No you don't. A basic understanding of our shared history is important for the proper functioning of a democratic society. An understanding of math beyond what is needed to balance a checkbook (or national budget) is not. 90% of us never use algebra, even once, after leaving school. It is basically pointless for non-techies.
It's a very good analogy, actually. You're right that a basic understanding of our shared history is important, but the vast majority of people rarely if ever use more than what they'd learned of history by the end of middle school.
I'd argue that that doesn't make it worthless to teach more in high school--although only maybe 5% will benefit directly from taking more advanced history classes, you don't know which 5% it is and you're handicapping your culture slightly by not providing that knowledge. It's exactly the same argument that I'd make about algebra, except more people are likely to use algebra at some point.
And in both cases I'd argue that even if you don't directly pull some history facts out now and again or have to figure out how much each of those 10 cars cost pre-tex, you still benefit indirectly by taking those classes--for one thing, it's a lot easier to forget the final levels of coursework than it is the stuff that you used again in later years of school. So if you want people to know, say, algebra I and 7th grade history for the rest of their life, then a very good way to help do that is to make sure they take algebra II and 8th grade history--not only did they just learn the former out of the book, but they practiced it as part of something else for another year.
I was a Data room tech, field engineer, service tech, systems administrator, and second level support tech (not in that order) for over thirty years, and while I had taken (and done well at) algebra, calculus and geometry/trigonometry etc I don't remember ever actually using it on the job
I work as a programmer. I took and did well in Spanish, geography, history, chemistry, physics, biology, sexual education, art, wood shop and gym classes in high school, but I don't remember ever using them on the job.
The idea that education should be reduced to voc-tech work is bizarre.
You realize that initial covering of this price was required by their underwriter, Morgan Stanley. They were basically sustaining the price at $38 all of opening day
Facebook's goal is to sell their IPO at the highest possible price. They don't really care whether it was individual investors or a large bank or huge corporate investors who is paying the money--as long as someone out there is paying $38/share that's good for them.
Now, Morgan Stanley may have done something stupid by agreeing to prop things up at a required price or something, but that's a separate issue--Zuckerberg/Facebook would be negligent to its stakeholders and employees if they knew Morgan Stanley (or whoever) was offering to pay $38/share and they said "no, that's too much, we only want $25". Calling it Zuckerberg's mistake is just sour grapes on the part of those--be they Morgan Stanley or individuals or whatever--who shelled out $38/share only to see it drop after the fact.
Thou are speaketh in a comical form and manor. I say art thoueth a baffoon or comical jester? You simply must continue on. Doesth thou produce plays? Pray tell where doest one sign up to get tickets? I wish to submit myself to thy whimsicle banter!
Art thou a foreigner still attempting to apprehend the vagaries of the English tongue?
Thy grammar is weak; though switchest betwixt second person singular and plural without reason, and though fabricatest idiomatic chimerae such as "thoueth" without regard for linguistic merit. Thy conjugation is often off: thou dost use the third person indicative form "speaketh" when thou speakest directly to a second person. Thine usage of helping verbs is often spurious, and thine orthological butcherings of "dost" and "doth" are most droll.
A run from NY to LA would run you several billion dollars just to get started and several hundred million every year after that for maintenance and repair. So, the real question is: is there enough traffic between NY and LA (for example) to recuperate the cost of construction and operations. I highly, highly doubt the answer is yes.
If it were that cheap it'd be "yes, absolutely, and we're going to hook up every major city as well."
They're talking about spending over $150 billion for the high-speed rail from San Francisco to Los Angeles, and Amtrak's discussing $100 million in track improvements to get TGV-level speeds from Boston to Washington, DC.
Those are the nicest train routes in the country, but they're peanuts compared to how profitable a NY-to-LA in under an hour route would be if it only cost a few billion to get going and several hundred million a year to operate.
I could define an extension to C where the + operator works between a string and an integer as follows:
mystr + myint is defined such that it returns a string of the same length as mystr, but where each element of the return value is myint less than the equivalent char in mystr. So "dddd" + 2 returns "bbbb", "zzzzz" + 3 returns "wwwww", etc.
This language is more expressive than C, but using the new feature is liable to confuse and obfuscate more than it is to help out.
This is not the fault of the language, however, but of the developer violating the principle of least surprise. Why would I expect an addition operator to subtract from elements of a string? Pragmatically, unless one has a very special application domain that requires it, one wouldn't normally do this.
No, the C-with-extended-addition language (which defines the + operator to work as in the example) is at fault as a language--in this case, it's the language itself that violates the principle of least surprise. (Note that the example is not of a language with operator overloading where a programmer defined a poor + operator; it's of a language that defines the + operator as given).
It's an obviously broken feature to illustrate the fact that more expressiveness in a language does not automatically mean that the language is better. Now, obviously the developer can work around a broken language feature by not using it (and many successful C++ projects are successful in large part by defining an acceptable subset of the language to use), but the need to do so is a sign that there's at the very least a bad smell in the language itself.
Speaking purely from anecdote, I've worked multiple C++ projects, both large and small, and have had a couple disasters, and more successes. The difference was in that the people and projects who successfully used C++ a) took the time to understand what C++ could do for their domain, and b) implemented their domain using the parts of the language that were well understood, as you point out. As an example, exceptions were often put to the wayside, as they weren't generally well understood by the respective teams. So they weren't used.
Yes, it's far from impossible to write good code in C++. But in my experience some languages tend to lead you more naturally toward writing better code and others do the opposite. All of them require some effort, and as you note a reasonable eye for maintainable design. But C++ has, on the projects I've worked on, been one of the languages where we've spent more time fighting the language than normal.
You technically could...but why would you? If you want to write object oriented C, there's C++. What's the benefit?
C++ is a more expressive language than C. That's not necessarily better, though--I could define an extension to C where the + operator works between a string and an integer as follows:
mystr + myint is defined such that it returns a string of the same length as mystr, but where each element of the return value is myint less than the equivalent char in mystr. So "dddd" + 2 returns "bbbb", "zzzzz" + 3 returns "wwwww", etc.
This language is more expressive than C, but using the new feature is liable to confuse and obfuscate more than it is to help out.
C++ has a huge helping of extended features over C. For whatever reason, it seems to fall into a class of languages (along with PHP, Perl, C#, and Java) that seem more expressive up front but wind up creating less understandable and less maintainable code in the long term than using alternatives (e.g. C, Python, ML, Lisp) has at the places I've worked. Programming language is really a low-level human interface problem, so it's really tough to know until you use a language for a long time in a large group whether it's likely to be a big advance or not--there are some (e.g. Ruby, Scheme, Objective C) that I don't feel I have a grasp on at all yet.
But given pretty considerable experience using both C++ and C, I'm fairly confident that using C++ is a far worse idea than using C for new projects; that's not to say that defining some subset of C++ and using it on a well-defined project might not be a good idea, but essentially to answer your question: Because C++ in practice turns out to be horrifically unmaintainable in most cases compared to straight C, despite seemingly requiring less contortions to write decent OO code.
And studies like TIOBE sort of help quantify this and reassure me that it's not just personal anecdote, it's by and large the experience that most of the programming world has had over the last 15+ years.
OO is *always* the best choice, depending on your definition of OO.
Believing this is a tremendously bad mistake to make. There are plenty of other high-level programming paradigms that for some tasks can be vastly superior to OO--e.g. straight functional programming in ML or Dylan, or logic programming in Prolog, or explicitly stack-based programming in Forth or Postscript. There is no "one size fits all" when it comes to programming.
"for gaming" being the key point--the vast majority of desktops aren't for gaming, and using a gaming rig to measure power draw is going to generate a spurious answer if you're trying to figure out what a normal desktop draws.
I did. Any "normal" fast rig, is gonna pull 150W min while idle. I've got an Athlon that can idle ~90W, screens off.
Pick one. That's a reasonable idle figure for a fast rig (e.g. a gaming rig with an outlandish GPU), but it's way, way more than a normal desktop computer draws at idle. We measure the desktops at work occasionally and they've hovered around the 70-90W when idle range (with some ups and downs) for the last decade despite various vendors and different iterations of CPUs, graphics cards, etc.
The first film is great, not action oriented but really epic. It's a true science fiction film. And because of this it's the perfect film for someone that is not a trekkie.
The first film's basically a drawn-out version of one of the original episodes (the Changeling) that starts pretty well and ends strong but has tons of dead space in the middle because the plot's too thin to support more than 45-50 minutes. You'd be better off watching the original episode or many of TOS episodes instead.
MacKenzie et al. and Meyer et al. merely show general Fitts law relationships for mice
Now I'm confused about what your objection is--Fitts' law is exactly the topic of discussion.
So, looks like I was right: you haven't read any studies actually showing that Mac menus are faster.
And the combination of these two shows me that you don't understand Fitts' law.
you cannot generalize results from that to large screens or multi-screen setups.
It's entirely possible that things are different on different systems. I suspect large monitors don't make much difference unless you have your mouse speed dialed way down (I have a 27" screen but it's still less than 3" of mouse movement to traverse it), but multi-screen setups probably do (at the very least a menubar per screen is probably faster than a single menubar).
MacKenzie, I. Scott, Sellen, Abigail and Buxton, Bill (1991): A Comparison of Input Devices in Elemental Pointing and Dragging Tasks. In: Robertson, Scott P., Olson, Gary M. and Olson, Judith S. (eds.) Proceedings of the ACM CHI 91 Human Factors in Computing Systems Conference April 28 - June 5, 1991, New Orleans, Louisiana. pp. 161-166.
Single menu is the perfect example here. It is 100% pure form over function.
Quite the opposite, actually; it's not as good-looking as having a menubar in each window, but every UI study I've seen shows it's faster to operate than one-per-window. Yes, it's slightly further away, but that's completely offset by the fact that it's a) in a fixed position, so you don't need to find it to know where to move; and b) it's at a screen edge, so you can simply move the mouse up and left without worrying about overshooting it rather than having to target a specific area (and, apropos of another ongoing topic, you can even open the leftmost menu blind).
It also saves screen real estate, which is a functional matter for many people.
I personally think it's ugly enough that I stick with menu-per-window layouts myself, but I don't pretend that's anything other than an aesthetic and what-I'm-accustomed-to decision.
Having said that, there's a reason why Washington was nicknamed "Cincinnatus". Strong leaders who would voluntarily give up power when it's no longer needed have always been hard to find.
It's a terrible nickname, in general, though. Cincinnatus stepped down from an essentially military position at the end of the war, but he didn't similarly give up his truly governmental position as Senator. And he spent his entire career opposing the sort of Constitution that Washington fought for, and opposing equality for the people.
In fact, he served as one of the most widely hated politicians in Roman history well into his 80s--he was beloved by the patrician noble class who wrote numerous hagiographies, but as a Senator and politician he was a horribly classist ruler who spent his entire career opposing the right of the lower and middle classes to involve themselves in governance. Even where the lower classes had the legal right to form tribunes with governmental power, he'd send his son Caeso Quinctius with troops to break them up and prevent them from reaching quorum (thereby preventing any general legal reform that might benefit the majority).
Indeed, he was so despised as a ruler that the Law of the Twelve Tables and the Roman Constitution were largely written as a backlash to his unmitigated pro-aristocratic draconianism. Plebian leader Terentilius opposed him for decades and advocated the passage of codified laws protecting all Romans and allowing plebs to run for Consul; eventually he managed to organize a secessio plebis (essentially a general strike) while the army was away and unable to put it down, thereby outmaneuvering Cincinatus' pro-patrician bloc and forcing the passage of the Law of Twelve Tables.
It's also worth noting that the dictatorship that he's so famous for giving up was a six-month term as Magister Populi ("Master of the People"), which was created for the express purpose of putting down the Aequi and other tribes in revolt. It's not like he was absolute ruler for life and gave it up.
And, as noted, there are many kinds of mean (geometric and arithmetic means being the most common, but there are harmonic mean, quadratic means, Lehmer means, contraharmonic means, etc). Some of those can be generalized into power means.
1. Temple is a special case when it comes to medicine; it's the home of Scott & White Medical Center, a VA hospital, and a huge Texas A&M med school campus--at least when my dad was there, it had more doctors per capita than any other city in the US.
2. Your POV is skewed by living in Texas. Temple's over 60,000 people in population, hardly a small city*. Portland, ME is the size of Temple and has a nationally acclaimed pediatric hospital. Burlington, VT is even smaller (45,000ish) and has one, too (both are the largest city in their respective states).
*Even in Texas most cities are closer in size to Lampasas or Marble Falls than to Temple--Temple's in the top 10% of Texas cities by population.
Maybe I will prove myself to be in the lower half, and not being a mathematician myself, I was lead to believe that the average was the sum of scores divided by the number of samples.
That's the arithmetic mean, which is one form of average. Other common ones Geometric mean--the nth root of the product of the scores, where n is the number of samples. In your example, that's approximately 110.37 The median (rank the samples in order and take the middle one, or the mean of the middle two if there are an even number of samples)--in your example, that's 120. The mode (the most commonly occuring value(s) in the set)--in your example, there are 5 modes (50, 75, 120, 130, and 280). If you change your example set to 50, 50, 130, 145, and 280, then there is only one mode: 50.
But, yeah, the statement you're objecting to is either wrong or under-specified (it'd be basically correct if it had specified the median).
"Never underestimate the bandwidth of a station wagon full of tapes hurtling down the highway." â"Tanenbaum, Andrew S. (1996)
I hope Tanenbaum cited his source; That quote was on USENET as far back as 1985 and is obviously considered old at the time. It's most likely Dennis Ritchie, but Tom Reidel, Warren Jackson, and occasionally Bob Sutterfield are also candidates.
D&D and AD&D had several versions alongside each other (they were separate games developed in parallel by TSR). After Wizards of the Coast bought TSR, they merged them into a single line that was named D&D but was more like TSR's AD&D rules. Consequently there are 2 different things called D&D 3rd Edition, D&D 4th Edition--to avoid confusion, Wizards of the Coast refers to the old TSR-released ones as "D&D Version 3" and reserves the name "3rd Edition" for the post-WotC merged game. But historically the TSR one was also called D&D 3rd Edition.
The timeline was something like: D&D 1st Edition/Chainmail rules D&D 1st Edition/Greyhawk rules D&D 2nd Edition
AD&D 1st Edition D&D 3rd Edition D&D 4th Edition
AD&D 2nd Edition D&D 5th Edition (Wizards of the Coast buys them out here)
D&D 3rd Edition
D&D 3.5th Edition
D&D 4th Edition
Wizard of the Coast's D&D 3rd Edition and later are evolutions of the AD&D rules more than of the D&D rules Unofficially the later years of AD&D 2nd Edition are called the 2.5th edition sometimes.
The original 1st edition of D&D you had to have the Chainmail table-top game rules to resolve combat; that changed when the Greyhawk supplement was released, giving D&D its own combat rules. So a lot of people consider the change from Chainmail to Greyhawk rules to be as significant as an official new edition.
As an ancient D&D player, I must say you are wrong. The Three Sided Die is shaped like a football with three ridges. The football shape keeps it from standing on either end, and you read the top ridge.
The ones I saw were all a football shape where you read the number off the bottom of the roll (similar to how you read a 4-sided die). No ridges or top reading.
They were always pretty rare, though, and don't give you any advantage over the d6. I'd say after the d4/6/8/10/12/20, the only other ones that were somewhat common were the the d30, and the d100, though a couple of my European friends had d34 that are apparently used for some bingo-type game over there.
Aside from those the only other one I've seen that got much traction was essentially an opaque d10 nested inside a transparent d10: another attempt at a d100.
I've seen at least log-shaped d5/d7 and more normal d14, d18, d24, too, but those are all basically novelties.
And then obviously all the dice with other stuff printed on the sides, but that's a whole other conversation.
You're absolutely right that there's strict differentiation between classes in D&D compared to other systems (or at least there was, before 3E's skills and feats). There are pluses and minuses to both mechanisms, IMO, but forgetting that distinction is why the post 2nd-Edition D&D rules have all sucked: either you want to play Dungeons and Dragons, in which case you want strong class delineation, or you want a skill-based game a la GURPS. 3E tried to blend the two with just plain ugly results.
Everyone I know played 3rd, 4th but eventually consolidated on Pathfinder. . Really? The 3rd Editions's the one that accelerated level advancement a lot; the game works best and is most fun for low-mid level characters, and that one change made it much tougher to run an ongoing regular campaign for more than a year or two. And the skills and feats changes made it feel less like D&D and more like a generic GURPsy fantasy RPG. Almost everyone I know settled on the 2nd Edition eventually.
Slashdot Mirror
No you don't. A basic understanding of our shared history is important for the proper functioning of a democratic society. An understanding of math beyond what is needed to balance a checkbook (or national budget) is not. 90% of us never use algebra, even once, after leaving school. It is basically pointless for non-techies.
It's a very good analogy, actually. You're right that a basic understanding of our shared history is important, but the vast majority of people rarely if ever use more than what they'd learned of history by the end of middle school.
I'd argue that that doesn't make it worthless to teach more in high school--although only maybe 5% will benefit directly from taking more advanced history classes, you don't know which 5% it is and you're handicapping your culture slightly by not providing that knowledge. It's exactly the same argument that I'd make about algebra, except more people are likely to use algebra at some point.
And in both cases I'd argue that even if you don't directly pull some history facts out now and again or have to figure out how much each of those 10 cars cost pre-tex, you still benefit indirectly by taking those classes--for one thing, it's a lot easier to forget the final levels of coursework than it is the stuff that you used again in later years of school. So if you want people to know, say, algebra I and 7th grade history for the rest of their life, then a very good way to help do that is to make sure they take algebra II and 8th grade history--not only did they just learn the former out of the book, but they practiced it as part of something else for another year.
I was a Data room tech, field engineer, service tech, systems administrator, and second level support tech (not in that order) for over thirty years, and while I had taken (and done well at) algebra, calculus and geometry/trigonometry etc I don't remember ever actually using it on the job
I work as a programmer. I took and did well in Spanish, geography, history, chemistry, physics, biology, sexual education, art, wood shop and gym classes in high school, but I don't remember ever using them on the job.
The idea that education should be reduced to voc-tech work is bizarre.
You realize that initial covering of this price was required by their underwriter, Morgan Stanley. They were basically sustaining the price at $38 all of opening day
Facebook's goal is to sell their IPO at the highest possible price. They don't really care whether it was individual investors or a large bank or huge corporate investors who is paying the money--as long as someone out there is paying $38/share that's good for them.
Now, Morgan Stanley may have done something stupid by agreeing to prop things up at a required price or something, but that's a separate issue--Zuckerberg/Facebook would be negligent to its stakeholders and employees if they knew Morgan Stanley (or whoever) was offering to pay $38/share and they said "no, that's too much, we only want $25". Calling it Zuckerberg's mistake is just sour grapes on the part of those--be they Morgan Stanley or individuals or whatever--who shelled out $38/share only to see it drop after the fact.
Thou are speaketh in a comical form and manor. I say art thoueth a baffoon or comical jester?
You simply must continue on. Doesth thou produce plays? Pray tell where doest one sign up to get tickets? I wish to submit myself to thy whimsicle banter!
Art thou a foreigner still attempting to apprehend the vagaries of the English tongue?
Thy grammar is weak; though switchest betwixt second person singular and plural without reason, and though fabricatest idiomatic chimerae such as "thoueth" without regard for linguistic merit. Thy conjugation is often off: thou dost use the third person indicative form "speaketh" when thou speakest directly to a second person. Thine usage of helping verbs is often spurious, and thine orthological butcherings of "dost" and "doth" are most droll.
A run from NY to LA would run you several billion dollars just to get started and several hundred million every year after that for maintenance and repair. So, the real question is: is there enough traffic between NY and LA (for example) to recuperate the cost of construction and operations. I highly, highly doubt the answer is yes.
If it were that cheap it'd be "yes, absolutely, and we're going to hook up every major city as well."
They're talking about spending over $150 billion for the high-speed rail from San Francisco to Los Angeles, and Amtrak's discussing $100 million in track improvements to get TGV-level speeds from Boston to Washington, DC.
Those are the nicest train routes in the country, but they're peanuts compared to how profitable a NY-to-LA in under an hour route would be if it only cost a few billion to get going and several hundred million a year to operate.
This is not the fault of the language, however, but of the developer violating the principle of least surprise. Why would I expect an addition operator to subtract from elements of a string? Pragmatically, unless one has a very special application domain that requires it, one wouldn't normally do this.
No, the C-with-extended-addition language (which defines the + operator to work as in the example) is at fault as a language--in this case, it's the language itself that violates the principle of least surprise. (Note that the example is not of a language with operator overloading where a programmer defined a poor + operator; it's of a language that defines the + operator as given).
It's an obviously broken feature to illustrate the fact that more expressiveness in a language does not automatically mean that the language is better. Now, obviously the developer can work around a broken language feature by not using it (and many successful C++ projects are successful in large part by defining an acceptable subset of the language to use), but the need to do so is a sign that there's at the very least a bad smell in the language itself.
Speaking purely from anecdote, I've worked multiple C++ projects, both large and small, and have had a couple disasters, and more successes. The difference was in that the people and projects who successfully used C++ a) took the time to understand what C++ could do for their domain, and b) implemented their domain using the parts of the language that were well understood, as you point out. As an example, exceptions were often put to the wayside, as they weren't generally well understood by the respective teams. So they weren't used.
Yes, it's far from impossible to write good code in C++. But in my experience some languages tend to lead you more naturally toward writing better code and others do the opposite. All of them require some effort, and as you note a reasonable eye for maintainable design. But C++ has, on the projects I've worked on, been one of the languages where we've spent more time fighting the language than normal.
Absolutely, and it's an excellent technique to apply in a lot of cases. Just not always.
You technically could...but why would you? If you want to write object oriented C, there's C++. What's the benefit?
C++ is a more expressive language than C. That's not necessarily better, though--I could define an extension to C where the + operator works between a string and an integer as follows:
mystr + myint is defined such that it returns a string of the same length as mystr, but where each element of the return value is myint less than the equivalent char in mystr. So "dddd" + 2 returns "bbbb", "zzzzz" + 3 returns "wwwww", etc.
This language is more expressive than C, but using the new feature is liable to confuse and obfuscate more than it is to help out.
C++ has a huge helping of extended features over C. For whatever reason, it seems to fall into a class of languages (along with PHP, Perl, C#, and Java) that seem more expressive up front but wind up creating less understandable and less maintainable code in the long term than using alternatives (e.g. C, Python, ML, Lisp) has at the places I've worked. Programming language is really a low-level human interface problem, so it's really tough to know until you use a language for a long time in a large group whether it's likely to be a big advance or not--there are some (e.g. Ruby, Scheme, Objective C) that I don't feel I have a grasp on at all yet.
But given pretty considerable experience using both C++ and C, I'm fairly confident that using C++ is a far worse idea than using C for new projects; that's not to say that defining some subset of C++ and using it on a well-defined project might not be a good idea, but essentially to answer your question: Because C++ in practice turns out to be horrifically unmaintainable in most cases compared to straight C, despite seemingly requiring less contortions to write decent OO code.
And studies like TIOBE sort of help quantify this and reassure me that it's not just personal anecdote, it's by and large the experience that most of the programming world has had over the last 15+ years.
OO is *always* the best choice, depending on your definition of OO.
Believing this is a tremendously bad mistake to make. There are plenty of other high-level programming paradigms that for some tasks can be vastly superior to OO--e.g. straight functional programming in ML or Dylan, or logic programming in Prolog, or explicitly stack-based programming in Forth or Postscript. There is no "one size fits all" when it comes to programming.
"for gaming" being the key point--the vast majority of desktops aren't for gaming, and using a gaming rig to measure power draw is going to generate a spurious answer if you're trying to figure out what a normal desktop draws.
Hence the first 2 sentences of my last reply.
I did. Any "normal" fast rig, is gonna pull 150W min while idle.
I've got an Athlon that can idle ~90W, screens off.
Pick one. That's a reasonable idle figure for a fast rig (e.g. a gaming rig with an outlandish GPU), but it's way, way more than a normal desktop computer draws at idle. We measure the desktops at work occasionally and they've hovered around the 70-90W when idle range (with some ups and downs) for the last decade despite various vendors and different iterations of CPUs, graphics cards, etc.
The first film is great, not action oriented but really epic. It's a true science fiction film.
And because of this it's the perfect film for someone that is not a trekkie.
The first film's basically a drawn-out version of one of the original episodes (the Changeling) that starts pretty well and ends strong but has tons of dead space in the middle because the plot's too thin to support more than 45-50 minutes. You'd be better off watching the original episode or many of TOS episodes instead.
But you claimed to have read studies that show that Mac menus are faster than Windows menus
At no point did I mention "Mac menus" or "Windows menus". I don't have a horse in the Windows vs. Mac race.
MacKenzie et al. and Meyer et al. merely show general Fitts law relationships for mice
Now I'm confused about what your objection is--Fitts' law is exactly the topic of discussion.
So, looks like I was right: you haven't read any studies actually showing that Mac menus are faster.
And the combination of these two shows me that you don't understand Fitts' law.
you cannot generalize results from that to large screens or multi-screen setups.
It's entirely possible that things are different on different systems. I suspect large monitors don't make much difference unless you have your mouse speed dialed way down (I have a 27" screen but it's still less than 3" of mouse movement to traverse it), but multi-screen setups probably do (at the very least a menubar per screen is probably faster than a single menubar).
The best is:
Unfortunately it's not available online.
http://www.umich.edu/~bcalab/documents/MeyerSmithKornblumAW1990.pdf is freely available and somewhat related, as are:
http://www.mackido.com/Interface/menu_target.html
http://www.codinghorror.com/blog/2006/08/fitts-law-and-infinite-width.html
Macs actually started off with a menu-per-window but moved to the current model after doing such studies; you can see images of the earlier implementation here:
http://folklore.org/projects/Macintosh/images/polaroids/polaroids.14.jpg
Single menu is the perfect example here. It is 100% pure form over function.
Quite the opposite, actually; it's not as good-looking as having a menubar in each window, but every UI study I've seen shows it's faster to operate than one-per-window. Yes, it's slightly further away, but that's completely offset by the fact that it's a) in a fixed position, so you don't need to find it to know where to move; and b) it's at a screen edge, so you can simply move the mouse up and left without worrying about overshooting it rather than having to target a specific area (and, apropos of another ongoing topic, you can even open the leftmost menu blind).
It also saves screen real estate, which is a functional matter for many people.
I personally think it's ugly enough that I stick with menu-per-window layouts myself, but I don't pretend that's anything other than an aesthetic and what-I'm-accustomed-to decision.
Having said that, there's a reason why Washington was nicknamed "Cincinnatus". Strong leaders who would voluntarily give up power when it's no longer needed have always been hard to find.
It's a terrible nickname, in general, though. Cincinnatus stepped down from an essentially military position at the end of the war, but he didn't similarly give up his truly governmental position as Senator. And he spent his entire career opposing the sort of Constitution that Washington fought for, and opposing equality for the people.
In fact, he served as one of the most widely hated politicians in Roman history well into his 80s--he was beloved by the patrician noble class who wrote numerous hagiographies, but as a Senator and politician he was a horribly classist ruler who spent his entire career opposing the right of the lower and middle classes to involve themselves in governance. Even where the lower classes had the legal right to form tribunes with governmental power, he'd send his son Caeso Quinctius with troops to break them up and prevent them from reaching quorum (thereby preventing any general legal reform that might benefit the majority).
Indeed, he was so despised as a ruler that the Law of the Twelve Tables and the Roman Constitution were largely written as a backlash to his unmitigated pro-aristocratic draconianism. Plebian leader Terentilius opposed him for decades and advocated the passage of codified laws protecting all Romans and allowing plebs to run for Consul; eventually he managed to organize a secessio plebis (essentially a general strike) while the army was away and unable to put it down, thereby outmaneuvering Cincinatus' pro-patrician bloc and forcing the passage of the Law of Twelve Tables.
It's also worth noting that the dictatorship that he's so famous for giving up was a six-month term as Magister Populi ("Master of the People"), which was created for the express purpose of putting down the Aequi and other tribes in revolt. It's not like he was absolute ruler for life and gave it up.
And, as noted, there are many kinds of mean (geometric and arithmetic means being the most common, but there are harmonic mean, quadratic means, Lehmer means, contraharmonic means, etc). Some of those can be generalized into power means.
1. Temple is a special case when it comes to medicine; it's the home of Scott & White Medical Center, a VA hospital, and a huge Texas A&M med school campus--at least when my dad was there, it had more doctors per capita than any other city in the US.
2. Your POV is skewed by living in Texas. Temple's over 60,000 people in population, hardly a small city*. Portland, ME is the size of Temple and has a nationally acclaimed pediatric hospital. Burlington, VT is even smaller (45,000ish) and has one, too (both are the largest city in their respective states).
*Even in Texas most cities are closer in size to Lampasas or Marble Falls than to Temple--Temple's in the top 10% of Texas cities by population.
Maybe I will prove myself to be in the lower half, and not being a mathematician myself, I was lead to believe that the average was the sum of scores divided by the number of samples.
That's the arithmetic mean, which is one form of average. Other common ones
Geometric mean--the nth root of the product of the scores, where n is the number of samples. In your example, that's approximately 110.37
The median (rank the samples in order and take the middle one, or the mean of the middle two if there are an even number of samples)--in your example, that's 120.
The mode (the most commonly occuring value(s) in the set)--in your example, there are 5 modes (50, 75, 120, 130, and 280). If you change your example set to 50, 50, 130, 145, and 280, then there is only one mode: 50.
But, yeah, the statement you're objecting to is either wrong or under-specified (it'd be basically correct if it had specified the median).
"Never underestimate the bandwidth of a station wagon full of tapes hurtling down the highway." â"Tanenbaum, Andrew S. (1996)
I hope Tanenbaum cited his source; That quote was on USENET as far back as 1985 and is obviously considered old at the time. It's most likely Dennis Ritchie, but Tom Reidel, Warren Jackson, and occasionally Bob Sutterfield are also candidates.
D&D and AD&D had several versions alongside each other (they were separate games developed in parallel by TSR). After Wizards of the Coast bought TSR, they merged them into a single line that was named D&D but was more like TSR's AD&D rules. Consequently there are 2 different things called D&D 3rd Edition, D&D 4th Edition--to avoid confusion, Wizards of the Coast refers to the old TSR-released ones as "D&D Version 3" and reserves the name "3rd Edition" for the post-WotC merged game. But historically the TSR one was also called D&D 3rd Edition.
The timeline was something like:
D&D 1st Edition/Chainmail rules
D&D 1st Edition/Greyhawk rules
D&D 2nd Edition
AD&D 1st Edition
D&D 3rd Edition
D&D 4th Edition
AD&D 2nd Edition
D&D 5th Edition
(Wizards of the Coast buys them out here)
D&D 3rd Edition
D&D 3.5th Edition
D&D 4th Edition
Wizard of the Coast's D&D 3rd Edition and later are evolutions of the AD&D rules more than of the D&D rules
Unofficially the later years of AD&D 2nd Edition are called the 2.5th edition sometimes.
The original 1st edition of D&D you had to have the Chainmail table-top game rules to resolve combat; that changed when the Greyhawk supplement was released, giving D&D its own combat rules. So a lot of people consider the change from Chainmail to Greyhawk rules to be as significant as an official new edition.
As an ancient D&D player, I must say you are wrong. The Three Sided Die is shaped like a football with three ridges. The football shape keeps it from standing on either end, and you read the top ridge.
The ones I saw were all a football shape where you read the number off the bottom of the roll (similar to how you read a 4-sided die). No ridges or top reading.
They were always pretty rare, though, and don't give you any advantage over the d6. I'd say after the d4/6/8/10/12/20, the only other ones that were somewhat common were the the d30, and the d100, though a couple of my European friends had d34 that are apparently used for some bingo-type game over there.
Aside from those the only other one I've seen that got much traction was essentially an opaque d10 nested inside a transparent d10: another attempt at a d100.
I've seen at least log-shaped d5/d7 and more normal d14, d18, d24, too, but those are all basically novelties.
And then obviously all the dice with other stuff printed on the sides, but that's a whole other conversation.
You're absolutely right that there's strict differentiation between classes in D&D compared to other systems (or at least there was, before 3E's skills and feats). There are pluses and minuses to both mechanisms, IMO, but forgetting that distinction is why the post 2nd-Edition D&D rules have all sucked: either you want to play Dungeons and Dragons, in which case you want strong class delineation, or you want a skill-based game a la GURPS. 3E tried to blend the two with just plain ugly results.
Everyone I know played 3rd, 4th but eventually consolidated on Pathfinder.
.
Really? The 3rd Editions's the one that accelerated level advancement a lot; the game works best and is most fun for low-mid level characters, and that one change made it much tougher to run an ongoing regular campaign for more than a year or two. And the skills and feats changes made it feel less like D&D and more like a generic GURPsy fantasy RPG. Almost everyone I know settled on the 2nd Edition eventually.