Powershell was very unreasonable in security terms. It relied on accessing, and locally mounting, with Administrative privileges, the hidden C$ from every powershell controlled host from every client running the powershell remote commands. That share is very dangerous to permit such direct CIFS mount access with, and is very difficult to disable without blocking the CIFS ports at your local firewalls. It''s a very powerful, but extremely dangerous tool to leave active by default. But turning off the C$ share can be quite difficult in most Windows configurations.:Like activating a VNC server on every Windows installation, and putting a common password on all of them "so that the admins can access the serves", it's a terrible practice from a security standpoint.
I'm afraid that without the tremendous wear, tear, and turbulence of an active heart pumping the approximate 20 liters of blood per minute of a human heart, the test is interesting but hardly complete. Turbulence can trigger blood clots, which are one of the main risks of transplants. Another risk of cardiac transplants is the failure of the connections to original veins and arteries. Until and unless those are tested under significant load, the experiments remain very incomplete.
Also, given the compatibility issues of pig hearts, I'm quite startled that human hearts of incorrect tissue matches are not more viable. I'm aware that few hearts are harvested in good condition, but I'm surprised that this difficult and risky xenotransplant is serously considered. If it were merely skin grafts, or blood vessel grafts for repairs, I'd understand it better.
Many FDA regulations are *insanely* confusing. Try doing *anything* that involves human nervous systems, such as research into sensory nerves or artificial vision, and you run into incredible amounts of what one "cannot" do and no acknowledge of what one *can* do. It's even worse for anything politically sensitive, such as revolution on human/simian comparative physiology, which offends the anti-evolution lobbies in Florida, artificial hearing, which offends the sign language deaf community, or Yahoo-Wahoo forbid, anything that mentions "stem cells", whether fetal or adult stem cells, due to the scare mongering about baby harvesting by these twits http://www.usatoday.com/story/...
You would also not *believe* the regulatory schizophrenia about homeopathy and Scientology. The FDA refuses, under lots of lobbying pressure, to call them outright frauds. But it also refuses to allow them to make medical claims, so practitioners of both frauds make the claims by implication or by "personal testimonial", not by official advertising, and the FDA continues to not act against them.
There are very, very good reasons for that. Subversion suffered profoundly from the inability to delete disastrous commits with binaries, mistaken directories, and completely obsolete and deleted projects. This made re-organization of content, even into another repository, very awkward and error prone. Even CVS and professional tools like Perforce were able to completely discard idle directories. The insistence on preserving deleted content also made sanitization of repositories to clear out passwords and password history, or clearing out accidental commits of bulky binaries, very difficult.
If you need a "reference" git repository with all history, certainly create one. But few projects actually need this.
> Also, it is now possible to tell Git not to guess your identity, which, instead, forces you to add a user.name and user.email before doing any commits.
This is invaluable. I occasionally want to source control configuration directories, with branches and merges among them for different environments. But when the working local repository is owned by "root" or "apache" and multiple administrators may need to configure the host, the need to reset the log associated with a commit to take individual credit or blame for it has been awkward. That has been the _only_ feature from the Subversion based clients that I've genuinely missed.
> That's why Sun was actually an investment worth making.
That is also what people said about SCO Group for a while after they tried to sue people for basing Linux software on UNIX. That did not work out well for SCO Group. But I'm afraid Sun did not have any products left with a profitable future. Freeware versions of Java were available, especially OpenJDK, and they'd profoundly missed the boat on the move to x86 based hardware. And inexpensive laptops, desktops, and "pizza box" servers easily replaced the formerly invaluable Sun workstations and servers. Coupled with the disastrous migration from SunOS to Solaris, there was little reason to use Sun operating systems anymore.
Their laptop for children used grayscale only for sunlight, for obvious power consumption reasons, and an effective low power color display for night use. I frankly wish most modern cell phones would use the same technolgy. I have no need to see pretty colors for a dozen icons on my cell phone, or for fancy borders on text messaging, email, or phone interfaces.
> It is when you're talking about the government's own use.
I beg to differ. Different departments of the US government want complete access to the data of other departments, and to be able to recover even secure data from their own records if it is accidentally or deliberately locked away. There may or not be legal justification or appropriate court orders involved, but they certainly want the access to even confidential, internal documents.
> refuse to give their diabetic children insulin on religious grounds.
I'd not realized there were real cases of this: there apparently have been. I'm shocked: giving insulin to a Type 1 diabetic is as clear a case of lifesaving medicine as putting a tourniquet on a severed limb. It's one of the few cases where a state or community, aware of a family refusal to treat the condition, would be on solid legal grounds to override the family and insist on treatment. And diagnosis is pretty easy: the excessive urination and sweet smell, and taste if you're willing to taste, of the urine have been well documented since ancient Greek times.
Some years ago, I actually got exhausted with an acquaintance who kept worrying about their risk of diabetes and kept failing to make it to a doctor's appointment. I eventually got fed up with them, picked up a glucometer at the local drug store for them, and got a diabetic at work to sit down with us and walk us through some quick tests. A few tests before and after lunch for my worried friend verified that they were probably borderline diabetic, and we got on the phone with their doctor's office to get them in two days, with a glucose tolerance test scheduled the next morning. It was us or the company nurse, and they _did not_ want to deal with the company nurse.
A good household distillation system is a very good solution to fecal contamination of drinking water. But it's still a burden for businesses who manage email, and for people whose mailboxes are occasionally overwhelmed by spam, for legitimate business traffic blocked as spam, and for people whose filters are not quite as good and get overwhelmed or defrauded by spammers.
I've not gotten so much telemarketing lately. I _am_ getting a lot of recruiter calls from fools in India who've seen a few keywords on my resume. Those calls are useless, and pointless, and get politely hung up on very quickly. A few times, people from actual startups have reached out to me unsolicited due to public open source work of mine. Since they've actually made an effort, I do spend a bit of time with them listening. Once or twice in the last year, I've actually been able to point them to someone in the field they might want to hire. And 3 times, I've explained why their business didn't work the last time it was tried, once pointing them to a research project from the 1960's that didn't work, either. But it was a bit of fun to educate youngsters in doing their background research before they try to get money for that business plan written on a napkin.
The US CAN-SPAM law was designed to permit "spam", unsolicited bulk communications, UBC, or "spam" as it was originally and very carefully defined. The law _protects_ spam, by setting an extremely low threshold for spam to be considered legal under US federal law, and by enforcing a US federal policy of "opt-out" rather than "opt-in" being the standard to avoid prosecution or civil suit for spam. It also prevents the publication and use of a "Do Not Spam" list for all bulk advertisers. Moreover, most commercial spammers consider any lists of "do not spam" addresses to be a very useful list of spam targets to use for their next company or next client, which they can legally do because of the "opt-out" structure of the CAN-SPAM law.
Spoofing phone numbers is a very distinct issue, and is now replaceable by using throwaway VOIP contact addresses from around the world, and throwaway phone numbers. I'd not expect this to get better.
I'm concerned that they'll teach coding the same way that many schools teach math. Reinvent the paradigms every few years, require extensive retraining of all the latest teachers in the latest paradigm, and care more about the fad than about the basic skills.
For reference, I've linked to Tom Lehrer's "New Math" song:
No, it's mostly VMS. Take a look at the extensive lawsuits when David Cutler was hired from DEC, and took a lot of his old VMS developer team with him to create the kernel for Windows NT.
It's illusory wealth, like his business "successes". It's coming down in a few years with the end of the currently expanding business bubble.
Be very, very frightened of the collapse. I'm genuinely frighted of the wars he's going to get us into, we still haven't paid off the debt and internatianal karma debt of the last pair of Republican "business sense" wars.
> I disagree. Codeweavers are the main developers behind open source wine
Which has been consistently disappointing. A simple review of the ratings of various packages show that of over 13,000 rated packages, over 4000 of them are rated as "garbage" under Wine operations, and well over 2000 are rated as "bronze". Taht's roughly half of the entire set of rated packages. And when I've tested silver or gold packages, I've found the ratings to be optimistically generous.
For a very few well defined and supported packages, like Microsoft Office, Codeweavers' work can be useful. But for financial applications, CAD, or games, it's behavior has been too unreliable for me to waste the effort using it, even when I tested it again last year.
To avoid systemd, I suspect. Since systemd only works with the Linux kernel, and so far has produced a great deal of difficulty in return for its very aggressive re-engineering of the entire Linux back end infrastructure, it seems very very reasonable to try simply replacing the kernel to get a clean divorce from the systemd infrastructure.
Except for graphics tablets, many mice and game controllers, laptop video and audio chipsets, laptop power control systems, and bleeding edge graphics cards. It's why FreeBSD was a good choice for the Apple kernel: they had control of the hardware and could judiciously invest driver development in only those components they actually used.
> You don't need to replace existing infrastructure, just keep the cables you already have!
Except that many if not most wall jacks don't have enough slack left to re-terminate, especially poorly done home installations. Even if there is enough slack left, inevitable mistakes are going to force recabling of up to 10% of all reterminated wire. That adds a great deal to the cost.
Data centers are not mostly built of high capacity fiber optic switches. For most, the large majority of the hardware is hosted business grade hardware, which so far has no need or desire for 100 GigE, they use far more economical GigE or perhaps 10 GigE.
The RJ-45 connector supports GigE quite easily and well, as long as some fool doesn't use low grade CAT5 cabling that's been abused and terminates it carefully, I cannot easily count the number of times I've found people who "economised" by hand-terminating cables, badly, and failing to use good quality connectors and failing to correctly apply the strain relief crimp, making the cables only barely long enough, and inevitably pulling wires out of the connector. I still remember the very long, very painful discussions about how much money he was saving making the cables only barely long enough with no slack whatsoever in them, and how much better the whole system behaved when I replaced his entire densely woven cable panels with commercial crade cables of standard sizes with a bit of slack in them, and using Velcro instead of rightly pulled plastic ty-wraps.
We eventually got rid of that legacy employee: their "economies" were killing our reliability. Cleaning up after people like that can be a nightmare.
Slashdot Mirror
Powershell was very unreasonable in security terms. It relied on accessing, and locally mounting, with Administrative privileges, the hidden C$ from every powershell controlled host from every client running the powershell remote commands. That share is very dangerous to permit such direct CIFS mount access with, and is very difficult to disable without blocking the CIFS ports at your local firewalls. It''s a very powerful, but extremely dangerous tool to leave active by default. But turning off the C$ share can be quite difficult in most Windows configurations. :Like activating a VNC server on every Windows installation, and putting a common password on all of them "so that the admins can access the serves", it's a terrible practice from a security standpoint.
I'm afraid that without the tremendous wear, tear, and turbulence of an active heart pumping the approximate 20 liters of blood per minute of a human heart, the test is interesting but hardly complete. Turbulence can trigger blood clots, which are one of the main risks of transplants. Another risk of cardiac transplants is the failure of the connections to original veins and arteries. Until and unless those are tested under significant load, the experiments remain very incomplete.
Also, given the compatibility issues of pig hearts, I'm quite startled that human hearts of incorrect tissue matches are not more viable. I'm aware that few hearts are harvested in good condition, but I'm surprised that this difficult and risky xenotransplant is serously considered. If it were merely skin grafts, or blood vessel grafts for repairs, I'd understand it better.
> SEC regulations can be maddeningly vague
Many FDA regulations are *insanely* confusing. Try doing *anything* that involves human nervous systems, such as research into sensory nerves or artificial vision, and you run into incredible amounts of what one "cannot" do and no acknowledge of what one *can* do. It's even worse for anything politically sensitive, such as revolution on human/simian comparative physiology, which offends the anti-evolution lobbies in Florida, artificial hearing, which offends the sign language deaf community, or Yahoo-Wahoo forbid, anything that mentions "stem cells", whether fetal or adult stem cells, due to the scare mongering about baby harvesting by these twits http://www.usatoday.com/story/...
You would also not *believe* the regulatory schizophrenia about homeopathy and Scientology. The FDA refuses, under lots of lobbying pressure, to call them outright frauds. But it also refuses to allow them to make medical claims, so practitioners of both frauds make the claims by implication or by "personal testimonial", not by official advertising, and the FDA continues to not act against them.
There are very, very good reasons for that. Subversion suffered profoundly from the inability to delete disastrous commits with binaries, mistaken directories, and completely obsolete and deleted projects. This made re-organization of content, even into another repository, very awkward and error prone. Even CVS and professional tools like Perforce were able to completely discard idle directories. The insistence on preserving deleted content also made sanitization of repositories to clear out passwords and password history, or clearing out accidental commits of bulky binaries, very difficult.
If you need a "reference" git repository with all history, certainly create one. But few projects actually need this.
> Also, it is now possible to tell Git not to guess your identity, which, instead, forces you to add a user.name and user.email before doing any commits.
This is invaluable. I occasionally want to source control configuration directories, with branches and merges among them for different environments. But when the working local repository is owned by "root" or "apache" and multiple administrators may need to configure the host, the need to reset the log associated with a commit to take individual credit or blame for it has been awkward. That has been the _only_ feature from the Subversion based clients that I've genuinely missed.
> That's why Sun was actually an investment worth making.
That is also what people said about SCO Group for a while after they tried to sue people for basing Linux software on UNIX. That did not work out well for SCO Group. But I'm afraid Sun did not have any products left with a profitable future. Freeware versions of Java were available, especially OpenJDK, and they'd profoundly missed the boat on the move to x86 based hardware. And inexpensive laptops, desktops, and "pizza box" servers easily replaced the formerly invaluable Sun workstations and servers. Coupled with the disastrous migration from SunOS to Solaris, there was little reason to use Sun operating systems anymore.
Their laptop for children used grayscale only for sunlight, for obvious power consumption reasons, and an effective low power color display for night use. I frankly wish most modern cell phones would use the same technolgy. I have no need to see pretty colors for a dozen icons on my cell phone, or for fancy borders on text messaging, email, or phone interfaces.
> It is when you're talking about the government's own use.
I beg to differ. Different departments of the US government want complete access to the data of other departments, and to be able to recover even secure data from their own records if it is accidentally or deliberately locked away. There may or not be legal justification or appropriate court orders involved, but they certainly want the access to even confidential, internal documents.
> refuse to give their diabetic children insulin on religious grounds.
I'd not realized there were real cases of this: there apparently have been. I'm shocked: giving insulin to a Type 1 diabetic is as clear a case of lifesaving medicine as putting a tourniquet on a severed limb. It's one of the few cases where a state or community, aware of a family refusal to treat the condition, would be on solid legal grounds to override the family and insist on treatment. And diagnosis is pretty easy: the excessive urination and sweet smell, and taste if you're willing to taste, of the urine have been well documented since ancient Greek times.
Some years ago, I actually got exhausted with an acquaintance who kept worrying about their risk of diabetes and kept failing to make it to a doctor's appointment. I eventually got fed up with them, picked up a glucometer at the local drug store for them, and got a diabetic at work to sit down with us and walk us through some quick tests. A few tests before and after lunch for my worried friend verified that they were probably borderline diabetic, and we got on the phone with their doctor's office to get them in two days, with a glucose tolerance test scheduled the next morning. It was us or the company nurse, and they _did not_ want to deal with the company nurse.
The NSA has no enforcement power: they cannot prosecute or file charges against anyone.
Do you really think it won't be repealed before the end of the year? Or another glaring hole opened up in it?
A good household distillation system is a very good solution to fecal contamination of drinking water. But it's still a burden for businesses who manage email, and for people whose mailboxes are occasionally overwhelmed by spam, for legitimate business traffic blocked as spam, and for people whose filters are not quite as good and get overwhelmed or defrauded by spammers.
I've not gotten so much telemarketing lately. I _am_ getting a lot of recruiter calls from fools in India who've seen a few keywords on my resume. Those calls are useless, and pointless, and get politely hung up on very quickly. A few times, people from actual startups have reached out to me unsolicited due to public open source work of mine. Since they've actually made an effort, I do spend a bit of time with them listening. Once or twice in the last year, I've actually been able to point them to someone in the field they might want to hire. And 3 times, I've explained why their business didn't work the last time it was tried, once pointing them to a research project from the 1960's that didn't work, either. But it was a bit of fun to educate youngsters in doing their background research before they try to get money for that business plan written on a napkin.
The US CAN-SPAM law was designed to permit "spam", unsolicited bulk communications, UBC, or "spam" as it was originally and very carefully defined. The law _protects_ spam, by setting an extremely low threshold for spam to be considered legal under US federal law, and by enforcing a US federal policy of "opt-out" rather than "opt-in" being the standard to avoid prosecution or civil suit for spam. It also prevents the publication and use of a "Do Not Spam" list for all bulk advertisers. Moreover, most commercial spammers consider any lists of "do not spam" addresses to be a very useful list of spam targets to use for their next company or next client, which they can legally do because of the "opt-out" structure of the CAN-SPAM law.
Spoofing phone numbers is a very distinct issue, and is now replaceable by using throwaway VOIP contact addresses from around the world, and throwaway phone numbers. I'd not expect this to get better.
I'm concerned that they'll teach coding the same way that many schools teach math. Reinvent the paradigms every few years, require extensive retraining of all the latest teachers in the latest paradigm, and care more about the fad than about the basic skills.
For reference, I've linked to Tom Lehrer's "New Math" song:
https://www.youtube.com/watch?...
And one must remember:
The important thing is to understand what you're doing, and not to get the right answer.
I'm afraid I've been dealing with the results of changing fads in math, and in programming classes, for decades.
> This is unbelievably stupid. I know, social engineering and all, but why the f$#%k would you click ok to a UAC warning to read a CV?! C
Because they're HR. The field has high turnover and is noted for poor security practices "in order to get their job done".
No, it's mostly VMS. Take a look at the extensive lawsuits when David Cutler was hired from DEC, and took a lot of his old VMS developer team with him to create the kernel for Windows NT.
> He's already one of the richest men in America,
It's illusory wealth, like his business "successes". It's coming down in a few years with the end of the currently expanding business bubble.
Be very, very frightened of the collapse. I'm genuinely frighted of the wars he's going to get us into, we still haven't paid off the debt and internatianal karma debt of the last pair of Republican "business sense" wars.
> I disagree. Codeweavers are the main developers behind open source wine
Which has been consistently disappointing. A simple review of the ratings of various packages show that of over 13,000 rated packages, over 4000 of them are rated as "garbage" under Wine operations, and well over 2000 are rated as "bronze". Taht's roughly half of the entire set of rated packages. And when I've tested silver or gold packages, I've found the ratings to be optimistically generous.
For a very few well defined and supported packages, like Microsoft Office, Codeweavers' work can be useful. But for financial applications, CAD, or games, it's behavior has been too unreliable for me to waste the effort using it, even when I tested it again last year.
To avoid systemd, I suspect. Since systemd only works with the Linux kernel, and so far has produced a great deal of difficulty in return for its very aggressive re-engineering of the entire Linux back end infrastructure, it seems very very reasonable to try simply replacing the kernel to get a clean divorce from the systemd infrastructure.
Except for graphics tablets, many mice and game controllers, laptop video and audio chipsets, laptop power control systems, and bleeding edge graphics cards. It's why FreeBSD was a good choice for the Apple kernel: they had control of the hardware and could judiciously invest driver development in only those components they actually used.
Look again. The GIMP for Windows had one heck of a time trying to re-establish control, because they wanted it _deleted_.
Those have tended to be very fragile. I cannot recommend them.
> You don't need to replace existing infrastructure, just keep the cables you already have!
Except that many if not most wall jacks don't have enough slack left to re-terminate, especially poorly done home installations. Even if there is enough slack left, inevitable mistakes are going to force recabling of up to 10% of all reterminated wire. That adds a great deal to the cost.
Data centers are not mostly built of high capacity fiber optic switches. For most, the large majority of the hardware is hosted business grade hardware, which so far has no need or desire for 100 GigE, they use far more economical GigE or perhaps 10 GigE.
The RJ-45 connector supports GigE quite easily and well, as long as some fool doesn't use low grade CAT5 cabling that's been abused and terminates it carefully, I cannot easily count the number of times I've found people who "economised" by hand-terminating cables, badly, and failing to use good quality connectors and failing to correctly apply the strain relief crimp, making the cables only barely long enough, and inevitably pulling wires out of the connector. I still remember the very long, very painful discussions about how much money he was saving making the cables only barely long enough with no slack whatsoever in them, and how much better the whole system behaved when I replaced his entire densely woven cable panels with commercial crade cables of standard sizes with a bit of slack in them, and using Velcro instead of rightly pulled plastic ty-wraps.
We eventually got rid of that legacy employee: their "economies" were killing our reliability. Cleaning up after people like that can be a nightmare.
Not 42?
https://en.wikipedia.org/wiki/...