Damon, I think you took this too personally. It was not meant to be offensive. I am sorry if it came across that way. Of course when I said "everything is programmed" I was generalizing for compactness and effect. There are exceptions. I have colleagues who are CS PhDs who use techniques such as theorem proving, tools such as Anvil, and who know how to create state diagrams; but the realm that I work in is the business realm, and the level of quality that I see is horrible. No wonder most web sites have vulnerabilities and problems. Most programmers don't design anymore. You might, but then you are in the minority. The "snootiness" that you perceived was actually frustration at the state of my own industry. When I say that "OSs don't work", etc., I am referring to the obvious poor reliability of seemingly everything nowadays - everything that is programmed that is, from routers (how many home routers don't have to be restarted occasionally?) to OSs (blue screen!) to phones (one now sometimes has to reboot a phone!!) and on and on and on. I blame the lack of rigor in the way these programmed systems are designed, in general. Nor do I claim that EEs are superior: they are people and no different; but their profession demands - by its very nature - a focus on concurrency and modeling states, and so they have largely solved those problems, and the lessons of the EE world go unused by most (not all) of the CS practitioner community.
Yes. Apple uses Objective C. But I was including OO programs in the "procedural" bucket, because OO languages use imperative (procedural) coding to implement algorithms. Semantics!
I will also point out though that systems should be simple to operate, hence Apple for example would never think of having more than two positions for an on/off switch: but in order to achieve that, the system has to be engineered to be truly robust. (I am not saying that Apple equipment is.)
It used to be that equipment had well-defined states, but nowadays everything is programmed using procedural code, and nothing works right anymore.
Electrical engineers are trained in how to design things that really work: they assume asynchronous behavior and concurrency from the outset, and they have design methodologies to create a system that has well-defined states. Procedural code has indeterminate states, unless one uses design paradigms that pair those states, and simulation to test the design. Programmers don't use these techniques: generally speaking, procedural code is hacked together, and so we have laptops OSs that freeze, cellphones that lock up, and airplanes that crash.
The software that exists today is by and large all crap. Procedural programming is appropriate for business apps, but for a reliable real-time system you need an asynchronous design methodology, and you need to prove correctness for critical functions. This is not always done, in aerospace and even for spacecraft software.
Today's programmers don't even have a culture any longer that espouses design and design verification, as opposed to hacking together "code". In their purported quest for "clean code" they have culturally inculcated an obsolete and broken approach.
Hi. You misunderstand what I said. I meant that if 80% of the news becomes garbage because NY Times and what-not become indistinguishable from People, people might see the lack of good news and then might be willing to pay for quality news. We might have to go through a painful transition, in which for a time there is no quality news.
I personally would be very willing to pay for quality news. For example, each month I buy Scientific American magazine, because to me it is quality news.
People used to routinely pay for news, by buying the newspaper. If the newspapers go away, and the online sources degenerate into garbage, then I think people will be willing to pay for non-garbage.
I think what is missing is a trustworthy and ubiquitous e-commerce framework to make it convenient and safe to buy news online. People are loathe to sign up for so many sites, each with an account and a password, giving up their credit card number each time. They will do this for something they materially want, but for news they are a little more reluctant because they can do without it. But if there were better e-commerce infrastructure I think that paying for news would be a no-brainer. What do you think?
But AP and Reuters are unique in that they actually do reporting. They are the source of the vast majority of actual news. They then sell that laboriously gathered news to newspapers, radio, and TV stations.
People might be willing to pay for that if 80% of the news eventually becomes unreliable garbage.
It will be interesting to see what happens over the next few years.
AP and Reuters are two of the few actual content providers. They SHOULD charge. After all, they charge newspapers for their content. They have live trained reporters around the world, many of them risking their lives. This has substantial value. They deserve to be paid.
Yes, citizen journalism has its place, but there is no substitute for trained professional reporters.
No, it was your comment,
"Make a real argument and someone will take you seriously - so far, no one seems have done so."
And your response shows just how childish you are.
Yes. Mainframes are important, still. And centralized load generation is important and will continue to be, for base load.
But it cannot continue to remain our only source. Instead of thinking about how to grow it, we need to be thinking about how to replace it wherever possible with a more secure, more decentralized, and more flexible approach. Decentralization is inherently more secure because if you take out one source you don't take down an entire region. This is a core assurance principle, articulated in my book on application security. Decentralization also returns control to you and me with regard to the forms of power that we choose. Today we are beholden to the very powerful energy companies, and they have dictated our Middle East policies. Continuing with that approach could very well bring down civilization.
My dismissal of centralized generation is not a "dismissal": I acknowledge the need for centralized generation. And it is not based on fantasy. It represents the current emerging consensus, in that we will need a mix of solutions, not one: are you perhaps only reading trade journals from your own industry?
With regard to "build me an example supporting 100,000 people", you should have listened to Al Gore's presentation to a Senate panel a few days ago. It was covered by C-SPAN. Al, a recipient of the Nobel Prize, did provide many such examples.
By the way, I am trying to have a dialog here. Please don't insult me with aspersions toward me or my thoughts. I have an open mind: the fact that when I was a nuclear engineer I listened to the anti-nuclear movement and ultimately left that profession as a result proves that I have an open mind. Only juveniles engage in "flaming" and "trolling" behavior. Such behavior fosters division, not consensus-building. Thank you.
I agree with all of your comments. I also feel that coal needs to be off the table from a strategic, long-term perspective, just as nuclear. I do feel though that in that timeframe solar will be viable. I feel it is viable today: it is perceived as less viable because coal and nuclear appear cheaper (when one does not include the systemic damage). But let's consider this: if we did not have coal or nuclear to consider, what would we do? I am sure we would be fine, as there are options, even though they have higher direct costs. We would not be "out in the cold".
Regarding centralized power generation, I think that needs to be part of the picture, but not the whole picture. Base load is optimal if it comes from highly efficient sources, but there is no reason why power cannot be augmented substantially by local sources. Having a local capability to augment base load is simply more optimal. It is also more secure, because it means that the whole power generation system would be more resilient to centralized failures.
Finally, and this is perhaps the most important point of all: these large energy companies rule us today. Just look at the Middle East: we are so interested in that whole region because of energy. We have to become energy independent, not just from other countries, but from these special interests (global corporations). It is these companies that promote these centralized technologies like nuclear power. They are not thinking about our safety: they are thinking about opportunities to build billion dollar power plants. We have to be careful not to re-introduce the nuclear economy that we were creating in the 70s. It took a huge movement to overturn that. Let's not waste that achievement!
Those are the wrong requirements. Those are requirements for a big centralized power generation plant. We need to move away from that approach, just as we have moved away from mainframes.
I completely agree with you on some things. I also think that our biggest risks are not from nuclear energy; but I do think it is a big risk. Private industry has used it in a relatively safe manner, but what I don't want to see is an escalation of that industry. It is dirty, and dangerous. And nuclear fuel is a limited resource, just as fossil fuels are. There is not that much uranium, and so that means that we have to produce plutonium, which is massively toxic, and can much more easily be refined into fissile material to make bombs.
I see the huge cars driving around, and all the unnecessary airline travel - in an age when teleconferencing is feasible - and I just don't think that we need nuclear energy to be prosperous.
I normally would not respond to such an impolite post. You must be poorly raised. But I will respond for the benefit of others.
Also, I am a former nuclear engineer, so be careful about your own claims or counter-arguments.
The issues are complex, but the complexities are really immaterial. In the end, there will be some (marginal) lifecycle cost difference between nuclear and its alternatives. However, one must account for the expected cost of the side effects. For example, a single terrorist nuclear incident would probably result in marshal law within the US, an escalation of our military, and a massive retrenchment of the world economy. Even if this as a 1% chance, the cost is so great that one must account for it. One percent of 100 trillion dollars in cumulative loss over the ensuing 50 years (not even accounting for loss of life) is $1T.
Further, nuclear energy economic models assume centralized power generation. That is a poor approach, an industrial age approach. We need to get away from that and move power generation closer to where power is used. This is starting to happen with IT. If you do that, you find that it is a matter of changing the traditional way that infrastructure is created. If you build solar generation into buildings and homes from the outset, the cost is built-in and not an add-on. You have to build a roof no matter what; if you design the roof to have solar generation embedded in it, the cost is dramatically lower that adding solar on afterwards.
If power generation is moved closer to its use, then transmission is less of an issue. Even so, high voltage transmission is very efficient: power can be sent across the country with only 1% loss if it is sent at 1Mv; so base load can be located in sunny regions.
We also have to remember that nuclear energy requires nuclear infrastructure. That is big business. Exxon is one of the most prominent suppliers of nuclear fuel. We are just cementing ourselves into the old regime of a small number of large energy companies controlling our destiny. And I can tell you that private industry is NOT up to the task of operating dangerous materials. There is too much cost pressure.
I don't feel that nuclear is a choice. It is too dangerous. In my opinion, it needs to be off the table as an option. If it did not exist, we would find a way and we would be fine.
Now Mr. "inviolet", if you respond to this and have any desire to have me respond in turn, be more polite and stick to the issues being discussed instead of insulting me.
There is no silver bullet for energy, and nuclear energy is not that silver bullet either. The question is whether our energy mix needs to include nuclear. I do not believe that it does. Instead of a mix of fossil fuels and nuclear, I would prefer to see a mix of fossil fuels and solar, among others. Even if it costs more in terms of short term direct costs. If you would like me to enumerate the dangers and ugliness of nuclear energy and the infrastructure that supports it, I can do that, as I have worked in that industry.
This is different. Cannons are not nuclear weapons. As a former nuclear engineer, I am well aware of the immense energy in nuclear material. We do NOT want to encourage the establishment of a nuclear infrastructure in commercial hands. It will make us all very unsafe. I am therefore not in favor of the use of nuclear energy for commercial energy production. I am only in favor of it as a possible means of future space propulsion, but only under extremely strict control and oversight. For commercial energy production, we do not NEED nuclear energy. There are safer alternatives. It is a needless risk.
Re:The bells and whistles nobody uses...
on
Less Is Moore
·
· Score: 1
I completely agree. It is a mystery to me how it can be that a spreadsheet program can require hundreds of Mb. Visicalc was obviously less that 640K - probably much less - does anyone know? True, the GUIs take up space, but does anyone recall how much space X-windows added to a typical app back in the minicomputer days? Something is really, really, really dysfunctional about the way that software is being written. I think that it is perhaps the huge, bloated general purpose API libraries. Doesn't the.Net framework in Windows have something like 80,000 methods? And that sits on top of what, the Windows API, which has something like 40,000 functions? And then one links in other libraries from third-party components. In any case, something is wrong somewhere. It is a crisis of increasing complexity of diminishing value. These vendors are not addressing the complexity problem. And they are not the only ones. It seems like in every camp, we are ending up with increasing complexity of diminishing value. Just look at the OASIS Web service standards and the W3C specs. And the plethora of IETF RFCs that preceded those. It is chaos.
Today's average PC or Mac is probably a million times more powerful than a mainframe of the early 70s, and those were used to allow hundreds of people to run autocad-like programs simultaneously. Something is really wrong.
I am a very experienced user of both Word and Open Office Writer. I am also very experienced with Powerpoint and Impress, as well as Open Office Draw, and somewhat familiar with Visio.
I find that OpenOffice is as usable and capable as the MS products. It is definitely ready for prime time, and indeed I prefer the OO programs. However, there are still some compatibility issues. The ones that I have personally encountered are:
1. Word cross references: Cross references created in Word do not transfer well. If your document uses lots of cross references, especially references to headings or numbered items, then expect problems.
2. Powerpoint fonts: Fonts are sometimes changed when going between Powerpoint and Impress.
3. Captions: Word has a figure caption mechanism that does not convert well to OO. MS Word models captions as paragraphs with no page break allowed between the figure and the caption paragraph; OO converts these to a floating block that contains both the figure and the caption. I find that I usually need to re-do captions after converting.
4. Alignment of bullets, and bullet characters: Bullets are often changed when converting, in terms of their positioning and the character used to represent the bullet itself.
There are other issues, but these are the ones that have been the biggest headache for me. I have used these programs to create very large documents (some are 800 pages), so both can handle large and complex projects. For large documents, it is not practical to go back and forth multiple times: expect to convert once and then stay there.
I myself have not had trouble with vector graphics as some here have.
You are right, but the reason is subtle. The key is to require that ISPs cannot prioritize based on sender. All senders (and receivers) should be treated equally. Then, if QOS is not sufficient, the ISP's media customers will all be equally affected. With such a policy, yes, we would ultimately be paying for the QOS because the cost would propagate to the endpoints.
I see the confusion. Their proposal sounds fair on the surface, but consider the ramification: they are trying to establish and end to end performance level based on their relationship with their ISP (the provider that connects the BBC network to the rest of the Internet). I feel that packets should not be prioritized based on who sent them. That is what they are trying to do, even though their purpose if reasonable. It sets a bad precedent. Imagine if Disney did the same thing: then Disney's ISP would start to give higher priority to Disney high QOS packets, OVER that given to other (non-Disney) sites. Very, very bad precedent.
I understand the QOS issue. With a packet network such as the Internet, you cannot guarantee QOS. All you can do is promise to prioritize packets, and provide a certain bandwidth within the network that you control. Out in the cloud, one can try to set up special arrangements, but as you know, nothing is for sure. One can always lose or delay a packet if traffic is heavy.
Thus, there really isn't a technical solution beyond what IP6 provides - which is not a guarantee.
What I am saying is only that I do not want QOS to be managed primarily by ISPs who deal with the deep pockets. I want the ISPs instead to attempt to treat each packet without regard to where it is from, and deal with the QOS service issue by providing enough bandwidth to satisfy their customers, without playing favorites.
At the consumer endpoint, the consumer should have the ability to improve performance by buying more bandwidth, but you are right, that if there is insufficient bandwidth at some point along the way the traffic will be choked. But if that occurs, I want it to occur evenly and fairly to all of the customers of the ISP that is causing the choking. No favorites.
That is the only way that we will ensure that players with big wallets will not hog the Internet and cause response time for other sites (perhaps ones with more open content) to be accessible.
I want to be in charge of the QOS I receive. I disapprove of any model in which the content provider pays the ISP for more QOS. That leads to a Disney and Coca-Cola Internet.
The consumer should be the one to choose (and pay) for QOS. And payment should be to the ISP, not the content provider, which would end up as a kickback to the content provider's ISP.
Only in this way can we hope to ensure that the Internet is not filtered by the content providers with the largest pockets, and by the ISPs themselves.
Slashdot Mirror
Damon, I think you took this too personally. It was not meant to be offensive. I am sorry if it came across that way. Of course when I said "everything is programmed" I was generalizing for compactness and effect. There are exceptions. I have colleagues who are CS PhDs who use techniques such as theorem proving, tools such as Anvil, and who know how to create state diagrams; but the realm that I work in is the business realm, and the level of quality that I see is horrible. No wonder most web sites have vulnerabilities and problems. Most programmers don't design anymore. You might, but then you are in the minority. The "snootiness" that you perceived was actually frustration at the state of my own industry. When I say that "OSs don't work", etc., I am referring to the obvious poor reliability of seemingly everything nowadays - everything that is programmed that is, from routers (how many home routers don't have to be restarted occasionally?) to OSs (blue screen!) to phones (one now sometimes has to reboot a phone!!) and on and on and on. I blame the lack of rigor in the way these programmed systems are designed, in general. Nor do I claim that EEs are superior: they are people and no different; but their profession demands - by its very nature - a focus on concurrency and modeling states, and so they have largely solved those problems, and the lessons of the EE world go unused by most (not all) of the CS practitioner community.
Yes. Apple uses Objective C. But I was including OO programs in the "procedural" bucket, because OO languages use imperative (procedural) coding to implement algorithms. Semantics!
Good points.
I will also point out though that systems should be simple to operate, hence Apple for example would never think of having more than two positions for an on/off switch: but in order to achieve that, the system has to be engineered to be truly robust. (I am not saying that Apple equipment is.)
It used to be that equipment had well-defined states, but nowadays everything is programmed using procedural code, and nothing works right anymore.
Electrical engineers are trained in how to design things that really work: they assume asynchronous behavior and concurrency from the outset, and they have design methodologies to create a system that has well-defined states. Procedural code has indeterminate states, unless one uses design paradigms that pair those states, and simulation to test the design. Programmers don't use these techniques: generally speaking, procedural code is hacked together, and so we have laptops OSs that freeze, cellphones that lock up, and airplanes that crash.
The software that exists today is by and large all crap. Procedural programming is appropriate for business apps, but for a reliable real-time system you need an asynchronous design methodology, and you need to prove correctness for critical functions. This is not always done, in aerospace and even for spacecraft software.
Today's programmers don't even have a culture any longer that espouses design and design verification, as opposed to hacking together "code". In their purported quest for "clean code" they have culturally inculcated an obsolete and broken approach.
Yes, and there is something very nice about a bookshelf full of one's favorite books and magazines....
Hi. You misunderstand what I said. I meant that if 80% of the news becomes garbage because NY Times and what-not become indistinguishable from People, people might see the lack of good news and then might be willing to pay for quality news. We might have to go through a painful transition, in which for a time there is no quality news.
I personally would be very willing to pay for quality news. For example, each month I buy Scientific American magazine, because to me it is quality news.
People used to routinely pay for news, by buying the newspaper. If the newspapers go away, and the online sources degenerate into garbage, then I think people will be willing to pay for non-garbage.
I think what is missing is a trustworthy and ubiquitous e-commerce framework to make it convenient and safe to buy news online. People are loathe to sign up for so many sites, each with an account and a password, giving up their credit card number each time. They will do this for something they materially want, but for news they are a little more reluctant because they can do without it. But if there were better e-commerce infrastructure I think that paying for news would be a no-brainer. What do you think?
Yes, perhaps.
At least for awhile.
But AP and Reuters are unique in that they actually do reporting. They are the source of the vast majority of actual news. They then sell that laboriously gathered news to newspapers, radio, and TV stations.
People might be willing to pay for that if 80% of the news eventually becomes unreliable garbage.
It will be interesting to see what happens over the next few years.
- Cliff
I agree with you.
AP and Reuters are two of the few actual content providers. They SHOULD charge. After all, they charge newspapers for their content. They have live trained reporters around the world, many of them risking their lives. This has substantial value. They deserve to be paid.
Yes, citizen journalism has its place, but there is no substitute for trained professional reporters.
That way I avoid the Flash adds. I can enable it when I need it, such as to use a site that has useful functionality written in Flash.
No, it was your comment, "Make a real argument and someone will take you seriously - so far, no one seems have done so." And your response shows just how childish you are.
Yes. Mainframes are important, still. And centralized load generation is important and will continue to be, for base load.
But it cannot continue to remain our only source. Instead of thinking about how to grow it, we need to be thinking about how to replace it wherever possible with a more secure, more decentralized, and more flexible approach. Decentralization is inherently more secure because if you take out one source you don't take down an entire region. This is a core assurance principle, articulated in my book on application security. Decentralization also returns control to you and me with regard to the forms of power that we choose. Today we are beholden to the very powerful energy companies, and they have dictated our Middle East policies. Continuing with that approach could very well bring down civilization.
My dismissal of centralized generation is not a "dismissal": I acknowledge the need for centralized generation. And it is not based on fantasy. It represents the current emerging consensus, in that we will need a mix of solutions, not one: are you perhaps only reading trade journals from your own industry?
With regard to "build me an example supporting 100,000 people", you should have listened to Al Gore's presentation to a Senate panel a few days ago. It was covered by C-SPAN. Al, a recipient of the Nobel Prize, did provide many such examples.
By the way, I am trying to have a dialog here. Please don't insult me with aspersions toward me or my thoughts. I have an open mind: the fact that when I was a nuclear engineer I listened to the anti-nuclear movement and ultimately left that profession as a result proves that I have an open mind. Only juveniles engage in "flaming" and "trolling" behavior. Such behavior fosters division, not consensus-building. Thank you.
You are a nasty person, making personal attacks. I won't reply to any of the substance of your comment (because there is none).
Yes, but they would have to have nuclear projectiles to shoot. Those are hard to make or obtain. Let's not make it easier!
Hi,
I agree with all of your comments. I also feel that coal needs to be off the table from a strategic, long-term perspective, just as nuclear. I do feel though that in that timeframe solar will be viable. I feel it is viable today: it is perceived as less viable because coal and nuclear appear cheaper (when one does not include the systemic damage). But let's consider this: if we did not have coal or nuclear to consider, what would we do? I am sure we would be fine, as there are options, even though they have higher direct costs. We would not be "out in the cold".
Regarding centralized power generation, I think that needs to be part of the picture, but not the whole picture. Base load is optimal if it comes from highly efficient sources, but there is no reason why power cannot be augmented substantially by local sources. Having a local capability to augment base load is simply more optimal. It is also more secure, because it means that the whole power generation system would be more resilient to centralized failures.
Finally, and this is perhaps the most important point of all: these large energy companies rule us today. Just look at the Middle East: we are so interested in that whole region because of energy. We have to become energy independent, not just from other countries, but from these special interests (global corporations). It is these companies that promote these centralized technologies like nuclear power. They are not thinking about our safety: they are thinking about opportunities to build billion dollar power plants. We have to be careful not to re-introduce the nuclear economy that we were creating in the 70s. It took a huge movement to overturn that. Let's not waste that achievement!
Those are the wrong requirements. Those are requirements for a big centralized power generation plant. We need to move away from that approach, just as we have moved away from mainframes.
An intelligent post. Thank you.
I completely agree with you on some things. I also think that our biggest risks are not from nuclear energy; but I do think it is a big risk. Private industry has used it in a relatively safe manner, but what I don't want to see is an escalation of that industry. It is dirty, and dangerous. And nuclear fuel is a limited resource, just as fossil fuels are. There is not that much uranium, and so that means that we have to produce plutonium, which is massively toxic, and can much more easily be refined into fissile material to make bombs.
I see the huge cars driving around, and all the unnecessary airline travel - in an age when teleconferencing is feasible - and I just don't think that we need nuclear energy to be prosperous.
The military is far more capable of protecting nuclear technology than private industry. We can debate that, but that has been my observation.
What a nasty response.
I normally would not respond to such an impolite post. You must be poorly raised. But I will respond for the benefit of others.
Also, I am a former nuclear engineer, so be careful about your own claims or counter-arguments.
The issues are complex, but the complexities are really immaterial. In the end, there will be some (marginal) lifecycle cost difference between nuclear and its alternatives. However, one must account for the expected cost of the side effects. For example, a single terrorist nuclear incident would probably result in marshal law within the US, an escalation of our military, and a massive retrenchment of the world economy. Even if this as a 1% chance, the cost is so great that one must account for it. One percent of 100 trillion dollars in cumulative loss over the ensuing 50 years (not even accounting for loss of life) is $1T.
Further, nuclear energy economic models assume centralized power generation. That is a poor approach, an industrial age approach. We need to get away from that and move power generation closer to where power is used. This is starting to happen with IT. If you do that, you find that it is a matter of changing the traditional way that infrastructure is created. If you build solar generation into buildings and homes from the outset, the cost is built-in and not an add-on. You have to build a roof no matter what; if you design the roof to have solar generation embedded in it, the cost is dramatically lower that adding solar on afterwards.
If power generation is moved closer to its use, then transmission is less of an issue. Even so, high voltage transmission is very efficient: power can be sent across the country with only 1% loss if it is sent at 1Mv; so base load can be located in sunny regions.
We also have to remember that nuclear energy requires nuclear infrastructure. That is big business. Exxon is one of the most prominent suppliers of nuclear fuel. We are just cementing ourselves into the old regime of a small number of large energy companies controlling our destiny. And I can tell you that private industry is NOT up to the task of operating dangerous materials. There is too much cost pressure.
I don't feel that nuclear is a choice. It is too dangerous. In my opinion, it needs to be off the table as an option. If it did not exist, we would find a way and we would be fine.
Now Mr. "inviolet", if you respond to this and have any desire to have me respond in turn, be more polite and stick to the issues being discussed instead of insulting me.
There is no silver bullet for energy, and nuclear energy is not that silver bullet either. The question is whether our energy mix needs to include nuclear. I do not believe that it does. Instead of a mix of fossil fuels and nuclear, I would prefer to see a mix of fossil fuels and solar, among others. Even if it costs more in terms of short term direct costs. If you would like me to enumerate the dangers and ugliness of nuclear energy and the infrastructure that supports it, I can do that, as I have worked in that industry.
This is different. Cannons are not nuclear weapons. As a former nuclear engineer, I am well aware of the immense energy in nuclear material. We do NOT want to encourage the establishment of a nuclear infrastructure in commercial hands. It will make us all very unsafe. I am therefore not in favor of the use of nuclear energy for commercial energy production. I am only in favor of it as a possible means of future space propulsion, but only under extremely strict control and oversight. For commercial energy production, we do not NEED nuclear energy. There are safer alternatives. It is a needless risk.
Today's average PC or Mac is probably a million times more powerful than a mainframe of the early 70s, and those were used to allow hundreds of people to run autocad-like programs simultaneously. Something is really wrong.
I find that OpenOffice is as usable and capable as the MS products. It is definitely ready for prime time, and indeed I prefer the OO programs. However, there are still some compatibility issues. The ones that I have personally encountered are:
1. Word cross references: Cross references created in Word do not transfer well. If your document uses lots of cross references, especially references to headings or numbered items, then expect problems.
2. Powerpoint fonts: Fonts are sometimes changed when going between Powerpoint and Impress.
3. Captions: Word has a figure caption mechanism that does not convert well to OO. MS Word models captions as paragraphs with no page break allowed between the figure and the caption paragraph; OO converts these to a floating block that contains both the figure and the caption. I find that I usually need to re-do captions after converting.
4. Alignment of bullets, and bullet characters: Bullets are often changed when converting, in terms of their positioning and the character used to represent the bullet itself.
There are other issues, but these are the ones that have been the biggest headache for me. I have used these programs to create very large documents (some are 800 pages), so both can handle large and complex projects. For large documents, it is not practical to go back and forth multiple times: expect to convert once and then stay there.
I myself have not had trouble with vector graphics as some here have.
- Cliff
You are right, but the reason is subtle. The key is to require that ISPs cannot prioritize based on sender. All senders (and receivers) should be treated equally. Then, if QOS is not sufficient, the ISP's media customers will all be equally affected. With such a policy, yes, we would ultimately be paying for the QOS because the cost would propagate to the endpoints.
I see the confusion. Their proposal sounds fair on the surface, but consider the ramification: they are trying to establish and end to end performance level based on their relationship with their ISP (the provider that connects the BBC network to the rest of the Internet). I feel that packets should not be prioritized based on who sent them. That is what they are trying to do, even though their purpose if reasonable. It sets a bad precedent. Imagine if Disney did the same thing: then Disney's ISP would start to give higher priority to Disney high QOS packets, OVER that given to other (non-Disney) sites. Very, very bad precedent.
I understand the QOS issue. With a packet network such as the Internet, you cannot guarantee QOS. All you can do is promise to prioritize packets, and provide a certain bandwidth within the network that you control. Out in the cloud, one can try to set up special arrangements, but as you know, nothing is for sure. One can always lose or delay a packet if traffic is heavy.
Thus, there really isn't a technical solution beyond what IP6 provides - which is not a guarantee.
What I am saying is only that I do not want QOS to be managed primarily by ISPs who deal with the deep pockets. I want the ISPs instead to attempt to treat each packet without regard to where it is from, and deal with the QOS service issue by providing enough bandwidth to satisfy their customers, without playing favorites.
At the consumer endpoint, the consumer should have the ability to improve performance by buying more bandwidth, but you are right, that if there is insufficient bandwidth at some point along the way the traffic will be choked. But if that occurs, I want it to occur evenly and fairly to all of the customers of the ISP that is causing the choking. No favorites.
That is the only way that we will ensure that players with big wallets will not hog the Internet and cause response time for other sites (perhaps ones with more open content) to be accessible.
Good.
I want to be in charge of the QOS I receive. I disapprove of any model in which the content provider pays the ISP for more QOS. That leads to a Disney and Coca-Cola Internet.
The consumer should be the one to choose (and pay) for QOS. And payment should be to the ISP, not the content provider, which would end up as a kickback to the content provider's ISP.
Only in this way can we hope to ensure that the Internet is not filtered by the content providers with the largest pockets, and by the ISPs themselves.