achieve optimum airflow.
Of course, standby power in electrically heated buildings is less of a problem than in electrically cooled ones. In that case you have waste power for standby and waste heat from standby that must be handled by the AC, causing even more waste.
Isn't the/. audience as a general entity supposed to have left Windows (if they ever used it) while the command interpreter was still called command.com?
As the mitochondria are the main producers of ATP (like 30 out of the 36 ATP in total per glucose molecule), this sounds kind of wrong. We also have to remember that most of the mitochondrial activity IS requiring oxygen (hence aerobic), while the anaerobic processes take place in the cytoplasm. Then, the question becomes: will there be enough oxygen, just being depleted at a very slow rate, or will the anaerobic process be enough for a prolonged time, thanks to everything being slow?
Yes, Explorer.exe will normally load mshtml.dll to render the info pane for folder contents. Yes, you can still turn off that and use classic folder view. In that case, Explorer.exe doesn't use the rendering engine of IE (unless you use HTML-based Active Desktop, but NOT web folders, a somewhat surprising combination). It's as simple as that. As another comment noted, the common controls were updated with IE and with IE as a recommended way to redistribute the new DLL.
Also, if an administrative user logs in, it will be with the admin profile. There is no immediate reason that someone only using Explorer.exe to browse the HD, even with web folders active, will somehow pick up a known exploit for Trident/MSHTML.
Autorendering of HTML mail has historically been a much worse decision than the use of HTML in the user interface of some local apps. Still, that is a decision that makes some sense, at least if one accepts the idea of people wanting formatted mail at all.
Actually, I think we should know these from the probes. Compare with the Pioneer effect and the overall trajectory calculations for the probes to the outer planets. Also observe the minimal number of failures during actual travel (but rather during takeoff/entering orbit/landing).
Almost any organism not actually evolving with current Earth life would probably be killed, rather than killing. And, if it was in fact some pathogen that has left the Earth previously, it might be a bit more dangerous, but on the other hand, we could risk finding those in almost any geological survey.
Well, what I like is the possibility of better range, power efficiency and reliability by using several signals through several antennas and processing those signals relative to each other to get much higher accuracy. An analogy would be the effect of having two eyes, compensating each others flaws, and perceiving 3D.
If you just let ping loop for a while on a 802.11 connection today, you will most probably see a few packets with very high turnaround. Anything that could rule those out, even in noisy settings, is well worth it, at least to me.
This would be true, if you avoid the fact that Steve Gibson take pride in coding his Windows apps in assembler. The normal release-compiled assembler x86 isn't that hard to grok, if you really claim to be a wiz at decoding it yourself. The debug symbols are out there for you to grab, too, so you get it nicely split up into functions.
Yes, but you don't have to emulate the whole system, just user mode down to libraries. For GUI refresh (especially a compositing GUI), you can get that done in native code, if done right.
But the fact that it even presents itself as a P6 and surely has several performance characteristics similar to P6 means that it is simply a development of P6. P6 was never exceedingly hot and had, compared to the architectures of today, high IPC. That Pentium Pro from high school is haunting you.
The point is that all SP2 binaries use "canaries" to prevent unknown buffer overruns to go crazy. This means that some unknown issues, that were critical in SP1, never affected SP2, at least not with any obvious exploit. The canaries are nothing fancy, just put a random number of the stack and check it before you read EIP back from the stack. It's all over the place in SP2 code, to some performance penalty, but it's a feature that is impossible just to crack, unless you have a remote exploit to get the canary value for a specific module (it isn't rerandomized all the time) and then craft your exploit to that specific value.
SYSENTER/SYSEXIT to speed up silly benchmarks doing unbuffered intensive character I/O (or other things basically consisting of kernel roundtrips) 300 % faster!
Fun fact: It confused me as hell when I decided to see Serenity at my last visit in the U.S. There were a few different codes on the ticket. I really tried to figure out which one might be the seat. After a while, I decided that it was hopefully free seating. (Of course, the seating is sometimes free in Sweden, in really small cinemas or unofficial screenings.)
Well, I think that referrer is much more privacy-invasive than this one. The outgoing link is part of the actual content on the site, but what keywords I searched for or what intranet site linked to the competitor's web page is far more dangerous.
"you'll see javascript to overwrite the status line so it doesn't show it's tracking you... and there are hundreds, if not thousands or millions of other sites that do the same." Like this one? Ok, the href might be valid, but the normal way to use the link is still tracked by going to a different URL.
Ever heard of cross-site scripting? It's not cross-site scripting to open a popup, or load an image, or whatever, with the URL of another site. Cross-site scripting is only relevant when such a frame or popup is able to change/read the URL or content of the launching site. This can also cause security issues, if the different sites have varying settings (or simply cookie reading exploits). None of that is present here.
Spreading information in a way intended by the creator of the site you visit may be a valid concern, but I haven't seen the widely deployed ban for images from other servers yet, for example.
The 64-bit switch is exactly the kind of "WTF, why should we support gen 1" thinking that could be going on, stopping future updates. The same would possibly apply to everything; Office, Photoshop, whatever -- if quartz and other UI components are also made 64-bit, 32-bit Intel Mac applications will be nothing more than a fad.
Considering that 64-bit UI isn't present on current PowerPC Macs, this doesn't have to happen, but I can understand the worries against joining the 1st-gen bandwagon right now. In this situation, even a PowerPC Mac might be supported in a more consistent manner in a few years time.
Slashdot Mirror
The registry isn't ditched, nor is full-screen OpenGL. Windowed OpenGL is a quite sad story.
Cared to check yet which ones are even reappearing from the list for Dothan/Banias? That would make the number of 34 in so-and-so many days even more pointless.
As all the DOM in IE is exposed as COM objects to friendly BHOs (Google Toolbar and the like) and the less friendly ones (countless spyware), and the scripting on a web page is manipulating the very same objects, it's more than likely that the compatibility syntax will just be a new way to access the very same MSXML COM object.
Nope, the point is that the user will be able to turn off creation off "marked safe for scripting" ActiveX objects by ProgID. Any J(ava)Script/DOM object in IE is an ActiveX object, more precisely a IDispatch object. This has some technical consequences, as it means that reference counting, not garbage collection, is used.
Maybe we should outlaw everything with a battery to, due to the losses in charging and discharging. By the way, a DC standard wiring would certainly be nice and encourage use of reasonably efficient DC-DC-conversion if they need another voltage. That would save a lot in standby (as AC-DC conversion tends to get quite inefficient when the drain is far lower than the normal drain the system is designed for), but also in use, with the simple assumption that the main transformer in the circuit would be high-quality.
You might save time, of course, there is no denial of that. Saving energy by the process is a kind of weird question. Will your saved time result in the machine staying in sleep mode for one minute longer, or will you do actual work for one more minute? The shift in power usage for an idle and active desktop system is not that significant, at least not when the shift won't involve heavy duty for the GPU in either case.
On the other hand, the sleep mode will also induce almost all of the material fatigue in different components that turning off would give. The HD will stop and so on.
Long live Suspend-to-disk, no matter what OS it is. Yes, it will take longer to resume than suspend-to-RAM, but it's still often quicker than a clean boot, and certainly quicker than a clean boot + resuming work where it was.
The real problem with this reasoning is that the generator of the original electricity was possibly going around drive by a turbine that was driven by heat. The efficiency of that transfer is far below 50 %. Only if your house is electrically heated, without employing a phase-change heat exchange (a reversed fridge for the air leaving the building, making the outside a little cooler) it's equivalent and one can still argue about how to achieve optimum airflow.
Of course, standby power in electrically heated buildings is less of a problem than in electrically cooled ones. In that case you have waste power for standby and waste heat from standby that must be handled by the AC, causing even more waste.
Isn't the /. audience as a general entity supposed to have left Windows (if they ever used it) while the command interpreter was still called command.com?
As the mitochondria are the main producers of ATP (like 30 out of the 36 ATP in total per glucose molecule), this sounds kind of wrong. We also have to remember that most of the mitochondrial activity IS requiring oxygen (hence aerobic), while the anaerobic processes take place in the cytoplasm. Then, the question becomes: will there be enough oxygen, just being depleted at a very slow rate, or will the anaerobic process be enough for a prolonged time, thanks to everything being slow?
Yes, Explorer.exe will normally load mshtml.dll to render the info pane for folder contents. Yes, you can still turn off that and use classic folder view. In that case, Explorer.exe doesn't use the rendering engine of IE (unless you use HTML-based Active Desktop, but NOT web folders, a somewhat surprising combination). It's as simple as that. As another comment noted, the common controls were updated with IE and with IE as a recommended way to redistribute the new DLL.
Also, if an administrative user logs in, it will be with the admin profile. There is no immediate reason that someone only using Explorer.exe to browse the HD, even with web folders active, will somehow pick up a known exploit for Trident/MSHTML.
Autorendering of HTML mail has historically been a much worse decision than the use of HTML in the user interface of some local apps. Still, that is a decision that makes some sense, at least if one accepts the idea of people wanting formatted mail at all.
Actually, I think we should know these from the probes. Compare with the Pioneer effect and the overall trajectory calculations for the probes to the outer planets. Also observe the minimal number of failures during actual travel (but rather during takeoff/entering orbit/landing).
Almost any organism not actually evolving with current Earth life would probably be killed, rather than killing. And, if it was in fact some pathogen that has left the Earth previously, it might be a bit more dangerous, but on the other hand, we could risk finding those in almost any geological survey.
If you just let ping loop for a while on a 802.11 connection today, you will most probably see a few packets with very high turnaround. Anything that could rule those out, even in noisy settings, is well worth it, at least to me.
This would be true, if you avoid the fact that Steve Gibson take pride in coding his Windows apps in assembler. The normal release-compiled assembler x86 isn't that hard to grok, if you really claim to be a wiz at decoding it yourself. The debug symbols are out there for you to grab, too, so you get it nicely split up into functions.
It's been that way in Stockholm for at least 6 years now. Now, all the fuss is about extending UMTS coverage in the tunnels. 384 kbps!
Yes, but you don't have to emulate the whole system, just user mode down to libraries. For GUI refresh (especially a compositing GUI), you can get that done in native code, if done right.
But the fact that it even presents itself as a P6 and surely has several performance characteristics similar to P6 means that it is simply a development of P6. P6 was never exceedingly hot and had, compared to the architectures of today, high IPC. That Pentium Pro from high school is haunting you.
The point is that all SP2 binaries use "canaries" to prevent unknown buffer overruns to go crazy. This means that some unknown issues, that were critical in SP1, never affected SP2, at least not with any obvious exploit. The canaries are nothing fancy, just put a random number of the stack and check it before you read EIP back from the stack. It's all over the place in SP2 code, to some performance penalty, but it's a feature that is impossible just to crack, unless you have a remote exploit to get the canary value for a specific module (it isn't rerandomized all the time) and then craft your exploit to that specific value.
SYSENTER/SYSEXIT to speed up silly benchmarks doing unbuffered intensive character I/O (or other things basically consisting of kernel roundtrips) 300 % faster!
Thanks to Yahoo for giving me the free tickets!
Well, I think that referrer is much more privacy-invasive than this one. The outgoing link is part of the actual content on the site, but what keywords I searched for or what intranet site linked to the competitor's web page is far more dangerous.
"you'll see javascript to overwrite the status line so it doesn't show it's tracking you ... and there are hundreds, if not thousands or millions of other sites that do the same." Like this one? Ok, the href might be valid, but the normal way to use the link is still tracked by going to a different URL.
Spreading information in a way intended by the creator of the site you visit may be a valid concern, but I haven't seen the widely deployed ban for images from other servers yet, for example.
Considering that 64-bit UI isn't present on current PowerPC Macs, this doesn't have to happen, but I can understand the worries against joining the 1st-gen bandwagon right now. In this situation, even a PowerPC Mac might be supported in a more consistent manner in a few years time.
Or a volume deal where the GM is used in some of the other models.