Yeah, 100 years of solitude is an incredible novel. But I would not put Marquez down as the father of magical realism - I think that honour has to go to Bulgarkov - "The Master and Margarita" is IMHO one of the greatest works of literature. "The Heart of a Dog" blends Sci-Fi with Political Satire, and some of his other stuff (esp "The White Guard") is just as good.
I think Rushdie will be read for a long time. I think a lot of his stuff is over-rated, but "The Satanic Verses" is just marvellous, although, that said, I can understand why many people find it offensive, although I do not myself.
I'd also reccomend "The King David Report" by Stefan Heym -history as written by the winners.
I'm just yabbering here, but...
In the solutions to this problem that I have seen, where is the check to ensure that SOME_NUMBER*sizeof(int) does not overflow the size_t type, and cause an area of memory of the wrong size to be returned?
Of course, this could be caught in the error handling of the/* do something */, but surely it would be better to be caught and flagged up earlier rather than later... where it was obvious that this was the problem, rather than having an array access (for example) die in the middle of a valid range.
Erm, not (quite) necessarily
There are two reasons why these companies may have patented extensions to VXML 1.0.
1. (If you are an optimist) These are defensive patents to stop them beeing screwed. They have no intention of enforcing them, but it does mean they can't be forced into licensing these bits of the technology themselves.
2. (If you are paranoid) The idea is to create an effective cartel between the companies owning the patents by the use of cross licensing of the relevant patents, thus raising the financial bar on any new entrants to the market. This has been doen in the past with GSM.
Alan Turing also wrote a paper called "The chemical basis of morphogenesis" which is one of the key papers in explaining how morphogenesis can produce differnt types of cells and hence organisms with differntiated parts (arms, legs, tentacles), or patterned coats (zebra stripes, cow splodges, cheetah spots) etc, on the basis of sets of chemical reactions.
radio isn't just handsets - there is the base statio side as well. I agree that the Motorola handsets are crap - I think they brought in a bunch of American designers who just don't understand the European handset market - that is why Nokia is THE brand. However, I would stand by my point that Motorola do the best radio engineering, especially on the infrastructure side.
I have a Motorola phone and it has never crashed (but then neither did my old Mitsubishi). But it is ugly, the navigation system is poor, and as for the documentation... don't even go there!
3G will happen. There is far, far too much pride and capital at stake for it to fail (and if it does it will make the present dot.com crash look like a minor hiccup). But it will probably happen like Concorde... limited roll out and a long time to make any sort of profit. And other solutions (2.5G, i.e. GPRS) will serve the mass market for a long tim to come.
The market shake out has already started. Ericsson have been really shafted by the sell-off of their handset business to Sony. This will mean that they only have access to one side of the radio stack - not a good position to be in. I suspect Nokia will survive, as will Motorola probably - Nokia has the best branding, and Motorola, while their branding is poor, have the best engineered radio systems, and they have huge amounts of capital. How the other players will fare is tricky. Lucent are in the poo already, and Alcatel are looking weak. Probably we will see more tie ups - Alcatel plus Fujitsu, Siemens plus NEC, Nortel plus who-ever.
A word of warning. I was once involved in a business where we kept backups of everything. Yess-sir! We had a grandfather father - son system, where one backup was kept on the machine, in the office on disk , and a previous backup disk at home. Every week, we brought in the old (home) backup, wrote a new backup on it, then took the old office backup home. This was done every friday.
The building burnt down on a friday, so we lost *all* the backups!
As some-one who has recently discovered Ruby after a 5 years c++, a bit of java, and having taught myself the rudiments of Perl, I can only say that ruby does seem to do OO far better than any of the above, IMHO. Quite frankly, I have never come across a language that I have become as quickly productive in.
The strength of Ruby is not only in its implementation of OO (which is very sweet - especially the iterators), but in that it lends itself to the writing of clean, easy of understand code. So when I go back to my code a few months later, it is *much* easier to understand -even with good comments in place.
Quite frankly, Ruby is now my language of choice if there are no other important factors to consider. If I need speed, I will go to C or C++. But until I am convinced of that, my number one choice for a language will be Ruby.
What about Steve Bell? Maybe his humour doesn't cross the atlantic, but the guy is so wickedly funny it is untrue. Very political, and so far to the left that evryone hets it just as hard...
regards
treefrog
Re:Hannibal isn't a sequel; it's third in a series
on
Hannibal's Return
·
· Score: 1
For those in the UK, Manhunter is on the telly tonight (monday) at 10pm.
Yeah, but the comments on scripting and testing are not wide of the mark - they are bang on. I may want to automate my output, and run things automatically. Ensuring that the GUI is just one possible way of controlling a program (a command line might be another) is just good sense, and practice. It does sit very well with the MVC (model view controller) idea.
It means that I can run tests scripts automatically to ensure that the internal functioning of a program is correct.
I can also automate data processing (just imagine taking those hundreds of simulation runs, doing the data analysis and plotting the results using a point and click interface). At least with a shell script, or perl, or ruby (my fav) you can automate this process - link the simulation tools to the analysis tools to the plotting tools. Even MS does this with VB (although it hust isn't as pretty!).
For those who don't know much about him. Vinnie Jones is an ex UK professional soccer player (Captain of Wimbledon in the "Crazy Gang" years, and captain of Wales) who had a hell of a reputation on the pitch as a bit of a thug. He holds the record for the fastest ever sending off - 5 seconds after the start of the match!
So, it may be typecasting, but the mere sight of Vinnie as a film character makes one think - hard, mean bloke (if you are a UK viewer).
As a digression, he was recently asked in an interview whether he ever got in touch with his feminine side. His reply was something along the lines of "Are you taking the piss...I'll fuckin' twat you!"
The wireless internet is coming along nicely as a piece of technology. TCP/IP works quite well across wireless. Big questions are
How well will mobile IP addresses work in practice?
Have operators (esp in the UK) paid too much for the 3G licences? Probably so. Vodaphone has certainly lost its gold plated credit rating as a result. This leads to the next question
Is there going to be a bloodbath in the telcos? The operators are desperate to claw back the licence fees. I read recently that some operators are looking for revenues of $250/month average per user. That is up from an average of probably $50 a month at present. So there are some stupid penalty clauses around to the manufacturers...who are desperate to win contracts because of their position on Wall St. It may well all end up in tears...
Actually, EDGE (which is an enhancement of GPRS, which is in turn an upgrade of GSM for packet data) can be overlaid on an IS 136 network. So yes, it is a GSM type system, but it is one hell of a tweak (like the difference between your average Ford and the Rally version)
This might actually be a smart move. GSM is the dominant 2G standard across the world, and it makes sense to buy into it. EDGE gives pretty much 3G performance. Whether UMTS (the big new 3G standard) turns out to be the biggest turkey in history is a moot point. No-one is sure at the moment whether they have done their sums right. It could be a bloodbath if the market turns out not to be there
regards,
treefrog
Re:Should be possible but by an algorithm?
on
Does P = NP?
·
· Score: 1
Let's just revise what we mean by P and NP
A P-time algorithm is one which will always produce an answer in P-time, ie the number of computational steps required is bound from above by a polynomial.
NP does not mean "not P-time". It means non-deterministic polynomial. The underlying idea is that an NP algorithm is one for which, if a solution can be found, the correctness of that solution can be verified by a P-time algorithm
An NP-complete algorithm is an NP algorithm for which a P-time transformation algorithm exists between it and every other NP algorithm (different P time transformation in every case, obviously).
An NP-hard algorithm is like an NP-complete algorithm, except it doesn't have to be NP itself, there just has to be a P-time reduction to evrery NP algorithm.
P==NP is at present a conjecture, which has been (generally) considered untrue. Now, I hav'nt had the time to look at this algorithm closely, but, if it is a non-crackpot attempt at proving P==NP, it will be of at least the importance of the Fermat's Last Theorem Proof. If proved this would have major implications for cryptography, because many of the crypto algorithms are based on the fact that finding the key is hard (usually NP-hard). If this algorithm is correct, then a P-time (ie quick) algorithm probably exists to convert your hard key discovery problem into an easy (P-time) problem such as the graph algorithm discussed in this paper.
Actually, there is an entire ETSI European Telecomunications Standards Institute) effort devoted to this. It's called TIPHON, and it is devoted to the merging of data and telephony networks. One of it's working groups (WG5) is devoted exclusively to looking at Quality of Service of voice over IP. I sit on this group, and I can tell you that while there are a lot of problems, there are also a lot of people trying to solve them. All the major telco manufacturers and many operators are represented, because the advantages of providing a VoIP network are huge. I also believe that one of the long term goals of 3gpp is to provide mobile IP and VoIP on UMTS mobile terminals. Which is kind of cool...
It all comes down to the old chestnut that IP is a best effort service, and the need to find a scalable solution that allows the reservation of bandwidth (Diffserv is scalable, but does not allow bandwidth reservation, only relative QoS, wheras RSVP allows bandwidth reservation but does not scale well).
The BBC has a piece on Jose Bove here . Jose Bove is a small French farmer, who, like many in Europe farms a small area of land. In his case he makes a fine Roquefort Cheese. His livelihood has been threatened by economic sanctions (punitive import taxes) placed on EU goods by the USA. So he has taken direct action against a US Multinational which is perceived as an icon of US Cultural Imperialism.
This form of direct action is actually quite common in France, it has not been long since French farmers stopped lorries carrying lamb entering France from the UK and burnt the (already slaughtered) contents.
At the moment it is not just french peasant farmers that are suffering. In the UK small upland farmers are going bankrupt at an amazing rate. One of the problems is increased new legislation, which I believe has been heavily lobbied for by the multinationals and large producers, which is forcing smaller abatoirs to close. Small farmers have to then take their produce a long way to get it slaughtered, and they cannot afford this.
There is also a problem with the buying power of the supermarkets and multinationals. You sell at our rates on our terms or you don't sell at all.
It is not all doom and gloom though. Organic produce is more popular than ever, and the rise of farmers markets in many towns is throwing a lifeline to the smaller producers.
I've already read many comments on this topic that seem to basically say "This guy is an anticapitalist nutter. Fuck him". This is a topic that many people in the EU (including myself) care passioately about. I visit the states regularly on business, and I have seen what unconstrained capitalism can do. Jose, I salute you and say "Vive le Roquefort".
And one more thing - beer guy, this is one thread where you will be so on-topic it will hurt!
Re:Substandard C++ compiler, libraries, etc...
on
Why Develop On Linux?
·
· Score: 1
Re the Borland Compiler - does it still use the RogueWave STL libraries. OK, it was a year ago, but back then when I was using Borland C++builder and VC++6, we gave up on Borland because it wouldn't compile its own ANSI libraries (ie STL) as ANSI strict. Since we wanted to use the software we were developing (financial prediction and monitoring stuff) cross platform, we used VC++6, as it seemed much better.
In fact, I actually think the MS compilers aren't bad. The libraries may be shite, but they obviously have thrown a lot of good bright people at the compilers.
Let's face it, there are not that many really good programmers around, so thank God that MS has actually thrown its talent at something important. Which would you rather be a good piece of software; the compiler or Word?
This is on the same lines as an apocryphal tale regarding a PhD viva (defence in the US):
The candidate comes into the room, is introduced the examiners and sits down. Before the exam starts he pullls out a bottle of gin and places it on the table in front of him.
The exam takes place, and after several hours of intense questioning, the candidate is told that he has passed. The examiners have one final question though, why the bottle of gin?
"If you read the footnote at the bottom of page 137", replies the candidiate, you will see that it reads "If you read this, you may claim a bottle of gin from me".
what are the chances of getting C++ (egcs maybe), or better still something like CC++ to compile under Plan 9. And would it work well, or just be a dog?
Might one be better off with a C++ to C translator running on top of the Plan 9 C compiler?
From what I understand of the whole MP3/Napster/GNutella etc debate, many of the supporters within the artistic community tend to be the young unsigned bands looking for a break, whereas many of the opponents tend to be established (and quite frankly rich) artists.
Do you think that as these younger artists get signed (or at least some of them will), the support for MP3 etc will become more widespread, or do you think that the artists will sell out (nothing like a good rock'n'roll cliche!).
Slashdot Mirror
Yeah, 100 years of solitude is an incredible novel. But I would not put Marquez down as the father of magical realism - I think that honour has to go to Bulgarkov - "The Master and Margarita" is IMHO one of the greatest works of literature. "The Heart of a Dog" blends Sci-Fi with Political Satire, and some of his other stuff (esp "The White Guard") is just as good.
I think Rushdie will be read for a long time. I think a lot of his stuff is over-rated, but "The Satanic Verses" is just marvellous, although, that said, I can understand why many people find it offensive, although I do not myself.
I'd also reccomend "The King David Report" by Stefan Heym -history as written by the winners.
Best regards
Treefrog
I'm just yabbering here, but...
/* do something */, but surely it would be better to be caught and flagged up earlier rather than later... where it was obvious that this was the problem, rather than having an array access (for example) die in the middle of a valid range.
In the solutions to this problem that I have seen, where is the check to ensure that SOME_NUMBER*sizeof(int) does not overflow the size_t type, and cause an area of memory of the wrong size to be returned?
Of course, this could be caught in the error handling of the
Best regards
Tree_frog
Erm, not (quite) necessarily
There are two reasons why these companies may have patented extensions to VXML 1.0.
1. (If you are an optimist) These are defensive patents to stop them beeing screwed. They have no intention of enforcing them, but it does mean they can't be forced into licensing these bits of the technology themselves.
2. (If you are paranoid) The idea is to create an effective cartel between the companies owning the patents by the use of cross licensing of the relevant patents, thus raising the financial bar on any new entrants to the market. This has been doen in the past with GSM.
Best regards
treefrog
Alan Turing also wrote a paper called "The chemical basis of morphogenesis" which is one of the key papers in explaining how morphogenesis can produce differnt types of cells and hence organisms with differntiated parts (arms, legs, tentacles), or patterned coats (zebra stripes, cow splodges, cheetah spots) etc, on the basis of sets of chemical reactions.
A genius spanning several fields....
Regards
Treefrog.
Please mod this up - this guy actually knows what he is talking about!
radio isn't just handsets - there is the base statio side as well. I agree that the Motorola handsets are crap - I think they brought in a bunch of American designers who just don't understand the European handset market - that is why Nokia is THE brand. However, I would stand by my point that Motorola do the best radio engineering, especially on the infrastructure side.
... don't even go there!
I have a Motorola phone and it has never crashed (but then neither did my old Mitsubishi). But it is ugly, the navigation system is poor, and as for the documentation
best regards
treefrog
3G will happen. There is far, far too much pride and capital at stake for it to fail (and if it does it will make the present dot.com crash look like a minor hiccup). But it will probably happen like Concorde ... limited roll out and a long time to make any sort of profit. And other solutions (2.5G, i.e. GPRS) will serve the mass market for a long tim to come.
The market shake out has already started. Ericsson have been really shafted by the sell-off of their handset business to Sony. This will mean that they only have access to one side of the radio stack - not a good position to be in. I suspect Nokia will survive, as will Motorola probably - Nokia has the best branding, and Motorola, while their branding is poor, have the best engineered radio systems, and they have huge amounts of capital. How the other players will fare is tricky. Lucent are in the poo already, and Alcatel are looking weak. Probably we will see more tie ups - Alcatel plus Fujitsu, Siemens plus NEC, Nortel plus who-ever.
Hold onto your hats, its going to be bumpy!
A word of warning. I was once involved in a business where we kept backups of everything. Yess-sir! We had a grandfather father - son system, where one backup was kept on the machine, in the office on disk , and a previous backup disk at home. Every week, we brought in the old (home) backup, wrote a new backup on it, then took the old office backup home. This was done every friday. The building burnt down on a friday, so we lost *all* the backups!
As some-one who has recently discovered Ruby after a 5 years c++, a bit of java, and having taught myself the rudiments of Perl, I can only say that ruby does seem to do OO far better than any of the above, IMHO. Quite frankly, I have never come across a language that I have become as quickly productive in.
The strength of Ruby is not only in its implementation of OO (which is very sweet - especially the iterators), but in that it lends itself to the writing of clean, easy of understand code. So when I go back to my code a few months later, it is *much* easier to understand -even with good comments in place.
Quite frankly, Ruby is now my language of choice if there are no other important factors to consider. If I need speed, I will go to C or C++. But until I am convinced of that, my number one choice for a language will be Ruby.
Best regards
treefrog
What about Steve Bell? Maybe his humour doesn't cross the atlantic, but the guy is so wickedly funny it is untrue. Very political, and so far to the left that evryone hets it just as hard...
regards
treefrog
For those in the UK, Manhunter is on the telly tonight (monday) at 10pm.
regards,
tree_frog
Yeah, but the comments on scripting and testing are not wide of the mark - they are bang on. I may want to automate my output, and run things automatically. Ensuring that the GUI is just one possible way of controlling a program (a command line might be another) is just good sense, and practice. It does sit very well with the MVC (model view controller) idea.
It means that I can run tests scripts automatically to ensure that the internal functioning of a program is correct.
I can also automate data processing (just imagine taking those hundreds of simulation runs, doing the data analysis and plotting the results using a point and click interface). At least with a shell script, or perl, or ruby (my fav) you can automate this process - link the simulation tools to the analysis tools to the plotting tools. Even MS does this with VB (although it hust isn't as pretty!).
regards
tree_frog
For those who don't know much about him. Vinnie Jones is an ex UK professional soccer player (Captain of Wimbledon in the "Crazy Gang" years, and captain of Wales) who had a hell of a reputation on the pitch as a bit of a thug. He holds the record for the fastest ever sending off - 5 seconds after the start of the match!
So, it may be typecasting, but the mere sight of Vinnie as a film character makes one think - hard, mean bloke (if you are a UK viewer).
As a digression, he was recently asked in an interview whether he ever got in touch with his feminine side. His reply was something along the lines of "Are you taking the piss...I'll fuckin' twat you!"
Treefrog
- How well will mobile IP addresses work in practice?
- Have operators (esp in the UK) paid too much for the 3G licences? Probably so. Vodaphone has certainly lost its gold plated credit rating as a result. This leads to the next question
- Is there going to be a bloodbath in the telcos? The operators are desperate to claw back the licence fees. I read recently that some operators are looking for revenues of $250/month average per user. That is up from an average of probably $50 a month at present. So there are some stupid penalty clauses around to the manufacturers...who are desperate to win contracts because of their position on Wall St. It may well all end up in tears...
Keep that resume sharp... Tree_frogThis might actually be a smart move. GSM is the dominant 2G standard across the world, and it makes sense to buy into it. EDGE gives pretty much 3G performance. Whether UMTS (the big new 3G standard) turns out to be the biggest turkey in history is a moot point. No-one is sure at the moment whether they have done their sums right. It could be a bloodbath if the market turns out not to be there
regards,
treefrog
A P-time algorithm is one which will always produce an answer in P-time, ie the number of computational steps required is bound from above by a polynomial.
NP does not mean "not P-time". It means non-deterministic polynomial. The underlying idea is that an NP algorithm is one for which, if a solution can be found, the correctness of that solution can be verified by a P-time algorithm
An NP-complete algorithm is an NP algorithm for which a P-time transformation algorithm exists between it and every other NP algorithm (different P time transformation in every case, obviously).
An NP-hard algorithm is like an NP-complete algorithm, except it doesn't have to be NP itself, there just has to be a P-time reduction to evrery NP algorithm.
P==NP is at present a conjecture, which has been (generally) considered untrue. Now, I hav'nt had the time to look at this algorithm closely, but, if it is a non-crackpot attempt at proving P==NP, it will be of at least the importance of the Fermat's Last Theorem Proof. If proved this would have major implications for cryptography, because many of the crypto algorithms are based on the fact that finding the key is hard (usually NP-hard). If this algorithm is correct, then a P-time (ie quick) algorithm probably exists to convert your hard key discovery problem into an easy (P-time) problem such as the graph algorithm discussed in this paper.
best regards,
tree frog
[Kevin Kline] You know, Bub, as in Bub Deelan
[Meg Ryan] Oh, you mean Baaaaahhb!
Regards, Treefrog
It all comes down to the old chestnut that IP is a best effort service, and the need to find a scalable solution that allows the reservation of bandwidth (Diffserv is scalable, but does not allow bandwidth reservation, only relative QoS, wheras RSVP allows bandwidth reservation but does not scale well).
regards, treefrog
This form of direct action is actually quite common in France, it has not been long since French farmers stopped lorries carrying lamb entering France from the UK and burnt the (already slaughtered) contents.
At the moment it is not just french peasant farmers that are suffering. In the UK small upland farmers are going bankrupt at an amazing rate. One of the problems is increased new legislation, which I believe has been heavily lobbied for by the multinationals and large producers, which is forcing smaller abatoirs to close. Small farmers have to then take their produce a long way to get it slaughtered, and they cannot afford this.
There is also a problem with the buying power of the supermarkets and multinationals. You sell at our rates on our terms or you don't sell at all.
It is not all doom and gloom though. Organic produce is more popular than ever, and the rise of farmers markets in many towns is throwing a lifeline to the smaller producers.
I've already read many comments on this topic that seem to basically say "This guy is an anticapitalist nutter. Fuck him". This is a topic that many people in the EU (including myself) care passioately about. I visit the states regularly on business, and I have seen what unconstrained capitalism can do. Jose, I salute you and say "Vive le Roquefort".
And one more thing - beer guy, this is one thread where you will be so on-topic it will hurt!
regards,
treefrog
regards, treefrog
In fact, I actually think the MS compilers aren't bad. The libraries may be shite, but they obviously have thrown a lot of good bright people at the compilers.
Let's face it, there are not that many really good programmers around, so thank God that MS has actually thrown its talent at something important. Which would you rather be a good piece of software; the compiler or Word?
regards, treefrog
Of course, the down side of this is that we may become extinct if we meet an alien species....
regards,
treefrog
The candidate comes into the room, is introduced the examiners and sits down. Before the exam starts he pullls out a bottle of gin and places it on the table in front of him.
The exam takes place, and after several hours of intense questioning, the candidate is told that he has passed. The examiners have one final question though, why the bottle of gin?
"If you read the footnote at the bottom of page 137", replies the candidiate, you will see that it reads "If you read this, you may claim a bottle of gin from me".
regards, treefrog
what are the chances of getting C++ (egcs maybe), or better still something like CC++ to compile under Plan 9. And would it work well, or just be a dog?
Might one be better off with a C++ to C translator running on top of the Plan 9 C compiler?
amphibiously affectionate
treefrog
Do you think that as these younger artists get signed (or at least some of them will), the support for MP3 etc will become more widespread, or do you think that the artists will sell out (nothing like a good rock'n'roll cliche!).
regards
tree_frog