It's a small MacBook that uses the iPhone as the touchpad and for its radios. There is an app that displays the iPhone screen on the larger display. The PhoneBook runs regular MacOSX so all regular Mac apps work and all iPhone apps work, too.
Connectivity in the PhoneBook complements what is in the iPhone, and there are slots for 16MB+ SD cards to expand storage.
You can undock the iPhone to use it when you don't want or need the functionality of the entire PhoneBook.
There are standards of identity (SOIs) for different types of chocolate, including white, milk, and sweet (which covers both semisweet and bittersweet although there is no legal distinction between the two). The minimum COCOA CONTENT in sweet chocolate is specified as "15 percent by weight of chocolate liquor" and for semi and bittersweet as "not less than 35 percent by weight of chocolate liquor." Chocolate liquor is ground up cocoa beans whose fat content is between 45-55% depending on the species of bean and where the bean is grown.
Cocoa content is NEVER specified in terms of solids (cocoa powder) content -- which is where the original poster got his/her facts wrong (this is a very common misunderstanding). Cocoa content refers to the total amount of cocoa solids and cocoa butter in a chocolate product. If you look closely at the ingredients label you'll see that most chocolates have cocoa liquor (or chocolate liquor or cocoa mass) plus added cocoa butter. Cocoa content is expressed as a percentage, by weight, of the combination of those two ingredients.
The FDA SOIs for all chocolates (white, milk, sweet, and liquor) allow manufacturers to include dairy fats and other milk ingredients (including whey) in their products. This is done to reduce the cost of manufacturing (which is what the FDA petition is ** really ** all about) as well as stabilize the cocoa butter crystals to extend the shelf life of the chocolate without adding preservatives.
There are a number of profound ironies here. Irrespective of the subjective taste impressions of American vs European chocolate, the fact is that American standards for what manufacturers can put in chocolate -- and still call it chocolate -- are stricter here than in the EU. The EU SOIs for chocolate allow manufacturers to replace up to 5% of the cocoa butter in chocolate with 'cocoa butter equivalent' and/or 'cocoa butter replacement' fats and still call it chocolate.
Manufacturers can't do that in the US but that is what the CMA and its fellow lobbying organizations are petitioning the FDA to do -- but they want to go much, much, further. They want the ability to replace 100% of the cocoa butter with other fats and still call it chocolate.
It is also important to recognize the difference, in this debate, between chocolate manufacturers and chocolatiers and candy makers. A chocolate manufacturers makes chocolate products (chocolate liquor, cocoa powder, cocoa butter, and finished chocolate) by processing cocoa beans. Chocolatiers buy chocolate from chocolate manufacturers and melt it down to make their finished products. There are relatively few companies in the world that both manufacture chocolate and are also chocolatiers though there are more chocolate manufacturers that also make candy.
If you look at the ingredients on a candy bar or on a box of truffles you'll see that chocolate is listed as an ingredient and -- if it's "good" (or "pure" or "real") chocolate, the only fat in the chocolate will be cocoa butter. All those other ingredients are not in the chocolate -- they're what's inside the outer chocolate coating.
Clay Gordon Editor and publisher www.chocophile.com Author Discover Chocolate (Gotham Books, October 2007)
The micromirrors they are talking about are probably already in production: think of the micromirror array chips that TI sells for video displays and projection -- DLP.
TI's DLP chips can actually be used both ways (as projectors as well as part of a sensor) and I know of a company that has experimented with using them in a tunable sensor that can survey a scene in different spectra sequentially in order to obtain a very accurate profile of what's in front of the sensor.
Think homeland security applications for remote sensing for potential water contamination, satellite surveillance, etc.:SI
The comparison about prices misses the mark because of the differences in form factor. There are places that the memory spot chip can be embedded where a smart card cannot. Done correctly, the memory spot chips will not draw attention to themselves, either. Obfuscation is an often-ignored aspect of securing data.
The FDA has a standard that's part of SRC 21 for medical device compliance -- 510(k). At its most basic level, it's a documentation process -- you have to document every market requirement, part spec, manufacturing prtocess, QA process, as well as the history of every part that goes into a device including who ordered it, when it was ordered, who it was ordered from, who received it on the loading dock, who accepted it, who did testing against spec, what device it went into, if there was ever a complaint filed on the device, if a part was ever returned for repair, and the disposition of all returned parts, etc, and so and so forth ad infinitum.
In case someone ever gets hurt by the device in use it's possible to figure out who okayed the decision(s) that lead to the alleged defect(s) in the product.
There is a COTS evaluation process, but it's pretty straightforward compared with the process for designing and developing the software that goes into a medical device.
How do I know this? I worked on the development of a software system to automate 510(k) compliance for a medical device manufacturer. I was responsible for translating all of the procedures in the Quality System into functional specs and UI designs. Along the way we learned that most companies do not have automated documentation systems for 510(k) compliance because it's far too complicated and there is no FDA spec for designing and implementing an electronic 510(k) compliance system. There used to be one but it was withdrawn because it was too complex.
What I can say is that the Software Development process covered in our Quality manual requires a level of detail I have never seen in any previous software development management system. Not only must every module and its inputs and outputs be documented, but algorithms must be documented. It must be possible (in case of an FDA audit) to trace customer requirements through to market specs to functional specs to technical specs to code, complete with test scenarios and procedures to ensure that the software sctually does what the design documents say it should do -- and that any deviation from specs discovered during testing are resolved.
The irony of all this is that the computer system we developed has too much information in it and FDA auditors aren't used to auditing the content in electronic systems. So, the system outputs paper reports that are put into file folders for the FDA to review if it's ever necessary.
We are in the midst of a complete reno of our 90 year-old house and did exactly this. We're not living through the reno and in the rental house we're in we're using 11g connnections because we can't do any installation that will mar walls. The WiFi connections are turning out to be surprisingly unreliable - won't connect consistently and the speed of the connection is wildly variable.
So, while the walls were open we decided to bite the bullet and run cable; 2 Cat5E and two video coax to 17 different locations in and outside the house. 10/100 switch plus ethernet and video patch bays in the basement. More reliable, more secure, too, and I can run video over the coax and phone signals over the second ethernet pair to any place in the house. Old computers are being used for media servers and hardware firewalls. Mac Mini with Elgato TV-Eye for PVR/iTunes for audio.
I agree with taking pictures and also suggest an overlay on the 'prints to help locate things. We're a test neighborhood for Verizon fiber to the home and when it's installed we're talking about a 20+mb downstream connection.
One of the reasons that EMR (Electronic Medical Record) systems have failed to achieve significant market acceptance is that they are not designed to be used by physicians in the clinical setting, e.g., in the exam room.
In virtually all instances, it looks like the systems have been designed to make it easy to submit insurance claims. The doctor needs to know about and enter too much extraneous data and can't concentrate on important stuff, such as how the technology will improve the quality of patient care and doctor/patient interaction.
It's my sense that the big companies in the consortium don't care one hoot for really improving the quality of medical care. It's been estimated that it will take something close to $234 billion over the next 10 years to implement the system. Even Bill takes notice when $234 Billion is at stake.
Where's the money for that going to come from with no new taxes? I know! My insurance premiums are going to go up -- and still the system won't work, not because it's not interoperable but because it changes the way the doctor works. Of course the insurance companies love it because it pushes even more responsibility for claims processing over to the doctor.
I think we all know the probability of success of a system that forces someone to change the way they work in order to use it - especially when there is little or no perceived benefit.:SI
Based on two years' experience in HD production in the early '90s, the data rate they are talking about is being recorded to tape, not to disk, which does make a real difference as there are no portable 3.3GB/sec sustained data rate disk arrays that you can connect to a camera and use in real production situations. (In 91, there were no portable HD VTRs either, and this is a lesson that they will not overlook as long this time.)
In addition to any DCT compression, they are probably also using color space conversion which provides compression but in a different way that is based on the way the eye sees and not on image blocks.. The cameras are certainly 3-chip RGB but the color space conversion would be similar to YPbPr (y= luminance, or green which is recorded at full bandwidth with the Pb and Pr color difference (y-b, y-r) signals recorded at lower bandwidth). When all three colors are recorded full bandwidth it's denoted 4:4:4; typical color space conversion with compression (before any compression using a DCT) would be denoted 4:2:2. The 4s and 2s refer to the sampling frequency as a multiple of the base frequency. (NTSC color conversion is called YUV, which is another algorithm, but the y is still luminance.)
Ever wonder why 59.94/60 was chosen in the first place? Remember this was back in the 1930s.
Well, it turns out that this is the frequency of the current coming out of wall sockets. Everything was slaved to that frequency to make things easier, electonically speaking.
The 4:3 aspect ratio of video is similarly mired in tradition. Turns out that the original 16mm silent films were shot 4:3 so when video came around they needed material to transfer and film was it.
The 3:2 pulldown in non-drop transfer from video/film is a pain but well understood. Harder to manage was non-drop in 60/59.94 transfers from HD to NTSC.
BTW. HDTV is really an origination medium or for theatrical distribution. The average consumer in the home would be very happy with widescreen D1 (component) and to make it very large, a line-doubler is perfect. The roadblocks to HD distribution are political (i.e., who benefits from the free spectrum and how), not technical. Widescreen D1 could be broadcast over existing towers and transmitters with only very mild compression, no new expensive distribution infrastructure required.
This is news? I remember visiting the Visible Language Workshop at MIT in 1983 and seing a far more capable device built by Professor Ron MacNeil. IIRC, they were programming everything in PL/1 in those days.
In this day and age, building your own counting app is a very naive suggestion. One of the reasons companies use third parties is that their methodology is "independent" and "consistent" across sites. Nobody knows how your homegrown app works, what it counts, or how it counts it. Therefore, nobody will believe your numbers, even if you can "prove" they are correct. Certainly no ad agency media planner who has to justify spending the dollars on the buy -- and that's the gatekeeper, some young kid protecting their job who is not going to go out on a limb because you think you can code a better app and save your company some bucks; that is not the point. It's sort of like the Nielsen ratings. everybody knows the methodology is flawed, but the flaws are equally advantageous and disadvantageous to everyone, so they use it because they can't agree on anything better.
What I want to know is what Amazon does with the commissions they are no longer paying to Morpheus. We have to assume that the actual affiliate referrer is lost in the process. (Maybe this is how Amazon is able to post a profit! - Hah!)
I suggest putting the money into an escrow account and a) distributing it to affiliates on some sort of basis, or, b) donating it to an organization that will take the perps to court.
The observation that 480 widescreen is "good enough" for most viewers in most viewing situations is right on the mark (as I said in this earlier comment on this subject).
What's missing from the original post is the understanding that the greatest contributor to increased picture quality is not the increased resolution HDTV affords (especially on screens smaller than about 8 feet diagonal), but the change in color space. NTSC was designed in the 1950s to enable black & white television sets to display a black & white image even if the signal being broadcast had color encoded in it. In order to do this, the color information has the bejeezus compressed out of it which is why it looks so lousy.
The single most important change that can be made to improve the quality of broadcast television is NOT to increase resolution, but to start broadcasting a component (e.g., YPrPb) signal while ensuring that the entire production chain, from origination through production to distribution and reception is component end to end.
When coupled with a widescreen aspect (a feature of most modern professional cameras), the component signal can be easily broadcast over existing equipment, or with minor and comparatively inexpensive transmitter upgrades. Most importantly, there is no incremental increase in cost to produce programming in widescreen D1 as there is in HDTV. Finally, monitors/receivers/decoders are much, much, cheaper.
But -- even this is not the issue. It's not about (and never was about) making it easy for consumers. It's about broadcasters wanting free spectrum without the onerous requirement of "wasting" it by having to broadcast HDTV all day long. The spectrum allotted for HDTV broadcast is enough to simultaneously broadcast 6, widescreen D1 streams. Now, instead of having one station in a market, a broadcaster can have 6 -- or rent one or more of the channels to others for other uses.
It's politics, always has been. Probably always will be. {sigh}
Clay
HDTV is a production medium, not for distribution
on
To HDTV or Not to HDTV?
·
· Score: 3, Insightful
I spent 1991-2 working for an HDTV production company here in the US, and then consulted for several years, writing the first book on HDTV production (The Guide to High Definition Video Production: Preparing for a Widescreen World; Focal Press 1996, ISBN-0-240-80265-9).
HDTV was designed to deliver equivalent resolution, to the screen, as the best 35mm projection. (Although film starts out with much higher resolution, by the time it goes through all of the processing steps, each with mechanical transports, and then projection, the final resolution is quite low.)
A minimum 36" diagonal monitor is required to display every pixel in the image. However, the full effect of HDTV is not apparent until the image is at least 8 feet (diagonal measure). Furthermore, the original color space demands very low light level conditions in order to appreciate fully. Finally, the minimum viewing distance should be 3x the height of the screen, in order to make the pixels/grain disappear, so you'd need to sit at least 15 feet away from the image.
There are very few viewing environments that meet these conditions -- the average American home is certainly not one.
From a production viewpoint, HDTV requires film-level production standards. The sloppiness that NTSC and PAL encourage (and hide) is clearly visible to HDTV production processes. As my old boss, Barry Rebo, used to ask, "What does HDTV bring to 'Wheel of Fortune'?" The answer is absolutely nothing, except increased production costs.
The political stalemate in Washington is a blind for the desires of broadcasters to capitalize on the offer of free spectrum. The technical arguments are about how a broadcast signal will degrade in fringe reception areas. This is a smokescreen designed to delay introduction until terms are more favorable to broadcasters.
In fact, there is a production medium today that can be broadcast today over existing equipment with only marginal incremental costs to broadcasters: widescreen D1. It's the appropriate aspect ratio (16:9) and is a component system from beginning to end, which is the major contributor to quality.
This format could be delivered more or less right away and would deliver picture quality far in excess of what people expect today, and is, in fact, perfectly appropriate to the broadest range of viewing environments. Importantly, the incremental cost of production in D1 is virtually zero making it very attractive from that standpoint.
Finally, it is very easy to downconvert an HDTV-originated image to D1, especially because the timebase is the same (60 frames (or fields)/second as opposed to 29.94 frames/sec for NTSC).
HDTV is a perfect production medium and is great for projection in controlled viewing environments. For mass distribution D1 (or, more particularly, widescreen component) is far more appropriate, with or without line-doubling.
John Dewey, the pre-eminent American philospher wrote in the 1930s that "art" is the process that the maker goes through in making "art objects" as well as the relationship that viewer has, with and in his or her own experiences, with the art object. Art objects (the objects of the artistic process), have no intrinsic value. Their value is only as a signpost of the expression of experience that the creator has, and any response(s) that a viewer has. The fact that art is possible at all is because of the common, shared, experiences between the creator and the viewer.
What this means is that "art" is whatever the creator calls "art" because, not of their intent or their motivation, but because the process of creation generates aesthetic satisfaction for the maker. That's all that is needed for something to be called art.
At the same time, art is also the response a viewer has with an object. If a person has a genuine aesthetic response to an object, it it not for another person to deride or minimize that response, based on the object itself, or any other knowledge that they might have. It's art to the viewer because it moved him or her.
What does this mean for Aaron? I think it fails the first, or "personal aesthetic satisfaction" test. I don't remember (I read the book in the mod-1980s and there's a copy in storage somewhere) that Aaron is motivated to make drawings because it feels good about the drawings it makes, or that, through some feedback mechanism, it feels "better" about some drawings than others because they more accurately represent its feelings than others.
However, if the drawings that Aaron generates evoke genuine aesthetic responses in viewers, then those responses can be called art. Aaron, or course, is completely oblivious to these responses and can't use them to "improve" its art.
At least, so far. It's an interesting experiment. I read Asimov's story many years ago and have always been struck by its irony and message -- that the best art is made by people who need a little adjustment.
What the author of the parent to this post did not include in his(?) comment was that Model204 (from Computer Corporate of America only runs on IBM S390 mainframes. While there is no way I would trust mission critical data to MySQL (because it does not pass the ACID test), I can't help but wonder what kind of combined throughput I could wring out of MySQL would if I was running a couple of thousand copies of Linux on an S390?
How about EnFuzion from TurboLinux? I've never had the opportunity to use it, but I came across it while doing related research for a financial services client who needed to perform some pretty hairy analyses. Said analyses could wait until after the markets closed and so could take advantage of literally thousands of otherwise idle machines. (The client ultimately decided to outsource the analyses, so we never pursued this approach.)
Key features of EnFuzion (these are lifted from a product marketing page so I can't comment on their veracity) include:
No need to rewrite or modify code.
Maximize system performance with multiple load monitoring and resource-sharing options.
Highly secure: EnFuzion's advanced security features bolster the standard security provided by Unix and Windows NT systems.
Easily integrate EnFuzion into your applications with EnFuzion's application programming interface (API).
Monitor and control EnFuzion execution using a familiar language such as C, Java, Perl, Bash, Tcl/Tk, etc.
Easily integrate EnFuzion with any shell or scripting language.
On the surface this would seem to fit wiredog's needs.
Juno Can Require Your Computer Be On 24/7?
on
Juno And Privacy
·
· Score: 2
Could this be the root cause of California's energy crisis? All those Juno screen savers calculating away 24 by 7? We already know the answer and it's 42. Bad Juno.
The poster missed one of the main points of the disposable phone, which is that, in order to manufacture it, an entirely different process is used. You simply can't use surface-mount components in this application, so there shouldn't be much, if any impact on the demand for conventional components. There's more information on (what I think is) the manufacturing process at the Rolltronics site. Rolltronics has patents on what they call roll-to-roll manufacturing technology. They appear to be the similar techniques.
A colleague and friend of mine who authored role-playing games in the late '80s told me that there are only two story lines: 1) A stranger comes to town, and 2) You go on a trip. While simplistic, there is an element of truth in this.
More academically, the dramatic theorist Georges Polti posited that there are only 36 Dramatic Situations, and all of literature and drama can be decomposed in a combination of these. The book is out of print, but here's a link to Amazon to find out more if you're interested. A Google search turns up some interesting sources, for those who want to know more.
As in Cover Your Ass. I was the tech director of an ad rep firm for two years, and not once, in all that time, did I encounter an advertiser/marketer who had invested enough in infrastructure on their web sites to determine the true effectiveness of their advertising campaigns. They did not know which creative drove the most traffic to their sites that resulted in measurable business activity, e.g., downloading information, requesting a brochure, asking for someone to call, or actually purchasing something. All they knew was which piece of creative, on which web site, drove the most traffic to a site. By relying on the click-through metric, all responsibility for the performance of the creative rests on the site on which the creative appeared -- allowing the advertiser/marketer and agency to lay blame elsewhere. Pay-per-click and similar variations on the theme are still other ways to shift responsibility away from the advertiser/marketer and on to the publisher/site owner. To start, publishers have to start accepting responsibility for their campaigns and have mechanisms in place to determine the effectiveness of ad campaigns in "conversion" terms, i.e., how much tangible business did this piece of creative, in this context, generate? Until they get serious on this, it's really just a huge waste of time and money.
With all due respect to Jakob, there's a much more serious parallel problem -- which is that the economics just don't work, and very few site owners have done the math on this topic. Let's say that I run "mycontentisking.com" and I make an ad buy across a selection of sites to generate traffic. I purchase 2 million impressions at an average $10/CPM ($20,000). Let's say my ads perform typically, and I get a response rate of.05% (10,000 clickthroughs). Let's say that I know that an average visit to my site generates five page views, resulting in 50,000 impressions. I try to "monetize" (what an ugly word) those page views by selling advertising. Let's say that I have multiple ad positions on a page and because of various factors my site commands an aggregate of $40/CPM across all page views. I gross $2K for those 50,000 pages I spent $20,000 to acquire. The challenge is that I, as the site owner, need to have a much more sophisticated plan for extracting revenue from visitors than advertising. Otherwise, it's an MLM pyramid and I am on the short end of somebody else's downstream.
Slashdot Mirror
The PhoneBook - of course.
It's a small MacBook that uses the iPhone as the touchpad and for its radios. There is an app that displays the iPhone screen on the larger display. The PhoneBook runs regular MacOSX so all regular Mac apps work and all iPhone apps work, too.
Connectivity in the PhoneBook complements what is in the iPhone, and there are slots for 16MB+ SD cards to expand storage.
You can undock the iPhone to use it when you don't want or need the functionality of the entire PhoneBook.
I am afraid MSBLACK has got his/her facts wrong.
/ cfcfr/CFRSearch.cfm?CFRPart=163.
The FDA Standards of Identity for chocolate are set out in Title 21 - Part 163 - Subpart B "Requirements for Specific Standardized Cacao Products" which can be found at the following URL http://www.accessdata.fda.gov/scripts/cdrh/cfdocs
There are standards of identity (SOIs) for different types of chocolate, including white, milk, and sweet (which covers both semisweet and bittersweet although there is no legal distinction between the two). The minimum COCOA CONTENT in sweet chocolate is specified as "15 percent by weight of chocolate liquor" and for semi and bittersweet as "not less than 35 percent by weight of chocolate liquor." Chocolate liquor is ground up cocoa beans whose fat content is between 45-55% depending on the species of bean and where the bean is grown.
Cocoa content is NEVER specified in terms of solids (cocoa powder) content -- which is where the original poster got his/her facts wrong (this is a very common misunderstanding). Cocoa content refers to the total amount of cocoa solids and cocoa butter in a chocolate product. If you look closely at the ingredients label you'll see that most chocolates have cocoa liquor (or chocolate liquor or cocoa mass) plus added cocoa butter. Cocoa content is expressed as a percentage, by weight, of the combination of those two ingredients.
The FDA SOIs for all chocolates (white, milk, sweet, and liquor) allow manufacturers to include dairy fats and other milk ingredients (including whey) in their products. This is done to reduce the cost of manufacturing (which is what the FDA petition is ** really ** all about) as well as stabilize the cocoa butter crystals to extend the shelf life of the chocolate without adding preservatives.
There are a number of profound ironies here. Irrespective of the subjective taste impressions of American vs European chocolate, the fact is that American standards for what manufacturers can put in chocolate -- and still call it chocolate -- are stricter here than in the EU. The EU SOIs for chocolate allow manufacturers to replace up to 5% of the cocoa butter in chocolate with 'cocoa butter equivalent' and/or 'cocoa butter replacement' fats and still call it chocolate.
Manufacturers can't do that in the US but that is what the CMA and its fellow lobbying organizations are petitioning the FDA to do -- but they want to go much, much, further. They want the ability to replace 100% of the cocoa butter with other fats and still call it chocolate.
It is also important to recognize the difference, in this debate, between chocolate manufacturers and chocolatiers and candy makers. A chocolate manufacturers makes chocolate products (chocolate liquor, cocoa powder, cocoa butter, and finished chocolate) by processing cocoa beans. Chocolatiers buy chocolate from chocolate manufacturers and melt it down to make their finished products. There are relatively few companies in the world that both manufacture chocolate and are also chocolatiers though there are more chocolate manufacturers that also make candy.
If you look at the ingredients on a candy bar or on a box of truffles you'll see that chocolate is listed as an ingredient and -- if it's "good" (or "pure" or "real") chocolate, the only fat in the chocolate will be cocoa butter. All those other ingredients are not in the chocolate -- they're what's inside the outer chocolate coating.
Clay Gordon
Editor and publisher
www.chocophile.com
Author Discover Chocolate (Gotham Books, October 2007)
The micromirrors they are talking about are probably already in production: think of the micromirror array chips that TI sells for video displays and projection -- DLP.
:SI
TI's DLP chips can actually be used both ways (as projectors as well as part of a sensor) and I know of a company that has experimented with using them in a tunable sensor that can survey a scene in different spectra sequentially in order to obtain a very accurate profile of what's in front of the sensor.
Think homeland security applications for remote sensing for potential water contamination, satellite surveillance, etc.
The comparison about prices misses the mark because of the differences in form factor. There are places that the memory spot chip can be embedded where a smart card cannot. Done correctly, the memory spot chips will not draw attention to themselves, either. Obfuscation is an often-ignored aspect of securing data.
The FDA has a standard that's part of SRC 21 for medical device compliance -- 510(k). At its most basic level, it's a documentation process -- you have to document every market requirement, part spec, manufacturing prtocess, QA process, as well as the history of every part that goes into a device including who ordered it, when it was ordered, who it was ordered from, who received it on the loading dock, who accepted it, who did testing against spec, what device it went into, if there was ever a complaint filed on the device, if a part was ever returned for repair, and the disposition of all returned parts, etc, and so and so forth ad infinitum.
In case someone ever gets hurt by the device in use it's possible to figure out who okayed the decision(s) that lead to the alleged defect(s) in the product.
There is a COTS evaluation process, but it's pretty straightforward compared with the process for designing and developing the software that goes into a medical device.
How do I know this? I worked on the development of a software system to automate 510(k) compliance for a medical device manufacturer. I was responsible for translating all of the procedures in the Quality System into functional specs and UI designs. Along the way we learned that most companies do not have automated documentation systems for 510(k) compliance because it's far too complicated and there is no FDA spec for designing and implementing an electronic 510(k) compliance system. There used to be one but it was withdrawn because it was too complex.
What I can say is that the Software Development process covered in our Quality manual requires a level of detail I have never seen in any previous software development management system. Not only must every module and its inputs and outputs be documented, but algorithms must be documented. It must be possible (in case of an FDA audit) to trace customer requirements through to market specs to functional specs to technical specs to code, complete with test scenarios and procedures to ensure that the software sctually does what the design documents say it should do -- and that any deviation from specs discovered during testing are resolved.
The irony of all this is that the computer system we developed has too much information in it and FDA auditors aren't used to auditing the content in electronic systems. So, the system outputs paper reports that are put into file folders for the FDA to review if it's ever necessary.
Sigh.
-- Si
And I thought the "VC" stood for venture capital. Silly me.
... specifically written to target this system?
clusterf*ck ?
SI (which stands for shark infested) Waters
We are in the midst of a complete reno of our 90 year-old house and did exactly this. We're not living through the reno and in the rental house we're in we're using 11g connnections because we can't do any installation that will mar walls. The WiFi connections are turning out to be surprisingly unreliable - won't connect consistently and the speed of the connection is wildly variable.
So, while the walls were open we decided to bite the bullet and run cable; 2 Cat5E and two video coax to 17 different locations in and outside the house. 10/100 switch plus ethernet and video patch bays in the basement. More reliable, more secure, too, and I can run video over the coax and phone signals over the second ethernet pair to any place in the house. Old computers are being used for media servers and hardware firewalls. Mac Mini with Elgato TV-Eye for PVR/iTunes for audio.
I agree with taking pictures and also suggest an overlay on the 'prints to help locate things. We're a test neighborhood for Verizon fiber to the home and when it's installed we're talking about a 20+mb downstream connection.
Are we having fun yet?
One of the reasons that EMR (Electronic Medical Record) systems have failed to achieve significant market acceptance is that they are not designed to be used by physicians in the clinical setting, e.g., in the exam room.
:SI
In virtually all instances, it looks like the systems have been designed to make it easy to submit insurance claims. The doctor needs to know about and enter too much extraneous data and can't concentrate on important stuff, such as how the technology will improve the quality of patient care and doctor/patient interaction.
It's my sense that the big companies in the consortium don't care one hoot for really improving the quality of medical care. It's been estimated that it will take something close to $234 billion over the next 10 years to implement the system. Even Bill takes notice when $234 Billion is at stake.
Where's the money for that going to come from with no new taxes? I know! My insurance premiums are going to go up -- and still the system won't work, not because it's not interoperable but because it changes the way the doctor works. Of course the insurance companies love it because it pushes even more responsibility for claims processing over to the doctor.
I think we all know the probability of success of a system that forces someone to change the way they work in order to use it - especially when there is little or no perceived benefit.
Have you not read Caliban? There is a "zeroeth" law, now: "A robot may not injure humanity, or, through inaction, allow humanity to come to harm."
Makes ya kinda misty eyed, don't it?
Based on two years' experience in HD production in the early '90s, the data rate they are talking about is being recorded to tape, not to disk, which does make a real difference as there are no portable 3.3GB/sec sustained data rate disk arrays that you can connect to a camera and use in real production situations. (In 91, there were no portable HD VTRs either, and this is a lesson that they will not overlook as long this time.)
In addition to any DCT compression, they are probably also using color space conversion which provides compression but in a different way that is based on the way the eye sees and not on image blocks.. The cameras are certainly 3-chip RGB but the color space conversion would be similar to YPbPr (y= luminance, or green which is recorded at full bandwidth with the Pb and Pr color difference (y-b, y-r) signals recorded at lower bandwidth). When all three colors are recorded full bandwidth it's denoted 4:4:4; typical color space conversion with compression (before any compression using a DCT) would be denoted 4:2:2. The 4s and 2s refer to the sampling frequency as a multiple of the base frequency. (NTSC color conversion is called YUV, which is another algorithm, but the y is still luminance.)
SI
Ever wonder why 59.94/60 was chosen in the first place? Remember this was back in the 1930s.
Well, it turns out that this is the frequency of the current coming out of wall sockets. Everything was slaved to that frequency to make things easier, electonically speaking.
The 4:3 aspect ratio of video is similarly mired in tradition. Turns out that the original 16mm silent films were shot 4:3 so when video came around they needed material to transfer and film was it.
The 3:2 pulldown in non-drop transfer from video/film is a pain but well understood. Harder to manage was non-drop in 60/59.94 transfers from HD to NTSC.
BTW. HDTV is really an origination medium or for theatrical distribution. The average consumer in the home would be very happy with widescreen D1 (component) and to make it very large, a line-doubler is perfect. The roadblocks to HD distribution are political (i.e., who benefits from the free spectrum and how), not technical. Widescreen D1 could be broadcast over existing towers and transmitters with only very mild compression, no new expensive distribution infrastructure required.
SI
This is news? I remember visiting the Visible Language Workshop at MIT in 1983 and seing a far more capable device built by Professor Ron MacNeil. IIRC, they were programming everything in PL/1 in those days.
SIW
In this day and age, building your own counting app is a very naive suggestion. One of the reasons companies use third parties is that their methodology is "independent" and "consistent" across sites. Nobody knows how your homegrown app works, what it counts, or how it counts it. Therefore, nobody will believe your numbers, even if you can "prove" they are correct. Certainly no ad agency media planner who has to justify spending the dollars on the buy -- and that's the gatekeeper, some young kid protecting their job who is not going to go out on a limb because you think you can code a better app and save your company some bucks; that is not the point. It's sort of like the Nielsen ratings. everybody knows the methodology is flawed, but the flaws are equally advantageous and disadvantageous to everyone, so they use it because they can't agree on anything better.
Clay
What I want to know is what Amazon does with the commissions they are no longer paying to Morpheus. We have to assume that the actual affiliate referrer is lost in the process. (Maybe this is how Amazon is able to post a profit! - Hah!)
I suggest putting the money into an escrow account and a) distributing it to affiliates on some sort of basis, or, b) donating it to an organization that will take the perps to court.
What's the right organization for this? EFF?
What's missing from the original post is the understanding that the greatest contributor to increased picture quality is not the increased resolution HDTV affords (especially on screens smaller than about 8 feet diagonal), but the change in color space. NTSC was designed in the 1950s to enable black & white television sets to display a black & white image even if the signal being broadcast had color encoded in it. In order to do this, the color information has the bejeezus compressed out of it which is why it looks so lousy.
The single most important change that can be made to improve the quality of broadcast television is NOT to increase resolution, but to start broadcasting a component (e.g., YPrPb) signal while ensuring that the entire production chain, from origination through production to distribution and reception is component end to end.
When coupled with a widescreen aspect (a feature of most modern professional cameras), the component signal can be easily broadcast over existing equipment, or with minor and comparatively inexpensive transmitter upgrades. Most importantly, there is no incremental increase in cost to produce programming in widescreen D1 as there is in HDTV. Finally, monitors/receivers/decoders are much, much, cheaper.
But -- even this is not the issue. It's not about (and never was about) making it easy for consumers. It's about broadcasters wanting free spectrum without the onerous requirement of "wasting" it by having to broadcast HDTV all day long. The spectrum allotted for HDTV broadcast is enough to simultaneously broadcast 6, widescreen D1 streams. Now, instead of having one station in a market, a broadcaster can have 6 -- or rent one or more of the channels to others for other uses.
It's politics, always has been. Probably always will be. {sigh}
Clay
HDTV was designed to deliver equivalent resolution, to the screen, as the best 35mm projection. (Although film starts out with much higher resolution, by the time it goes through all of the processing steps, each with mechanical transports, and then projection, the final resolution is quite low.)
A minimum 36" diagonal monitor is required to display every pixel in the image. However, the full effect of HDTV is not apparent until the image is at least 8 feet (diagonal measure). Furthermore, the original color space demands very low light level conditions in order to appreciate fully. Finally, the minimum viewing distance should be 3x the height of the screen, in order to make the pixels/grain disappear, so you'd need to sit at least 15 feet away from the image.
There are very few viewing environments that meet these conditions -- the average American home is certainly not one.
From a production viewpoint, HDTV requires film-level production standards. The sloppiness that NTSC and PAL encourage (and hide) is clearly visible to HDTV production processes. As my old boss, Barry Rebo, used to ask, "What does HDTV bring to 'Wheel of Fortune'?" The answer is absolutely nothing, except increased production costs.
The political stalemate in Washington is a blind for the desires of broadcasters to capitalize on the offer of free spectrum. The technical arguments are about how a broadcast signal will degrade in fringe reception areas. This is a smokescreen designed to delay introduction until terms are more favorable to broadcasters.
In fact, there is a production medium today that can be broadcast today over existing equipment with only marginal incremental costs to broadcasters: widescreen D1. It's the appropriate aspect ratio (16:9) and is a component system from beginning to end, which is the major contributor to quality.
This format could be delivered more or less right away and would deliver picture quality far in excess of what people expect today, and is, in fact, perfectly appropriate to the broadest range of viewing environments. Importantly, the incremental cost of production in D1 is virtually zero making it very attractive from that standpoint.
Finally, it is very easy to downconvert an HDTV-originated image to D1, especially because the timebase is the same (60 frames (or fields)/second as opposed to 29.94 frames/sec for NTSC).
HDTV is a perfect production medium and is great for projection in controlled viewing environments. For mass distribution D1 (or, more particularly, widescreen component) is far more appropriate, with or without line-doubling.
-- Clay
John Dewey, the pre-eminent American philospher wrote in the 1930s that "art" is the process that the maker goes through in making "art objects" as well as the relationship that viewer has, with and in his or her own experiences, with the art object. Art objects (the objects of the artistic process), have no intrinsic value. Their value is only as a signpost of the expression of experience that the creator has, and any response(s) that a viewer has. The fact that art is possible at all is because of the common, shared, experiences between the creator and the viewer.
What this means is that "art" is whatever the creator calls "art" because, not of their intent or their motivation, but because the process of creation generates aesthetic satisfaction for the maker. That's all that is needed for something to be called art.
At the same time, art is also the response a viewer has with an object. If a person has a genuine aesthetic response to an object, it it not for another person to deride or minimize that response, based on the object itself, or any other knowledge that they might have. It's art to the viewer because it moved him or her.
What does this mean for Aaron? I think it fails the first, or "personal aesthetic satisfaction" test. I don't remember (I read the book in the mod-1980s and there's a copy in storage somewhere) that Aaron is motivated to make drawings because it feels good about the drawings it makes, or that, through some feedback mechanism, it feels "better" about some drawings than others because they more accurately represent its feelings than others.
However, if the drawings that Aaron generates evoke genuine aesthetic responses in viewers, then those responses can be called art. Aaron, or course, is completely oblivious to these responses and can't use them to "improve" its art.
At least, so far. It's an interesting experiment. I read Asimov's story many years ago and have always been struck by its irony and message -- that the best art is made by people who need a little adjustment.
SIKey features of EnFuzion (these are lifted from a product marketing page so I can't comment on their veracity) include:
- No need to rewrite or modify code.
- Maximize system performance with multiple load monitoring and resource-sharing options.
- Highly secure: EnFuzion's advanced security features bolster the standard security provided by Unix and Windows NT systems.
- Easily integrate EnFuzion into your applications with EnFuzion's application programming interface (API).
- Monitor and control EnFuzion execution using a familiar language such as C, Java, Perl, Bash, Tcl/Tk, etc.
- Easily integrate EnFuzion with any shell or scripting language.
On the surface this would seem to fit wiredog's needs.More academically, the dramatic theorist Georges Polti posited that there are only 36 Dramatic Situations, and all of literature and drama can be decomposed in a combination of these. The book is out of print, but here's a link to Amazon to find out more if you're interested. A Google search turns up some interesting sources, for those who want to know more.
With all due respect to Jakob, there's a much more serious parallel problem -- which is that the economics just don't work, and very few site owners have done the math on this topic. Let's say that I run "mycontentisking.com" and I make an ad buy across a selection of sites to generate traffic. I purchase 2 million impressions at an average $10/CPM ($20,000). Let's say my ads perform typically, and I get a response rate of .05% (10,000 clickthroughs). Let's say that I know that an average visit to my site generates five page views, resulting in 50,000 impressions. I try to "monetize" (what an ugly word) those page views by selling advertising. Let's say that I have multiple ad positions on a page and because of various factors my site commands an aggregate of $40/CPM across all page views. I gross $2K for those 50,000 pages I spent $20,000 to acquire. The challenge is that I, as the site owner, need to have a much more sophisticated plan for extracting revenue from visitors than advertising. Otherwise, it's an MLM pyramid and I am on the short end of somebody else's downstream.