> Nothing is preventing anybody to re-implement the external API's of systemd on any OS (well need D-Bus).
D-Bus does not exist for any other kernel. Attempting to replicate the API's of such a constantly growing, expanding, and feature increasing API that replaces stable internal daemon, network, mounting, and security features would be far more destabilizing that systemd's own growth.
The result is that cross-platform support now requires publication of several sets of init tools, typically SysV scripts for other platforms and systemd for modern Linux distributions. I'm afraid it's discouraging cross-platform work, which is unfortunate.
>> The default was changed to enabled in the systemd source code in release 230. Unless distributions or developers manually patch their configurations, this unrequested behavior is now on by default.
> There is a configure flag. Anyway, systemd needs to be integrated inside the distributions, it's definitely not advised for random users to compile it themselves without using the distro patches.
The default for the configure flag was changed. Previously working builds would now actively break common user tools, such as screen, tmux, and nohup, without leaving any systemd log of when they were killing such processes.
Like many other systemd architectural changes, this broke numerous long stable tools for no particular benefit. The normal way to kill long-running processes belonging to absent users is a cron job that emails the adminisator. Such tasks normally also allow exceptions for adminstrative users, and have even been written with alternative policies for unattended use or with warning notes to the particular users before killing their tasks, along with a log of the worst offenders.
> The fact that they are disenfranchising their voters means nothing.
Not at all. The ability to vote the way _some_ voters of their districts selected, rather than the current "winner-take-all" common practices, would indeed enfranchise many voters from their districts. It's also supported by the Constitution. Even a casual reading of Article II shows that the electors can use their judgment to select a President.
> Well the goal of systemd is to be a base system for GNU/Linux distributions to reduce the gaps between them.
Please note the "Linux only" focus. It's not compatible with _any_ other free software or open source based operating system. And "reducing the gaps between them" has been breaking many stable, long-working tools without significant benefit.
>> such as the default enabled "KillUserProcess" command
> No distribution that I know enable that by default
The default was changed to enabled in the systemd source code in release 230. Unless distributions or developers manually patch their configurations, this unrequested behavior is now on by default.
> There is no reason, even with the price of electricity in Europe to shut your computer down, hence boot time is moot
There is with Virtual Machines. You pay for uptime, a reboot is often necessary to add more storage or memory successfully to an existing VM, and even for local development VM's they take significant server RAM for their active kernels. For many VM's, proper nightly backup snapshotting also requires a short shutdown. Also, after activation of new services, it's often useful to reboot the system to ensure complete activation of services when and if they *are* rebooted unattended.
> To that end, the only real interaction you normally have with systemd is to start or stop a service, and view the associated logs if some service is misbehaving.
Systemd has also taken over network configuration with an unnecessary DHCP service, which it should _not_ have touched, automounting, and is now attempting to manage user processes with misfeatures that kill user processes silently, such as the default enabled "KillUserProcess" command. Please be clear that systemd is not attempting to _manage_ processes. It is attempting to directly manage almost _all_ system services, many of them by direct replacement with dangerously incompatible and modified systems.
> Here's the funny thing, we had gasoline engine cars in the 90's that would easily get 50mpg,
One change is that they tests became much more stringent, and more accurate. This lowered many reported fuel efficiencies considerably. have become more true as the fraud became more noticed. Another change is emissions standards, which can play havoc with fuel economy. Some older, lighter cars with no catalytic converters had notably better fuel economy than more recent cars of similar size and model. Another change is increased safety requirements: Crumple zones and air bags add mass, cost, drag and in some cases all three to vehicles.
> Now the way you get tell fakes from the genuine product is because they were made better and lasted longer than the genuine product
Generally, no. The fake sneakers tend to be made with poor quality vinylized fabric rather than leather, the soles are thin and wear out quickly, and the stitches are more sparse and seams done more with glue than with stronger stitching. And since they're fake, the chances of getting a refund from the manufacturer are very poor, much as they are with other goods.
The fake name brand problem also includes Cisco and other quite high end network equipment, cheap phones that have been "skinned" to look like iPhones, tools and toys made of poor quality tin instead of the expected steel, and children's toys painted and made with toxic levels of heavy metals. It's not merely a matter of defrauding trademarked vendors, but of selling dangerous objects to unsuspecting purchasers.
The L1 visas are _for_ the top managers. They allow international companies with personnel like Donald Trump himself to move to, live in, and be paid in whatever nation is to their personal and professional benefit.
The H1B visa workers help foster profits for the top mangers of the companies that employ them. Donald Trump's policies are _all_ aimed at putting power and money in the hands of the wealthy. It should be trivial, as the president and with a Rep8ublican congress, to "reduce H1B visas" for his supporting voters but leave in large loopholes to protect their broad corporate use.
> It should be apparent if you think about it for a moment. A speaker is a transducer
Electromechanically, it's apparent. In terms of feedback that can be read by any sensory circuitry on the PC itself, it is not. A headphone or speaker circuit need have no _sensors_ that can be read or recorded by the signal generator. I'm afraid it's the introduction of simple chip solutions, designed to connect different electrical jacks to different programmable signals, and the introduction of A/D circuitry for noise cancellation and microphones that allows the cross connection of what is normally an output circuit to an input circuit.
Such features help reduce costs of circuitry for computer motherboards by providing single well designed, well understood chips for both functions. But it's not a design requirement.
>> HFT relies heavily on the positive feedback loops it starts, and rides like a surfboard, to deal hundreds or thousands of incremental trades to a new price
> No. If it did work that way it would quickly be shut down as illegal.
No more than insider trading is normally shut down.
> provided it does not involve market manipulation or other illegal activity.
HFT relies heavily on the positive feedback loops it starts, and rides like a surfboard, to deal hundreds or thousands of incremental trades to a new price. The volume of trading _itself_ generates positive feedback.
> BTW, now that pretty much 100% of traders are on a "high frequency" platform now,
I'm afraid you're overstating its popularity. HFT peaked in roughly 2009, and has fallen to less than 50% of trading volume in the last few years. Revenue from it has also dropped profoundly in the last 5 years.
"Market Manipulation" is the basis of high frequency trading. I wasn't claiming that what Sarao was doing was, itself, high frequency trading.
Much of the profit in high frequency trading comes _precisely_ from providing small feedback loops, and cycling through them all the way to where the stock bottoms out or peaks, then riding those same cycles the other way as the stock returns from any over-response. And yes, the high-frequency generates small positive feedback loops that drive stocks more quickly towards their equilibrium, and sometimes beyond. Much of the subtlety of the management of HFT is to detect exactly where the inflection points are, and to avoid overstimulating the stocks to a level where the SEC will admit the problem and insist on effective delays to trading, delays which would drain HFT's advantage.
Illegal or not, it's the basis of high frequency trading. The instantaneous sales, _themselves_, cause direct feedback and tip the market in various directions, which the high frequency traders pre-analyze and step out at _precisely_ the moment to maximize their own profits. It's built into the system.
> She was being practical and mistakeingly cut a few corners. No real harm was done,
As an IT professional who handles confidential data, and data security, that act was not "cutting a few corners". The leaks show that she was using the private server for direct activity as the Secretary of State as a matter of course. I've helped fire people who were pulling such abuses, especially for sensitive data such as HIPAA protected medical documentation or FERPA protected educational records.
Like failing to scrub a computer for handing it over to another employee, or leaving your passphrases on sticky pads on your desk, it's a very foolish and normally unnecessary act. Please don't ignore the very real misbehavior revealed by these leaks and charges.
That said, the woman is _far_ less dangerous than Mr. Trump, whose insistence on gambling at every level of personal and professional life is legendary and constitutes a _much_ larger danger.
> Why wouldn't Java be something a forward thinking CTO would be using? Java's been the intro CS language for
Because many intro CS majors learn actively dangerous practices which take years to unlearn. Excessive recursion, excessive object oriented programming, a blanket refusal to look below or above the designated layer of abstraction for the particular class assignment, code that does _not_ run correctly on even slightly different versions of Java due to hidden dependencies and the consistently hand-managed list of local libraries keeping production from matching productioni or test environments, and the deadly insistence on relying on out of band "garbage collection" to recover resources all burden Java and help prevent it from being a high performance or high reliability language.
I could go on with the foibles of CamelCaseForEverything programming and the dangers of compilation environments that randomly select the latest versions of all components from poorly maintained 3rd party repositories such as those used by Ant, Maven, and Gradle. But I suspect you've gotten the idea. Unlearning all those dangerous practices can take decades.
> The extent and type of surveillance that a Clinton administration would undertake is speculation at this point,
I'm afraid that former Secretary Clinton's party beliefs have been fairly clear, we saw them during her husband's administration. We also saw, with her email on a private server, that she's willing to ignore existing law, basic security practice, and a concern for setting safe precedents for IT in the name of political expediency.
I expect Mr. Trump to be _much worse_ because he's a political "big vision" leader, who squanders enormous resources on theater, and he is personally and professionally addicted to gambling with _enormous_ stakes in the name of his visions. That is not safe for IT, and it's not safe for handling surveillance policy, civil rights, or the enormous amounts of personal data accessible with or without warrants in the USA.
There are whole companies dedicated to precisely this. It's called "high-frequency trading". I'm afraid what this gentleman failed to do was to "pay his cut" to the others in the business, and to make it too apparent that the "liquidity" of US markets is designed precisely to allow skimming money from the system through arbitrage.
> when it can easily be simulated by replacing it with a safe, inert, and cheap substance.
When manufacturing uranium, one creates a _very_ large amount of depleted uranium. It would make a great deal of sense to let the weapon makers practice casting, molding, and shaping with the much safer depleted uranium. After all that practice, I'm sure several complete or nearly complete mockups were left over, and could very easily be repurposed for cargo testing.
_Modest_ virtualization, and modest graphics, can work well on reasonably modern laptops. Note that I said "Do _not_ expect to run extensive virtualized environments or multiple docker containers". I was referring to environments I've personally worked with where people set up one VM as a highly secure firewall for testing, another for database operations, another for web services, another for credentials management, etc., etc.
If you start running each of those VM's with Gnome active, expect to run out of resources quite quickly.
It depends on the distribution I'm using. For extreme lightweight, such as X in a virtual environment, I normally use twm for light weight and to avoid the dangerous instabilities I've found with Gnome, and more recently with KDE. I've recently had good success with Cinnamon.
Note that I'm often _hammering_ my laptop with work related communications, video conferencing, and lengthy local software tasks.
Thinkpads have been solid workhorses for me. Do _not_ expect to run extensive virtualized environments or multiple docker containers, and under _no_ circumstances use the default Gnome or KDE window managers common to modern Linux distributions. There are many better window managers, far more stable and far less resource gobbling window managers. Since the underlying graphical environment is one of the most memory, disk, and battery consuming features of modern Linux operating systems, keeping the environment lean will make it perform far better and extend its physical lifespan.
> The feeling at the time though, amongst young adult partiers, was that things were treatable
The feeling amongst young adults is often that they are immortal and invincible. But let's not echo the nonsensical claim that you "just took some penicillin" for STD's besides AIDS. Some people may have, and indeed I've met fools in my life who believed it. They were mistaken, and many of them have since died for many foolish reasons.
> Bias is irrelevant. The methodology is either valid or invalid in relation to the hypothesis and the results. Are the findings supported by the evidence?
Bias can, and does, profoundly skew direct results and their interpretation. This is why double blind studies are so useful, to help avoid experimental bias. Thee was an excellent example of this in the 1970's, involving Jacques Benveniste, a supporter of homeopathic medicine in his immunological research. His positive results could not be replicated when the experiments were done by a review committee with better double blind methods.
The same can be seen in almost every business plan and employee evaluation.
> STDs that people knew about you just took some penicillin
This is nonsense. Hepatitis C is devastating and difficult to treat, genital herpes is not curable, gonorrhea and syphilis are treatable but can be lethal if undetected.
Slashdot Mirror
> Nothing is preventing anybody to re-implement the external API's of systemd on any OS (well need D-Bus).
D-Bus does not exist for any other kernel. Attempting to replicate the API's of such a constantly growing, expanding, and feature increasing API that replaces stable internal daemon, network, mounting, and security features would be far more destabilizing that systemd's own growth.
The result is that cross-platform support now requires publication of several sets of init tools, typically SysV scripts for other platforms and systemd for modern Linux distributions. I'm afraid it's discouraging cross-platform work, which is unfortunate.
>> The default was changed to enabled in the systemd source code in release 230. Unless distributions or developers manually patch their configurations, this unrequested behavior is now on by default.
> There is a configure flag. Anyway, systemd needs to be integrated inside the distributions, it's definitely not advised for random users to compile it themselves without using the distro patches.
The default for the configure flag was changed. Previously working builds would now actively break common user tools, such as screen, tmux, and nohup, without leaving any systemd log of when they were killing such processes.
Like many other systemd architectural changes, this broke numerous long stable tools for no particular benefit. The normal way to kill long-running processes belonging to absent users is a cron job that emails the adminisator. Such tasks normally also allow exceptions for adminstrative users, and have even been written with alternative policies for unattended use or with warning notes to the particular users before killing their tasks, along with a log of the worst offenders.
> The fact that they are disenfranchising their voters means nothing.
Not at all. The ability to vote the way _some_ voters of their districts selected, rather than the current "winner-take-all" common practices, would indeed enfranchise many voters from their districts. It's also supported by the Constitution. Even a casual reading of Article II shows that the electors can use their judgment to select a President.
> Well the goal of systemd is to be a base system for GNU/Linux distributions to reduce the gaps between them.
Please note the "Linux only" focus. It's not compatible with _any_ other free software or open source based operating system. And "reducing the gaps between them" has been breaking many stable, long-working tools without significant benefit.
>> such as the default enabled "KillUserProcess" command
> No distribution that I know enable that by default
The default was changed to enabled in the systemd source code in release 230. Unless distributions or developers manually patch their configurations, this unrequested behavior is now on by default.
> There is no reason, even with the price of electricity in Europe to shut your computer down, hence boot time is moot
There is with Virtual Machines. You pay for uptime, a reboot is often necessary to add more storage or memory successfully to an existing VM, and even for local development VM's they take significant server RAM for their active kernels. For many VM's, proper nightly backup snapshotting also requires a short shutdown. Also, after activation of new services, it's often useful to reboot the system to ensure complete activation of services when and if they *are* rebooted unattended.
> To that end, the only real interaction you normally have with systemd is to start or stop a service, and view the associated logs if some service is misbehaving.
Systemd has also taken over network configuration with an unnecessary DHCP service, which it should _not_ have touched, automounting, and is now attempting to manage user processes with misfeatures that kill user processes silently, such as the default enabled "KillUserProcess" command. Please be clear that systemd is not attempting to _manage_ processes. It is attempting to directly manage almost _all_ system services, many of them by direct replacement with dangerously incompatible and modified systems.
> Here's the funny thing, we had gasoline engine cars in the 90's that would easily get 50mpg,
One change is that they tests became much more stringent, and more accurate. This lowered many reported fuel efficiencies considerably. have become more true as the fraud became more noticed. Another change is emissions standards, which can play havoc with fuel economy. Some older, lighter cars with no catalytic converters had notably better fuel economy than more recent cars of similar size and model. Another change is increased safety requirements: Crumple zones and air bags add mass, cost, drag and in some cases all three to vehicles.
> Now the way you get tell fakes from the genuine product is because they were made better and lasted longer than the genuine product
Generally, no. The fake sneakers tend to be made with poor quality vinylized fabric rather than leather, the soles are thin and wear out quickly, and the stitches are more sparse and seams done more with glue than with stronger stitching. And since they're fake, the chances of getting a refund from the manufacturer are very poor, much as they are with other goods.
The fake name brand problem also includes Cisco and other quite high end network equipment, cheap phones that have been "skinned" to look like iPhones, tools and toys made of poor quality tin instead of the expected steel, and children's toys painted and made with toxic levels of heavy metals. It's not merely a matter of defrauding trademarked vendors, but of selling dangerous objects to unsuspecting purchasers.
The L1 visas are _for_ the top managers. They allow international companies with personnel like Donald Trump himself to move to, live in, and be paid in whatever nation is to their personal and professional benefit.
The H1B visa workers help foster profits for the top mangers of the companies that employ them. Donald Trump's policies are _all_ aimed at putting power and money in the hands of the wealthy. It should be trivial, as the president and with a Rep8ublican congress, to "reduce H1B visas" for his supporting voters but leave in large loopholes to protect their broad corporate use.
> It should be apparent if you think about it for a moment. A speaker is a transducer
Electromechanically, it's apparent. In terms of feedback that can be read by any sensory circuitry on the PC itself, it is not. A headphone or speaker circuit need have no _sensors_ that can be read or recorded by the signal generator. I'm afraid it's the introduction of simple chip solutions, designed to connect different electrical jacks to different programmable signals, and the introduction of A/D circuitry for noise cancellation and microphones that allows the cross connection of what is normally an output circuit to an input circuit.
Such features help reduce costs of circuitry for computer motherboards by providing single well designed, well understood chips for both functions. But it's not a design requirement.
>> HFT relies heavily on the positive feedback loops it starts, and rides like a surfboard, to deal hundreds or thousands of incremental trades to a new price
> No. If it did work that way it would quickly be shut down as illegal.
No more than insider trading is normally shut down.
> provided it does not involve market manipulation or other illegal activity.
HFT relies heavily on the positive feedback loops it starts, and rides like a surfboard, to deal hundreds or thousands of incremental trades to a new price. The volume of trading _itself_ generates positive feedback.
> BTW, now that pretty much 100% of traders are on a "high frequency" platform now,
I'm afraid you're overstating its popularity. HFT peaked in roughly 2009, and has fallen to less than 50% of trading volume in the last few years. Revenue from it has also dropped profoundly in the last 5 years.
"Market Manipulation" is the basis of high frequency trading. I wasn't claiming that what Sarao was doing was, itself, high frequency trading.
Much of the profit in high frequency trading comes _precisely_ from providing small feedback loops, and cycling through them all the way to where the stock bottoms out or peaks, then riding those same cycles the other way as the stock returns from any over-response. And yes, the high-frequency generates small positive feedback loops that drive stocks more quickly towards their equilibrium, and sometimes beyond. Much of the subtlety of the management of HFT is to detect exactly where the inflection points are, and to avoid overstimulating the stocks to a level where the SEC will admit the problem and insist on effective delays to trading, delays which would drain HFT's advantage.
> No, it is called "market manipulation"
Illegal or not, it's the basis of high frequency trading. The instantaneous sales, _themselves_, cause direct feedback and tip the market in various directions, which the high frequency traders pre-analyze and step out at _precisely_ the moment to maximize their own profits. It's built into the system.
> She was being practical and mistakeingly cut a few corners. No real harm was done,
As an IT professional who handles confidential data, and data security, that act was not "cutting a few corners". The leaks show that she was using the private server for direct activity as the Secretary of State as a matter of course. I've helped fire people who were pulling such abuses, especially for sensitive data such as HIPAA protected medical documentation or FERPA protected educational records.
Like failing to scrub a computer for handing it over to another employee, or leaving your passphrases on sticky pads on your desk, it's a very foolish and normally unnecessary act. Please don't ignore the very real misbehavior revealed by these leaks and charges.
That said, the woman is _far_ less dangerous than Mr. Trump, whose insistence on gambling at every level of personal and professional life is legendary and constitutes a _much_ larger danger.
> Why wouldn't Java be something a forward thinking CTO would be using? Java's been the intro CS language for
Because many intro CS majors learn actively dangerous practices which take years to unlearn. Excessive recursion, excessive object oriented programming, a blanket refusal to look below or above the designated layer of abstraction for the particular class assignment, code that does _not_ run correctly on even slightly different versions of Java due to hidden dependencies and the consistently hand-managed list of local libraries keeping production from matching productioni or test environments, and the deadly insistence on relying on out of band "garbage collection" to recover resources all burden Java and help prevent it from being a high performance or high reliability language.
I could go on with the foibles of CamelCaseForEverything programming and the dangers of compilation environments that randomly select the latest versions of all components from poorly maintained 3rd party repositories such as those used by Ant, Maven, and Gradle. But I suspect you've gotten the idea. Unlearning all those dangerous practices can take decades.
> The extent and type of surveillance that a Clinton administration would undertake is speculation at this point,
I'm afraid that former Secretary Clinton's party beliefs have been fairly clear, we saw them during her husband's administration. We also saw, with her email on a private server, that she's willing to ignore existing law, basic security practice, and a concern for setting safe precedents for IT in the name of political expediency.
I expect Mr. Trump to be _much worse_ because he's a political "big vision" leader, who squanders enormous resources on theater, and he is personally and professionally addicted to gambling with _enormous_ stakes in the name of his visions. That is not safe for IT, and it's not safe for handling surveillance policy, civil rights, or the enormous amounts of personal data accessible with or without warrants in the USA.
> If one guy can cause this,
There are whole companies dedicated to precisely this. It's called "high-frequency trading". I'm afraid what this gentleman failed to do was to "pay his cut" to the others in the business, and to make it too apparent that the "liquidity" of US markets is designed precisely to allow skimming money from the system through arbitrage.
> when it can easily be simulated by replacing it with a safe, inert, and cheap substance.
When manufacturing uranium, one creates a _very_ large amount of depleted uranium. It would make a great deal of sense to let the weapon makers practice casting, molding, and shaping with the much safer depleted uranium. After all that practice, I'm sure several complete or nearly complete mockups were left over, and could very easily be repurposed for cargo testing.
_Modest_ virtualization, and modest graphics, can work well on reasonably modern laptops. Note that I said "Do _not_ expect to run extensive virtualized environments or multiple docker containers". I was referring to environments I've personally worked with where people set up one VM as a highly secure firewall for testing, another for database operations, another for web services, another for credentials management, etc., etc.
If you start running each of those VM's with Gnome active, expect to run out of resources quite quickly.
It depends on the distribution I'm using. For extreme lightweight, such as X in a virtual environment, I normally use twm for light weight and to avoid the dangerous instabilities I've found with Gnome, and more recently with KDE. I've recently had good success with Cinnamon.
Note that I'm often _hammering_ my laptop with work related communications, video conferencing, and lengthy local software tasks.
Thinkpads have been solid workhorses for me. Do _not_ expect to run extensive virtualized environments or multiple docker containers, and under _no_ circumstances use the default Gnome or KDE window managers common to modern Linux distributions. There are many better window managers, far more stable and far less resource gobbling window managers. Since the underlying graphical environment is one of the most memory, disk, and battery consuming features of modern Linux operating systems, keeping the environment lean will make it perform far better and extend its physical lifespan.
> The feeling at the time though, amongst young adult partiers, was that things were treatable
The feeling amongst young adults is often that they are immortal and invincible. But let's not echo the nonsensical claim that you "just took some penicillin" for STD's besides AIDS. Some people may have, and indeed I've met fools in my life who believed it. They were mistaken, and many of them have since died for many foolish reasons.
> Bias is irrelevant. The methodology is either valid or invalid in relation to the hypothesis and the results. Are the findings supported by the evidence?
Bias can, and does, profoundly skew direct results and their interpretation. This is why double blind studies are so useful, to help avoid experimental bias. Thee was an excellent example of this in the 1970's, involving Jacques Benveniste, a supporter of homeopathic medicine in his immunological research. His positive results could not be replicated when the experiments were done by a review committee with better double blind methods.
The same can be seen in almost every business plan and employee evaluation.
> STDs that people knew about you just took some penicillin
This is nonsense. Hepatitis C is devastating and difficult to treat, genital herpes is not curable, gonorrhea and syphilis are treatable but can be lethal if undetected.