From the "Homomorphic Encryption" page linked from the article: "Only in 2009 did Craig Gentry of IBM publish a mathematical proof showing fully homomorphic encryption was possible."
In the past (in a very hand-wavy kind of way) I've argued that it should be possible to "prove" that homomorphic encryption isn't feasible... because, in order to implement multiplication and addition of integers, I need a total-ordering over my data... and if I have a total ordering, my data is (effectively) decrypted. This, of course, doesn't preclude obfuscation and scrambling - but I (used to) believe strong homomorphic encryption was not worth investigation.
Does anyone have a reference to this proof? Has anyone read it? Can anyone summarise what Gentry concludes?
I've been thinking about reliability of connectivity quite a bit recently... Using a 3G dongle as a backup is one option - but with this option attracting either a noticable monthly charge or requiring a pre-pay to be renewed every 1 or 3 months... it is a bit frustrating... for a service I hope I never need to use. I'm currently wavering on the brink of taking the plunge - the clincher will be if I find time to convince myself that I can configure automatic fail-over satisfactorily.
The first interesting idea that springs to mind is this: if 0.5mb/s is free, what's stopping me buying N wi-fi dongles and channel-bonding their connections to give a ~N/2 mb/s connection, also for free?
Another interesting idea is that if Virgin had this service when I moved in, they'd have me as a customer... The way things actually panned out, I paid a deposit - they jerked me about and gave me absolutely no clue when I'd be given service - so I told them to sling-their-hook and went with Sky (who proved similarly useless - but eventually provided a DSL line.)
The real losers will surely be the telephone companies. Why bother with a pay-go mobile for texts if you can be connected to the web at 0.5mbps everywhere you go?
What really cripples things is when process is deemed a substitute for understanding the specifics of individual situations - where a one-size-fits-all-problems approach is adopted and imposed - usually by people who have no practical experience with the processes they espouse.
If software development could be successfully reduced to a process, I'd have automated it. Where there's a considerable burden of process, either the process is inappropriate - or developing the software itself is inappropriate as it amounts merely to re-inventing the wheel... an exercise in task creation that benefits no-one.
We should think of software development techniques and apply them judiciously - and the more techniques a developer masters, the wider their skill-set and the better they will adapt to new challenges. The critical question that needs to be asked is this: why is a technique being used and is it providing tangible benefits? If this question can't be adequately answered, everyone involved is wasting their time.
Seagate and Samsung are my favourite two drive manufacturers at the moment... I'd have preferred they remain separate.
If I'm thinking about my data, I want - above all - for it to be reliably stored. With the best will in the world, eventually every drive fails... So... I tend to buy different makes of drives in pairs - from different suppliers... the logic is that it is far less likely that both drives will fail simultaneously - leaving my raid-1 data intact.
If Seagate and Samsung share manufacturing/storage/distribution, then the independence of Seagate and Samsung drives vanishes... forcing me to go to another less-preferred vendor.
I wonder when these consolidations will stop being a good idea? I definitely hope that it will be possible to buy independently manufactured drives in future.
While I accept that the fastest modes of transport are disappearing, I think the mean speed at which people travel during a year is almost certainly higher now than it has been in the past. The change is that a lot more people are travelling pretty fast far more frequently... and, arguably, this is far more useful than a handful of people travelling very fast very occasionally.
I am broadly in favour of nuclear energy - in principle. In practice, I have faith in neither corporate compliance nor government monitoring. Neither entity is equipped, or motivated, to appropriately manage long term risk... and that means you can good as guarantee failures. Hysteria about nuclear contamination, IMHO, has made matters worse - encouraging officials to focus exclusively upon reassuring the public that there is "no risk" at the expense of a focus on restricting and mitigating the consequences of the (ultimately inevitable) eventual accident.
I must take issue with the claim that "The only people who hate on SQL are the people who don't understand databases."
I think you should take some time and read the extensive publications of Chris Date (of Codd and Date International fame) - which, without exception, are extremely critical of SQL (while praising the relational model.)
SQL has its place - but it is far from perfect... and, the 'shackle of compatibility' prevents most of what is counter-productive about SQL from being fixed. NOSQL, while an infant technology, at least shows promise... only by starting afresh can we establish better DBMS strategies. No, I don't think NO-SQL is an alternative to SQL - it is an entirely different approach... one that should be judged on merit in each individual context.
In reality, the situation is far worse than you suggest.
When I discover some scientific principle indirectly (for example, in a book; paper or lecture) it isn't sufficient for me to consider only whether the principle is 'factual'... I need to consider if the language used to describe the principle means the same to me as it does to the author - we necessarily interpret the idea in different contexts. This potential miscommunication, of course, applies not only where I consider others' work... their work, in turn will be significantly influenced by their indirect research and education. At each step there's a very real risk of misunderstanding. As an analogy, in mathematics we now know that truth can only be evaluated reliably subject to unquestionable axioms... at some level, all reasoning about science must at an axiomatic level - and I, for one, am extremely sceptical of any claim that current science adopts appropriate axioms. I think there is a real problem with the relevance of aspects of conventional scientific wisdom that we assume self-evident.
While I commend the effort to find a way around the inevitable onslaught of spit, I note that there are technical drawbacks. The web of trust sounds like a great idea - until you try to use it... then most people find themselves ham-strung by human nature which finds 'vouching' to be an onerous responsibility - especially when the full implications of making errors is not understood. Conversely, a centralised directory almost completely undermines the value of using a decentralised mesh network over conventional technology. Finally - the e-cash suggestion is especially amusing for a project called Freecall.:)
I see value in having software by which groups can self-organise - with their own contact lists - but that is a different proposition to replacing telephones or skype.
The problem I foresee with a distributed system (as I understand Freecall is intended to be) is nuisance calls - the so-called SPIT (SPam over Internet Telephony...)
Traditional phones mainly relied upon cost - and the likes of Skype have the benefit of infrastructure under their control.
Are there any plans to address this (orthogonal) issue, I wonder?
While I don't want to argue that PBKDF2 is unsound, I don't believe you about "zero entropy loss" - if this were true, then general purpose hash functions could all be proven to be perfect hashes when recursively applied to their own output. I don't believe this is the case, but - of course - if you've any evidence, I'd love to see some...
A placebo effect* doesn't fix anything,ever. It makes people feel better subjectively.
There was a time when I'd have thought something similar. Recently, though not from an interest in placebos, but from an interest in allergies, my default opinion has changed.
I'm currently very interested in how the body's immune system is regulated. I was particularly interested in how Cortisol (an anti-inflammatory and immune suppressant hormone) is regulated by the body. According to Wikipedia:
Under normal unstressed conditions, the human adrenal glands produce the equivalent of 35–40 mg of cortisone acetate per day. In contrast to the direct innervation of the medulla, the cortex is regulated by neuroendocrine hormones secreted by the pituitary gland and hypothalamus, as well as by the renin-angiotensin system.
Please forgive my ignorance on these matters, but if the body's immune system is regulated by the hypothalamus - then, perhaps what influences the hypothalamus influences the immune system - allowing a tiny change to have a significant, eventual, biochemical consequence... thus -potentially - linking thought and recovery.
BTW - if anyone can provide a more thorough reference (for a non-specialist) on what influences adrenal function, then I'd be very grateful.
I *really* want to find a good quality contact management program. By this, I don't mean an address book, per-se, but - rather - an application to help me keep in touch with diverse contacts (private and professional) who, otherwise, I'd quickly forget about.
I need, for example, to keep logs of interactions, so - if I contact them - I can be sure what we last talked about, and when. I'd like it to integrate with my iCal calendar so that I can schedule tasks to follow-up, when - otherwise - I'd forget.
It's extremely frustrating... I can't seem to find anything... and this is one of the very few bits of software I'd willingly pay for, because - to me - it would be invaluable. Can't find anything suitable though...
Thankfully, the Guardian, which has far superior journalism and doesn't seek to ram politics down everyone's throats in "news" stories like News International's papers do
Erm, I think this statement belies your own politics. The guardian is staunchly "New Labour" and I find the vast majority of its reporting to be extremely politically biased. This shouldn't surprise anyone who considers the volume of public sector advertising in this paper.
I don't think you'll ever find a useful comparison chart - because, while charts exist, they won't focus on what your application needs. Your application is the issue - not the DBMS choice; you need to think about what interactions you need (want) from an application perspective - then look for a solution that meets that specification. A lot depends upon granularity of transactions; on the volumes of data per transaction; on the volumes of data over time; on acceptable latency; on acceptable redundancy and fault tolerance - etc. etc.
The package I'd advise you to look at, is Oracle-Embedded (which has an open source licence, as well as a commercial one) which moves you out of the realm of relational databases all together. If this is a suitable solution, it suggests an entirely new way to think about what sort of back end you really need.
In summary, no-one else's comparison of RDBMS is likely to be relevant to your domain. All such comparisons are extremely fragile in the context of application demand.
I'll never support them and I'll continue to explain to those that ask my advice (I'm one of the go-to technical people for a lot of friends) exactly why I don't like them and suggest competitors.
Go on then - who offers a reliable, honest, cost-effective UK broadband service (preferably with a static IP)?
I'm certain I'd be best served by "a text book" - the question, of course, is which one? Browsing at my local bookshops has turned up nothing worth buying - and I get no further with Amazon et al.
I'd really like a definitive reference book on statistical methods... one that assumes its reader is mathematically able, but only occasionally engaged in statistical work.:)
5) If you know the basics, you know just enough to be dangerous and really piss off people who know what they're doing. Being able to run an anova or determine correlation makes some people think they actually know what's going on because, hey, it's math. But a lot of people who just do the basic stuff think their results are more meaningful than they actually are - falling prey to the whole "it's statistically significant therefore it must be IMPORTANT" fallacy (when you can certainly have things that are "statistically significant" but actually have virtually no impact on the outcome.
Here, I think, you've mentioned a critical issue with statistics. The field of statistics is treated as if it were an omniscient black box by the majority of the population, and the more complex the calculation and the less people understand about the process, the more weight the statistic tends to have on decision making. One idea that interests me is that this might be valid for some 'real life' situations... especially where the people are utterly bamboozled by the statistics but use them to guide their behaviours in the absence of credible alternatives... essentially turning arbitrary choices into self-fulfilling prophecies. I am intrigued as to whether or not such subtlety would allow one to devise statistics that have subversive systemic effects.
I would love to receive a recommendation for a solid statistics reference book. All the books I can find infuriate me - they talk down to the reader and appear to be written for someone who is innumerate... labouring the trivial and obvious - while skimming the non-obvious dismissively. Another thing that frustrates me about statistics texts is that they frequently focus on 'case' studies - encouraging the reader to assume that they can re-apply techniques when they encounter similar-looking situations in real life. This is utterly bonkers - since, without a full understanding of the mechanics of the calculations, it is impossible to determine if the similarity between two situations is an irrelevant detail or crucial to the technique. I'm not interested in 'learning by example' - so I can copy a bunch of people whose views I distrust... I'd like a reference that allows me to use a typical approach in my statistical reasoning rather than re-inventing the wheel myself. I'd like it to be comprehensive and compact - I don't want it to aim to indoctrinate me to think in a particular way in my analyses.
Pray, how do you know that? I've never bought a tablet - and wouldn't buy from any to-date because they don't meet my minimum expectations. Don't be fooled, however, I'd love to use a good tablet PC.
Sticking points for me are : * High resolution A4-ish screen (to read PDFs page-at-a-time) * Long battery life (at least a day) * Wi-fi connectivity and cheap 3G (perhaps via a bluetooth phone) for generic Web access
When technology meets my demands (and I'm sure it eventually will) I'll buy one - as, I suspect, will the rest of the world.
That seems quite an important extension - any idea when (or if) it will be supported by TB3?
To me, it seems like an error of judgement to mainstream release a new version when key addons have not been satisfactorily updated. For the likes of Lightening, it isn't just eye-candy... and, for many, I suspect, breaking existing (addon) functionality will be unacceptable.
That said, I'm looking forward to 'conversation' view - and I've craved an improved address book for years... though what I saw when I last took a peek at the Beta wasn't much better than in TB2.
So, not all change is bad, man-made change can sometimes be really good...but I don't think that's the case on balance here. YMMV, though. Did you have anything concrete that you'd like to point out to support what you were saying? I would be interested to hear it (and not just as a rhetorical point, either).
I'm afraid my arguments are not of the form that would facilitate my explaining future consequences - and justifying in advance why they are desirable. My objection to climate-change-fundamentalism is somewhat more abstract.
A considerable proportion of replies to my post centred on the idea that I might personally prefer different weather in a particular location. This is definitely not what I meant! I reject the idea that the world should be coerced away from (or penalised for using) conventional power sources. There will be a natural shift away from fossil fuels when superior alternatives are available - assuming no half-baked political interventions.
I'm from the UK - and there's a good chance that, if sea levels rise by a few meters, London will flood... and almost 10% of the population live there. Will that be a bad thing? I'm not so sure... perhaps it will prompt a less insane geographic concentration of business - leading to improved quality of life for all? It's very difficult to decide - in advance.
To pnot, I recognise that your (rather banal) suggestion, that I advocate mass starvation, is a fallacious argument. People starve today - but it is not because there is any inherent lack of food - but, rather, a combined issues of politics and finance... issues which are not addressed by climate hysterics.
I'm all for establishing clean, efficient energy sources, but I believe that if change is desirable, it will arise from technological progress and not from repression or efforts to engineer guilt - an effort which only has political motives - motives with which I strongly disagree. Technological progress is best fostered through freedom and transparency - it is most hindered by politicisation and centralisation; by ill-conceived targets and divisive manipulation of objective observations to serve vested interests.
I think the whole 'climate change' debate is bizarre, and I am deeply sceptical about those who use climate change to push tenuously related political agendas.
If prompted for my strongest climate-sceptic view, it is this: while I'm willing to accept that substantial data shows that mankind's activities have resulted in atmospheric change - and while I'm willing to believe that this influences climate, I've one key question: is the change for the worse? Really? How can you be so sure?
I dislike the doom-laden 'climate change will wreck our environment' crowd for one key reason: they can't provide any evidence that I wouldn't prefer the climate after it has changed. Lots of things have been affected by mankind - and, frankly, I prefer to live in the world in which these changes have been made.
While I applaud being economical - and dislike pollution as much as the next sane person... I also think mankind belongs on earth... and I'm not willing to blindly jump on the change implies disaster bandwagon. I'd like the scientific debate to be, erm, more scientific... science can't tell us what we should chose for our future - it only illuminates mechanisms... if we want to engage in a debate about what influence we should exert on our own futures, maybe we need to bring in philosophy and ethics. All I can assure you is that I expect no clear cut answers.
Slashdot Mirror
From the "Homomorphic Encryption" page linked from the article:
"Only in 2009 did Craig Gentry of IBM publish a mathematical proof showing fully homomorphic encryption was possible."
In the past (in a very hand-wavy kind of way) I've argued that it should be possible to "prove" that homomorphic encryption isn't feasible... because, in order to implement multiplication and addition of integers, I need a total-ordering over my data... and if I have a total ordering, my data is (effectively) decrypted. This, of course, doesn't preclude obfuscation and scrambling - but I (used to) believe strong homomorphic encryption was not worth investigation.
Does anyone have a reference to this proof? Has anyone read it? Can anyone summarise what Gentry concludes?
I've been thinking about reliability of connectivity quite a bit recently... Using a 3G dongle as a backup is one option - but with this option attracting either a noticable monthly charge or requiring a pre-pay to be renewed every 1 or 3 months... it is a bit frustrating... for a service I hope I never need to use. I'm currently wavering on the brink of taking the plunge - the clincher will be if I find time to convince myself that I can configure automatic fail-over satisfactorily.
The first interesting idea that springs to mind is this: if 0.5mb/s is free, what's stopping me buying N wi-fi dongles and channel-bonding their connections to give a ~N/2 mb/s connection, also for free?
Another interesting idea is that if Virgin had this service when I moved in, they'd have me as a customer... The way things actually panned out, I paid a deposit - they jerked me about and gave me absolutely no clue when I'd be given service - so I told them to sling-their-hook and went with Sky (who proved similarly useless - but eventually provided a DSL line.)
The real losers will surely be the telephone companies. Why bother with a pay-go mobile for texts if you can be connected to the web at 0.5mbps everywhere you go?
What really cripples things is when process is deemed a substitute for understanding the specifics of individual situations - where a one-size-fits-all-problems approach is adopted and imposed - usually by people who have no practical experience with the processes they espouse.
If software development could be successfully reduced to a process, I'd have automated it. Where there's a considerable burden of process, either the process is inappropriate - or developing the software itself is inappropriate as it amounts merely to re-inventing the wheel... an exercise in task creation that benefits no-one.
We should think of software development techniques and apply them judiciously - and the more techniques a developer masters, the wider their skill-set and the better they will adapt to new challenges. The critical question that needs to be asked is this: why is a technique being used and is it providing tangible benefits? If this question can't be adequately answered, everyone involved is wasting their time.
Seagate and Samsung are my favourite two drive manufacturers at the moment... I'd have preferred they remain separate.
If I'm thinking about my data, I want - above all - for it to be reliably stored. With the best will in the world, eventually every drive fails... So... I tend to buy different makes of drives in pairs - from different suppliers... the logic is that it is far less likely that both drives will fail simultaneously - leaving my raid-1 data intact.
If Seagate and Samsung share manufacturing/storage/distribution, then the independence of Seagate and Samsung drives vanishes... forcing me to go to another less-preferred vendor.
I wonder when these consolidations will stop being a good idea? I definitely hope that it will be possible to buy independently manufactured drives in future.
While I accept that the fastest modes of transport are disappearing, I think the mean speed at which people travel during a year is almost certainly higher now than it has been in the past. The change is that a lot more people are travelling pretty fast far more frequently... and, arguably, this is far more useful than a handful of people travelling very fast very occasionally.
I am broadly in favour of nuclear energy - in principle. In practice, I have faith in neither corporate compliance nor government monitoring. Neither entity is equipped, or motivated, to appropriately manage long term risk... and that means you can good as guarantee failures. Hysteria about nuclear contamination, IMHO, has made matters worse - encouraging officials to focus exclusively upon reassuring the public that there is "no risk" at the expense of a focus on restricting and mitigating the consequences of the (ultimately inevitable) eventual accident.
I must take issue with the claim that "The only people who hate on SQL are the people who don't understand databases."
I think you should take some time and read the extensive publications of Chris Date (of Codd and Date International fame) - which, without exception, are extremely critical of SQL (while praising the relational model.)
SQL has its place - but it is far from perfect... and, the 'shackle of compatibility' prevents most of what is counter-productive about SQL from being fixed. NOSQL, while an infant technology, at least shows promise... only by starting afresh can we establish better DBMS strategies. No, I don't think NO-SQL is an alternative to SQL - it is an entirely different approach... one that should be judged on merit in each individual context.
In reality, the situation is far worse than you suggest.
When I discover some scientific principle indirectly (for example, in a book; paper or lecture) it isn't sufficient for me to consider only whether the principle is 'factual'... I need to consider if the language used to describe the principle means the same to me as it does to the author - we necessarily interpret the idea in different contexts. This potential miscommunication, of course, applies not only where I consider others' work... their work, in turn will be significantly influenced by their indirect research and education. At each step there's a very real risk of misunderstanding. As an analogy, in mathematics we now know that truth can only be evaluated reliably subject to unquestionable axioms... at some level, all reasoning about science must at an axiomatic level - and I, for one, am extremely sceptical of any claim that current science adopts appropriate axioms. I think there is a real problem with the relevance of aspects of conventional scientific wisdom that we assume self-evident.
I wholeheartedly agree with your assessment of BT incompetence.
While I commend the effort to find a way around the inevitable onslaught of spit, I note that there are technical drawbacks. The web of trust sounds like a great idea - until you try to use it... then most people find themselves ham-strung by human nature which finds 'vouching' to be an onerous responsibility - especially when the full implications of making errors is not understood. Conversely, a centralised directory almost completely undermines the value of using a decentralised mesh network over conventional technology. Finally - the e-cash suggestion is especially amusing for a project called Freecall. :)
I see value in having software by which groups can self-organise - with their own contact lists - but that is a different proposition to replacing telephones or skype.
The problem I foresee with a distributed system (as I understand Freecall is intended to be) is nuisance calls - the so-called SPIT (SPam over Internet Telephony...)
Traditional phones mainly relied upon cost - and the likes of Skype have the benefit of infrastructure under their control.
Are there any plans to address this (orthogonal) issue, I wonder?
While I don't want to argue that PBKDF2 is unsound, I don't believe you about "zero entropy loss" - if this were true, then general purpose hash functions could all be proven to be perfect hashes when recursively applied to their own output. I don't believe this is the case, but - of course - if you've any evidence, I'd love to see some...
A placebo effect* doesn't fix anything,ever. It makes people feel better subjectively.
There was a time when I'd have thought something similar. Recently, though not from an interest in placebos, but from an interest in allergies, my default opinion has changed.
I'm currently very interested in how the body's immune system is regulated. I was particularly interested in how Cortisol (an anti-inflammatory and immune suppressant hormone) is regulated by the body. According to Wikipedia:
Under normal unstressed conditions, the human adrenal glands produce the equivalent of 35–40 mg of cortisone acetate per day. In contrast to the direct innervation of the medulla, the cortex is regulated by neuroendocrine hormones secreted by the pituitary gland and hypothalamus, as well as by the renin-angiotensin system.
Please forgive my ignorance on these matters, but if the body's immune system is regulated by the hypothalamus - then, perhaps what influences the hypothalamus influences the immune system - allowing a tiny change to have a significant, eventual, biochemical consequence... thus -potentially - linking thought and recovery.
BTW - if anyone can provide a more thorough reference (for a non-specialist) on what influences adrenal function, then I'd be very grateful.
Are there ANY open source key/value stores that support prefix compression?
I *really* want to find a good quality contact management program. By this, I don't mean an address book, per-se, but - rather - an application to help me keep in touch with diverse contacts (private and professional) who, otherwise, I'd quickly forget about.
I need, for example, to keep logs of interactions, so - if I contact them - I can be sure what we last talked about, and when. I'd like it to integrate with my iCal calendar so that I can schedule tasks to follow-up, when - otherwise - I'd forget.
It's extremely frustrating... I can't seem to find anything... and this is one of the very few bits of software I'd willingly pay for, because - to me - it would be invaluable. Can't find anything suitable though...
Thankfully, the Guardian, which has far superior journalism and doesn't seek to ram politics down everyone's throats in "news" stories like News International's papers do
Erm, I think this statement belies your own politics. The guardian is staunchly "New Labour" and I find the vast majority of its reporting to be extremely politically biased. This shouldn't surprise anyone who considers the volume of public sector advertising in this paper.
Second, there's nothing wrong with SQL as a language
I beg to differ - SQL is preposterously baroque!
That said, if you're problem is of a particular kind, it is a perfectly reasonable, practical, solution to many problems.
My two-pence...
I don't think you'll ever find a useful comparison chart - because, while charts exist, they won't focus on what your application needs. Your application is the issue - not the DBMS choice; you need to think about what interactions you need (want) from an application perspective - then look for a solution that meets that specification. A lot depends upon granularity of transactions; on the volumes of data per transaction; on the volumes of data over time; on acceptable latency; on acceptable redundancy and fault tolerance - etc. etc.
The package I'd advise you to look at, is Oracle-Embedded (which has an open source licence, as well as a commercial one) which moves you out of the realm of relational databases all together. If this is a suitable solution, it suggests an entirely new way to think about what sort of back end you really need.
In summary, no-one else's comparison of RDBMS is likely to be relevant to your domain. All such comparisons are extremely fragile in the context of application demand.
I'll never support them and I'll continue to explain to those that ask my advice (I'm one of the go-to technical people for a lot of friends) exactly why I don't like them and suggest competitors.
Go on then - who offers a reliable, honest, cost-effective UK broadband service (preferably with a static IP)?
I'm certain I'd be best served by "a text book" - the question, of course, is which one? Browsing at my local bookshops has turned up nothing worth buying - and I get no further with Amazon et al.
I'd really like a definitive reference book on statistical methods... one that assumes its reader is mathematically able, but only occasionally engaged in statistical work. :)
5) If you know the basics, you know just enough to be dangerous and really piss off people who know what they're doing. Being able to run an anova or determine correlation makes some people think they actually know what's going on because, hey, it's math. But a lot of people who just do the basic stuff think their results are more meaningful than they actually are - falling prey to the whole "it's statistically significant therefore it must be IMPORTANT" fallacy (when you can certainly have things that are "statistically significant" but actually have virtually no impact on the outcome.
Here, I think, you've mentioned a critical issue with statistics. The field of statistics is treated as if it were an omniscient black box by the majority of the population, and the more complex the calculation and the less people understand about the process, the more weight the statistic tends to have on decision making. One idea that interests me is that this might be valid for some 'real life' situations... especially where the people are utterly bamboozled by the statistics but use them to guide their behaviours in the absence of credible alternatives... essentially turning arbitrary choices into self-fulfilling prophecies. I am intrigued as to whether or not such subtlety would allow one to devise statistics that have subversive systemic effects.
I would love to receive a recommendation for a solid statistics reference book. All the books I can find infuriate me - they talk down to the reader and appear to be written for someone who is innumerate... labouring the trivial and obvious - while skimming the non-obvious dismissively. Another thing that frustrates me about statistics texts is that they frequently focus on 'case' studies - encouraging the reader to assume that they can re-apply techniques when they encounter similar-looking situations in real life. This is utterly bonkers - since, without a full understanding of the mechanics of the calculations, it is impossible to determine if the similarity between two situations is an irrelevant detail or crucial to the technique. I'm not interested in 'learning by example' - so I can copy a bunch of people whose views I distrust... I'd like a reference that allows me to use a typical approach in my statistical reasoning rather than re-inventing the wheel myself. I'd like it to be comprehensive and compact - I don't want it to aim to indoctrinate me to think in a particular way in my analyses.
(Can anyone recommend such a book?)
people don't like tablets.
Pray, how do you know that? I've never bought a tablet - and wouldn't buy from any to-date because they don't meet my minimum expectations. Don't be fooled, however, I'd love to use a good tablet PC.
Sticking points for me are :
* High resolution A4-ish screen (to read PDFs page-at-a-time)
* Long battery life (at least a day)
* Wi-fi connectivity and cheap 3G (perhaps via a bluetooth phone) for generic Web access
When technology meets my demands (and I'm sure it eventually will) I'll buy one - as, I suspect, will the rest of the world.
That seems quite an important extension - any idea when (or if) it will be supported by TB3?
To me, it seems like an error of judgement to mainstream release a new version when key addons have not been satisfactorily updated. For the likes of Lightening, it isn't just eye-candy... and, for many, I suspect, breaking existing (addon) functionality will be unacceptable.
That said, I'm looking forward to 'conversation' view - and I've craved an improved address book for years... though what I saw when I last took a peek at the Beta wasn't much better than in TB2.
So, not all change is bad, man-made change can sometimes be really good...but I don't think that's the case on balance here. YMMV, though. Did you have anything concrete that you'd like to point out to support what you were saying? I would be interested to hear it (and not just as a rhetorical point, either).
I'm afraid my arguments are not of the form that would facilitate my explaining future consequences - and justifying in advance why they are desirable. My objection to climate-change-fundamentalism is somewhat more abstract.
A considerable proportion of replies to my post centred on the idea that I might personally prefer different weather in a particular location. This is definitely not what I meant! I reject the idea that the world should be coerced away from (or penalised for using) conventional power sources. There will be a natural shift away from fossil fuels when superior alternatives are available - assuming no half-baked political interventions.
I'm from the UK - and there's a good chance that, if sea levels rise by a few meters, London will flood... and almost 10% of the population live there. Will that be a bad thing? I'm not so sure... perhaps it will prompt a less insane geographic concentration of business - leading to improved quality of life for all? It's very difficult to decide - in advance.
To pnot, I recognise that your (rather banal) suggestion, that I advocate mass starvation, is a fallacious argument. People starve today - but it is not because there is any inherent lack of food - but, rather, a combined issues of politics and finance... issues which are not addressed by climate hysterics.
I'm all for establishing clean, efficient energy sources, but I believe that if change is desirable, it will arise from technological progress and not from repression or efforts to engineer guilt - an effort which only has political motives - motives with which I strongly disagree. Technological progress is best fostered through freedom and transparency - it is most hindered by politicisation and centralisation; by ill-conceived targets and divisive manipulation of objective observations to serve vested interests.
I think the whole 'climate change' debate is bizarre, and I am deeply sceptical about those who use climate change to push tenuously related political agendas.
If prompted for my strongest climate-sceptic view, it is this: while I'm willing to accept that substantial data shows that mankind's activities have resulted in atmospheric change - and while I'm willing to believe that this influences climate, I've one key question: is the change for the worse? Really? How can you be so sure?
I dislike the doom-laden 'climate change will wreck our environment' crowd for one key reason: they can't provide any evidence that I wouldn't prefer the climate after it has changed. Lots of things have been affected by mankind - and, frankly, I prefer to live in the world in which these changes have been made.
While I applaud being economical - and dislike pollution as much as the next sane person... I also think mankind belongs on earth... and I'm not willing to blindly jump on the change implies disaster bandwagon. I'd like the scientific debate to be, erm, more scientific... science can't tell us what we should chose for our future - it only illuminates mechanisms... if we want to engage in a debate about what influence we should exert on our own futures, maybe we need to bring in philosophy and ethics. All I can assure you is that I expect no clear cut answers.