..and you could see a vending machine which would only sell coke light/diet/zero to anyone with a BMI over 30. After all, a company that can create a machine which can sell me calorie-laden drinks and determine my vital statistics, can surely be sued when I develop diabetes and suffer a heart attack 10 years later?
Or maybe in 10 years time individuals will be expected to take responsibility for their own actions?
Sounds an awful lot like the uk data protection act of 1984, which applied to all data, written and electronic, held on an individual.
"Personal information may be kept for no longer than is necessary and must be kept up to date."
"Data must not be disclosed to other parties without the consent of the individual about whom it is about..."
"Entities holding personal information are required to have adequate security measures in place. Those include technical measures (such as firewalls) and organisational measures (such as staff training)."
Looking through the main points of the Act, it makes you wonder why you don't hear more about nefarious data-collecting companies being taken to the courts here in the U.K.
On the recommendation of those here, I have just started trying out org-mode to see if it will help with managing my jobs. It looks promising, but I have a question - how do you get rid of old tasks? I.E. I am using headings for each of the projects I need doing, with sub-headings to divide up the project (as in GTD). I'm labelling the headings as 'TODO's and there are further TODO's inside these headings. When I'm done with a project, how do I archive it?
Yep - similar experiences with my dad and sister - I was expecting loads of trouble, and was surprised to hear nothing from them. A ring a week later and it's 'no problems, you?'.
..that scientific base research (for new pills, procedures, etc.) is not the issue here. This is about studies, i.e. field testing of large numbers of patients, and the (wrong, causation != correlation, etc.) interpretations that are made public afterwards...
Two points: 1. That's why there are different levels of the quality of evidence in medical research
* Ia: systematic review or meta-analysis of RCTs (randomised controlled trials).
* Ib: at least one RCT.
* IIa: at least one well-designed controlled study without randomisation.
* IIb: at least one well-designed quasi-experimental study, such as a cohort study.
* III: well-designed non-experimental descriptive studies, such as comparative studies, correlation studies, case-control studies and case series.
* IV: expert committee reports, opinions and/or clinical experience of respected authorities.
-note that causation/correlation is not an issue with RCTs, whereas they can be problem, obviously, in correlation studies.
2. Your comment suggests that 'scientific based research' is separate to 'field testing'. I would say that testing of the medical intervention is probably the most important part of 'scientific based research'. A medication which works in theory, often can actually cause harm in practice, which is only shown in the medical research. An excellent example is beta-carotene. This anti-oxidant was believed to reduce cancer rates in smokers by neutralising free-radicals. It wasn't until a randomised studies actually showed that it increased cancer in smokers, by a mechanism which is still not fully elucidated. (see J Nutr. 2004 Jan;134(1):262S-268S."The enigma of beta-carotene in carcinogenesis: what can be learned from animal studies.")
Funny enough, until recently, criticising the official results of medical studies was seen as conspiracy theory by those in power in medical circles.
I don't know what this statement is actually based on, but at least where I trained, all doctors were taught to be able to critically review medical research, and most doctors will tell you that they continue to critique papers and present them regularly at local meetings. Also, there is no such thing as 'the official results' - it is the same as any other scientific paper - their conclusions are based on their results, and open to criticism (as they often are).
Well I'm running 10.04 on a 6 year old Thinkpad T40 (1.3Ghz centrino, 32Mb ATI radeon), and it's never been so good!
I will grant you that Win XP is very snappy when newly installed, but 10.04 is still fast and slick, boots up in half the time, and more importantly, remains running fast after 6 months of use, by which point Win XP is crawling towards it's reinstall.
Hard-drive grinding for no apparent reason seems to be a feature of Windows XP and 7 which I've happily lived without for the past two years.
Any chance you could tell me where to find the second of the nouveau driver problems? Have had a quick google around, but can't see it, and I think it might be affecting a Dell Precision with 10.04.
It seems so far to be pretty good. It feels like a slight evolution of 10.04 rather than the huge leap 10.04 was from 9.10. There isn't a huge amount different, but that might explain why it seems much more stable than 9.10 and 10.04 were when they were initially released.
The nicest new feature is the beautiful ubuntu font.
If not Canonical, then I think someone needs to press the EU into pushing open standards for software.
In the UK it seems a lot of pressure has been put on the mobile phone and broadband companies to allow users to easily switch between service providers without suffering a huge penalty. The same goes for gas and electric supplies. I'm sure most people would agree this has been hugely beneficial, both for the service provided, and for allowing other businesses into the market.
The main reason for lack of linux uptake amongst my peers, is simply because it isn't compatible with the industry standard - the closed-source.doc.xls and.ppt files. And no, openoffice.org, while good, does not have seemless compatibility, and is useless for such things as collaborating on files due to the slight differences in formatting.
The fact that one also has to pay Microsoft to buy any mainstream laptop is also, surely, anti-competitive. What it means is that no matter what operating system I choose, Microsoft will receive my money, which will no doubt fund a business practice I don't approve of, and an operating system I don't rate.
Thank goodness for open standards such as html, www, email; twitter, facebook, gmail, googlechat protocols etc etc, or else Linux, for the home user, would be almost useless!
It's not the same thing - on the one hand you're receiving data from the internet, in the other case you are sending data which you are asking the network to notify a gsm phone to receive this data, and to keep trying until the mobile is able to receive it.
I'm actually undecided on whether hunting for food is good or not. I don't have much experience of it myself, apart from having shot and cooked a rabbit once. A friend of mine does not enjoying eating fish if the head is visible - because he doesn't like to remember that he is eating what was once an animal. I very much disagree with this point of view, and believe that as a meat-eater you are responsible for the death (and often life) of your food.
On the one hand it appeals to me that an animal can run free, reproduce and avoid predators, even if they may come to an untimely end from a bullet from a gun. On the other hand we risk lethally injuring an animal - a shameful waste and cause of suffering. However, farming has it's own downsides - we often play god with their reproduction, we remove their selection pressures (so they can't survive in the wild), they live in captivity, and they have no hope of not being killed.
Hunting for pest/population control depends on whether there are humane ways of performing the same function. I do not see anything wrong with the deer hunters employed in the uk to keep numbers down. They will also eat the meat, which is a bonus.
When fox-hunting was banned in the UK, there was a pretty huge outcry from the hunting community, with modest support from locals/country folk etc, and ambivalence from most of the rest of us. However, it quickly became apparent that the only real / main reason to keep hunting going was to continue the 'tradition' and to keep a few people employed. I think at this point public opinion swung in support of the government, and I think most of us haven't looked back since.
I'm all for hunting for food, but hunting for sport just seems gratuitous and disrespectful to nature.
These studies are not new, and doctors have known about the 'J' shaped curve for a while - where abstaining is correlated with high all-cause mortality than moderate alcohol intake, but heavy intake is associated with increased mortality too. However, these studies have only shown a correlation, not a causation, and until we have a randomised control trial (which is unlikely ever to happen) this is unlikely to be proven clearly. There is some suggestion that the correlation may be to do with a confounding factor - which is social drinking. Those drinking 3/day every other day, likely have a good social environment, which is independently linked to improved health. On the other hand, the cardiologist at our local hospital believes it is a true effect of alcohol and prescribes two glasses of wine to all patients admitted with a myocardial infarction/heart attack.
You might be right, but I wouldn't hold your breath:
"Lets talk about image quality, right from the first time I got to take samples home from a D30 shooting session (back in August) I was stunned and amazed at the purity of the images, they're so clean and smooth yet not lacking in detail, this grainless look goes well beyond film quality to a new arena of high quality "scene digitization" which captures every detail of the scene without any noise or stray artifacts. Colour rendition is very good as is pure resolution (as measured by our test charts)" dpreview 2000
That was back in 2000 on a 3MP camera from dpreviews review on the Canon D30.
And interestingly, it seems like the website has edited their original review on the Nikon D1, which I'm pretty sure concluded at the time that the D1 produced images of higher quality than scanned film. I imagined they edited this to save the continued embarrassment from the continued linking to it.
We'll see what happens, but I must admit that I'll be annoyed if a lack of research into film, and future advanced film types is stunted by stupid conclusions about digital image quality.
Well I'm one of those 'still clinging onto that idea'.
I'll take your example - can you show me an artistic photograph that uses this photo-stacking technique? As a keen photographer myself, I do not see the point of it - it would require using a tripod in order that the images are perfectly aligned, if that is the case, then a long shutter speed combined with a small aperture would achieve the same, and without any artefacts due to the slight changes in focal length seen in many lenses when refocused.
I think the public perception of film is stunted by the cheap ways film used to be dealt with in your local processing lab - processed in a non-dust free environment, and scanned by poor quality machines with poor quality operators. In actual fact 'flim' itself has a tremendous capacity to capture information, and if one is willing to take a small amount of effort to maximise information obtained from the film, one would find very high resolution (35mm captures around the equivalent of a 24mp dslr - see link below), excellent dynamic range, which has a curved shoulder allow colours to fade smoothly into white when overexposued, tonality (see the 7D versus fuji velvia - it's not just the resolution, but also the colour accuracy and colour resolution).
Now please do not reply with one of the comparisons which confirmed in a large number of peoples minds that digital was superior, without 1st quoting what scanner was used to do the comparison, and if it isn't a drum scanner, you are already standing on shaky ground.
Duncan.
P.S. Of course buying a Nikon D3X is more convenient and probably cheaper than using a 35mm film camera and sending your photos off for drum scanning, but that is not what we're discussing. I have no problem with people stating that digital is cheaper and more convenient with quality nearly up there with film.
It depends whether you are talking about media designed to represent art or information. Of course, if you were using film to make copies of documents, then it's behaviour beyond it's resolution limit is of no interest, as that is information lost.
However, lets take a photograph taken on a 35mm format camera, at, say ISO 1600, enlarged to 20"x20" - most photographs would say that this is technically beyond what film or digital at this size can do. But here is such a photograph by Testino, which was shown at this size at the National Portrait Gallery:
You can see from even this small image the amount of grain in the photograph - imagine how much grain the fullsize print had. However, this print was one of the most impressive in a room full of portraits by Testino, Rankin, etc..
When you take photographs at the limit of the media's capability, and then enlarge it to large sizes, the artefacts become crystal clear, which is a problem if they are ugly.
With digital expect to see: moire (an interference pattern created by the precise pattern of the sensor), bayer artefacts (similar to moire, except in two dimensions - caused by the algorithm used to interpolate brightness values from the RGB sensors), noise reduction artefacts (seen as a smoothing of areas with little detail - the problem occurs with natural subjects, where the decision to smooth is not intelligent, and you lose information in order to lower the 'noise' at high ISOs), sharpening artefacts (to counteract the softening of noise reduction - except it won't know the size you are enlarging to, so it is often unhelpful), noise (of a multi-colour sort, seen by most people as undesirable). I won't go on, but I left out chromatic aberrations and blooming. Admittedly, with better sensors, these are improving.
Here is an example I have taken where grain is an integral part of the image. Apart from the grain, this image suffers from no other artefacts that I can see:
If you go back to my original posting, you'll see that I state right off that many comparisons compare digital with scanned film - the one you have linked is a good example. At most you can say that film scanned with a Polaroid Sprintscan 4 has worse dynamic range than digital. But that is not what we're discussing here.
Clarkvision's site has been around for a while, so I am well aware of his tests. I think some of them have some important flaws. One of his most linked graphs compares resolution with ISO of both film and digital, unfortunately he measures resolution in MP, rather than tested resolution, therefore concluding that a 20MP camera at 3200 has as much resolution as the same sensor at 100. I'm not sure if this has been updated. I also cannot remember how the film was scanned.
Here is a link comparing film scanned with a half-decent scanner (Nikon Coolscan), though still not professional grade (i.e. not a Heidelberg Tango drum scanner) with digital:
You keep saying that we are discussing the technical limits of the media here, but if you are comparing media, you need to take all aspects of it into consideration, and I think how a medium performs beyond its limits is important aswell.
Slashdot Mirror
..and you could see a vending machine which would only sell coke light/diet/zero to anyone with a BMI over 30. After all, a company that can create a machine which can sell me calorie-laden drinks and determine my vital statistics, can surely be sued when I develop diabetes and suffer a heart attack 10 years later?
Or maybe in 10 years time individuals will be expected to take responsibility for their own actions?
Sounds an awful lot like the uk data protection act of 1984, which applied to all data, written and electronic, held on an individual.
"Personal information may be kept for no longer than is necessary and must be kept up to date."
"Data must not be disclosed to other parties without the consent of the individual about whom it is about..."
"Entities holding personal information are required to have adequate security measures in place. Those include technical measures (such as firewalls) and organisational measures (such as staff training)."
http://en.wikipedia.org/wiki/Data_Protection_Act_1984#History
Looking through the main points of the Act, it makes you wonder why you don't hear more about nefarious data-collecting companies being taken to the courts here in the U.K.
On the recommendation of those here, I have just started trying out org-mode to see if it will help with managing my jobs. It looks promising, but I have a question - how do you get rid of old tasks? I.E. I am using headings for each of the projects I need doing, with sub-headings to divide up the project (as in GTD). I'm labelling the headings as 'TODO's and there are further TODO's inside these headings. When I'm done with a project, how do I archive it?
That is brilliant. I'm replying to it so that I have a copy of it somewhere on my account.
Yep - similar experiences with my dad and sister - I was expecting loads of trouble, and was surprised to hear nothing from them. A ring a week later and it's 'no problems, you?'.
..that scientific base research (for new pills, procedures, etc.) is not the issue here. This is about studies, i.e. field testing of large numbers of patients, and the (wrong, causation != correlation, etc.) interpretations that are made public afterwards...
Two points: 1. That's why there are different levels of the quality of evidence in medical research
* Ia: systematic review or meta-analysis of RCTs (randomised controlled trials).
* Ib: at least one RCT.
* IIa: at least one well-designed controlled study without randomisation.
* IIb: at least one well-designed quasi-experimental study, such as a cohort study.
* III: well-designed non-experimental descriptive studies, such as comparative studies, correlation studies, case-control studies and case series.
* IV: expert committee reports, opinions and/or clinical experience of respected authorities.
-note that causation/correlation is not an issue with RCTs, whereas they can be problem, obviously, in correlation studies.
2. Your comment suggests that 'scientific based research' is separate to 'field testing'. I would say that testing of the medical intervention is probably the most important part of 'scientific based research'. A medication which works in theory, often can actually cause harm in practice, which is only shown in the medical research. An excellent example is beta-carotene. This anti-oxidant was believed to reduce cancer rates in smokers by neutralising free-radicals. It wasn't until a randomised studies actually showed that it increased cancer in smokers, by a mechanism which is still not fully elucidated. (see J Nutr. 2004 Jan;134(1):262S-268S."The enigma of beta-carotene in carcinogenesis: what can be learned from animal studies.")
Funny enough, until recently, criticising the official results of medical studies was seen as conspiracy theory by those in power in medical circles.
I don't know what this statement is actually based on, but at least where I trained, all doctors were taught to be able to critically review medical research, and most doctors will tell you that they continue to critique papers and present them regularly at local meetings. Also, there is no such thing as 'the official results' - it is the same as any other scientific paper - their conclusions are based on their results, and open to criticism (as they often are).
Er... And Amiga OS?
Whether this also affects Linux users, and secondly, how does one configure java/flash/their browser etc, to prevent this happening?
Well I'm running 10.04 on a 6 year old Thinkpad T40 (1.3Ghz centrino, 32Mb ATI radeon), and it's never been so good!
I will grant you that Win XP is very snappy when newly installed, but 10.04 is still fast and slick, boots up in half the time, and more importantly, remains running fast after 6 months of use, by which point Win XP is crawling towards it's reinstall.
Hard-drive grinding for no apparent reason seems to be a feature of Windows XP and 7 which I've happily lived without for the past two years.
YMMV
Any chance you could tell me where to find the second of the nouveau driver problems? Have had a quick google around, but can't see it, and I think it might be affecting a Dell Precision with 10.04.
Thanks
It seems so far to be pretty good. It feels like a slight evolution of 10.04 rather than the huge leap 10.04 was from 9.10. There isn't a huge amount different, but that might explain why it seems much more stable than 9.10 and 10.04 were when they were initially released.
The nicest new feature is the beautiful ubuntu font.
If not Canonical, then I think someone needs to press the EU into pushing open standards for software.
In the UK it seems a lot of pressure has been put on the mobile phone and broadband companies to allow users to easily switch between service providers without suffering a huge penalty. The same goes for gas and electric supplies. I'm sure most people would agree this has been hugely beneficial, both for the service provided, and for allowing other businesses into the market.
The main reason for lack of linux uptake amongst my peers, is simply because it isn't compatible with the industry standard - the closed-source .doc .xls and .ppt files. And no, openoffice.org, while good, does not have seemless compatibility, and is useless for such things as collaborating on files due to the slight differences in formatting.
The fact that one also has to pay Microsoft to buy any mainstream laptop is also, surely, anti-competitive. What it means is that no matter what operating system I choose, Microsoft will receive my money, which will no doubt fund a business practice I don't approve of, and an operating system I don't rate.
Thank goodness for open standards such as html, www, email; twitter, facebook, gmail, googlechat protocols etc etc, or else Linux, for the home user, would be almost useless!
Duncan
It's not the same thing - on the one hand you're receiving data from the internet, in the other case you are sending data which you are asking the network to notify a gsm phone to receive this data, and to keep trying until the mobile is able to receive it.
Just to clarify my slightly on-the-side remark...
I'm actually undecided on whether hunting for food is good or not. I don't have much experience of it myself, apart from having shot and cooked a rabbit once. A friend of mine does not enjoying eating fish if the head is visible - because he doesn't like to remember that he is eating what was once an animal. I very much disagree with this point of view, and believe that as a meat-eater you are responsible for the death (and often life) of your food.
On the one hand it appeals to me that an animal can run free, reproduce and avoid predators, even if they may come to an untimely end from a bullet from a gun. On the other hand we risk lethally injuring an animal - a shameful waste and cause of suffering. However, farming has it's own downsides - we often play god with their reproduction, we remove their selection pressures (so they can't survive in the wild), they live in captivity, and they have no hope of not being killed.
Hunting for pest/population control depends on whether there are humane ways of performing the same function. I do not see anything wrong with the deer hunters employed in the uk to keep numbers down. They will also eat the meat, which is a bonus.
All the above sentiments.
When fox-hunting was banned in the UK, there was a pretty huge outcry from the hunting community, with modest support from locals/country folk etc, and ambivalence from most of the rest of us. However, it quickly became apparent that the only real / main reason to keep hunting going was to continue the 'tradition' and to keep a few people employed. I think at this point public opinion swung in support of the government, and I think most of us haven't looked back since.
I'm all for hunting for food, but hunting for sport just seems gratuitous and disrespectful to nature.
Duncan.
Well, at 3 weeks prior to the most important professional exam of my career, I appear to be posting on Slashdot.
I hereby donate my brain to medical science so that the lesion present in my prefrontal cortex can help pinpoint this area more precisely.
I would say you are enjoying the benefits of herd-immunity. You might not find life to be so rosy if everyone were to act in the same way.
chances are with a few minutes of research you are smarter than your doctor...
A little bit of knowledge can be a dangerous thing...
These studies are not new, and doctors have known about the 'J' shaped curve for a while - where abstaining is correlated with high all-cause mortality than moderate alcohol intake, but heavy intake is associated with increased mortality too. However, these studies have only shown a correlation, not a causation, and until we have a randomised control trial (which is unlikely ever to happen) this is unlikely to be proven clearly. There is some suggestion that the correlation may be to do with a confounding factor - which is social drinking. Those drinking 3/day every other day, likely have a good social environment, which is independently linked to improved health. On the other hand, the cardiologist at our local hospital believes it is a true effect of alcohol and prescribes two glasses of wine to all patients admitted with a myocardial infarction/heart attack.
You might be right, but I wouldn't hold your breath:
"Lets talk about image quality, right from the first time I got to take samples home from a D30 shooting session (back in August) I was stunned and amazed at the purity of the images, they're so clean and smooth yet not lacking in detail, this grainless look goes well beyond film quality to a new arena of high quality "scene digitization" which captures every detail of the scene without any noise or stray artifacts. Colour rendition is very good as is pure resolution (as measured by our test charts)"
dpreview 2000
That was back in 2000 on a 3MP camera from dpreviews review on the Canon D30.
And interestingly, it seems like the website has edited their original review on the Nikon D1, which I'm pretty sure concluded at the time that the D1 produced images of higher quality than scanned film. I imagined they edited this to save the continued embarrassment from the continued linking to it.
We'll see what happens, but I must admit that I'll be annoyed if a lack of research into film, and future advanced film types is stunted by stupid conclusions about digital image quality.
Well I'm one of those 'still clinging onto that idea'.
I'll take your example - can you show me an artistic photograph that uses this photo-stacking technique? As a keen photographer myself, I do not see the point of it - it would require using a tripod in order that the images are perfectly aligned, if that is the case, then a long shutter speed combined with a small aperture would achieve the same, and without any artefacts due to the slight changes in focal length seen in many lenses when refocused.
I think the public perception of film is stunted by the cheap ways film used to be dealt with in your local processing lab - processed in a non-dust free environment, and scanned by poor quality machines with poor quality operators. In actual fact 'flim' itself has a tremendous capacity to capture information, and if one is willing to take a small amount of effort to maximise information obtained from the film, one would find very high resolution (35mm captures around the equivalent of a 24mp dslr - see link below), excellent dynamic range, which has a curved shoulder allow colours to fade smoothly into white when overexposued, tonality (see the 7D versus fuji velvia - it's not just the resolution, but also the colour accuracy and colour resolution).
http://photo.net/film-and-processing-forum/00WErk?start=200
This article looks more at the non-resolution aspects of film:
http://www.twinlenslife.com/2009/05/digital-vs-film-real-deal-nikon-d300-vs.html
And Ken Rockwell, as much as he says things clearly thinking it through, has an excellent article with many more valid points here:
http://www.kenrockwell.com/tech/why-we-love-film.htm
Now please do not reply with one of the comparisons which confirmed in a large number of peoples minds that digital was superior, without 1st quoting what scanner was used to do the comparison, and if it isn't a drum scanner, you are already standing on shaky ground.
Duncan.
P.S. Of course buying a Nikon D3X is more convenient and probably cheaper than using a 35mm film camera and sending your photos off for drum scanning, but that is not what we're discussing. I have no problem with people stating that digital is cheaper and more convenient with quality nearly up there with film.
Looks like we're seeing a new loss of confidence in Java, much like the loss of confidence in mono, for which patent concerns stunted its uptake.
So where to next?
And where is my replacement for open office?
Muprhy's Law in action...
It depends whether you are talking about media designed to represent art or information. Of course, if you were using film to make copies of documents, then it's behaviour beyond it's resolution limit is of no interest, as that is information lost.
However, lets take a photograph taken on a 35mm format camera, at, say ISO 1600, enlarged to 20"x20" - most photographs would say that this is technically beyond what film or digital at this size can do. But here is such a photograph by Testino, which was shown at this size at the National Portrait Gallery:
http://www.pbase.com/omoses/image/103309345
You can see from even this small image the amount of grain in the photograph - imagine how much grain the fullsize print had. However, this print was one of the most impressive in a room full of portraits by Testino, Rankin, etc..
When you take photographs at the limit of the media's capability, and then enlarge it to large sizes, the artefacts become crystal clear, which is a problem if they are ugly.
With digital expect to see: moire (an interference pattern created by the precise pattern of the sensor), bayer artefacts (similar to moire, except in two dimensions - caused by the algorithm used to interpolate brightness values from the RGB sensors), noise reduction artefacts (seen as a smoothing of areas with little detail - the problem occurs with natural subjects, where the decision to smooth is not intelligent, and you lose information in order to lower the 'noise' at high ISOs), sharpening artefacts (to counteract the softening of noise reduction - except it won't know the size you are enlarging to, so it is often unhelpful), noise (of a multi-colour sort, seen by most people as undesirable). I won't go on, but I left out chromatic aberrations and blooming. Admittedly, with better sensors, these are improving.
Here is an example I have taken where grain is an integral part of the image. Apart from the grain, this image suffers from no other artefacts that I can see:
http://photo.net/photodb/photo?photo_id=11324330
If you go back to my original posting, you'll see that I state right off that many comparisons compare digital with scanned film - the one you have linked is a good example. At most you can say that film scanned with a Polaroid Sprintscan 4 has worse dynamic range than digital. But that is not what we're discussing here.
Clarkvision's site has been around for a while, so I am well aware of his tests. I think some of them have some important flaws. One of his most linked graphs compares resolution with ISO of both film and digital, unfortunately he measures resolution in MP, rather than tested resolution, therefore concluding that a 20MP camera at 3200 has as much resolution as the same sensor at 100. I'm not sure if this has been updated. I also cannot remember how the film was scanned.
Here is a link comparing film scanned with a half-decent scanner (Nikon Coolscan), though still not professional grade (i.e. not a Heidelberg Tango drum scanner) with digital:
http://photo.net/film-and-processing-forum/00Rb8S
You keep saying that we are discussing the technical limits of the media here, but if you are comparing media, you need to take all aspects of it into consideration, and I think how a medium performs beyond its limits is important aswell.