The signal is extremely weak (the output was only something like 10 watt, similar to a mobile phone). As I understood it, they will only be able to reconstruct the carrier signal, and not be able to retrieve the actual data sent on that carrier signal. They are going to use frequency variations in the carrier signal to reconstruct the path of the probe. (These frequency variations are refered to as the Doppler effect.) The huygens probe at a set of detectors (some of them also using doppler effects) to measure its trajectory through the athmosphere. I am not sure whether the Doppler effect in the carrier signal from huygens to cassini was measured.
Sorry, to say, but the article referenced does not give any new information. The data is lost and an attempt is being made to reconstruct the path of the huygens probe using the doppler shift of the signal picked up by several radio telescopes on earth. This require huge computational efforts which can take several months to be completed.
Apart from the question if the climate change is real, people all over the world will just continue to burn fossile fuels as long as they are the cheapest form of energy available to them. And it seems that this will still be the fact for a very long time. History has learned that it is natural to regard the short-time effects (food and comfort) more important then the long-term effects (climate change in 50 years from now).
First of all space is not a vacum, although from our point of view (being in a dense atmosphere) it is often considered as such. Secondly, at the time the waves occured, space was much smaller and thus also denser and hotter. What they have observed is just the traces of those waves, not the waves themselves. It like those sand patterns on the beach that you see when the tide is low.
It's easier to parse than virtually anything else whatsoever. Why? Because 99% of programmers don't parse it. They use a pre-existing XML parser.
Yes, that is true. But those parser are complex, and each increasement in complexity also has an impact on parsing speed.
XML Schema and NG have already solved this.
The fact is that XML Schema is hardly used by any one. It could have been in the XML standard, but because XML comes from a domain where people work with textual documents, nobody thought about the fact that you might want to have a proper representation for other datatypes, such as integers, times, dates. With little effort it could have been put in XML. Please remember that this topic is about design flaws. Things that could have been done correctly in the first place, but where not done properly.
Why is this a problem?
There is an essential difference between sets and lists. One implies an ordering, the other not. XML is claiming to be a self explainatory representation of data. Because XML does not make this distinction, it is a flaw.
ID does this
There can be more than one identifying property. Another flaw of XML is that it does that you often have the choice between representing data as contents or as attributes. There is a logical distinction between contents and attributes, but the problem is that the kind of values that you can represent in attributes are rather limited. This means that you have to revert to using additional tags for representing "complex" attributes. BTW, I never knew that the ID attribute had some special meaning. And I can also not know of any XML parser that do anything special with this.
IDREF and IDREFS does this.
Again this is not commonly used. And again it is a after the fact fix of a design flaw. It is almost always a design flaw when certain "names" are given a special meaning and semantic.
No. Although I have to admit that the syntax of LISP is much easier to parse than XML and that it is basically the same. LISP is also a little more compact than XML.
But using LISP instead of XML syntax would not solve any of the mentioned problems.
Why not try to build cities on the bottom of the Oceans? I guess people like this really have no idea of the amouth of resouces that would be needed to colonize Mars. It would be much cheaper to colonize the deep seas. But nobody thinks about doing this. If we are not even able to preserve the climate on Earth how are we going to think we can colonize Mars.
Consider to use a parser that is easier and more powerful than yacc and lex. Have a look at
IParse, a simple, small interpretting parser. The whole source is in a single 92 Kbyte file.
Forget about using a Virtual Machine. That is nice for speed, but it requires a lot of work. Beter make an interpretter that interprets the abstract program tree. I once started doing this for a JavaScript interpretter, but I never came to finish it.
"God does not exist" is also a believe and not a fact.
If you really think about it, many things that are stated as "fact", are not things you can be absolutely sure about. Just think about "The Matrix". There is not much we can be sure of, not even that we will die. It basically comes down to that we are aware that something exists.
Faith and science (including the idea of our world around us) move in completly different dimensions and they cannot be compared.
This is exactly what falls under "political reasons". Microsoft was using its position in the market to push forward a standard that would strengthen their position in the market. That is what every company would do, even small ones for the matter. As soon as companies switch to "open standards", it means that their customers can easily switch to an alternative. As a company, you always want to bound your customers to you. Using a propertary file format or standard is a very effective way of doing this.
Yes, there are indeed too many standards. On reason is that standards are used for political reasons (in the sense of gaining influence), both by companies and government organisations.
The sad thing about standards is that often the less good standards win. We saw this with the video tape format. And we see it with XML (which, believe it or not, is really not that good at all).
The most effective techniques for finding defects is still code review. It seems that one of the pair programming is a very good way of doing code review.
However the greatest problem with writing good software is still in the marketing. In order to sell/license software it needs to have features, and the lack of defect often does not count as "features".
Okay that makes some sense. If I understand it correctly, it means that in between the bolts and the ring were replaced by another ring for another satelite, and that afterwards these people removed that ring and replaced it by the original ring, but did not put on the bolts again.
Yet, again, I could raise the question why they did not use one TOC per satelite, and keep it mounted on it all the time? What's the cost of a TOC compared to the satelite? I also wonder how much time is wasted with putting the satelites on/off the TOC's. From the document it seems that 8-10 people are needed and that these procedures take half a day. Swapping satelites ten times would cost about 3 men months of work.
I would not consider it data-driven, as it is being advertized. Simply retrieving data from a server is not data-driven. A true data-driven rich client would at least implement a kind of transaction mechanism that would allow you to edit the data.
I studied the report, but nowhere I found any mention of who took out the 24 bolts. I figure this is the typical example of trying to save some bucks by having one set of high grade bolts being used for two TOC units. That would be a typical management type of decision that I have seen made so often, trying to save on small things not realizing that they could cause large costs elsewhere. But the funny thing is that those type of decisions never end up in this type of reports, because they reveal the deeper cause of the problem being a bad "management" attitude, due to lack of technical insight.
SQL is not a specification language, it is a query and data manipulation language, and does not have any transactional properties. It is not what I was thinking about. Most people don't realize that relations databases and object-oriented models are just "implementations" of more general data structures.
Every so often a new language is proposed as the solution for the problem of developing applications. However, most of these languages are really generic programming languages. What we really need is a platform that centers around persistant data and views, with built in transaction mechanisms. Instead of a language to program in, we need a language to specify data centric applications.
He suffered from an aggressive streptococ/stafylococ bacterial infection and had to undergo surgery more than once to remove damaged tissue from his feet. He has had a fever for about a week. And although people around him are saying that he seems to be very alert and in good mental health, others have expressed their doubts whether he will be able to resume his normal (very busy) working schedule in the coming weeks, and that he might require some months to recover from it completely.
Most interesting is the fact that the Dutch parlement is trying to force Minister Brinkhorst to change his vote and that he continues to refuse to do so. The only reason he seems to be able to get away with this is that it is not a political issue, because the Dutch media not understanding software patents is not giving it any attention. The infection in the feet of our prime minister is far more interesting. (The latest rumours are that it was a rather serious infection, which might have killed him.)
If we would give all power to those that are in control of the money, democracy and freedom of speech would be gone soon. Governments are there for promoting the well being of all, and in some cases this means controling the power that the rich (people and companies) have. They also should prevent corruption (the misuse of money for the purpose of executing power). And BTW, there are no software patents yet in Europe yet.
Interesting how he describes how the introduction of each new medium (like painting, drama, novels) produced it own geniuses that were never surpassed by anyone that came later. I think this is also true for computers, but I guess that we are already behind the stage of those geniuses with respect to computers. So the time of the "Great Hackers" is already behind us, especially if you think that there are milions trying to write computer programs for their daily living. I have written programs for more than a quarter of a century, and I can tell you, that I haven't seen any new and great ideas in the past fifteen years. (And this language "arc", some syntactical sugar on top of LISP, is really BS. They syntax of LISP does not need any improvement. Twenty five years ago, I could count paratheses, simply by reading LISP. I never have understood what was the point.)
When HTTP was designed, it was intended to become something like WebDAV, because it does have a PUT method. For some reason most HTTP servers never implemented.
Slashdot Mirror
The signal is extremely weak (the output was only something like 10 watt, similar to a mobile phone). As I understood it, they will only be able to reconstruct the carrier signal, and not be able to retrieve the actual data sent on that carrier signal. They are going to use frequency variations in the carrier signal to reconstruct the path of the probe. (These frequency variations are refered to as the Doppler effect.) The huygens probe at a set of detectors (some of them also using doppler effects) to measure its trajectory through the athmosphere. I am not sure whether the Doppler effect in the carrier signal from huygens to cassini was measured.
Sorry, to say, but the article referenced does not give any new information. The data is lost and an attempt is being made to reconstruct the path of the huygens probe using the doppler shift of the signal picked up by several radio telescopes on earth. This require huge computational efforts which can take several months to be completed.
Apart from the question if the climate change is real, people all over the world will just continue to burn fossile fuels as long as they are the cheapest form of energy available to them. And it seems that this will still be the fact for a very long time. History has learned that it is natural to regard the short-time effects (food and comfort) more important then the long-term effects (climate change in 50 years from now).
First of all space is not a vacum, although from our point of view (being in a dense atmosphere) it is often considered as such. Secondly, at the time the waves occured, space was much smaller and thus also denser and hotter. What they have observed is just the traces of those waves, not the waves themselves. It like those sand patterns on the beach that you see when the tide is low.
Yes, that is true. But those parser are complex, and each increasement in complexity also has an impact on parsing speed.
XML Schema and NG have already solved this.
The fact is that XML Schema is hardly used by any one. It could have been in the XML standard, but because XML comes from a domain where people work with textual documents, nobody thought about the fact that you might want to have a proper representation for other datatypes, such as integers, times, dates. With little effort it could have been put in XML. Please remember that this topic is about design flaws. Things that could have been done correctly in the first place, but where not done properly.
Why is this a problem?
There is an essential difference between sets and lists. One implies an ordering, the other not. XML is claiming to be a self explainatory representation of data. Because XML does not make this distinction, it is a flaw.
ID does this
There can be more than one identifying property. Another flaw of XML is that it does that you often have the choice between representing data as contents or as attributes. There is a logical distinction between contents and attributes, but the problem is that the kind of values that you can represent in attributes are rather limited. This means that you have to revert to using additional tags for representing "complex" attributes. BTW, I never knew that the ID attribute had some special meaning. And I can also not know of any XML parser that do anything special with this.
IDREF and IDREFS does this.
Again this is not commonly used. And again it is a after the fact fix of a design flaw. It is almost always a design flaw when certain "names" are given a special meaning and semantic.
But using LISP instead of XML syntax would not solve any of the mentioned problems.
Why not try to build cities on the bottom of the Oceans? I guess people like this really have no idea of the amouth of resouces that would be needed to colonize Mars. It would be much cheaper to colonize the deep seas. But nobody thinks about doing this. If we are not even able to preserve the climate on Earth how are we going to think we can colonize Mars.
Forget about using a Virtual Machine. That is nice for speed, but it requires a lot of work. Beter make an interpretter that interprets the abstract program tree. I once started doing this for a JavaScript interpretter, but I never came to finish it.
If you really think about it, many things that are stated as "fact", are not things you can be absolutely sure about. Just think about "The Matrix". There is not much we can be sure of, not even that we will die. It basically comes down to that we are aware that something exists.
Faith and science (including the idea of our world around us) move in completly different dimensions and they cannot be compared.
I thought that quote in the sig was from Yoda not from Spock.
This is exactly what falls under "political reasons". Microsoft was using its position in the market to push forward a standard that would strengthen their position in the market. That is what every company would do, even small ones for the matter. As soon as companies switch to "open standards", it means that their customers can easily switch to an alternative. As a company, you always want to bound your customers to you. Using a propertary file format or standard is a very effective way of doing this.
The sad thing about standards is that often the less good standards win. We saw this with the video tape format. And we see it with XML (which, believe it or not, is really not that good at all).
However the greatest problem with writing good software is still in the marketing. In order to sell/license software it needs to have features, and the lack of defect often does not count as "features".
Yet, again, I could raise the question why they did not use one TOC per satelite, and keep it mounted on it all the time? What's the cost of a TOC compared to the satelite? I also wonder how much time is wasted with putting the satelites on/off the TOC's. From the document it seems that 8-10 people are needed and that these procedures take half a day. Swapping satelites ten times would cost about 3 men months of work.
I would not consider it data-driven, as it is being advertized. Simply retrieving data from a server is not data-driven. A true data-driven rich client would at least implement a kind of transaction mechanism that would allow you to edit the data.
I studied the report, but nowhere I found any mention of who took out the 24 bolts. I figure this is the typical example of trying to save some bucks by having one set of high grade bolts being used for two TOC units. That would be a typical management type of decision that I have seen made so often, trying to save on small things not realizing that they could cause large costs elsewhere. But the funny thing is that those type of decisions never end up in this type of reports, because they reveal the deeper cause of the problem being a bad "management" attitude, due to lack of technical insight.
SQL is not a specification language, it is a query and data manipulation language, and does not have any transactional properties. It is not what I was thinking about. Most people don't realize that relations databases and object-oriented models are just "implementations" of more general data structures.
Every so often a new language is proposed as the solution for the problem of developing applications. However, most of these languages are really generic programming languages. What we really need is a platform that centers around persistant data and views, with built in transaction mechanisms. Instead of a language to program in, we need a language to specify data centric applications.
He suffered from an aggressive streptococ/stafylococ bacterial infection and had to undergo surgery more than once to remove damaged tissue from his feet. He has had a fever for about a week. And although people around him are saying that he seems to be very alert and in good mental health, others have expressed their doubts whether he will be able to resume his normal (very busy) working schedule in the coming weeks, and that he might require some months to recover from it completely.
Most interesting is the fact that the Dutch parlement is trying to force Minister Brinkhorst to change his vote and that he continues to refuse to do so. The only reason he seems to be able to get away with this is that it is not a political issue, because the Dutch media not understanding software patents is not giving it any attention. The infection in the feet of our prime minister is far more interesting. (The latest rumours are that it was a rather serious infection, which might have killed him.)
If we would give all power to those that are in control of the money, democracy and freedom of speech would be gone soon. Governments are there for promoting the well being of all, and in some cases this means controling the power that the rich (people and companies) have. They also should prevent corruption (the misuse of money for the purpose of executing power). And BTW, there are no software patents yet in Europe yet.
Interesting how he describes how the introduction of each new medium (like painting, drama, novels) produced it own geniuses that were never surpassed by anyone that came later. I think this is also true for computers, but I guess that we are already behind the stage of those geniuses with respect to computers. So the time of the "Great Hackers" is already behind us, especially if you think that there are milions trying to write computer programs for their daily living. I have written programs for more than a quarter of a century, and I can tell you, that I haven't seen any new and great ideas in the past fifteen years. (And this language "arc", some syntactical sugar on top of LISP, is really BS. They syntax of LISP does not need any improvement. Twenty five years ago, I could count paratheses, simply by reading LISP. I never have understood what was the point.)
Yes, and people make spelling mistakes or invent new words all the time too. If I Google "misapostrophize" I get zero hits. Congratulations!!!
When HTTP was designed, it was intended to become something like WebDAV, because it does have a PUT method. For some reason most HTTP servers never implemented.