Well, of course all of that is true about X.400, but really, that's the minor stuff.
For starters, let's talk about addresses. You can forget about there being anything simple and usable like user@domain. In X.400 an address is a collection of attributes and corresponding values. (In X.400-1984 there were about 15 of these, X.400-1988 expanded the list to around 30, and if memory serves you could also define your own attributes identified by an OID.) And there was no, repeat NO, defined string format for writing these things down. (This was finally addressed in the early 90s when the problems with not having a way to write addresses in a consistent way on, say, a business card became too painful to ignore. Of course they picked a different format than the one that was in common use - "a1=v1; a2=v2;... aN=vN" versus "/a1=v1/a2=v2/.../aN=vN" - leading to even more confusion.)
So what are the attributes? Well, the stuff that corresponds, more or less, to the user part of user@domain is over-designed and overly restrictive but workable - surname, given-name, initials, and generational-qualifier are the main ones - but instead of a domain you have three attributes: prmd (primary domain), admd (administrative domain) and c (country). prmd is basically the name of your company or something similar. admd is the name of your service provider. See the problem? Every address is tied to your provider - no address portability. (Again, this was addressed at some point by allowing an admd of a single space, the idea being that there would be this gigantic network-wide white pages X.500 directory that would provide the missing information. Which of course never deployed and given privacy concerns never could have.)
Let's move on to routing. The assumption in X.400 was that everyone would simply direct connect to a service provider, so routing was the service provider's problem. And by some incredibly twisted logic that I never could follow that meant it could be left entirely unspecified! I quite literally had to invent a format for routing tables for our implementation. Everyone else did the same, so of course they were all different.
I'll skip the format for headers - lots of problems there, including no extensibility whatsoever in X.400-1984 (and very few ever bothered to upgrade to X.400-1988) - but most of them are rather complicated and would take too long to describe.
Message bodies consisted of a series of parts. The main one was essentially equivalent to "text/plain; charset=us-ascii". So what about other scripts? Well, there was a type of part based on T.61, where you use ISO-2022 escape sequences to switch into various other charsets, but it was wildly complex and never interoperated worth a damn. So in X.400-1988 they introduced a general-text body part that was slightly better, except that almost nobody implemented it. Seems they implemented a different body part that was only defined in a *draft* of the standard and later removed from the final version. (Since I didn't have a copy of the draft I had to reverse engineer it from dumps of the ASN.1.)
There were predefined parts for images (based on T.something-or-other, think lo-res FAX) and audio (audio/basic, nothing else), and a couple of others I forget. And there was a catch-all part equivalent to application/octet-stream, no parameters allowed. Want to send around arbitrary typed content? Sorry, not possible. (Again, this was addressed later on with the definition of the file-transfer-body-part, a hugely complex nightmare that labelled types with OIDs.)
There was also a message part that allowed for nested messages, but there was no other way to nest parts like MIME provides or any way to label a collection of parts. So you can forget about multipart/alternative, multipart/related, or anything similar.
Enough about the envelope and message format. Let's talk about the protocols. MTA to MTA transfers used something called RTS (later RTSE) that sits on top of the OSI stack (
I'm a co-creator of MIME. I was recently interviewed about it, and one of the questions that came up was why we didn't patent it?
The obvious answer is it simply didn't occur to us, but even if it had, I very much doubly we would have done it. Our primary purpose was to change the capabilities of existing and future email systems in a fundamental way. (A secondary purpose for me at least was to drive a stake through the heart of X.400 - having just done an implementation of that mess, I was deeply worried that its very limited multimedia capabilities would drive its deployment in place of the much simpler Internet email protocols, and the entire world would have been much worse off had that happened.)
But neither of those goals could possibly have been met by patented technology. To be successful we needed as many implementers to adopt the technology as soon as possible, and a patent would have prevented that. (One of the reasons MIME is a bit ugly in places is because we believed that being able to implement and deploy it on top of existing infrastructure trumped design niceties.)
And while I can't speak for Nathaniel or Keith, it was certainly possible for me to profit from our work without a patent. But I did it the old fashioned way - by building a good implementation and selling it.
Of course there are situations where patents make sense, like to protect small inventors who come up manufacturing process or whatever. But there are lots of cases where they don't, and right now things are canted much too much towards patenting everything, even in cases where it actively stifles innovation and progress.
Mine is an MFC-8860N, but my experience is the same. I frequently scan 20-30 page double sided documents and I can't recall the last time it jammed. The Mac software isn't especially great but it gets the job done.
The printer is actually the part that jams sometimes, but only when I'm printing a bunch of stuff on really thick paper.
I taught myself to program in high school, mostly by reading the PDP-8 Assembly Language Handbook over and over until I figured out what the hell they were talking about. The only CS or programming class I ever took was "Introduction to Computer Science" my first year in college, where the language used was, believe it or not, Algol 60. I hated it mostly because I was much more interested in working on the retargetable assembler/disassembler system I had designed and the classwork was a distraction from that. The year after that I was assigned to a project to write the back end of a FORTRAN compiler for a MIL-SPEC computer, where I designed and implemented the register allocator and part of the code generator. The year after that I was assigned to a different project, where I wrote the printer subsystem for a block diagram editor. That I didn't enjoy much because once again I was more interested my own work, which happened to be a symbolic algebra library that could be easily integrated with various numerical analysis and graphics libraries. That eventually morphed into a commercial product.
I've published several articles on computer science, and I also contributed one of the exercises that appears in TAOCP Volume 4A (which references one of my articles).
I'm also a coauthor of the MIME standard. These days I my main job is architect of a high-end MTA.
Of course the plural of anecdote is not data, but several of the best architects and coders I know have a similar appalling lack of credentials.
Makes sense. In my case the colectomy was because dysplastic cells had shown up, meaning cancer was inevitable. And I did have a temporary colostomy that was removed after a couple of months - they called it the "takedown" procedure.
Yeah, I was wondering about it's ability to implement wide margins myself. I don't know that much about stomach cancer, but with colon cancer there are good reasons why they do hemi or total colectomies. In fact if I remember correctly in ulcerative colitis cancer tends to show up in calm regions adjacent to the inflamed parts.
Really? Are you sure that's the case when they do a full ileo-anal anastamosis? (That's where they fashion a J-shaped pouch from the end of the intestine and sew it to the tissue just behind the anal sphincter.) My own procedure in 2003 took six and a half hours to complete, and at the time they told me five hours was the average. I'm sure it's gotten faster, but that would be HUGE reduction. (There's also lots of variability between patients - some procedures are much more difficult than others. And so, unfortunately, are the outcomes - something like 2% of the time it won't reach and you end up with a permanent colostomy bag.)
That said, I do agree about this being little more than an upgrade. They cut out polyps using a coloscope all the time. And as another example, they leave a stent in the newly-connected ureter after a kidney transplant, then after three weeks they go in with a cystoscope, locate it, and send in a little pincer thingie to grab it and pull it out. It's completely routine and actually pretty cool to watch. (Oh, and not all urologists have switched to using the scope. Some still use a stiff metal rod. Not speaking from personal experience or anything like that here, but my advice is if your doc gets out the rod to get the hell out of there and find a new urologist.)
Yeah, that's basically what I did. The high degree of consistency student-by-student with midterm grades made it easy to justify.
FWIW, I've found that in such classes the overall distribution of the grades is almost always bicameral, one hump indicating those who "got it" and the other those who did not.
As the newest math professor in the department, of course I was lowest of the low. I was informed that there was no classroom available in the classroom building and I had to choose between one in Animal Husbandry and Poultry Science. In a moment of true quantum stupidity I chose the one in Poultry Science because it was closer to my office.
The classroom sat adjacent to a room that contained hundreds of chickens, maybe more. You had to smell it to believe it. Of course the students complained but there was nothing I could do.
The class actually did quite well, that is, until the day of the final exam. When I got there to deliver the exam (which of course was being given at a different time) the door was locked and no key could be found. I was forced to walk the entire class over to the classroom building and give the exam in an empty classroom.
Checking the scores against the midterm, I found there had been a significant drop for almost every student. To this day I am convinced that the context change and the lack of that awful smell was as or more responsible for the difference than all the chaos leading up to taking the exam.
This isn't IPv6-only in any meaningful sense of the term. All you've done is move the dual part of the stack from the mobile device to the operator. In fact since the *overwhelming* majority of servers are reachable by IPv4 only, the NAT64 will be used for almost everything. And since the IPv4 address the device would get in a dual stack setup would almost be from a NAT as well, you haven't actually changed IPv4 address usage in any significant way.
This is one of those necessary steps that has to be taken on the road to an IPv6-only world and I am glad to see it happen, but it is one that offers fairly little direct benefit. And the really big problems remain: (1) The millions of home routers that aren't IPv6-capable and the failure of the vast majority of ISPs to offer IPv6 connectivity to their customers.
Currently have a LEAF, 2010 Prius, and a 1999 4Runner. The 4Runner was an OK car, but we're selling it next week. We're also on the waiting list for an Tesla S Series; when that arrives next year the Prius will be the one to go. I agree that the LEAF makes the Prius feel old fashioned, although how much of that is because I prefer the Nissan's overall design sensibilities to Toyota's is unclear.
Entirely relevant, actually. The reason I brought it up is because it has various been asserted that an ionizing effect in this situation is possible, and others have claimed it's impossible. I don't know about you, but when I want to consider whether or not something is possible in a given situation, I look at the actual physics and see what's involved in getting it to happen. And we both know, what's involved doesn't seem likely to exist (ok, that's an considerable understatement) as a combination of what a phone actually emits and what's between a persons's ears. But you don't know that until you look.
In other words, you now admit that my original point, which was that while it's possible to turn non-ionizing radiation
to ionization radiation, the chances of anything in a human body doing it are vanishingly small, was in fact correct.
First of all, there are ionizing and non-ionizing photons. The difference is the energy/frequency/wavelength. Visible light, infrared, radio are all relatively low energy and non-ionizing. X-rays and gamma rays, OTOH, are ionizing.
Second, your assertion that there are no processes that can convert non-ionizing radiation to ionizing radiation is false. I've done it myself in the lab - a 10Mw picosecond pulse neodymium YAG laser puts out light in the infrared (non-ionizing). But with frequency doubling optics you get green light. And you can then combine the green light and some of the infrared and get ultraviolet (ionizing) radiation out. (And incidentally, I was a physic postdoc at the time - that's how you get to play with such cool expensive toys.)
Third, getting this to work depends on the tremendously powerful (but extremely short) pulses the laser puts out. Without sufficient quantities of photons the changes of the same atom getting hit twice are extremely low. And the radiation produced by a cell phone isn't even close to having the necessary density, not to mention that the photons are so low energy you'd have to combine a lot of them to get to the ionizing range. But the phenomenon does exist.
Nonlinear effects are possible, like where two photons are absorbed then only one is emitted. So non-ionizing radiation could in theory interact in a way to produce ionizing energy. It's also possible that some structures are exquisitely sensitive to particular frequencies of radiation.
But a closer look shows just how unlikely such phenomena are. The probability of such interactions depends on there being sufficient energy density - you see them with megawatt lasers but not at the power levels where cell phones operate. As for some sensitive structure being present, if there was you'd think we would have found it by now.
I have a Leaf. My commute is just under 20 miles one way, most of it on the freeway at 65+Mph speeds and since this is summer in SoCal, the A/C is running pretty much all the time. When I leave the car usually says I have 90 miles of range; when I get home the car says I have somewhere between 40-50 miles of range left.
I haven't gotten to the point of "driving on electron fumes" yet, but I have gotten it down to about 15 miles of range left. That took about 75 miles of driving, so assuming there really was 15 miles of range left, the effective range on average seems to be around 90 miles. Which of course is less than 100 but more than 73.
A friend of mine does research on flare stars. For this you need to look at stars for long periods.
If you think about it, there's a major experiment underway that's already looking at lots of stars for long periods: The search for extrasolar planets. And even better, when one of these experiments finds a flare star, that data isn't especially interesting because they're looking for planets that can sustain life, and frequent solar flares are... unhelpful in that regard. As long as the data is properly attributed, sharing with the flare star folks means you get credit for data you collected but cannot use yourself.
So it's a win-win to share data in this case, but even so there are various hoops to jump through to get access. I suspect a lot of it boils down to, "We know you and we trust you will stay within your area of interest that doesn't overlap ours."
Some of the relevant cases here are National Bellas Hess, Inc. v. Department of Revenue of Illinois, Quill Corp. v. North Dakota, and Complete Auto Transit, Inc. v. Brady. In brief, in Complete Auto the Supreme Court created a four prong test. The prongs are: (1) Substantial nexus (I used the wrong term earlier, sorry), (2) Nondiscrimination, (3) Fair apportionment, and (4) Fair relationship to services provided by the state. Of these, it seems to me that only (1) is in play here.
Quill established that a company that only does mail-order business does not create a "substantial nexus". But Amazon and other similar retailers do a lot more in the California than mail stuff to people. Among other things, they partner with local California companies to do order fulfillment.
If this ends up before the Supreme Court - and I rather suspect it will - it's a situation where two different legal doctrines - the so-called Dormant Commerce Clause and Due Process, are in conflict. Historically the Court has tended to side with Dormant Commerce Clause, which would mean this bill is unconstitutional. But I rather suspect that the scale and scope of these retailer's activities will tend to push this towards a ruling based on due process considerations. But again, IANAL, YMMV, etc. etc.
Exactly. If you buy something in California from out of state, the vendor doesn't charge you California sales tax, and you proceed to directly use that item, you are supposed to pay use tax. That's the law; if you're buying stuff from Amazon or wherever and not paying those use-taxes you are in violation. (And yes, this means a HUGE number of people are in violation, which is sort of the point.)
All this measure does is require that vendors like Amazon collect the tax instead. Since Amazon already does all sorts of order fulfillment in California and collects sales taxes when it does; this should be trivial for them to do at least.
Speaking as a California resident, I welcome this because right now keeping track of use-tax is a royal pain. In fact if anything I will be strongly inclined to prefer vendors that collect the tax because it saves me so much trouble.
As for counting on some sort of "only the feds can do this stuff" ruling so you can continue to break the law, this gets down to how what the lawyers call a jurisdictional nexus is defined. The bill appears to be tweaking that a little, but I doubt it's tweaking it enough for there to be constitutional issue. But IANAL, and even if I were, this is very deep stuff and you really need an expert's read on it.
I don't recall the exact dates, but it was around 1982 that these utilities were created. ROLLOUT saved the entire VMS system state to disk and ROLLIN brought it back, not only restoring any long-running batch jobs and so on but also dropping the boot time down from 5-10 minutes to less than 30 seconds.
Although it was easy to get ahold of ROLLIN/ROLLOUT if you wanted them Digital would never condone their use. The reason for that was simple: It depended on disk state not changing whlie the system was down, and not only was that too fragile for Digital's taste, the clustering system Digital was busy designing at the time was going to make that a practical impossibility. So this capability was never used all that time, and then clusters came out and made it unworkable.
I have two sets of solar panels - one 10 year old set that's wired up conventionally to a single inverter, and another 8-month-old set that uses per-panel microinverters.
According to the very nice graphical display of power generation the microinverters provide via a built in web monitoring thingie, I average about four hours each day operating between 90% and peak generation.
Installation of the original set of panels was a major PITA because UL delisted the inverter after it was installed. Getting a massive piece of equipment down off the wall is easy, getting the replacement back up, not so much. But once that one startup issue was dealt with, the system has been 100% reliable and has required no repairs. As for the microinverters, it's early days yet, but they've been completely reliable so far.
As for washing the panels, yes, doing that more often increases output, but in my experience, not by that much. Around here the windows need to be washed twice a year so the panels get done as part of that. No big deal.
And as for all this "resetting after a power failure" - it appears your experience with grid-tie systems is seriously out of date. My 10 year old inverter handles power failures automatically. Aside from monitoring, I haven't had to touch the thing once the replacement inverter was installed. Ditto for the microinverters.
Finally, you appear to be conflating grid-tie and off-grid setups. I agree that a fully off-grid setup isn't easy. I have battery backup as part of my original system, but since the batteries are only used when there's a grid failure they haven't needed to be replaced. (And most grid-tie systems don't need them at all.) An off-grid setup that charges and discharges the batteries every day is going to require a lot more maintenance. And when solar is the only energy source the system has to be overbuilt in the fashion you describe (just not as much as you claim). And you probably care more about keeping the panels clean when they are your only power source.
But the vast majority of solar systems are grid-tied, not off-grid. So most of your issues simply don't apply.
Now, perhaps you'll say my experience is unusual. Yes, it's only anecdotal, but I know three people with similar solar setups in the area, and their systems have all worked flawlessly since they were installed.
Maybe he's using a custom clock app or something, but on my iPhone the built in clock app has four clearly labeled mode setting buttons at the bottom: "World Clock", "Alarm", "Stopwatch", and "Timer". Pressing the one called "Alarm" to set an alarm seems, well, obvious, and when you do that you get a screen saying "no alarms" and exactly one "+" button you can press, so unless you simply freeze up at that point I don't see how this can be so confusing. In particular, no clock face is displayed at this point so there is no possibility of, "Pressing the clock image takes you through to choices about how the clock is displayed, and it's not easy to get back again."
If you want to criticize the alarm and calendar stuff on the iPhone, a better place to start is the spinning dial thing used to enter times. (Which is what comes up once you press "+".) A lot of people dislike this and find it hard to use. I don't find it difficult personally, but I have to admit I'd prefer a numeric keypad.
I grew up in Oklahoma and attended Oklahoma public schools up until high school, when I went to Groton School. The difference was stark. For example, 8th grade math in Oklahoma was still focused on basic arithmetic. (I really wish I was making that up, but I'm not.) By the time I graduated from Groton I managed to get a 5 on the AP calculus exam.
Prior to attending Groton I coudn't write worth a damn. Groton fixed that too.
But most important thing I learned there was, basically, how to think - the self-discipline needed to properly organize and express my thoughts and ideas.
Now, as for your little list.Maybe I didn't take the right classes or something, but I'm fairly sure I didn't learn enough of that "tak the talk" thing at Groton to matter - when it came time to get serious about business relations for our startup company, we went outside and hired a "suit" to do that.
As for friends helping out, I cannot recall a single instance after graduating when I asked a fellow Groton alum for help.
And as for that self-importance thing - let me just point out Groton's motto: Cui servire est regnare.
Now, I have no doubt there are private schools that provide what you describe and little else. But not all of them are like that.
TFA makes a lot of assertions about unnecessary treatment and increased costs with no associated benefits, but doesn't present or link to a single piece of actual peer-reviewed data.
In the specific case of changing the cutoff of blood glucose levels from 140 to 130, the appropriate question to ask is whether or not treating the many side effects of diabetes sooner saves more than it costs.
Slashdot Mirror
Well, of course all of that is true about X.400, but really, that's the minor stuff.
... aN=vN" versus "/a1=v1/a2=v2/.../aN=vN" - leading to even more confusion.)
For starters, let's talk about addresses. You can forget about there being anything simple and usable like user@domain. In X.400 an address is a collection of attributes and corresponding values. (In X.400-1984 there were about 15 of these, X.400-1988 expanded the list to around 30, and if memory serves you could also define your own attributes identified by an OID.) And there was no, repeat NO, defined string format for writing these things down. (This was finally addressed in the early 90s when the problems with not having a way to write addresses in a consistent way on, say, a business card became too painful to ignore. Of course they picked a different format than the one that was in common use - "a1=v1; a2=v2;
So what are the attributes? Well, the stuff that corresponds, more or less, to the user part of user@domain is over-designed and overly restrictive but workable - surname, given-name, initials, and generational-qualifier are the main ones - but instead of a domain you have three attributes: prmd (primary domain), admd (administrative domain) and c (country). prmd is basically the name of your company or something similar. admd is the name of your service provider. See the problem? Every address is tied to your provider - no address portability. (Again, this was addressed at some point by allowing an admd of a single space, the idea being that there would be this gigantic network-wide white pages X.500 directory that would provide the missing information. Which of course never deployed and given privacy concerns never could have.)
Let's move on to routing. The assumption in X.400 was that everyone would simply direct connect to a service provider, so routing was the service provider's problem. And by some incredibly twisted logic that I never could follow that meant it could be left entirely unspecified! I quite literally had to invent a format for routing tables for our implementation. Everyone else did the same, so of course they were all different.
I'll skip the format for headers - lots of problems there, including no extensibility whatsoever in X.400-1984 (and very few ever bothered to upgrade to X.400-1988) - but most of them are rather complicated and would take too long to describe.
Message bodies consisted of a series of parts. The main one was essentially equivalent to "text/plain; charset=us-ascii". So what about other scripts? Well, there was a type of part based on T.61, where you use ISO-2022 escape sequences to switch into various other charsets, but it was wildly complex and never interoperated worth a damn. So in X.400-1988 they introduced a general-text body part that was slightly better, except that almost nobody implemented it. Seems they implemented a different body part that was only defined in a *draft* of the standard and later removed from the final version. (Since I didn't have a copy of the draft I had to reverse engineer it from dumps of the ASN.1.)
There were predefined parts for images (based on T.something-or-other, think lo-res FAX) and audio (audio/basic, nothing else), and a couple of others I forget. And there was a catch-all part equivalent to application/octet-stream, no parameters allowed. Want to send around arbitrary typed content? Sorry, not possible. (Again, this was addressed later on with the definition of the file-transfer-body-part, a hugely complex nightmare that labelled types with OIDs.)
There was also a message part that allowed for nested messages, but there was no other way to nest parts like MIME provides or any way to label a collection of parts. So you can forget about multipart/alternative, multipart/related, or anything similar.
Enough about the envelope and message format. Let's talk about the protocols. MTA to MTA transfers used something called RTS (later RTSE) that sits on top of the OSI stack (
I'm a co-creator of MIME. I was recently interviewed about it, and one of the questions that came up was why we didn't patent it?
The obvious answer is it simply didn't occur to us, but even if it had, I very much doubly we would have done it. Our primary purpose was to change the capabilities of existing and future email systems in a fundamental way. (A secondary purpose for me at least was to drive a stake through the heart of X.400 - having just done an implementation of that mess, I was deeply worried that its very limited multimedia capabilities would drive its deployment in place of the much simpler Internet email protocols, and the entire world would have been much worse off had that happened.)
But neither of those goals could possibly have been met by patented technology. To be successful we needed as many implementers to adopt the technology as soon as possible, and a patent would have prevented that. (One of the reasons MIME is a bit ugly in places is because we believed that being able to implement and deploy it on top of existing infrastructure trumped design niceties.)
And while I can't speak for Nathaniel or Keith, it was certainly possible for me to profit from our work without a patent. But I did it the old fashioned way - by building a good implementation and selling it.
Of course there are situations where patents make sense, like to protect small inventors who come up manufacturing process or whatever. But there are lots of cases where they don't, and right now things are canted much too much towards patenting everything, even in cases where it actively stifles innovation and progress.
Mine is an MFC-8860N, but my experience is the same. I frequently scan 20-30 page double sided documents and I can't recall the last time it jammed. The Mac software isn't especially great but it gets the job done.
The printer is actually the part that jams sometimes, but only when I'm printing a bunch of stuff on really thick paper.
Might be over-generalizing just a tad there...
I taught myself to program in high school, mostly by reading the PDP-8 Assembly Language Handbook over and over until I figured out what the hell they were talking about. The only CS or programming class I ever took was "Introduction to Computer Science" my first year in college, where the language used was, believe it or not, Algol 60. I hated it mostly because I was much more interested in working on the retargetable assembler/disassembler system I had designed and the classwork was a distraction from that. The year after that I was assigned to a project to write the back end of a FORTRAN compiler for a MIL-SPEC computer, where I designed and implemented the register allocator and part of the code generator. The year after that I was assigned to a different project, where I wrote the printer subsystem for a block diagram editor. That I didn't enjoy much because once again I was more interested my own work, which happened to be a symbolic algebra library that could be easily integrated with various numerical analysis and graphics libraries. That eventually morphed into a commercial product.
I've published several articles on computer science, and I also contributed one of the exercises that appears in TAOCP Volume 4A (which references one of my articles).
I'm also a coauthor of the MIME standard. These days I my main job is architect of a high-end MTA.
Of course the plural of anecdote is not data, but several of the best architects and coders I know have a similar appalling lack of credentials.
Makes sense. In my case the colectomy was because dysplastic cells had shown up, meaning cancer was inevitable. And I did have a temporary colostomy that was removed after a couple of months - they called it the "takedown" procedure.
Yeah, I was wondering about it's ability to implement wide margins myself. I don't know that much about stomach cancer, but with colon cancer there are good reasons why they do hemi or total colectomies. In fact if I remember correctly in ulcerative colitis cancer tends to show up in calm regions adjacent to the inflamed parts.
Really? Are you sure that's the case when they do a full ileo-anal anastamosis? (That's where they fashion a J-shaped pouch from the end of the intestine and sew it to the tissue just behind the anal sphincter.) My own procedure in 2003 took six and a half hours to complete, and at the time they told me five hours was the average. I'm sure it's gotten faster, but that would be HUGE reduction. (There's also lots of variability between patients - some procedures are much more difficult than others. And so, unfortunately, are the outcomes - something like 2% of the time it won't reach and you end up with a permanent colostomy bag.)
That said, I do agree about this being little more than an upgrade. They cut out polyps using a coloscope all the time. And as another example, they leave a stent in the newly-connected ureter after a kidney transplant, then after three weeks they go in with a cystoscope, locate it, and send in a little pincer thingie to grab it and pull it out. It's completely routine and actually pretty cool to watch. (Oh, and not all urologists have switched to using the scope. Some still use a stiff metal rod. Not speaking from personal experience or anything like that here, but my advice is if your doc gets out the rod to get the hell out of there and find a new urologist.)
Yeah, that's basically what I did. The high degree of consistency student-by-student with midterm grades made it easy to justify. FWIW, I've found that in such classes the overall distribution of the grades is almost always bicameral, one hump indicating those who "got it" and the other those who did not.
Could well be. A controlled experiment would be very interesting, but who in their right mind would volunteer?
As the newest math professor in the department, of course I was lowest of the low. I was informed that there was no classroom available in the classroom building and I had to choose between one in Animal Husbandry and Poultry Science. In a moment of true quantum stupidity I chose the one in Poultry Science because it was closer to my office.
The classroom sat adjacent to a room that contained hundreds of chickens, maybe more. You had to smell it to believe it. Of course the students complained but there was nothing I could do.
The class actually did quite well, that is, until the day of the final exam. When I got there to deliver the exam (which of course was being given at a different time) the door was locked and no key could be found. I was forced to walk the entire class over to the classroom building and give the exam in an empty classroom.
Checking the scores against the midterm, I found there had been a significant drop for almost every student. To this day I am convinced that the context change and the lack of that awful smell was as or more responsible for the difference than all the chaos leading up to taking the exam.
This isn't IPv6-only in any meaningful sense of the term. All you've done is move the dual part of the stack from the mobile device to the operator. In fact since the *overwhelming* majority of servers are reachable by IPv4 only, the NAT64 will be used for almost everything. And since the IPv4 address the device would get in a dual stack setup would almost be from a NAT as well, you haven't actually changed IPv4 address usage in any significant way.
This is one of those necessary steps that has to be taken on the road to an IPv6-only world and I am glad to see it happen, but it is one that offers fairly little direct benefit. And the really big problems remain: (1) The millions of home routers that aren't IPv6-capable and the failure of the vast majority of ISPs to offer IPv6 connectivity to their customers.
Currently have a LEAF, 2010 Prius, and a 1999 4Runner. The 4Runner was an OK car, but we're selling it next week. We're also on the waiting list for an Tesla S Series; when that arrives next year the Prius will be the one to go. I agree that the LEAF makes the Prius feel old fashioned, although how much of that is because I prefer the Nissan's overall design sensibilities to Toyota's is unclear.
Entirely relevant, actually. The reason I brought it up is because it has various been asserted that an ionizing effect in this situation is possible, and others have claimed it's impossible. I don't know about you, but when I want to consider whether or not something is possible in a given situation, I look at the actual physics and see what's involved in getting it to happen. And we both know, what's involved doesn't seem likely to exist (ok, that's an considerable understatement) as a combination of what a phone actually emits and what's between a persons's ears. But you don't know that until you look.
In other words, you now admit that my original point, which was that while it's possible to turn non-ionizing radiation to ionization radiation, the chances of anything in a human body doing it are vanishingly small, was in fact correct.
First of all, there are ionizing and non-ionizing photons. The difference is the energy/frequency/wavelength. Visible light, infrared, radio are all relatively low energy and non-ionizing. X-rays and gamma rays, OTOH, are ionizing.
Second, your assertion that there are no processes that can convert non-ionizing radiation to ionizing radiation is false. I've done it myself in the lab - a 10Mw picosecond pulse neodymium YAG laser puts out light in the infrared (non-ionizing). But with frequency doubling optics you get green light. And you can then combine the green light and some of the infrared and get ultraviolet (ionizing) radiation out. (And incidentally, I was a physic postdoc at the time - that's how you get to play with such cool expensive toys.)
Third, getting this to work depends on the tremendously powerful (but extremely short) pulses the laser puts out. Without sufficient quantities of photons the changes of the same atom getting hit twice are extremely low. And the radiation produced by a cell phone isn't even close to having the necessary density, not to mention that the photons are so low energy you'd have to combine a lot of them to get to the ionizing range. But the phenomenon does exist.
Nonlinear effects are possible, like where two photons are absorbed then only one is emitted. So non-ionizing radiation could in theory interact in a way to produce ionizing energy. It's also possible that some structures are exquisitely sensitive to particular frequencies of radiation.
But a closer look shows just how unlikely such phenomena are. The probability of such interactions depends on there being sufficient energy density - you see them with megawatt lasers but not at the power levels where cell phones operate. As for some sensitive structure being present, if there was you'd think we would have found it by now.
I have a Leaf. My commute is just under 20 miles one way, most of it on the freeway at 65+Mph speeds and since this is summer in SoCal, the A/C is running pretty much all the time. When I leave the car usually says I have 90 miles of range; when I get home the car says I have somewhere between 40-50 miles of range left.
I haven't gotten to the point of "driving on electron fumes" yet, but I have gotten it down to about 15 miles of range left. That took about 75 miles of driving, so assuming there really was 15 miles of range left, the effective range on average seems to be around 90 miles. Which of course is less than 100 but more than 73.
A friend of mine does research on flare stars. For this you need to look at stars for long periods.
If you think about it, there's a major experiment underway that's already looking at lots of stars for long periods: The search for extrasolar planets. And even better, when one of these experiments finds a flare star, that data isn't especially interesting because they're looking for planets that can sustain life, and frequent solar flares are ... unhelpful in that regard. As long as the data is properly attributed, sharing with the flare star folks means you get credit for data you collected but cannot use yourself.
So it's a win-win to share data in this case, but even so there are various hoops to jump through to get access. I suspect a lot of it boils down to, "We know you and we trust you will stay within your area of interest that doesn't overlap ours."
As usual, nothing in the law is this simple.
Some of the relevant cases here are National Bellas Hess, Inc. v. Department of Revenue of Illinois, Quill Corp. v. North Dakota, and Complete Auto Transit, Inc. v. Brady. In brief, in Complete Auto the Supreme Court created a four prong test. The prongs are: (1) Substantial nexus (I used the wrong term earlier, sorry), (2) Nondiscrimination, (3) Fair apportionment, and (4) Fair relationship to services provided by the state. Of these, it seems to me that only (1) is in play here.
Quill established that a company that only does mail-order business does not create a "substantial nexus". But Amazon and other similar retailers do a lot more in the California than mail stuff to people. Among other things, they partner with local California companies to do order fulfillment.
If this ends up before the Supreme Court - and I rather suspect it will - it's a situation where two different legal doctrines - the so-called Dormant Commerce Clause and Due Process, are in conflict. Historically the Court has tended to side with Dormant Commerce Clause, which would mean this bill is unconstitutional. But I rather suspect that the scale and scope of these retailer's activities will tend to push this towards a ruling based on due process considerations. But again, IANAL, YMMV, etc. etc.
Exactly. If you buy something in California from out of state, the vendor doesn't charge you California sales tax, and you proceed to directly use that item, you are supposed to pay use tax. That's the law; if you're buying stuff from Amazon or wherever and not paying those use-taxes you are in violation. (And yes, this means a HUGE number of people are in violation, which is sort of the point.)
All this measure does is require that vendors like Amazon collect the tax instead. Since Amazon already does all sorts of order fulfillment in California and collects sales taxes when it does; this should be trivial for them to do at least.
Speaking as a California resident, I welcome this because right now keeping track of use-tax is a royal pain. In fact if anything I will be strongly inclined to prefer vendors that collect the tax because it saves me so much trouble.
As for counting on some sort of "only the feds can do this stuff" ruling so you can continue to break the law, this gets down to how what the lawyers call a jurisdictional nexus is defined. The bill appears to be tweaking that a little, but I doubt it's tweaking it enough for there to be constitutional issue. But IANAL, and even if I were, this is very deep stuff and you really need an expert's read on it.
I don't recall the exact dates, but it was around 1982 that these utilities were created. ROLLOUT saved the entire VMS system state to disk and ROLLIN brought it back, not only restoring any long-running batch jobs and so on but also dropping the boot time down from 5-10 minutes to less than 30 seconds.
Although it was easy to get ahold of ROLLIN/ROLLOUT if you wanted them Digital would never condone their use. The reason for that was simple: It depended on disk state not changing whlie the system was down, and not only was that too fragile for Digital's taste, the clustering system Digital was busy designing at the time was going to make that a practical impossibility. So this capability was never used all that time, and then clusters came out and made it unworkable.
I have two sets of solar panels - one 10 year old set that's wired up conventionally to a single inverter, and another 8-month-old set that uses per-panel microinverters.
According to the very nice graphical display of power generation the microinverters provide via a built in web monitoring thingie, I average about four hours each day operating between 90% and peak generation.
Installation of the original set of panels was a major PITA because UL delisted the inverter after it was installed. Getting a massive piece of equipment down off the wall is easy, getting the replacement back up, not so much. But once that one startup issue was dealt with, the system has been 100% reliable and has required no repairs. As for the microinverters, it's early days yet, but they've been completely reliable so far.
As for washing the panels, yes, doing that more often increases output, but in my experience, not by that much. Around here the windows need to be washed twice a year so the panels get done as part of that. No big deal.
And as for all this "resetting after a power failure" - it appears your experience with grid-tie systems is seriously out of date. My 10 year old inverter handles power failures automatically. Aside from monitoring, I haven't had to touch the thing once the replacement inverter was installed. Ditto for the microinverters.
Finally, you appear to be conflating grid-tie and off-grid setups. I agree that a fully off-grid setup isn't easy. I have battery backup as part of my original system, but since the batteries are only used when there's a grid failure they haven't needed to be replaced. (And most grid-tie systems don't need them at all.) An off-grid setup that charges and discharges the batteries every day is going to require a lot more maintenance. And when solar is the only energy source the system has to be overbuilt in the fashion you describe (just not as much as you claim). And you probably care more about keeping the panels clean when they are your only power source.
But the vast majority of solar systems are grid-tied, not off-grid. So most of your issues simply don't apply.
Now, perhaps you'll say my experience is unusual. Yes, it's only anecdotal, but I know three people with similar solar setups in the area, and their systems have all worked flawlessly since they were installed.
Maybe he's using a custom clock app or something, but on my iPhone the built in clock app has four clearly labeled mode setting buttons at the bottom: "World Clock", "Alarm", "Stopwatch", and "Timer". Pressing the one called "Alarm" to set an alarm seems, well, obvious, and when you do that you get a screen saying "no alarms" and exactly one "+" button you can press, so unless you simply freeze up at that point I don't see how this can be so confusing. In particular, no clock face is displayed at this point so there is no possibility of, "Pressing the clock image takes you through to choices about how the clock is displayed, and it's not easy to get back again."
If you want to criticize the alarm and calendar stuff on the iPhone, a better place to start is the spinning dial thing used to enter times. (Which is what comes up once you press "+".) A lot of people dislike this and find it hard to use. I don't find it difficult personally, but I have to admit I'd prefer a numeric keypad.
I grew up in Oklahoma and attended Oklahoma public schools up until high school, when I went to Groton School. The difference was stark. For example, 8th grade math in Oklahoma was still focused on basic arithmetic. (I really wish I was making that up, but I'm not.) By the time I graduated from Groton I managed to get a 5 on the AP calculus exam.
Prior to attending Groton I coudn't write worth a damn. Groton fixed that too.
But most important thing I learned there was, basically, how to think - the self-discipline needed to properly organize and express my thoughts and ideas.
Now, as for your little list.Maybe I didn't take the right classes or something, but I'm fairly sure I didn't learn enough of that "tak the talk" thing at Groton to matter - when it came time to get serious about business relations for our startup company, we went outside and hired a "suit" to do that.
As for friends helping out, I cannot recall a single instance after graduating when I asked a fellow Groton alum for help.
And as for that self-importance thing - let me just point out Groton's motto: Cui servire est regnare.
Now, I have no doubt there are private schools that provide what you describe and little else. But not all of them are like that.
TFA makes a lot of assertions about unnecessary treatment and increased costs with no associated benefits, but doesn't present or link to a single piece of actual peer-reviewed data. In the specific case of changing the cutoff of blood glucose levels from 140 to 130, the appropriate question to ask is whether or not treating the many side effects of diabetes sooner saves more than it costs.