'The crew use fairly standard laptops running Microsoft Outlook (currently Outlook 2003) with Exchange Server as the email host...Because there's limited bandwidth up to the shuttle it's important to keep the OST fairly small so occasionally you'll hear NASA controllers ask the crew to clean out their Outlook files'
I say we take off and re-install the entire OS from orbit. It's the only way to be sure.
My guess: Google stops censoring itself, gaining credibility for its belatedly 'principled' stand against the Chinese government, while sending a message to China that hacking its servers is Not Polite. China predictably steps in to filter the search results using its own mechanisms, relieving Google of the burden. Google gets to keep its advertising revenue, while the users behind the Great Firewall get (at best) the same censorship as before. Now if Google really wants to make a point, with a genuine and serious risk of losing business, how about making google.cn an exclusively SSL site and seeing how fast China blocks it..?
'Text is one of the most expensive ways to communicate. What you can say in a 10 second conversation may take a multitude of texts back n forth.'
Certainly it can be a ripoff if you need to conduct a 'conversation' over multiple messages. But if I just want to say 'CU @ pub 8PM' it'll cost me 10p for the text, but 25p for a 1 minute phone call (on T-mobile UK PAYG, assuming the recipient is not also on T-mobile, which most people I know aren't). Or I could buy unlimited texts for 9 GBP per month. However, both look really expensive compared to internet access on the same network, which oddly enough is only 5 GBP per month, or 20 GBP for 6 months. So it's actually cheaper for me to use email or IM than text or voice. Unless, of course, I buy the Skype phone on the Three network, which would give me free mobile Skype-to-Skype without topups. For once, we actually get some pretty reasonable deals on this side of the pond.
'Texting and facebook updating is a leisure activity, and doesn't mix with work at all.'
I wonder how many of the other supposed differences are really down to the younger generation being, well, younger? A text message is probably cheaper than a voice call, which is handy if you're on a limited budget with a PAYG phone. A school or college age kid may have a wider social network than an older person in a full-time job, so online networking tools could be more useful. There be may less tendency to veg out in front of passive TV entertainment like an exhausted wage slave if you're out enjoying yourself all the time. Multitasking could be less difficult for a younger brain, etc. Of course, these are just the senile ramblings of an ageing mind, so take them with a pinch of salt. And get off my lawn.
'BTW, the Wikipedia entry shows that the "8%" number was known as long as 6 years ago.'
The news article is based on a rather careless reading of the new paper, where the 8% figure (which refers to endogenous retroviruses) is referenced in the Abstract but not presented as a new finding. The new result is that non-retroviral sequences from a completely different class of virus (Bornavirus) have also become incorporated into the genome. Bornavirus is an RNA virus that has the unusual property of setting up house in the nucleus. They think that a cellular reverse transcriptase (encoded by endogenous retrotransposons) could have been responsible for inserting DNA forms of Bornavirus genes into human (and other animal) genomes. This is different to the situation with retroviruses, which are able to integrate themselves into the genome using their own (viral) reverse transcriptases.
'Second, and more importantly, Can your $50 Walfart special transfer movies to your psp so you can watch it on the plane, Or do you have to take the disk with you and risk scratching it?'
On the other hand, when your plane lands in the 'wrong' country, don't even think about buying a regioned disk to take home to the PS3 (which is, like most players, Broken By Design and incapable of playing it). Of course, if you'd bought one of the cheapo supermarket models that has multi-region BD playback (they do exist), then...
'The problem here is that you're no longer dealing with a normal human tumor; you're selecting for tumor cells that grow in the artificial tissue culture environment.'
That's a good point. Both studies also used cell lines derived from metastases rather than the primary tumour, so we're already potentially dealing with a selected (though interesting) subclone even before the cells hit the culture dish. Not sure why they chose cell lines for the pilot study - perhaps because these lines were already well analysed by other methods, or because they could be confident that they were sequencing a pure population of tumour cells, or just to ensure they had lots of material to work with. In the lung cancer paper, it's mentioned that primary tumours of this subtype are rarely removed surgically, so it may be hard to get hold of good quality material.
'The second problem is that you're not sure what to compare the tumor sequence with. Due to privacy concerns, you almost never know who actually gave the tumor that was made into a cell line (as an aside, look up the HeLa cell line and its sordid history) so you have to compare to the human genome project. The problem here is that there are differences between people and you can't tell whether the "mutation" you see is just a normal variation or actually something in the tumor'
Both studies took account of this by sequencing normal DNA from the same donor, so they must have obtained appropriate samples and consent. For the reasons you mention, this is really the only sensible way to proceed at this point (lots of previously unknown polymorphisms are turning up in each new normal genome that's sequenced, so it would be hard to pick out tumour-specific changes without a matched normal reference).
'Show me the evidence that almost 100% of DNA errors in skin cells or skin cancer cells are caused by sun exposure...'
Not 100% perhaps, but from the paper:
'DNA damage due to ultraviolet light leads to the formation of covalent links between two adjacent pyrimidines. Consequently, C>T mutations due to ultraviolet light usually occur at dipyrimidine sequences. Therefore, to evaluate further the role of ultraviolet light in the pathogenesis of somatic mutations in COLO-829, we examined the sequence context of C>T substitutions...[Lots of technical stuff about the sequence context of the mutations with some impressive looking p-values]...Therefore, the mutation spectrum and sequence context indicate that most C>T/G>A somatic substitutions in COLO-829 are attributable to ultraviolet-light-induced DNA damage.'
'The ICGC will identify a list of approximately 50 cancer types and subtypes that are of clinical significance around the globe, aiming to study cancers of all major organs, including breast, ovary, prostate, lung and blood cancers...All the data generated will be made rapidly and freely available to the global research community. '
'It seems that they should do this with cancer cells from several different patients and compare them to find out which mutations actually trigger the cancer.'
Believe it or not, they have thought of this! An international consortium has been set up to use exactly this technology on a really large scale. See e.g.:
'Each ICGC member will conduct a comprehensive, high-resolution analysis of the full range of genomic changes in at least one specific type or subtype of cancer, with studies built around common standards of data collection and analysis. Each project will each require cancer specimens from 500 patients and have an estimated cost of US$20 million.'
In these particular studies, they're only looking at 'somatic mutations' (mutations confined to the tumour, and not found in the patient's normal cells). Anything they inherited that might have made them susceptible to cancer in the first place gets 'cancelled out' by comparing the tumour DNA to normal DNA (e.g. from blood). You have to do a different type of study to find susceptibility genes, e.g. by using a large collection of 'normal' DNA samples from a population and collecting their medical data. Right now, this is being done at a relatively low resolution using 'SNP arrays' that usually only look at a few hundred thousand DNA bases (a few million max). But because of genetic linkage, this can still give you very useful information about where the important genes are. When the genome sequencing technology gets _really_ cheap, we can except this sort of study to be done by sequencing too.
It's true that each patient is extremely likely to have a unique 'cancer genome', a specific combination of mutations found only in their tumour. But the vast majority of these will be 'passenger' mutations that aren't relevant to the progress of the tumour. The trick, as you suggest, is to home in on the 'driver' mutations that are really causing the disease. One way to get at these is to look first at the mutations in the coding sequences of known genes (and because of the human genome project and all the work that's followed it, we pretty much know where all the protein-coding genes are located).
I just had a quick look at both papers, and it turns out that in the lung cancer case, fewer than 100 of the tens of thousands of mutations actually cause an amino acid change in a protein sequence (for the melanoma, the figure is less than 200). This doesn't mean that there aren't other interesting needles to find in the haystack of mutations (e.g. changes in regulatory sequences), but they might as well go after the 'low hanging fruit' first. With current technology, it's very easy to sequence 100-200 genes in a pretty large set of samples from different patients. Any of these genes that turn out to be mutated in multiple tumours immediately become subjects for further study.
As the technology starts to ramp up and gets cheaper every year, we can begin to go after the less obvious changes. Each of these studies is in effect an entire human genome project (they haven't just done a low resolution map, they've completely sequenced the genomes). Pretty soon we're going to have a large collection of sequenced tumour samples to compare and use to find common alterations.
No idea about US delivery, customs, or pre-paid network availability, but you certainly won't be able to roam cheaply on the T-mobile US network with a UK SIM. I think the unlocking fee is under 20GBP (haven't checked this), so you could potentially use it on any network, but you'd have to check if the local 3G frequency is supported (it's quad band for 2G).
The 16Gb 3Gs on PAYG is 440 GBP (in comparison the 8Gb Touch is 149 GBP, the 16Gb is 229 GBP, the T-Mobile Pulse is 137 GBP, and an 8Gb micro SD is 10 GBP).
Here in the UK, you can currently pick up a T-mobile Pulse Android phone, and add an 8Gb micro SD card, for less than the price of the 8Gb Touch. And that's on a PAYG tariff, so there's no contract to service. Pre-paid 3G net access is only 20 GBP for 6 months on T-mobile, so there's no reason to restrict yourself to wifi either. With full phones at this sort of price, an Android PMP would have to be pretty cheap (or offer significant extra features, like a lot of storage) to be competitive. I wonder if the Touch would even exist if not for Apple's lucrative deals with the iPhone network providers? The price difference between the Touch and the PAYG iPhone (or contract iPhone with the monthly charges factored in) hardly seems justified based on the probable costs of the additional components alone.
'I'm sure there are those who feel that Google should be willing to "stand up" to the Chinese Government, but when you boil it down to the basics, there is nothing obliging Google as a company to engage in this fight.'
I wonder why Google doesn't disclose the search terms they do censor in China? Perhaps they 'don't want anyone to know' because they 'shouldn't be doing it in the first place.'...
'The other question is that there are no implementations of the standard yet...'
The Rakudo guys have now committed themselves to a useful/usable release (if not a complete implementation of everything in the standard) in Spring 2010 (the target is April):
"Perl 6 is a new language. Perl 5 and Perl 6 are two languages in the Perl family, but of different lineages. There is no current release schedule for Perl 6."
Some people, of course, may still find this confusing. These people should use Python:-)
A longer answer (together with several chapters of new Perl 6 book written by some of the developers) is here:
"Some might ask, 'Why call it Perl if it's a different language?' Perl is more than just the vagaries of syntax. Perl is philosophy (there's more than one way to do it; easy things easy, hard things possible); Perl is custom (unit testing); Perl is architectual edifice (Comprehensive Perl Archive Network); Perl is community (perl5porters, perl6-language). These are things that both Perl 5 and Perl 6 will share to varying degrees. Also, due to Perl's habit of stealing good ideas, Perl 5 and Perl 6 will converge in some areas as Perl 5 borrows ideas from Perl 6 and vice versa."
Perl 6 is a new language, not just an upgrade. Perl 5 has not been neglected, and continues to receive updates and new features (some of them originally developed for Perl 6). The plan is to continue Perl 5 support and development indefinitely, irrespective of the status of Perl 6. And of course Perl 5 has its own advent calendar, which this year focuses on interesting stuff you can do with various CPAN modules:
Slashdot Mirror
'The crew use fairly standard laptops running Microsoft Outlook (currently Outlook 2003) with Exchange Server as the email host...Because there's limited bandwidth up to the shuttle it's important to keep the OST fairly small so occasionally you'll hear NASA controllers ask the crew to clean out their Outlook files'
I say we take off and re-install the entire OS from orbit. It's the only way to be sure.
My guess: Google stops censoring itself, gaining credibility for its belatedly 'principled' stand against the Chinese government, while sending a message to China that hacking its servers is Not Polite. China predictably steps in to filter the search results using its own mechanisms, relieving Google of the burden. Google gets to keep its advertising revenue, while the users behind the Great Firewall get (at best) the same censorship as before. Now if Google really wants to make a point, with a genuine and serious risk of losing business, how about making google.cn an exclusively SSL site and seeing how fast China blocks it..?
'Text is one of the most expensive ways to communicate. What you can say in a 10 second conversation may take a multitude of texts back n forth.'
Certainly it can be a ripoff if you need to conduct a 'conversation' over multiple messages. But if I just want to say 'CU @ pub 8PM' it'll cost me 10p for the text, but 25p for a 1 minute phone call (on T-mobile UK PAYG, assuming the recipient is not also on T-mobile, which most people I know aren't). Or I could buy unlimited texts for 9 GBP per month. However, both look really expensive compared to internet access on the same network, which oddly enough is only 5 GBP per month, or 20 GBP for 6 months. So it's actually cheaper for me to use email or IM than text or voice. Unless, of course, I buy the Skype phone on the Three network, which would give me free mobile Skype-to-Skype without topups. For once, we actually get some pretty reasonable deals on this side of the pond.
'Texting and facebook updating is a leisure activity, and doesn't mix with work at all.'
I wonder how many of the other supposed differences are really down to the younger generation being, well, younger? A text message is probably cheaper than a voice call, which is handy if you're on a limited budget with a PAYG phone. A school or college age kid may have a wider social network than an older person in a full-time job, so online networking tools could be more useful. There be may less tendency to veg out in front of passive TV entertainment like an exhausted wage slave if you're out enjoying yourself all the time. Multitasking could be less difficult for a younger brain, etc. Of course, these are just the senile ramblings of an ageing mind, so take them with a pinch of salt. And get off my lawn.
Could be worse:
http://www.youtube.com/watch?v=SVFstSSUm1w
'BTW, the Wikipedia entry shows that the "8%" number was known as long as 6 years ago.'
The news article is based on a rather careless reading of the new paper, where the 8% figure (which refers to endogenous retroviruses) is referenced in the Abstract but not presented as a new finding. The new result is that non-retroviral sequences from a completely different class of virus (Bornavirus) have also become incorporated into the genome. Bornavirus is an RNA virus that has the unusual property of setting up house in the nucleus. They think that a cellular reverse transcriptase (encoded by endogenous retrotransposons) could have been responsible for inserting DNA forms of Bornavirus genes into human (and other animal) genomes. This is different to the situation with retroviruses, which are able to integrate themselves into the genome using their own (viral) reverse transcriptases.
'The other mitigating factor is just about every Region locked Movie I've run into is PAL.'
That's another win for (at least one of) the cheapo players. You get a built-in NTSC/PAL standards converter.
'Second, and more importantly, Can your $50 Walfart special transfer movies to your psp so you can watch it on the plane, Or do you have to take the disk with you and risk scratching it?'
On the other hand, when your plane lands in the 'wrong' country, don't even think about buying a regioned disk to take home to the PS3 (which is, like most players, Broken By Design and incapable of playing it). Of course, if you'd bought one of the cheapo supermarket models that has multi-region BD playback (they do exist), then...
'Now: a slastdot post with "i'm too lazy to find this XKCD link, someone else can post it" is an automatic +5'
Another XKCD strip (and mouseover) is probably relevant here:
http://xkcd.com/14/
Sounds like you might be interested in this exciting new media access concept!:
http://en.wikipedia.org/wiki/Library
'The problem here is that you're no longer dealing with a normal human tumor; you're selecting for tumor cells that grow in the artificial tissue culture environment.'
That's a good point. Both studies also used cell lines derived from metastases rather than the primary tumour, so we're already potentially dealing with a selected (though interesting) subclone even before the cells hit the culture dish. Not sure why they chose cell lines for the pilot study - perhaps because these lines were already well analysed by other methods, or because they could be confident that they were sequencing a pure population of tumour cells, or just to ensure they had lots of material to work with. In the lung cancer paper, it's mentioned that primary tumours of this subtype are rarely removed surgically, so it may be hard to get hold of good quality material.
'The second problem is that you're not sure what to compare the tumor sequence with. Due to privacy concerns, you almost never know who actually gave the tumor that was made into a cell line (as an aside, look up the HeLa cell line and its sordid history) so you have to compare to the human genome project. The problem here is that there are differences between people and you can't tell whether the "mutation" you see is just a normal variation or actually something in the tumor'
Both studies took account of this by sequencing normal DNA from the same donor, so they must have obtained appropriate samples and consent. For the reasons you mention, this is really the only sensible way to proceed at this point (lots of previously unknown polymorphisms are turning up in each new normal genome that's sequenced, so it would be hard to pick out tumour-specific changes without a matched normal reference).
'Show me the evidence that almost 100% of DNA errors in skin cells or skin cancer cells are caused by sun exposure...'
Not 100% perhaps, but from the paper:
'DNA damage due to ultraviolet light leads to the formation of covalent links between two adjacent pyrimidines. Consequently, C>T mutations due to ultraviolet light usually occur at dipyrimidine sequences. Therefore, to evaluate further the role of ultraviolet light in the pathogenesis of somatic mutations in COLO-829, we examined the sequence context of C>T substitutions...[Lots of technical stuff about the sequence context of the mutations with some impressive looking p-values] ...Therefore, the mutation spectrum and sequence context indicate that most C>T/G>A somatic substitutions in COLO-829 are attributable to ultraviolet-light-induced DNA damage.'
They started off with a couple of common cancers, but the plan is to do many more:
http://www.sanger.ac.uk/about/press/2008/080429.html
'The ICGC will identify a list of approximately 50 cancer types and subtypes that are of clinical significance around the globe, aiming to study cancers of all major organs, including breast, ovary, prostate, lung and blood cancers...All the data generated will be made rapidly and freely available to the global research community. '
'It seems that they should do this with cancer cells from several different patients and compare them to find out which mutations actually trigger the cancer.'
Believe it or not, they have thought of this! An international consortium has been set up to use exactly this technology on a really large scale. See e.g.:
http://www.sanger.ac.uk/about/press/2008/080429.html
'Each ICGC member will conduct a comprehensive, high-resolution analysis of the full range of genomic changes in at least one specific type or subtype of cancer, with studies built around common standards of data collection and analysis. Each project will each require cancer specimens from 500 patients and have an estimated cost of US$20 million.'
In these particular studies, they're only looking at 'somatic mutations' (mutations confined to the tumour, and not found in the patient's normal cells). Anything they inherited that might have made them susceptible to cancer in the first place gets 'cancelled out' by comparing the tumour DNA to normal DNA (e.g. from blood). You have to do a different type of study to find susceptibility genes, e.g. by using a large collection of 'normal' DNA samples from a population and collecting their medical data. Right now, this is being done at a relatively low resolution using 'SNP arrays' that usually only look at a few hundred thousand DNA bases (a few million max). But because of genetic linkage, this can still give you very useful information about where the important genes are. When the genome sequencing technology gets _really_ cheap, we can except this sort of study to be done by sequencing too.
It's true that each patient is extremely likely to have a unique 'cancer genome', a specific combination of mutations found only in their tumour. But the vast majority of these will be 'passenger' mutations that aren't relevant to the progress of the tumour. The trick, as you suggest, is to home in on the 'driver' mutations that are really causing the disease. One way to get at these is to look first at the mutations in the coding sequences of known genes (and because of the human genome project and all the work that's followed it, we pretty much know where all the protein-coding genes are located).
I just had a quick look at both papers, and it turns out that in the lung cancer case, fewer than 100 of the tens of thousands of mutations actually cause an amino acid change in a protein sequence (for the melanoma, the figure is less than 200). This doesn't mean that there aren't other interesting needles to find in the haystack of mutations (e.g. changes in regulatory sequences), but they might as well go after the 'low hanging fruit' first. With current technology, it's very easy to sequence 100-200 genes in a pretty large set of samples from different patients. Any of these genes that turn out to be mutated in multiple tumours immediately become subjects for further study.
As the technology starts to ramp up and gets cheaper every year, we can begin to go after the less obvious changes. Each of these studies is in effect an entire human genome project (they haven't just done a low resolution map, they've completely sequenced the genomes). Pretty soon we're going to have a large collection of sequenced tumour samples to compare and use to find common alterations.
'So in exchange for CCTV surveillance everywhere, UK residents get more reasonably priced mobile data service.'
Absolutely!. How else could Airstrip One get a Telescreen not just on every street corner, but in every pocket..?
Of course, in some areas we still have a lot of catching up to do:
http://www.eff.org/issues/nsa-spying
http://www.newyorker.com/archive/2005/02/14/050214fa_fact6
No idea about US delivery, customs, or pre-paid network availability, but you certainly won't be able to roam cheaply on the T-mobile US network with a UK SIM. I think the unlocking fee is under 20GBP (haven't checked this), so you could potentially use it on any network, but you'd have to check if the local 3G frequency is supported (it's quad band for 2G).
'the PAYG iPhone...Doesn't exist.'
Does here:
http://shop.o2.co.uk/promo/iphoneindex/Pay_And_Go/3G_S
The 16Gb 3Gs on PAYG is 440 GBP (in comparison the 8Gb Touch is 149 GBP, the 16Gb is 229 GBP, the T-Mobile Pulse is 137 GBP, and an 8Gb micro SD is 10 GBP).
Here in the UK, you can currently pick up a T-mobile Pulse Android phone, and add an 8Gb micro SD card, for less than the price of the 8Gb Touch. And that's on a PAYG tariff, so there's no contract to service. Pre-paid 3G net access is only 20 GBP for 6 months on T-mobile, so there's no reason to restrict yourself to wifi either. With full phones at this sort of price, an Android PMP would have to be pretty cheap (or offer significant extra features, like a lot of storage) to be competitive. I wonder if the Touch would even exist if not for Apple's lucrative deals with the iPhone network providers? The price difference between the Touch and the PAYG iPhone (or contract iPhone with the monthly charges factored in) hardly seems justified based on the probable costs of the additional components alone.
'I'm sure there are those who feel that Google should be willing to "stand up" to the Chinese Government, but when you boil it down to the basics, there is nothing obliging Google as a company to engage in this fight.'
I wonder why Google doesn't disclose the search terms they do censor in China? Perhaps they 'don't want anyone to know' because they 'shouldn't be doing it in the first place.'...
'The other question is that there are no implementations of the standard yet...'
The Rakudo guys have now committed themselves to a useful/usable release (if not a complete implementation of everything in the standard) in Spring 2010 (the target is April):
http://use.perl.org/~pmichaud/journal/39411
They intend this to be a release which 'application writers will feel comfortable enough to start using in their projects'.
This probably helped:
http://news.perlfoundation.org/2008/05/tpf_receives_large_donation_in.html
Here's where they are now:
http://rakudo.org/status
http://github.com/rakudo/rakudo/blob/master/docs/ROADMAP
Well, the first line of the first Google hit for 'Perl 6':
http://dev.perl.org/perl6/
says:
"Perl 6 is a new language. Perl 5 and Perl 6 are two languages in the Perl family, but of different lineages. There is no current release schedule for Perl 6."
Some people, of course, may still find this confusing. These people should use Python :-)
A longer answer (together with several chapters of new Perl 6 book written by some of the developers) is here:
http://cloud.github.com/downloads/perl6/book/book-2009-11.pdf
"Some might ask, 'Why call it Perl if it's a different language?' Perl is more than just the vagaries of syntax. Perl is philosophy (there's more than one way to do it; easy things easy, hard things possible); Perl is custom (unit testing); Perl is architectual edifice (Comprehensive Perl Archive Network); Perl is community (perl5porters, perl6-language). These are things that both Perl 5 and Perl 6 will share to varying degrees. Also, due to Perl's habit of stealing good ideas, Perl 5 and Perl 6 will converge in some areas as Perl 5 borrows ideas from Perl 6 and vice versa."
Perl 6 is a new language, not just an upgrade. Perl 5 has not been neglected, and continues to receive updates and new features (some of them originally developed for Perl 6). The plan is to continue Perl 5 support and development indefinitely, irrespective of the status of Perl 6. And of course Perl 5 has its own advent calendar, which this year focuses on interesting stuff you can do with various CPAN modules:
http://www.perladvent.org/2009/
'...including one teenager who then resold the software to another 240 people.'
Sounds the Toshiba guy should have used some sort of 'copy-protection' technology to safeguard his product.
'Turns out one survived -- and is now available on Amazon'
Or at least, that's what they want you to think...