> If you want to be good at multi-tasking, practice multi-tasking. > If you want to be good at focusing, practice focusing.
The problem is that research shows multi-tasking to be a given ability, you have it from birth, or you don't.
The question is still open for focusing. It depends a lot on physical processes within your body, but personally I think there is an element of practice to it.
The important thing to remember is that people who say they are good at focusing usually are, while people who say they are good at multi-tasking are usually just bad at focusing.
> No I wouldn't use a tiny case like this for a CAD workstation or gaming rig but for a small business PC or student PC it will be fine.
I would mind a tiny case for a workstation or gaming rig. But the problem is that despite technological progress, the heat generated by graphics cards needed to play recent games is going up! So in terms of consumer use, gaming PCs are getting worse and worse from year to year. There should be a way to change this, but it would involve reshuffling the whole industry.
> The consequences of that action is irrelevant; the action itself is bad.
The consequence are near nil. The plugin will get blacklisted by Firefox within days, and hopefully Microsoft takes the hint. I doubt they will play the cat and mouse game, so the only other options are to wise up or to give up.
And I don't understand why you cannot buy one big monitor instead of several small ones. The screen area is the same, which is main cost factor, so why are decent sized displays (1920x1200 and up) still so expensive?
> EFI is a much larger, more complex,[1][2]:4 replacement for the older BIOS firmware
Do we really want more complex? All that it should do get the hardware ready for basic tasks, and then load the boot loader and start it. If it can load more than one sector, that would be a bonus, or if it can offer different active partitions.
Further settings can be done by the OS once it is running, or by a special cut down OS that is loaded on demand. I never understood why there is only one (ugly) interface for the BIOS settings, or why you need to do those things globally at all. What if OS A wants different settings from OS B?
So I think this is a clear design of replacing a bad design by a more complicated one.
> In truth, akonadi and nepomuk are just a waste of system resources.
I agree completely. I switched to KDE 4 at 4.1.8 or so, and it was terrible. 4.2 was usable, but still not great.
Now I am playing with trinity, and while it looks a bit dated, it is really nice to use. Plus it responds a lot faster, even on a state of the art PC with 8 GB of RAM and binary 3D drivers.
And you cannot register a car with the ID card - again you need a passport or drivers license. It was a really pointless card, trying to detract from the fact that the real "value" is in the database.
> Actually VPro looks very cool. The question is how do you get it? Seems like it must be built into the system at the motherboard or bios level.
Yes, it is just like any other feature: you buy a PC that has it. My office PC for example has Intel AMT (nearly the same thing), but they wont tell me the password.
> That's one of the major flaws of the transition plan: not making IPv6 addresses just a superset of IPv4 addresses.
Funny enough they are. And you can (could) contact a IPv4 system from an IPv6 system, but the IPv4 cannot send the answer to an IPv6 address.
But I agree that this is going nowhere without IPv4/IPv6 compatibility. Yes, the problem is hard, but it is also very important! Instead of all the encapsulation protocols for IPv4 in IPv6, maybe somebody should have figured out application compatibility. As the world is today, getting http to work from an IPv6 client to an IPv4 server would be enough to get 90% of the internet working:-).
> So long as a user is actually allowed to execute what they want on a system
BTW, who even thought that was a good idea? Corporate users get a PC for a purpose, and all required applications should be provided. And even if not, a white list should cover 99% of all required software.
Of course as a user I know that things are not that simple. If the only provided browser is IE6 (actually IE7 since recently), Java, Flash, Acrobat, Quicktime and WinZip are all outdated, and the command line is disabled, then obviously I demand to install my own software. Or I just use PortableApps.
But from a security point of view users should not be able to execute their own files. That is only required for developers.
> McAfee's documentation specifically mentions turning it off because there is a high processor utilization bug still in it. Although you'd need to read the "read me" file that came with the patches.
And stupid me thought that high processor utilization is a "feature" of McAfee. Seriously, if it is bug, why has it been there for years if not decades?
> You can fully "undress" it, down to the bare basics, and it is incredibly stable.
But then again a stock Kubuntu Lucid in VirtualBox works quite well, and is set up in half an hour. I would recommend moving temporary files and swap to the local drive by creating a second virtual hard disk, but that is pretty much all you have to do.
> LTS release schedules are more stable and less work to maintain because they typically have all the software I need in their supported repositories.
I agree, the LTS release schedule looks very convenient. But we have to see how it works out, this is only the first time that Ubuntu supports LTS -> LTS upgrade directly without going through intermediate repositories. It worked quite well for me, so no complaints at this point.
> One business man bought a childs bike second hand just so he could board....idiotic rules for the ferry company to have.
Well, that happens if you hire idiots with no qualifications or training, and assume that "process" (read: strict adherence to rules) will sort it out.
> First, getters/setters are generated by your IDE of-course
No, I think that is exactly what the article is about: your IDE is the wrong level of abstraction. Adding generators does not change that, as long as you still view, save and edit the generated code. Code generation in general is usually a sign of things going wrong, especially if used as scaffolding.
The right level of abstraction is thinking about data, not state, and not code. Define your date, and please use a more modern approach than SQL, which is 90% focused on representation, and only 10% on meaning. Once you define your date, the relationships, the restrictions, one part is done. The next part is to define how to view it and how to edit it. It is classical MVC, actually, except that V and C can be specified together if you have a clever basis. Xopus does most of that, for example.
Whether you like it or not, the goal is to do simple business forms completely without programming. And I agree: the best code is code you don't need to write. SharePoint is doing quite a good job there, in that you get basic functionality without any code, you have quite a flexible data model, automatic or GUI design MVC, and the option to extend where necessary. It is the right level of abstraction (although not necessarily the best implementation).
> If a bunch of AG people and sheriffs descend on Amazon's offices with search warrants for "Any and all computers, disks, hardware, etc.", I think Amazon will take notice pretty quickly.
Interesting option. I would go one step further: since the attack has been committed from a virtual machine, it seems reasonable to confiscate for further analysis the virtual machine in question. Now this may not be as inconvenient for Amazon, but it also makes it more likely for them to cooperate.
The point being that the police or anybody could learn very little from the cloud hardware, I assume, because everything they need is in the software. So why not have a technically sound interface for investigating virtual machines? I think in the long term that will be inevitable for Amazon, if they want to avoid hardware being seized.
> At least their error messages are descriptive and informative.
Indeed. Accurate error messages are something that Microsoft never quite achieved, and Apple never even tried. "It does not work, please have a look at our website www.fuckandall.com for possible causes" - I hate that!
> Pretend that if an attempt to log into his account fails three times, his account is locked and requires a new password.
That is unless you get hold of the SAM. Then no limits apply, and even reasonably complex passwords are easily found with rainbow tables or one of the available online services.
Getting security right is hard. Not so much because it is difficult, but because you have to prevent a lot of well established bad practices. Eliminating these problems increases the security, password aging does not.
> The odds of the cracking succeeding does not change at all by password changing.
Very true. But the exposure duration of the cracked password would be reduced by changing it. So assuming that no backdoors are created, the cracked password would become useless very soon.
Is that better? Not by much I would say. And this being the best argument in favour of password aging, I agree that the whole idea is fundamentally flawed.
> Google had to, sooner or later, start fighting such a fight.
I agree, but I would have assumed Google to start the fight by demanding peering fees for access to youtube, not the other way round. After all, Google has the content, and the customers pay for internet. So while the videos flow from Google to the customer, the money flows from the customer to Google.
The telcos charging from both sides is just the good old dinosaur like approach. The world revolves around us, and we can charge for everything: making calls, receiving calls, not receiving calls, not making calls, connection fee, monthly fee, the lot!
In my view telcos are already "dump pipes", and that is a good thing. Now they just need to get cheap.
> So how quiet is the Phantom? 20 dB or less typically
That may sound low, but as geeks should now, dB is a unit of acoustic power, which is only a very rough indication how loud it is perceived.
If you are so much into technically details, why don't you give the sound level in Sone? That would make a lot more sense - and the number would be impressively low, I assume.
> What does that teach them? Don't do anything regardless of what it is unless you're "bribed".
Except that bribe is the wrong word. It would imply paying money for illegal services in your interest. Neither is the case here: learning is not illegal (at least not yet), and it is not ultimately in the teacher's interest either.
So it would teach them from an early age that honest hard work should be properly rewarded. Sounds fair to me.
Slashdot Mirror
> If you want to be good at multi-tasking, practice multi-tasking.
> If you want to be good at focusing, practice focusing.
The problem is that research shows multi-tasking to be a given ability, you have it from birth, or you don't.
The question is still open for focusing. It depends a lot on physical processes within your body, but personally I think there is an element of practice to it.
The important thing to remember is that people who say they are good at focusing usually are, while people who say they are good at multi-tasking are usually just bad at focusing.
> No I wouldn't use a tiny case like this for a CAD workstation or gaming rig but for a small business PC or student PC it will be fine.
I would mind a tiny case for a workstation or gaming rig. But the problem is that despite technological progress, the heat generated by graphics cards needed to play recent games is going up! So in terms of consumer use, gaming PCs are getting worse and worse from year to year. There should be a way to change this, but it would involve reshuffling the whole industry.
> The consequences of that action is irrelevant; the action itself is bad.
The consequence are near nil. The plugin will get blacklisted by Firefox within days, and hopefully Microsoft takes the hint. I doubt they will play the cat and mouse game, so the only other options are to wise up or to give up.
And I don't understand why you cannot buy one big monitor instead of several small ones. The screen area is the same, which is main cost factor, so why are decent sized displays (1920x1200 and up) still so expensive?
> EFI is a much larger, more complex,[1][2]:4 replacement for the older BIOS firmware
Do we really want more complex? All that it should do get the hardware ready for basic tasks, and then load the boot loader and start it. If it can load more than one sector, that would be a bonus, or if it can offer different active partitions.
Further settings can be done by the OS once it is running, or by a special cut down OS that is loaded on demand. I never understood why there is only one (ugly) interface for the BIOS settings, or why you need to do those things globally at all. What if OS A wants different settings from OS B?
So I think this is a clear design of replacing a bad design by a more complicated one.
> In truth, akonadi and nepomuk are just a waste of system resources.
I agree completely. I switched to KDE 4 at 4.1.8 or so, and it was terrible. 4.2 was usable, but still not great.
Now I am playing with trinity, and while it looks a bit dated, it is really nice to use. Plus it responds a lot faster, even on a state of the art PC with 8 GB of RAM and binary 3D drivers.
And you cannot register a car with the ID card - again you need a passport or drivers license. It was a really pointless card, trying to detract from the fact that the real "value" is in the database.
> Actually VPro looks very cool. The question is how do you get it? Seems like it must be built into the system at the motherboard or bios level.
Yes, it is just like any other feature: you buy a PC that has it. My office PC for example has Intel AMT (nearly the same thing), but they wont tell me the password.
> That's one of the major flaws of the transition plan: not making IPv6 addresses just a superset of IPv4 addresses.
Funny enough they are. And you can (could) contact a IPv4 system from an IPv6 system, but the IPv4 cannot send the answer to an IPv6 address.
But I agree that this is going nowhere without IPv4/IPv6 compatibility. Yes, the problem is hard, but it is also very important! Instead of all the encapsulation protocols for IPv4 in IPv6, maybe somebody should have figured out application compatibility. As the world is today, getting http to work from an IPv6 client to an IPv4 server would be enough to get 90% of the internet working :-).
> So long as a user is actually allowed to execute what they want on a system
BTW, who even thought that was a good idea? Corporate users get a PC for a purpose, and all required applications should be provided. And even if not, a white list should cover 99% of all required software.
Of course as a user I know that things are not that simple. If the only provided browser is IE6 (actually IE7 since recently), Java, Flash, Acrobat, Quicktime and WinZip are all outdated, and the command line is disabled, then obviously I demand to install my own software. Or I just use PortableApps.
But from a security point of view users should not be able to execute their own files. That is only required for developers.
> McAfee's documentation specifically mentions turning it off because there is a high processor utilization bug still in it. Although you'd need to read the "read me" file that came with the patches.
And stupid me thought that high processor utilization is a "feature" of McAfee. Seriously, if it is bug, why has it been there for years if not decades?
> Honestly - I don't know what the right answer for a corporate entity is...
Sophos is another good choice. But really any choice is better than Norton or McAfee. Avoid these at all costs.
> You can fully "undress" it, down to the bare basics, and it is incredibly stable.
But then again a stock Kubuntu Lucid in VirtualBox works quite well, and is set up in half an hour. I would recommend moving temporary files and swap to the local drive by creating a second virtual hard disk, but that is pretty much all you have to do.
> LTS release schedules are more stable and less work to maintain because they typically have all the software I need in their supported repositories.
I agree, the LTS release schedule looks very convenient. But we have to see how it works out, this is only the first time that Ubuntu supports LTS -> LTS upgrade directly without going through intermediate repositories. It worked quite well for me, so no complaints at this point.
And you have exactly 60 seconds to do that? :-)
> One business man bought a childs bike second hand just so he could board....idiotic rules for the ferry company to have.
Well, that happens if you hire idiots with no qualifications or training, and assume that "process" (read: strict adherence to rules) will sort it out.
> First, getters/setters are generated by your IDE of-course
No, I think that is exactly what the article is about: your IDE is the wrong level of abstraction. Adding generators does not change that, as long as you still view, save and edit the generated code. Code generation in general is usually a sign of things going wrong, especially if used as scaffolding.
The right level of abstraction is thinking about data, not state, and not code. Define your date, and please use a more modern approach than SQL, which is 90% focused on representation, and only 10% on meaning. Once you define your date, the relationships, the restrictions, one part is done. The next part is to define how to view it and how to edit it. It is classical MVC, actually, except that V and C can be specified together if you have a clever basis. Xopus does most of that, for example.
Whether you like it or not, the goal is to do simple business forms completely without programming. And I agree: the best code is code you don't need to write. SharePoint is doing quite a good job there, in that you get basic functionality without any code, you have quite a flexible data model, automatic or GUI design MVC, and the option to extend where necessary. It is the right level of abstraction (although not necessarily the best implementation).
> If a bunch of AG people and sheriffs descend on Amazon's offices with search warrants for "Any and all computers, disks, hardware, etc.", I think Amazon will take notice pretty quickly.
Interesting option. I would go one step further: since the attack has been committed from a virtual machine, it seems reasonable to confiscate for further analysis the virtual machine in question. Now this may not be as inconvenient for Amazon, but it also makes it more likely for them to cooperate.
The point being that the police or anybody could learn very little from the cloud hardware, I assume, because everything they need is in the software. So why not have a technically sound interface for investigating virtual machines? I think in the long term that will be inevitable for Amazon, if they want to avoid hardware being seized.
> At least their error messages are descriptive and informative.
Indeed. Accurate error messages are something that Microsoft never quite achieved, and Apple never even tried. "It does not work, please have a look at our website www.fuckandall.com for possible causes" - I hate that!
> Pretend that if an attempt to log into his account fails three times, his account is locked and requires a new password.
That is unless you get hold of the SAM. Then no limits apply, and even reasonably complex passwords are easily found with rainbow tables or one of the available online services.
Getting security right is hard. Not so much because it is difficult, but because you have to prevent a lot of well established bad practices. Eliminating these problems increases the security, password aging does not.
> The odds of the cracking succeeding does not change at all by password changing.
Very true. But the exposure duration of the cracked password would be reduced by changing it. So assuming that no backdoors are created, the cracked password would become useless very soon.
Is that better? Not by much I would say. And this being the best argument in favour of password aging, I agree that the whole idea is fundamentally flawed.
> Google had to, sooner or later, start fighting such a fight.
I agree, but I would have assumed Google to start the fight by demanding peering fees for access to youtube, not the other way round. After all, Google has the content, and the customers pay for internet. So while the videos flow from Google to the customer, the money flows from the customer to Google.
The telcos charging from both sides is just the good old dinosaur like approach. The world revolves around us, and we can charge for everything: making calls, receiving calls, not receiving calls, not making calls, connection fee, monthly fee, the lot!
In my view telcos are already "dump pipes", and that is a good thing. Now they just need to get cheap.
> So how quiet is the Phantom? 20 dB or less typically
That may sound low, but as geeks should now, dB is a unit of acoustic power, which is only a very rough indication how loud it is perceived.
If you are so much into technically details, why don't you give the sound level in Sone? That would make a lot more sense - and the number would be impressively low, I assume.
> What does that teach them? Don't do anything regardless of what it is unless you're "bribed".
Except that bribe is the wrong word. It would imply paying money for illegal services in your interest. Neither is the case here: learning is not illegal (at least not yet), and it is not ultimately in the teacher's interest either.
So it would teach them from an early age that honest hard work should be properly rewarded. Sounds fair to me.
> Windows NT 6.1 is called "seven" (or is it Mojave?).
You insensitive clot - I happen to be superstitious about the number after 6.