Look at the specifications for modern POSIX compliance and then ask that again.
Can you point me to the POSIX standard concerned with the init system? All I can find is standards concerned with the C library, system calls, how to express command line arguments, file system,...;-)
Besides, Linux never was fully POSIX compliant anyways.
I was now following this discussion thread for some minutes and I have to say, posts like this prove Lennart's point. Apparently many people are not able to argue without personal insult or abusive language:-(
The good thing with Open Source is: if sufficiently many people think there is something wrong, they can fix it themselves. Apparently, there are not sufficiently many people who think anything is wrong with systemd, only a minority that is very active in forums but not willing/capable of changing the situation themselves.
Personally, I liked SysV Init, but I have no problems learning or adapting to systemd. And I do like the fact my laptop and servers boot in a few seconds (even if this is just a minor feature of systemd...);-)
The "bounty" is mostly a marketing instrument, and not so much a reward (just a nice gesture). The rationale is: "our software is so good that we can afford to give out bounties for the few bugs that you will find". A message to the majority of users, not to potential bug hunters.:-)
There is so much myth and misinformation spread in this discussion, and the original post isn't very helpful either. The whole concept is described here and worked out by a scientific foundation called Desertec. They have thought much beyond the common objections found in this forum. Some common ones:
inefficient solar cells: the power plants do not use photovoltaic cells, they are thermal power plants using mirrors to heat up a special heating fluid, very efficient
energy storage over night: this problem is addressed by huge salt tanks that can store heat with minimum losses over longer periods of time
unstable region: this is North Africa, not Middle East. Also, plants will be distributed over several countries, and energy production is part of a larger plan that does not only involve solar power from Africa
single technology: as said, the solar plants are part of a larger concept that also involves many different forms of renewable energy involving a Europe-wide "grid" that connects also with wind power in different parts of Northern and Western Europe and hydro power in the alpine regions of Central Europe
If there is no such thing as radioactive waste, why are we then building very expensive and very complex ways of storing that waste for the next 10.000s of years?
No need to make me aware of other sources of death - dying from Chernobyl radiation in Germany is probably as likely as dying from a terrorist attack.;-) According to other sources (I think it was on Wikipedia), the radiation stemming from Chernobyl for the average European is about as high as the radiation stemming from nuclear bomb tests and as the radiation stemming from regular, non-accident nuclear plants. I am actually regularly buying mushrooms from White Russia, knowing that it has been affected most severely by the radiation after the meltdown. Everything from car accidents over smoking and household accidents is certainly much more likely. Radiation is just a bit more spooky, but there is certainly no reason to panic.
However, this does not change the issue that nuclear energy poses an additional danger to mankind that could be avoided if we instead invested more in other, more environment friendly ways to create energy. I am not in favour of immediately shutting down existing nuclear plants as one of my parent posts suggests ("lights going out..."). I am however much in favour of treating all fossil energy sources as outdated and putting considerable research efforts in renewable energy sources instead of betting on nuclear energy as the solution to our environment issues (this is IMHO more than stupid). Instead, we are still investing twice as much money in nuclear energy research than in renewable energy research, and we are spending many many times as much in trying to stabilise our oil sources. This is what I am criticising.
Going back to the original article: the environment friendly aspect of this approach is that it actually is a good complement for the renewable energy sources that do not provide a reliable stream of energy. At least for the moment. And therefore it is a nice thing to experiment with it.
Even though I am thousands of kilometers away, it is still recommended to not eat mushrooms more than a couple of times a year, and I want a better future for my own children.
Are you sure that recommendation is based on good science? Or is it like the Vaccine scare here in the USA about Thermisol? That has parents not vaccinating their kids even with thermisol free vaccines.
There is still a warning of the German Federal Office for Radiation Protection here and here discouraging the consumption of certain very popular types of mushrooms that still have a CS-137 contamination of over 1000 Becquerel per kilogram more than 20 years after the catastrophe and thousands of kilometers away from the site. A single serving of 200g mushrooms will result in a contamination of 0.01 milisievert, equalling a typical long-distance flight. The maximum allowed exposure in Germany is 1 mS alltogether, including X-ray treatment and natural sources.
Since you are working in resources, your interest in and lobbying for fossil/nuclear fuel and FUD on alternative energy sources seems obvious, doesn't it?;-)
Have you looked at the DESERTEC concept at all? It answers a lot of the issues you are raising with solar energy. True, it is visionary, but it is also backed by several studies and major institutions.
but i know you won't listen to reason, you've been spoon fed this nonsense for years. i'll just wait for your lights to go out.
Where is the "reason" you are offering? I could argue the same "spoon feeding" for your argumentation ("nuclear power is save"). Maybe we Europeans are more careful with such statements, being closer to Chernobyl. Even though I am thousands of kilometers away, it is still recommended to not eat mushrooms more than a couple of times a year, and I want a better future for my own children.
Regarding research spendings I could quickly find this resource, which has a really amazing chart: http://www.solarpowerrocks.com/solar-trends/a-sick-graph-iraq-war-spending-vs-spending-on-renewable-energy/, showing that US research spendings on solar energy are still only half of those on nuclear energy despite the fact that you claim that there is essentially no research on nuclear energy! ; figures are from National Council for Science and the Environment.
I cannot here this boring argument about solar, wind and wave being no alternative repeated again and again. The argument does not get better over time. Had we invested a fraction of the research funding that we have given to nuclear power industries into renewable energy research, we would probably already have most of our energy from renewalbe sources.
Nuclear power is inherently dangerous, we do not know how to deal with the waste, the nuclear fossil fuel will last only a couple of decades, and huge power plants are as inefficient as it gets because of the long distances electricity is transported. By contrast, distributed generation of electricity as proposed by the article is much more efficient, because it happens very close to the consumer.
Solar-based energy is technically possible for Europe even with a 24/7 load. The initiative "Desertec" is following this approach, and there are several studies showing the feasability, financed by the German Federal Ministry for the Environment, Nature Conversation and Nuclear Safety, see http://www.desertec.org/en/concept/studies/ . The reasons why this initiative might still fail are purely political, and for me, this is no excuse.
For the record; I am a researcher working in the Semantic Web area, and I am primary developer of the system IkeWiki and the reasoning language Xcerpt. Since this discussion seems to pop up again and again on Slashdot, I didn't want to add comments to the same issues (trust, search) again. But your comment might add something new to the discussion:
Without knowing the details of your circumstances, it sounds like, maybe, the real point is that what you want is an object oriented database rather than relational one. RDF allows for much more of an object oriented design than a traditional RDBMS does.
In principle, you are right. But there is an important difference between RDF and Object Oriented Databases: while OO DBMS require that the data always conforms to a strict, pre-defined schema, RDF data is semi-structured and can be very flexibly extended. To give an example: in an OO DBMS, it is a problem if a person is defined only by first name and last name, and someone else wants to add a "friend" relationship to this person that is not foreseen in the schema. With RDF, this is not an issue: programs and repositories that were designed just for first name/last name will equally well work in the presence of a "friend" property. In a Web environment, chaotic as it is, this is a crucial property.
Too much attention paid to a silly little thing like accuracy? Give me a fucking break. If Wikipedia wants to be taken seriously as a true encyclopedia, then accuracy is paramount. No, of course it is not going to be perfect, but the cavalier attitude taken towards accuracy by many is nauseating. All of this other stuff about it being a reflection of "how humanity understands itself" has interest, but that doesn't negate or even make less important the accuracy issue.
In many cases - I'd even argue in most - there is no absolute accuracy in the same way as there is no absolute truth. What is considered a fact depends very much on the perception of the observers. Keeping this in mind, a reflection of how humanity understands itself" is actually the most accurate way of viewing things because it takes into account differing perceptions and opinions.
While I agree with most of what you said, I'd like to comment on your following statement:
XML has no semantics whatsoever. None. It's a way of serialising and unserialising a tree of elements and attributes. It's markup languages that are built on top of XML that contain the semantics.
This is not true. XML has no less and no more "semantics" than OWL or RDF. The main difference is that OWL and RDF allow you to build more complicated structures (graphs in RDF, typed graphs in OWL) than XML. The Semantic Web essentially is only about structure, whose symbols are then given "semantics" by the programs processing them and, ultimately, by the humans sitting in front of the machine and looking at the presented content. Therefore I often rather call it the "Symbolic Web". In this respect, I consider XML as a Semantic Web language (and it is also part of the Semantic Web stack).
The really innovative thing about the Semantic Web is a set of standardised languages and standardised reasoning mechanisms for working with the elements of these languages, and that these languages and tools are "Web-capable".
Think less about "This bit of text is a paragraph" and more about "This is the relationship X has with Y".
But what about "This bit of text is the name of the author"? I'd say that this is very well in the scope of the "Semantic Web". It's just a matter of interpreting the structure.
It's trying to impose structure on something that is not very structured--human thought.
True. Unfortunately, "structure" is the only language that machines can work with. If we want the machines to help us, then we need to provide information in a structured way. The current situation (without the "Semantic Web") is in no way different, only that the structure currently only describes the layout and presentation of information. The goal of the Semantic Web is to also be able to provide structure that describes further aspects, like (but not limited to) the content.
One issue we (the Semantic Web community as a whole and I in particular) are working on is to provide tools that make the creation of structure as easy as possible - ultimately through natural language processing, but that does not work too well at the moment.
Why should I have to tag everything I read online? I don't tag things I see in real life.
This is a very valid question, and believe me, it is considered by many researchers. You should, because you get something out of it whose value is higher than the effort that you invested. Why do you mark up text in HTML? I don't do it in "real life", I just put my pen on paper.
If we want computers to be actually useful to us as assistants and not just stupid tools then they will need to begin to operate the same way. That is a very tough problem, yes.
One of the biggest failures in computer science history / AI was exactly that - to try to make computers behave in the same way. Computers are not humans, and they never will be. Computer intelligence is different, and has many strengths but also many weaknesses
compared to human intelligence. Computers are increadibly fast at computing, faster than any human can be, but they are extremely bad with vague information and intuitive judgements. So rather than trying to behave like humans, computers should complement humans, and they actually do.
I have read an article about AI (German) in a German newspaper, and it has a very interesting statement: the problem with AI is that all methods that are developed appear rather simple and structured afterwards, and therefore not "intelligent". For this reason, many people claim that AI is a failure. At the same time, we have gone an increadibly long way in AI research.
And no semantics anywhere (outside of humans) either, just sets of relationships attached to symbols stored in computers.
That's what I actually tried to propagate the "Symbolic Web" rather than the "Semantic Web". But as people are usually looking very confused at me on these occasions I am now (out of resignation) using the "Semantic Web" as everyone else does.;-)
The "Semantic Web" is not about search engines, as you and many other posters seem to believe. It is about representing Web content in a structured, formal way that is more easily accessed by machines, going beyond simple presentation. This can be used for searching, but also for many other applications, e.g. integration, exchange, personalisation,... .
Spam content on the Semantic Web is in no way different to spam content on the normal Web (well, except that it is formal). This also means that a search engine that is capable of working with Semantic Web data has exactly the same issues with trust as traditional search engines. Except that on the Semantic Web, trust can be expressed formally as well. Similar to the authorities in Google, whose outgoing links make a statement about the trustworthiness of other sites, an "authority" on the Semantic Web can make statements about the trustworthiness of other sites. However, these statements are explicit, and they could also be used to state that another site is *not* trustworthy.
Google has the right idea, automatic extraction of semantics from content.
Google does not extract any semantics from content. It merely analyses the linking between websites and connects that with keywords. No semantics here.
Generally, we say what we mean and we don't disguise it in a bunch of niceties or doublespeak.
Interestingly, we Europeans have always had the opposite impression. An example: why is the first thing an American asks if he meets you "how are you" if he expects this question to be answered by "fine" or even some superlative of "fine" instead of the plain truth?
Actually, the fact that American people don't say what they think is one of the points that is mentioned most as a reason for returning from a well-paid job in the US.
Most people actually do not die from the influenza itself, but rather from a superinfection with pneumonia. The reason is that influenza weakens the immune system (if you ever had it, you know how seriously it knocks you out) and hurts the lungs so much that it paves the way for a pneumonia infection. A healthy immune system would deal with this easily, but it is a serious problem for a weakend immune system. For this reason, you will usually also get antibiotics (that of course do not help against the virus) when you get the flu.
Personally, I have been programming with JDK 1.5 for quite some while now and find my productivity greatly improved. Using "generics", the Java container classes are now as powerful (or even more so) as C++. The extended for-loop is now as easy to use as foreach in many scripting languages (Python etc.), but much faster and strongly typed.
Indeed, the advent of nuclear weapons gave us a long period of relative peace. But we must not forget that in the last 50 years, the nuclear powers have also been ruled by mostly reasonable rulers that knew exactly what would happen in case of a nuclear war, and were not willing to pay that price. Also, this period was extraordinary in the fact that we had two super powers controlling most parts of the world and holding each other at balance. In this time, the world was much simpler than it is now.
It is, however, unlikely that this period continues. In fact, it has already ended, with only one super power left, with instable nations like Pakistan, or fundamentalists like Iran, gaining access to nuclear weapons, and with islamic fanatics bombing western cities disregarding the cost. Indeed, it is IMHO completely unlikely that nuclear weapons will *not* be used in wars in the next 100 or even 50 years, or that they will serve any deterrent in the future.
The best solution would be to get rid of all nuclear weapons immediately. Unfortunately, this goal will not be achieved in the near future, if it is achievable at all.
The concept of drag and drop is not mathematical. It is a concept. You are patenting the concept,
That's exactly what mathematics is about. Concepts. It is the science of concepts and abstraction, purely mental things (that, by coincidence, can be run on hardware). In this case, the highlevel concept of "drag and drop" is transformed by the programmer to the lowlevel concepts in a certain programming language. A purely mathematical process.
I know that CS research has a mathemetical basis, but this has NOTHING TO DO WITH SOFTWARE DEVELOPMENT.
So, what do you think is the topic of CS research then? I have tried to follow the thread, and forgive me, but you don't have a clue.
I happen to have designed a programming/query language for XML (and got a PhD for it - google for Xcerpt if you are interested), and believe me, this is indeed purely mathematics. As Edsger Dijkstra said: "Computer science is no more about computers than astronomy is about telescopes."
You give no arguments for that. Switzerland is surrounded by friends, after all.
They very rarely take more than a moment [...]
If you are Swiss, maybe. If you are a EU citizen, no. I'd say that if the border control takes more time than passing through the country (as in my case), you can no longer talk about "a moment".
As for Banking Secrecy being the basis of the Swiss, do you ever read GDP analyses; you would do better with chocolate and pharmaceuticals
If you read my post again, you'll see that I didn't say that (it was in a different post). Still, the Swiss economy largely depends on the EU countries.
How many of those who are expected to vote on that constitution have read it?
Definately not enough. One of the problems with the EU is a lack of "information dissemination", at least in the classic media. You'll find everything on the Internet, but the average citizen won't look.
given the way things are going, I think there are many current EU members, such as the UK and Denmark that are wondering how they can get out of it as fast as possible.
I don't think so. Even the "separatist" UK recognizes that they cannot exist outside the EU.
Maybe not the people - they are influenced by the yellow press - but the politicians, the economy, and research.
I will make very sure, by the democratic means of public initiative, aka privately initiated referendums, which we have in Switzerland, that Switzerland will never join that bastion of corruption.
Not joining the EU will life much harder for the Swiss, at least in the long term. For example, even now, the borders of member countries to Switzerland are already considered as "external border", similar to the borders to Belarus, Russia, and Ukraine.
Also, Switzerland is already part of that "bastion of corruption" in many areas (like funding research, taxes, etc), albeit not formally a member of the EU (and thus with less rights). If Switzerland would not cooperate with the EU in large parts, it would even now have serious economical problems as much of the foreign trade of Switzerland is with the EU.
Regarding software patents, if the EU decides to have them (hopefully not), then Switzerland will be pretty fast to adopt this directive.
Personally, even although there are still so many deficits in the European Union, I am glad to live inside it and not outside it, I am glad to have the Euro, and I am glad that the 200 years period of war seems to be over in Europe. I am also glad that although being German and Swedish, I have no problems whatsoever to take a position and live in France (or elsewhere).
And I tell you something, I once in my life tried to pass through Switzerland (I was invited to a position in Milan and live in Southern Germany, and the route through Switzerland saved me some 50km), and I'll never do it again until Switzerland joins the EU. The border control in Switzerland was so close-minded (closed-mindedness appears to be a property of many Swiss) that they simply would not believe me that I actually came from Germany to apply for a position in Milan. It would escape their minds that someone in Europe could actually do something like that. Instead, they thought I was smuggling drugs into Switzerland. They essentially disassembled my car, it took more than 60 minutes. The 50km saved would have taken 30 minutes had I passed through Austria with no border controls instead.
P.S. the European Parliament (and thus the democratic process) will be strengthend significantly when the new EU constitution is finally ratified by the member countries. Let's only hope that the referendums held in many countries are not used as a vote "over the European Union" but instead as a vote "over more democracy".
You can find everything at this introduction by ESA. Although made for children, it nonetheless explains the most important features of the mission, and you surely don't need a PhD to understand it.:-)
Interestingly, the probe passed close to the Sun and twice close to Venus to use their gravity for acceleration. The kind of precision they use for these calculations is truly fascinating - I mean you have to know the gravity and "course" of Venus pretty exactly to send a probe around Venus for acceleration!
Slashdot Mirror
Look at the specifications for modern POSIX compliance and then ask that again.
Can you point me to the POSIX standard concerned with the init system? All I can find is standards concerned with the C library, system calls, how to express command line arguments, file system, ... ;-)
Besides, Linux never was fully POSIX compliant anyways.
I was now following this discussion thread for some minutes and I have to say, posts like this prove Lennart's point. Apparently many people are not able to argue without personal insult or abusive language :-(
The good thing with Open Source is: if sufficiently many people think there is something wrong, they can fix it themselves. Apparently, there are not sufficiently many people who think anything is wrong with systemd, only a minority that is very active in forums but not willing/capable of changing the situation themselves.
Personally, I liked SysV Init, but I have no problems learning or adapting to systemd. And I do like the fact my laptop and servers boot in a few seconds (even if this is just a minor feature of systemd...) ;-)
The "bounty" is mostly a marketing instrument, and not so much a reward (just a nice gesture). The rationale is: "our software is so good that we can afford to give out bounties for the few bugs that you will find". A message to the majority of users, not to potential bug hunters. :-)
Lots of questions are also answered at Desertec's FAQ
Greetings, Sebastian
If there is no such thing as radioactive waste, why are we then building very expensive and very complex ways of storing that waste for the next 10.000s of years?
However, this does not change the issue that nuclear energy poses an additional danger to mankind that could be avoided if we instead invested more in other, more environment friendly ways to create energy. I am not in favour of immediately shutting down existing nuclear plants as one of my parent posts suggests ("lights going out ..."). I am however much in favour of treating all fossil energy sources as outdated and putting considerable research efforts in renewable energy sources instead of betting on nuclear energy as the solution to our environment issues (this is IMHO more than stupid). Instead, we are still investing twice as much money in nuclear energy research than in renewable energy research, and we are spending many many times as much in trying to stabilise our oil sources. This is what I am criticising.
Going back to the original article: the environment friendly aspect of this approach is that it actually is a good complement for the renewable energy sources that do not provide a reliable stream of energy. At least for the moment. And therefore it is a nice thing to experiment with it.
Even though I am thousands of kilometers away, it is still recommended to not eat mushrooms more than a couple of times a year, and I want a better future for my own children.
Are you sure that recommendation is based on good science? Or is it like the Vaccine scare here in the USA about Thermisol? That has parents not vaccinating their kids even with thermisol free vaccines.
There is still a warning of the German Federal Office for Radiation Protection here and here discouraging the consumption of certain very popular types of mushrooms that still have a CS-137 contamination of over 1000 Becquerel per kilogram more than 20 years after the catastrophe and thousands of kilometers away from the site. A single serving of 200g mushrooms will result in a contamination of 0.01 milisievert, equalling a typical long-distance flight. The maximum allowed exposure in Germany is 1 mS alltogether, including X-ray treatment and natural sources.
Have you looked at the DESERTEC concept at all? It answers a lot of the issues you are raising with solar energy. True, it is visionary, but it is also backed by several studies and major institutions.
but i know you won't listen to reason, you've been spoon fed this nonsense for years. i'll just wait for your lights to go out.
Where is the "reason" you are offering? I could argue the same "spoon feeding" for your argumentation ("nuclear power is save"). Maybe we Europeans are more careful with such statements, being closer to Chernobyl. Even though I am thousands of kilometers away, it is still recommended to not eat mushrooms more than a couple of times a year, and I want a better future for my own children.
Regarding research spendings I could quickly find this resource, which has a really amazing chart: http://www.solarpowerrocks.com/solar-trends/a-sick-graph-iraq-war-spending-vs-spending-on-renewable-energy/, showing that US research spendings on solar energy are still only half of those on nuclear energy despite the fact that you claim that there is essentially no research on nuclear energy! ; figures are from National Council for Science and the Environment.
Nuclear power is inherently dangerous, we do not know how to deal with the waste, the nuclear fossil fuel will last only a couple of decades, and huge power plants are as inefficient as it gets because of the long distances electricity is transported. By contrast, distributed generation of electricity as proposed by the article is much more efficient, because it happens very close to the consumer.
Solar-based energy is technically possible for Europe even with a 24/7 load. The initiative "Desertec" is following this approach, and there are several studies showing the feasability, financed by the German Federal Ministry for the Environment, Nature Conversation and Nuclear Safety, see http://www.desertec.org/en/concept/studies/ . The reasons why this initiative might still fail are purely political, and for me, this is no excuse.
For the record; I am a researcher working in the Semantic Web area, and I am primary developer of the system IkeWiki and the reasoning language Xcerpt. Since this discussion seems to pop up again and again on Slashdot, I didn't want to add comments to the same issues (trust, search) again. But your comment might add something new to the discussion:
Without knowing the details of your circumstances, it sounds like, maybe, the real point is that what you want is an object oriented database rather than relational one. RDF allows for much more of an object oriented design than a traditional RDBMS does.
In principle, you are right. But there is an important difference between RDF and Object Oriented Databases: while OO DBMS require that the data always conforms to a strict, pre-defined schema, RDF data is semi-structured and can be very flexibly extended. To give an example: in an OO DBMS, it is a problem if a person is defined only by first name and last name, and someone else wants to add a "friend" relationship to this person that is not foreseen in the schema. With RDF, this is not an issue: programs and repositories that were designed just for first name/last name will equally well work in the presence of a "friend" property. In a Web environment, chaotic as it is, this is a crucial property.
Greetings, Sebastian
Too much attention paid to a silly little thing like accuracy? Give me a fucking break. If Wikipedia wants to be taken seriously as a true encyclopedia, then accuracy is paramount. No, of course it is not going to be perfect, but the cavalier attitude taken towards accuracy by many is nauseating. All of this other stuff about it being a reflection of "how humanity understands itself" has interest, but that doesn't negate or even make less important the accuracy issue.
In many cases - I'd even argue in most - there is no absolute accuracy in the same way as there is no absolute truth. What is considered a fact depends very much on the perception of the observers. Keeping this in mind, a reflection of how humanity understands itself" is actually the most accurate way of viewing things because it takes into account differing perceptions and opinions.
Sebastian
XML has no semantics whatsoever. None. It's a way of serialising and unserialising a tree of elements and attributes. It's markup languages that are built on top of XML that contain the semantics.
This is not true. XML has no less and no more "semantics" than OWL or RDF. The main difference is that OWL and RDF allow you to build more complicated structures (graphs in RDF, typed graphs in OWL) than XML. The Semantic Web essentially is only about structure, whose symbols are then given "semantics" by the programs processing them and, ultimately, by the humans sitting in front of the machine and looking at the presented content. Therefore I often rather call it the "Symbolic Web". In this respect, I consider XML as a Semantic Web language (and it is also part of the Semantic Web stack).
The really innovative thing about the Semantic Web is a set of standardised languages and standardised reasoning mechanisms for working with the elements of these languages, and that these languages and tools are "Web-capable".
Think less about "This bit of text is a paragraph" and more about "This is the relationship X has with Y".
But what about "This bit of text is the name of the author"? I'd say that this is very well in the scope of the "Semantic Web". It's just a matter of interpreting the structure.
Sebastian
It's trying to impose structure on something that is not very structured--human thought.
True. Unfortunately, "structure" is the only language that machines can work with. If we want the machines to help us, then we need to provide information in a structured way. The current situation (without the "Semantic Web") is in no way different, only that the structure currently only describes the layout and presentation of information. The goal of the Semantic Web is to also be able to provide structure that describes further aspects, like (but not limited to) the content.
One issue we (the Semantic Web community as a whole and I in particular) are working on is to provide tools that make the creation of structure as easy as possible - ultimately through natural language processing, but that does not work too well at the moment.
Why should I have to tag everything I read online? I don't tag things I see in real life.
This is a very valid question, and believe me, it is considered by many researchers. You should, because you get something out of it whose value is higher than the effort that you invested. Why do you mark up text in HTML? I don't do it in "real life", I just put my pen on paper.
If we want computers to be actually useful to us as assistants and not just stupid tools then they will need to begin to operate the same way. That is a very tough problem, yes.
One of the biggest failures in computer science history / AI was exactly that - to try to make computers behave in the same way. Computers are not humans, and they never will be. Computer intelligence is different, and has many strengths but also many weaknesses compared to human intelligence. Computers are increadibly fast at computing, faster than any human can be, but they are extremely bad with vague information and intuitive judgements. So rather than trying to behave like humans, computers should complement humans, and they actually do.
I have read an article about AI (German) in a German newspaper, and it has a very interesting statement: the problem with AI is that all methods that are developed appear rather simple and structured afterwards, and therefore not "intelligent". For this reason, many people claim that AI is a failure. At the same time, we have gone an increadibly long way in AI research.
Sebastian
And no semantics anywhere (outside of humans) either, just sets of relationships attached to symbols stored in computers.
That's what I actually tried to propagate the "Symbolic Web" rather than the "Semantic Web". But as people are usually looking very confused at me on these occasions I am now (out of resignation) using the "Semantic Web" as everyone else does.;-)
SebastianThe "Semantic Web" is not about search engines, as you and many other posters seem to believe. It is about representing Web content in a structured, formal way that is more easily accessed by machines, going beyond simple presentation. This can be used for searching, but also for many other applications, e.g. integration, exchange, personalisation, ... .
Spam content on the Semantic Web is in no way different to spam content on the normal Web (well, except that it is formal). This also means that a search engine that is capable of working with Semantic Web data has exactly the same issues with trust as traditional search engines. Except that on the Semantic Web, trust can be expressed formally as well. Similar to the authorities in Google, whose outgoing links make a statement about the trustworthiness of other sites, an "authority" on the Semantic Web can make statements about the trustworthiness of other sites. However, these statements are explicit, and they could also be used to state that another site is *not* trustworthy.
Google has the right idea, automatic extraction of semantics from content.
Google does not extract any semantics from content. It merely analyses the linking between websites and connects that with keywords. No semantics here.
Sebastian
Generally, we say what we mean and we don't disguise it in a bunch of niceties or doublespeak.
Interestingly, we Europeans have always had the opposite impression. An example: why is the first thing an American asks if he meets you "how are you" if he expects this question to be answered by "fine" or even some superlative of "fine" instead of the plain truth?
Actually, the fact that American people don't say what they think is one of the points that is mentioned most as a reason for returning from a well-paid job in the US.
SebastianSebastian
Personally, I have been programming with JDK 1.5 for quite some while now and find my productivity greatly improved. Using "generics", the Java container classes are now as powerful (or even more so) as C++. The extended for-loop is now as easy to use as foreach in many scripting languages (Python etc.), but much faster and strongly typed.
Sebastian
It is, however, unlikely that this period continues. In fact, it has already ended, with only one super power left, with instable nations like Pakistan, or fundamentalists like Iran, gaining access to nuclear weapons, and with islamic fanatics bombing western cities disregarding the cost. Indeed, it is IMHO completely unlikely that nuclear weapons will *not* be used in wars in the next 100 or even 50 years, or that they will serve any deterrent in the future.
The best solution would be to get rid of all nuclear weapons immediately. Unfortunately, this goal will not be achieved in the near future, if it is achievable at all.
Sebastian
The concept of drag and drop is not mathematical. It is a concept. You are patenting the concept,
That's exactly what mathematics is about. Concepts. It is the science of concepts and abstraction, purely mental things (that, by coincidence, can be run on hardware). In this case, the highlevel concept of "drag and drop" is transformed by the programmer to the lowlevel concepts in a certain programming language. A purely mathematical process.
I know that CS research has a mathemetical basis, but this has NOTHING TO DO WITH SOFTWARE DEVELOPMENT.
So, what do you think is the topic of CS research then? I have tried to follow the thread, and forgive me, but you don't have a clue.
I happen to have designed a programming/query language for XML (and got a PhD for it - google for Xcerpt if you are interested), and believe me, this is indeed purely mathematics. As Edsger Dijkstra said: "Computer science is no more about computers than astronomy is about telescopes."
Sebastian
Sebastian
You give no arguments for that. Switzerland is surrounded by friends, after all.
They very rarely take more than a moment [...]
If you are Swiss, maybe. If you are a EU citizen, no. I'd say that if the border control takes more time than passing through the country (as in my case), you can no longer talk about "a moment".
As for Banking Secrecy being the basis of the Swiss, do you ever read GDP analyses; you would do better with chocolate and pharmaceuticals
If you read my post again, you'll see that I didn't say that (it was in a different post). Still, the Swiss economy largely depends on the EU countries.
Sebastian
Definately not enough. One of the problems with the EU is a lack of "information dissemination", at least in the classic media. You'll find everything on the Internet, but the average citizen won't look.
Sebastian
I don't think so. Even the "separatist" UK recognizes that they cannot exist outside the EU. Maybe not the people - they are influenced by the yellow press - but the politicians, the economy, and research.
I will make very sure, by the democratic means of public initiative, aka privately initiated referendums, which we have in Switzerland, that Switzerland will never join that bastion of corruption.
Not joining the EU will life much harder for the Swiss, at least in the long term. For example, even now, the borders of member countries to Switzerland are already considered as "external border", similar to the borders to Belarus, Russia, and Ukraine.
Also, Switzerland is already part of that "bastion of corruption" in many areas (like funding research, taxes, etc), albeit not formally a member of the EU (and thus with less rights). If Switzerland would not cooperate with the EU in large parts, it would even now have serious economical problems as much of the foreign trade of Switzerland is with the EU.
Regarding software patents, if the EU decides to have them (hopefully not), then Switzerland will be pretty fast to adopt this directive.
Personally, even although there are still so many deficits in the European Union, I am glad to live inside it and not outside it, I am glad to have the Euro, and I am glad that the 200 years period of war seems to be over in Europe. I am also glad that although being German and Swedish, I have no problems whatsoever to take a position and live in France (or elsewhere).
And I tell you something, I once in my life tried to pass through Switzerland (I was invited to a position in Milan and live in Southern Germany, and the route through Switzerland saved me some 50km), and I'll never do it again until Switzerland joins the EU. The border control in Switzerland was so close-minded (closed-mindedness appears to be a property of many Swiss) that they simply would not believe me that I actually came from Germany to apply for a position in Milan. It would escape their minds that someone in Europe could actually do something like that. Instead, they thought I was smuggling drugs into Switzerland. They essentially disassembled my car, it took more than 60 minutes. The 50km saved would have taken 30 minutes had I passed through Austria with no border controls instead.
P.S. the European Parliament (and thus the democratic process) will be strengthend significantly when the new EU constitution is finally ratified by the member countries. Let's only hope that the referendums held in many countries are not used as a vote "over the European Union" but instead as a vote "over more democracy".
Sebastian
Interestingly, the probe passed close to the Sun and twice close to Venus to use their gravity for acceleration. The kind of precision they use for these calculations is truly fascinating - I mean you have to know the gravity and "course" of Venus pretty exactly to send a probe around Venus for acceleration!
Sebastian