There is a turtle standing on silicon at some point. Get a micro-controller evaluation kit and twiddle some bits. Write a "Hello World", and no using that fancy printf stuff. Write the bits to the UART and manage the status flags.
Then just for fun, do some add with carry to a longer int type.
The things you list are excellent. Without understanding basic pitfalls and trade-offs, no amount of MIPS or RAM can save you. 6502 was very limiting; I'm grateful that we have cheap micros that can do 30 MIPS instead of 0.5 MIPS. So whole new classes of problems can be solved in real-time. But embedded programming is still a place were bytes and number of instructions are a big deal.
It boggles my mind how inconsiderate programmers have become where a news webpage can bring a multi-GHz 64 bit CPU to its knees.
Embedded systems may not get the attention of mobile apps or web development, but they are literally surrounding you. They are everything but a niche. Every entertainment device, appliance, furnace/AC, remote control, toys, car, cable modem, etc. And they all run on low-cost processors where bits and instruction cycles make a difference.
True. My only criticism of projects like ITER, NIF, etc. are that they are more academic than deliberate in achieving energy generation anytime this century. Granted, you have to work out the fusion part first, but then what? Do we wait another 50-100 years for the anointed project scientists to build a working power plant? We need tons of engineering going into this - now.
The ROI of fusion research is unimaginable if it ultimately works out. $20 billion _per year_ should be the least we could do. We should keep throwing ITER a bone _and_ triple whatever domestic (USA) programs we have currently. Get multiple projects in the pipeline, humanity cannot wait much longer.
Ending energy scarcity forever is _the_ most important thing humanity will have done up to this date. Wealth, poverty, hunger, and greed will have very little meaning when energy is virtually free. On the other hand, there is much political denial since it would certainly be a game changer for those in power.
The same effort that went into the Manhattan project, or the Apollo program should just be a given here. Being over-budget is rather meaningless; you could be over-budget by 10x, and I wouldn't bat an eye. The Manhattan project built multiple enormous facilities in parallel. Laying foundations for enrichment refineries that weren't even designed yet. That took some leap of faith.
You have to take a lot of false paths and then breakthroughs can happen unexpectedly. But we won't get there by not trying at all. Until it is proven to be impossible, the risk of dropping a few tens of billions a year to end energy scarcity forever is a reasonable gamble.
DC voltages would be blocked at the transformer. The miles-long transmission lines wouldn't carry a DC voltage into your house unless the protective gear on the pole failed somehow. The transformer itself could likely overheat leaving you without power later on.
Unfortunately, faulty compilers are more than just hypothetical. But besides that...
Even with a perfect compiler, the knowledge it takes to avoid these 'equal semantics' traps is beyond most experienced programmers (no matter how much our egos say otherwise.) But otherwise, I totally agree that most 'optimizer bugs' I've seen are just lack of understanding by the programmer. The compiler rightfully saw its solution as equivalent to what the source code stated.
I think there should be a rule of thumb: "Every year of programming experience gets you halfway closer to knowing all the gotchas"
So we can blame the programmer, but that isn't going to produce better code in the real world. The tools and languages need to meet with common expectations. If that means taking a more "secure by default" approach then so be it. The experts can choose to take off the training wheels with pragmas or additional compiler options.
I don't think any language is portable at that level, so you may as well use asm (my preference is linked instead of inline to ensure a clean and simple abstraction.) Every processor has different math status registers and different math instruction capabilities. I'm not sure how C can express these things.
Life is hardly based on facts - that information just isn't always available. Humans generally create good outcomes based on limited information. When school administrator have wildly bad judgement they usually get booted out by parents and/or school board.
Does gun control save lives in Mexico, Russia, or dozens of other places it is on the law books? Statistics do lie if you cherry pick the results you prefer. I'd promote gun control if I thought it had a strong correlation to reducing violence. But it doesn't, the results are lost in the statistical noise. I'd prefer to look at the issues of poverty, inequality, education, abuse, mental health, addiction, etc.
The USA is not and analogue to Canada, UK, or Australia. We have massive social problems more like a 3rd world country. We also like to make laws and then not follow them.
There may be innovative approaches to gun control, but simple bans, license, or registration are constantly ignored on the black market by criminals.
I have to agree that the CC company makes a difference, Capital One has always been the one to tell me when something bad has happened. Slight inconvenience to me. Somebody besides me ate the cost (probably Capital One.) So obviously their business model is profitable enough to not really worry too much.
That being said, it is about F'ing time that retailers and CC companies make the investment into chip and pin systems. Not perfect, but would basically shut down most causal card skimmers. The one-time card numbers for online transactions are a good compromise, especially considering it might be difficult to get the card reading devices at the home-user level.
I agree that many schools bend the rules for athletics, music, theater, chess club, whatever... But I've never seen how tenure stopped that injustice.
Hard teachers should be unpopular... good, effective teachers that help kids pass Regents, SAT, and AP tests are popular. If you live in a place that doesn't care about those things, then feel free to ignore this post.
Exactly. A university professor may have some argument for tenure when doing controversial research. But even then, maybe that professor should shop around for a university that is more sympathetic. But in primary and high schools, the curriculum is usually predefined. The school should absolutely have the tools to retain and reward the most effective teachers.
Employers (school boards) are not specifically all a bunch of fat cat a-holes walking around looking for people to fire. They will absolutely retain people who do their jobs well. What is the specific conflict of interest inherent to the teaching industry that requires tenure as compared to other industries?
justfacts is pretty good. We would be a much less ignorant people if people just got a reality check from sites like this after watching the evening news.
But it would also be good if schools made curricula such as 'logical reasoning', 'critical thinking', and 'statistical analysis' mandatory for all students. You don't have to be a math genius to understand these things. Perhaps people would be less gullible to politically motivated dialogs. But I'm not sure if schools are intended to produce critical-thinkers...
I wish more people asked questions like: "why do those stats look cherry-picked, what is the selection criteria?", or "how does the evening news choose stories, do they proportionally represent reality?", "What is the best investment strategy that will actually serve me and not Wall Street?", "Is a vote for politician A or B a false dichotomy?"
The whole misdirection to Bitlocker is probably a sarcastic joke pointing to a company far more likely to adhere to NSL's.. Bitlocker isn't even provided on Home editions of Windows, so it really isn't such an obvious alternative. Their directions literally go through steps to change the Windows product key. I would assume to do this legally you pay Microsoft, is that correct? So you are telling me that TrueCrypt as a free alternative for home users isn't still worth developing by someone?
There isn't any way they can give us confidence that they are playing nice either. This is what happens when you violate the trust of the US people and the rest of the world. People used to believe that the US Constitution was the fire-block that was stopping this same nonsense that you would expect from China or other authoritarian governments with no protection of human rights. Now its official, there is no difference.
US corporations have lost major credibility in the world technology market: "We promise this time we won't put secret back doors in our products that we won't tell you about because our government forced us to and we couldn't tell you. We promise this won't happen anymore."
Ah, yes... "If you aren't doing anything wrong, then what do you have to worry about" Except there are plenty of cases of persecution if you happen to be:
- Gay,
- A former member of the communist party,
- Union organizer,
- Whistle blower,
- Protester, objector, not in line with corporate America,
- Catholic, Jewish, Japanese, or anything else not favorable at the time... None of these people are terrorists, but clearly lost their liberties, reputation, or assets when they were "outed"
That sounds about right. I think the corollary to that is: "When a simpler approach is found, at least one person in the group will refuse to give up the needlessly complex bits."
I think people feel attached to XP because it was comfortable and stable (for the first time) for most people. People were accustomed to the Win95 look and feel, so it wasn't a culture shock. And it worked well in corporate domains and at home with a 'grown-up' security model unlike 95/98.
Everything since then has been a forced upgrade that can break existing applications and infrastructure. Most IT departments and individuals avoided Vista as much as possible. Windows 7 is useable. Windows 8 is confusing and radical for many people.
I feel like I've lost something going from XP to Win7, like the search dialog. The folder structure is re-arranged, but still arbitrary. The Registry is still as arbitrary, and the control panel is more confusing. The document library is good in theory, but most people I've helped with Windows 7 just get lost and try to avoid it, but it still gets in your face. So maybe Windows 7 offered as improvements the UAC security model and IPV6 built-in.
Redmond and Silicon Valley needs to wake up that this isn't 1999 anymore. The fact that Windows 7 and 8 came along 3 years apart is too soon. People don't want to upgrade until something actually breaks. And even then, they don't want to (or can't) spend another capital investment on software for the new OS, especially if it is tied to some type of equipment or machinery . Most businesses expect a 5-10 year amortization on basic equipment. There are even tax and accounting consequences to this if the recovery period doesn't meet expectations. Microsoft could probably make money selling XP (and sell support for critical flaws) for quite some time. I'm not sure why they need to set an arbitrary date to close it out. Does it really cost that much to maintain?
Slashdot Mirror
There is a turtle standing on silicon at some point. Get a micro-controller evaluation kit and twiddle some bits. Write a "Hello World", and no using that fancy printf stuff. Write the bits to the UART and manage the status flags.
Then just for fun, do some add with carry to a longer int type.
The things you list are excellent. Without understanding basic pitfalls and trade-offs, no amount of MIPS or RAM can save you.
6502 was very limiting; I'm grateful that we have cheap micros that can do 30 MIPS instead of 0.5 MIPS. So whole new classes of problems can be solved in real-time. But embedded programming is still a place were bytes and number of instructions are a big deal.
It boggles my mind how inconsiderate programmers have become where a news webpage can bring a multi-GHz 64 bit CPU to its knees.
Embedded systems may not get the attention of mobile apps or web development, but they are literally surrounding you. They are everything but a niche. Every entertainment device, appliance, furnace/AC, remote control, toys, car, cable modem, etc. And they all run on low-cost processors where bits and instruction cycles make a difference.
True. My only criticism of projects like ITER, NIF, etc. are that they are more academic than deliberate in achieving energy generation anytime this century. Granted, you have to work out the fusion part first, but then what? Do we wait another 50-100 years for the anointed project scientists to build a working power plant? We need tons of engineering going into this - now.
The ROI of fusion research is unimaginable if it ultimately works out. $20 billion _per year_ should be the least we could do. We should keep throwing ITER a bone _and_ triple whatever domestic (USA) programs we have currently. Get multiple projects in the pipeline, humanity cannot wait much longer.
Ending energy scarcity forever is _the_ most important thing humanity will have done up to this date. Wealth, poverty, hunger, and greed will have very little meaning when energy is virtually free. On the other hand, there is much political denial since it would certainly be a game changer for those in power.
The same effort that went into the Manhattan project, or the Apollo program should just be a given here. Being over-budget is rather meaningless; you could be over-budget by 10x, and I wouldn't bat an eye. The Manhattan project built multiple enormous facilities in parallel. Laying foundations for enrichment refineries that weren't even designed yet. That took some leap of faith.
You have to take a lot of false paths and then breakthroughs can happen unexpectedly. But we won't get there by not trying at all. Until it is proven to be impossible, the risk of dropping a few tens of billions a year to end energy scarcity forever is a reasonable gamble.
DC voltages would be blocked at the transformer. The miles-long transmission lines wouldn't carry a DC voltage into your house unless the protective gear on the pole failed somehow. The transformer itself could likely overheat leaving you without power later on.
The answer to just about all of our problems. Yet how much effort are we really putting in.
It's turtles all the way down...
resort to the long long long type?
Your statement is perfectly true.
Unfortunately, faulty compilers are more than just hypothetical. But besides that...
Even with a perfect compiler, the knowledge it takes to avoid these 'equal semantics' traps is beyond most experienced programmers (no matter how much our egos say otherwise.) But otherwise, I totally agree that most 'optimizer bugs' I've seen are just lack of understanding by the programmer. The compiler rightfully saw its solution as equivalent to what the source code stated.
I think there should be a rule of thumb: "Every year of programming experience gets you halfway closer to knowing all the gotchas"
So we can blame the programmer, but that isn't going to produce better code in the real world. The tools and languages need to meet with common expectations. If that means taking a more "secure by default" approach then so be it. The experts can choose to take off the training wheels with pragmas or additional compiler options.
I don't think any language is portable at that level, so you may as well use asm (my preference is linked instead of inline to ensure a clean and simple abstraction.) Every processor has different math status registers and different math instruction capabilities. I'm not sure how C can express these things.
Life is hardly based on facts - that information just isn't always available. Humans generally create good outcomes based on limited information. When school administrator have wildly bad judgement they usually get booted out by parents and/or school board.
It would only take a few motivated parents to vote in new school board members to remedy that.
Does gun control save lives in Mexico, Russia, or dozens of other places it is on the law books? Statistics do lie if you cherry pick the results you prefer. I'd promote gun control if I thought it had a strong correlation to reducing violence. But it doesn't, the results are lost in the statistical noise. I'd prefer to look at the issues of poverty, inequality, education, abuse, mental health, addiction, etc.
The USA is not and analogue to Canada, UK, or Australia. We have massive social problems more like a 3rd world country. We also like to make laws and then not follow them.
There may be innovative approaches to gun control, but simple bans, license, or registration are constantly ignored on the black market by criminals.
I have to agree that the CC company makes a difference, Capital One has always been the one to tell me when something bad has happened. Slight inconvenience to me. Somebody besides me ate the cost (probably Capital One.) So obviously their business model is profitable enough to not really worry too much.
That being said, it is about F'ing time that retailers and CC companies make the investment into chip and pin systems. Not perfect, but would basically shut down most causal card skimmers. The one-time card numbers for online transactions are a good compromise, especially considering it might be difficult to get the card reading devices at the home-user level.
I agree that many schools bend the rules for athletics, music, theater, chess club, whatever... But I've never seen how tenure stopped that injustice.
Hard teachers should be unpopular... good, effective teachers that help kids pass Regents, SAT, and AP tests are popular. If you live in a place that doesn't care about those things, then feel free to ignore this post.
Exactly. A university professor may have some argument for tenure when doing controversial research. But even then, maybe that professor should shop around for a university that is more sympathetic. But in primary and high schools, the curriculum is usually predefined. The school should absolutely have the tools to retain and reward the most effective teachers.
Employers (school boards) are not specifically all a bunch of fat cat a-holes walking around looking for people to fire. They will absolutely retain people who do their jobs well. What is the specific conflict of interest inherent to the teaching industry that requires tenure as compared to other industries?
justfacts is pretty good. We would be a much less ignorant people if people just got a reality check from sites like this after watching the evening news.
But it would also be good if schools made curricula such as 'logical reasoning', 'critical thinking', and 'statistical analysis' mandatory for all students. You don't have to be a math genius to understand these things. Perhaps people would be less gullible to politically motivated dialogs. But I'm not sure if schools are intended to produce critical-thinkers...
I wish more people asked questions like: "why do those stats look cherry-picked, what is the selection criteria?", or "how does the evening news choose stories, do they proportionally represent reality?", "What is the best investment strategy that will actually serve me and not Wall Street?", "Is a vote for politician A or B a false dichotomy?"
The whole misdirection to Bitlocker is probably a sarcastic joke pointing to a company far more likely to adhere to NSL's.. Bitlocker isn't even provided on Home editions of Windows, so it really isn't such an obvious alternative. Their directions literally go through steps to change the Windows product key. I would assume to do this legally you pay Microsoft, is that correct? So you are telling me that TrueCrypt as a free alternative for home users isn't still worth developing by someone?
There isn't any way they can give us confidence that they are playing nice either. This is what happens when you violate the trust of the US people and the rest of the world. People used to believe that the US Constitution was the fire-block that was stopping this same nonsense that you would expect from China or other authoritarian governments with no protection of human rights. Now its official, there is no difference.
US corporations have lost major credibility in the world technology market: "We promise this time we won't put secret back doors in our products that we won't tell you about because our government forced us to and we couldn't tell you. We promise this won't happen anymore."
Ah, yes... "If you aren't doing anything wrong, then what do you have to worry about"
Except there are plenty of cases of persecution if you happen to be:
- Gay,
- A former member of the communist party,
- Union organizer,
- Whistle blower,
- Protester, objector, not in line with corporate America,
- Catholic, Jewish, Japanese, or anything else not favorable at the time...
None of these people are terrorists, but clearly lost their liberties, reputation, or assets when they were "outed"
3. ???
4. Profit!!
That sounds about right. I think the corollary to that is: "When a simpler approach is found, at least one person in the group will refuse to give up the needlessly complex bits."
I think people feel attached to XP because it was comfortable and stable (for the first time) for most people. People were accustomed to the Win95 look and feel, so it wasn't a culture shock. And it worked well in corporate domains and at home with a 'grown-up' security model unlike 95/98.
Everything since then has been a forced upgrade that can break existing applications and infrastructure. Most IT departments and individuals avoided Vista as much as possible. Windows 7 is useable. Windows 8 is confusing and radical for many people.
I feel like I've lost something going from XP to Win7, like the search dialog. The folder structure is re-arranged, but still arbitrary. The Registry is still as arbitrary, and the control panel is more confusing. The document library is good in theory, but most people I've helped with Windows 7 just get lost and try to avoid it, but it still gets in your face. So maybe Windows 7 offered as improvements the UAC security model and IPV6 built-in.
Redmond and Silicon Valley needs to wake up that this isn't 1999 anymore. The fact that Windows 7 and 8 came along 3 years apart is too soon. People don't want to upgrade until something actually breaks. And even then, they don't want to (or can't) spend another capital investment on software for the new OS, especially if it is tied to some type of equipment or machinery . Most businesses expect a 5-10 year amortization on basic equipment. There are even tax and accounting consequences to this if the recovery period doesn't meet expectations. Microsoft could probably make money selling XP (and sell support for critical flaws) for quite some time. I'm not sure why they need to set an arbitrary date to close it out. Does it really cost that much to maintain?
This is often helpful. But other times, programs still do their own checks and refuse to run.
Somebody is making model T parts.