Apparently you do not know your history. I'm not an Opera user myself (except for occasional testing purposes), but its long-term servival is pretty much settled. Of the dozen or so browsers with anything approaching significant market share, Opera is the second-oldest, after Netscape. In that time, Opera has never lost any significant percentage of what market share it has. Granted, it has also never had a very large market share, compared to the top two players. OTOH, which other two browsers are the top two players, and in what order, has changed three or four times while Opera has stayed consistently in the top four. Opera's long-term survival, assuming we define "long-term" in a way that makes sense in the context of the web browser market, isn't really in question.
Whether it will ever climb into the top two is another matter.
> Geeze, here it is 2006 and people still think that the return address in unsigned > email means ANYTHING.
Well, yeah. I had to explain to two coworkers just last week that the scary messages they were getting weren't really from eBay, and they were quite surprised. (So I told them that if they were concerned that they might need to check their eBay accounts, to use the bookmarks they usually use to go there, because they would know that those really go to eBay. The link in this message only says it goes to eBay, and really it goes someplace else, to another site. Such gasps of outrage and astonishment as I then heard, you'd have thought I was telling them that their husbands lied about the business trip and were really with in Las Vegas with girlfriends.)
This is at least partly because of the way mailreaders present the data. Instead of showing the headers as part of the message (which is, essentially, how they're transmitted), most mail readers parse the headers and present certain pieces of data from them (the From address, for instance) separately from the message, as metadata. Well, yeah, it *is* metadata in a sense, but the way it's presented makes it appear, to the casual user, as if it's something the mailreader knows about the message, rather than something the message claims about itself. Other critical headers, such as Receives:, are not shown at all (unless the user specifically goes looking for them in a "Show All Headers" or "View Message Source" option or somesuch.
There are, of course, good solid usability reasons why these things are the way they are, but it doesn't take a doctor of psychology to tell you what people are going to think as a result.
Personally I'd like to see the information parsed out of the headers, especially the sender information, labelled just a little differently, e.g., "Claims to be From:". I'm not sure that would entirely solve the problem, but it might help a little. I'm also deeply annoyed that our ISP's mail server accepts HTML messages for delivery (if we had our own mail server in house it sure wouldn't), and that all the decent, deployable, user-friendly mail clients I can find happily render and display HTML mail. Even recent versions of Pegasus cannot, as near as I can determine, be configured to show the source or treat the HTML as an attachment.
> > As for Notepad, have you ever tried to edit a 6GB text file? > vi can open files of any size because it caches to disk and only has part in memory
Notepad, as I'm sure you're aware, is not vi.
> And you can get vi for windows
So, to clarify, are you saying that because it is possible to obtain *other* software for Windows, that therefore Windows should not ship with a text editor that can edit large files? Perhaps there is no reason for Notepad to be included at all?
> I do this sort of thing all of the time with (32-bit) Safari on a computer with 1 GB > of memory. It uses virtual memory to page out parts of memory to disk. I'd guess IE > does the same thing.
Now I think you may be (if I understand what you're saying) confusing physical memory with address space. I don't *think* you're implying that each application should handle virtual memory management itself (rather than having the OS do this), although it sort of sounds that way in the above-quoted paragraph. Either way, though, whether the app does the virtual memory management or whether, as is more usual, the OS handles that, the address size still limits how much memory, real or virtual, can be addressed. If the hardware and kernel are 64-bit, therefore, it makes sense for applications such as web browsers also to be 64-bit.
> As for semantics, every word processor I know supports it and encourages its use. > The 'right' way to use Word is to use Styles for formatting.
The word processors support that in *theory*, but in practice the interface doesn't work very well for that, and so vanishingly close to everyone ends up using word processors the ''wrong'' way, i.e., in a standard WYSIWYG semantics-are-irrelevant fashion.
> The font issue is not unique to web pages: If you e-mail a word processing document > to someone who lacks the right fonts, it will look different on their computer.
Word processing is designed for printing on paper. What happens when you send it by email is about as relevant as what happens when you try to display it on a dumb terminal. (Yes, I know there are people who send word processing documents by email. There are also people who send cellphone text messages while driving. People *do* all sorts of inane things.)
> Actually hand coding HTML is just not realistic for most users.
Perhaps you would like to explain why it is unrealistic?
Interesting possibility. When I first read the story I of course immediately thought, like most everyone else here I suspect, "Wow, there must be some kind of interdimensional portal, and it's all coming through that from an alien planet!" However, now that you mention it, I suppose after all it *could* just be coming from underground.
> This is one thing that has always bugged me about (most) religions. On the one hand > it is said that God (s/God/Deity of your choice) is infallible, and that the Bible > (s/Bible/Religous text of your choice) is the word of God. But on the other hand, > people say that their religion is flexible and adaptive.
Two different groups of people. "Flexible and adaptive" is the sort of thing the liberals and eccumenists say; conservatives (usually called "fundamentalists" if you're liberal) are the ones who believe in things like infallibility and inerrancy. There is very little overlap between the two groups.
Where it gets confusing is when you start trying to categorize denominations. In any given denomination, you usually have both sorts of people (although one or the other may dominate), so you hear both kinds of comments coming out of them.
ISTR that this is not the first year it "just" came out and also not the first year that was going to be the year BTX completely took over. I suspect 2005 will not be the last year that BTX fails to gain any significant traction in the market. Fundamentally, the considerations are like this. On the one hand, there's a major tech company or two that likes BTX and wants it to be the industry standard. On the other hand, the benefits are minor, and nobody else cares, least of all consumers, and most of the big OEMs would just as soon stick with the existing industry standard. I don't know if this reminds *you* of anything, but it makes *me* think of the MicroChannel expansion bus which, year after year, relentlessly, continued to *not* take over the expansion bus market.
> Notepad? IE? How could those ever need more than 4 GB of memory?
As for Notepad, have you ever tried to edit a 6GB text file? (Granted, Notepad is *not* what I would use for any serious editing job in any case...) As for IE,... well, okay, so let's imagine that IE got improved to the point where you could actually use it as your regular web browser. In that case, you could have half a dozen IE windows each with 20-30 tabs open, and you could leave some of those windows open for weeks on end and, yeah, I could see it potentially reaching 4GB of memory then, especially with cacheing and prefetching.
> is this a Windows XP flaw or is it just an Internet Explorer/Outlook flaw?
It is not an IE/OE flaw, although the flaw is significantly exacerbated by IE/OE's default behavior of automatically opening everything without prompting the user. However, the vulnerability is in the picture-viewer application that comes with Windows and is the default way to view several common image types, including.jpg and.wmf among others. Most of the current exploits are abusing the IE/OE automatic-display feature get the image displayed (and thus, the exploit code executed), but if you use the Win32 version of Lynx as your browser, save one of the malicious.wmf image in the My Documents folder, and try to open it in the picture viewer that you find under All Programs->Accessories (IIRC), you'll get infected.
> So is it IE or Windows that is home to the vulnerability?
We covered that the other day in the original article on the vulnerability. (This article is about how more sites are now exploiting it...) The vulnerability is in the picture viewer app/library that is included with Windows out of the box and is the default way to view certain kinds of images, including.wmf images.
Fill out a 1040. You know, with a pencil. Takes about ten minutes, plus another five or so for the Ohio IT-1040. I pick up the forms a the public library. Unless you've got something complicated going on, like self-employment or real-estate holdings, I just don't see the point in spending an extra $20 or more on special-purpose software that isn't useful for anything else and will have to be replaced with a new version next year. Maybe if the IRS distributed their own tax software, for free, I'd consider it, but then again frankly if it would probably have such an icky interface that I'd end up going with the paper forms anyhow.
Depends greatly on your target audience. If you're creating a site of AJAX resources for webmasters, it needs to handle resolutions from 1280x1024 up to 3200x2450 or so. Game tips and hits site? More like 800x600 to 1600x1200 or thereabouts.
If your site is of _general_ interest, however, you need to handle, at minimum, everything from 640x480 (lower if you care about cellphone users) up to at least 2048x1500 or so. At the lower end of that range, it's acceptable if the sidebar goes off the edge, but be very careful that the user doesn't have to scroll to read each line of any given column. At 800x600, try to make the entire layout fit within the width of the window. At the higher end, you'll be more concerned with how sparse the site looks and how the whitespace distributes.
Be aware also that higher resolutions are driving users to set minimum font sizes (e.g., nothing less than 16pt) and/or automatic text scaling (e.g., all font sizes 150% of what the page author specifies).
Images are the big problem. It would *really* help layouts if we could specify an image width in % or em units, but if you try that with a png (or any other bitmap format), it looks, in a word, bad. Webmasters at this point are desparately in need of widespread SVG support, but I suspect that it is still 5-10 years from being widespread enough that we can expect to work for enough users that we can use it as a replacement for almost all bitmapped graphics.
> When you say "usual moron who has never had kids" does that include morons that choose > not to have kids, but still has a firm grip on how to deal with a child?
It includes everyone. Parents who have set out to raise their children in a certain way and failed to have it come together in the way that they originally intended frequently think they know more about child-rearing than anyone else. (This is probably *why* they have failed to work things out the way they wanted.) They want to blame the laws of nature for the problems that they've had, rather than looking for what mistakes they may have made. They couldn't have made any mistakes, of course. They did everything perfect. The fact that their little angels are hellians just proves that parenting is a hopeless task, and that anyone who says otherwise has obviously never attempted to raise children. Any examples to the contrary are deceptive anomolies, or simply untrue.
It isn't worth arguing with these people. By the time they reach the point of calling anyone who hasn't raised children a moron, they've already finished with attempting to raise their own children, so it's pointless (and possibly cruel) to try to help them see their mistakes.
One thing I know about raising children: it's difficult difficult (and possibly impossible) to know that you've succeeded before the children are fully grown, but many people fail before the children can walk or talk, and those of us who have seen children raised well (e.g., younger siblings) can spot it immediately. When you see the three-year-old dictating to the parents what they will fix for him to eat (just for example), you know that these are parents who will grow up to make comments like the grandparent post.
You're thinking of Euler's Relation, one of the links in the article. And you have it slightly off; the way you've written it, it'd be -1 instead of 1. To have the positive one you have to move it over to the left side, leaving zero standing by itself on the right.
> What humans do when you ask them to make something random is an interesting subject.
Indeed.
> Doctor Plaster's strategy sounds like a good one. There's still some nonzero > random chance score
Yes, but it's 1/8 instead of the 1/2 you get with a regular true/false test, much deeper into flunking range. Nonetheless, *educated* guessing is useful. For instance, in my fake example, if you aren't certain whether:: is a syntactically valid name for a subroutine, you may still be able to determine that A is syntactically valid and B is not, leaving you with B and E as possible answers, for (at least) a 50% chance of answering correctly (more if you _think_ you might actually know about C but just aren't quite sure), which IMO is fair, although AFAIK nobody ever accused his tests of being *easy*.
Then there's the other extreme, my eighth-grade current events teacher, who in the late 80s liked to put true/false questions on his test like "Nobody has been killed in the West Bank or Gaza Strip since Christmas".
> The obvious question is why test prep folks have been telling people bogus information > for years? Either they are just bad at statistics (a scary thought, given the business > they're in), or there's some psychological benefit to convincing people that picking > the same letter will improve their score. Perhaps it's marginally faster during the > exam? Hard to believe that's significant.
There are several reasons, all having to do with human psychology. The first is that, as you noted, most tests are prepared by humans; thus, the answers are neither random nor uniformly distributed.
The second reason is nervousness on the part of the test-taker. Most people who attend test-prep sessions or buy test-prep materials are nervous about how they are going to perform on the test, and they tend to be the sort of people who fall apart under the pressure in an intense test situation. If somebody like me, who doesn't get nervous and performs well under pressure, wants to prepare for a test, he just studies the material that the test covers (or, if it's general like the SAT, goes to bed an hour early the night before). When somebody's nervous, concrete advice like "mark B" is easier to follow than something abstract like "pick an answer at random"; the latter is more likely to augment the nervousness.
Another reason is because the test-prep people sound more authoritative and reliable handing out concrete advice; saying "If you don't know, it doesn't matter what you mark" sounds like punting, even if it happens to be true. People would think, "Then why do we pay you?"
> Assuming test answer key is truly random and uniformly distributed
Umm, if it were truly random, it would almost certainly not be uniformly distributed. In particular, it's not possible to guarantee that a truly random selection will be uniformly distributed, and it usually won't be. Going the other way, uaranteeing uniform distribution absolutely precludes true randomness.
> So, either strategy works equally well.
Either strategy works equally well if the test is either uniformly distributed *or* random. However, in practice, most tests are *intended* to be either uniformly distributed or random, but because they are generally constructed by humans it seldom works out that way. Most teachers have a subconscious bias for or against certain letters, and it's also not unusual for a teacher to have a bias for or against certain sequences. Notably, the sequence "abc" is just as likely as any other sequence to occur at random, but many teachers will deliberately avoid it under the mistaken impression that they are somehow "improving" the randomness. Also, it is well documented that if there are five choices, a through e, b and d are statistically better bets than a, c, or e (though this of course can vary with specific teachers).
I still like Doctor Plaster's method for preventing statistical approaches from being of any value in taking his multiple choice tests. Every question was constructed something like this (although the following question per se would never appear because he doesn't teach computer stuff):
Which of the following lines of Perl code contains a syntax error under perl 5.8.7? A: print for grep { $_ lt $: } %foo; B: return map { sprintf "%05d" $_ } @bar; C: push @baz, sort {::($b) <=>::($a) } @quux; D: A and B E: B and C F: A and C G: All of the above H: None of the above
Basically, it's three closely related true-and-false questions rolled together in an all-or-nothing fashion, so that random guessing is not useful. (The answer, incidentally, in this case is B, since both A and C are syntactically valid, albeit not the most elegant lines ever written. The error in B is the omission of the comma after the first argument to sprintf.)
It's possible to miss questions and still score 800. The test is deliberately built with more discriminatory capacity than is required to create the scores, to allow for adjustments based on comparison with past editions of the tests. Also, the basic SAT includes an entire _section_ that does not count toward your score, that is used for the express purpose of statistically comparing performances across editions; I am not sure whether the more specific tests such as Math II have this entire-extra-section feature, though.
> This coming term, I will have the choice of studying > either Java or C# for my Object Oriented Programming > class. Now I'm a diehard Linux user, so I'm slightly > conflicted here. Which should I take?"
Assuming neither of these is a language that you really want to end up using after you're done with the course, take the C#. First, unless I've missed my guess, it means you'll be using MS Visual Studio.NET for the course, which is not something you'll likely be tempted to use after you finish the course. Yeah, there's mono, but it's not so similar to VS.NET as the somewhat-cross-platform JDK would be. So you won't be focused on the tools. Second, using C# for the course probably means a lot of the assignments will delve into the Win32 API, which has two benefits. On the one hand, as a Linux user, it gives you some insight into the other side of things, which is always advantageous; on the other hand, it means, again, that you'll be less tempted to continue using the stuff after the course. So you won't be focused on the nitty-gritty details of the language and the API. Not focused on the tools, not focused on the language or the API, means you'll be focused on the *concepts*, which is the whole point of the course anyhow.
This is all assuming you don't want to build a long-term relationship with either language. If you *do* want to do so, then by all means, go with Java.
Either way, you should, on the side, also expose yourself to another object-oriented language, one with a really good object model. My personal recommendation would be Inform (get the Inform Designer's Manual; it's excellent), but Smalltalk also has its proponents, and there are other options.
> I'm often asked for a simple WYSIWYG html editor
Just say no.
HTML was not designed to be created with a WYSIWYG editor, and there are fundamental reasons why it should not be.
Among other things, the semantics get lost with a WYSIWYG editor, as the WYSIWYG editor has no way of knowing, from the fact that you made the font such-and-such, whether the content of that section is a header, just another part of the document that happens to be emphasized, or what. You made this phrase italic, does that mean it's a title of a work that should be placed in a cite element, or are you using the italics just for emphasis, or perhaps because you like the way Georgia looks in italic? The software doesn't know. Did you indent this because it's a definition, or because it's a block quotation, or just because you wanted it layed out that way? The editor can't tell.
Furthermore, WYSIWYG implies things that the web is not designed to guarantee, things that in practice will break the instant somebody looks at the page on a different computer. Will the typeface really look like that? (Does the user even have a typeface *called* Arial? If so, is it from the same foundry? Is the hinting the same?) Will the line really wrap between these two words? (Haha.) Is what you see *really* what you get?
Now, an HTML editor with a *preview* feature would be okay. But true WYSIWYG, like in word processing, is fundamentally incompatible with the web.
> Lotus SmartSuite and a user interface that is quite enjoyable to me
Reportedly, it also gets rave reviews from a high percentage of its other users, possibly even from both of them.
I'm not saying that it shouldn't be looked at; plenty of obscure things have ideas in them that are worth looking at. BeOS, for instance, has some interesting features that other OSes *still* would do well to look at. ITS had (optional) file versioning built in at the filesystem level, a feature I would really like to have in Linux. TOPS-20 had a very innovative integrated help system that nothing since has really equalled. And so on.
However, while I'm sure Lotus has some nifty ideas in it, I think you also need to understand that if the developers of some other software don't implement these ideas, it's not because they have some kind of personal vendetta or blind spot regarding Lotus. It's just because Lotus is, in a word, obscure (except for 123 for DOS, of course, which is not so much obscure as just dated). And if you think that because they don't implement the ideas that you plead and beg for each release if means they have a personal vendetta or blind spot against you, then you obviously have no idea how many thousands of users plead and beg for features each release. You and the other three people pleading for features from Lotus are getting drowned out by tens of thousands of users pleading and begging for features found in relatively more common software, such as Emacs, to say nothing of the hundreds of thousands of braindead and inspecific comments to the effect of "Please, for the love of all that is sane, make it more like Word!" Then there's the "You simply must stop adding any more features until you reduce the memory footprint so it will run on my Pentium 60 with 3MB of RAM" crowd. Et cetera, ad infinitum, ad nauseam, a veritable deluge of pleading and begging. A vaccuum? Please. The developers are closer to operating in a wind tunnel than a vaccuum.
You can stop checking each release to see if it magically transformed into Lotus. It didn't. It won't, unless you and the other Lotus ObscureSuite user write the code. I'm not saying this is the ideal scenario, but it is a reality you must accept, unless you are willing to put in the work required to change it (which means writing the code).
> I have students that turn things in in the most obscure, dated > formats imaginable [...] Yep the Works files are the problem.
Wait, MS Works formats are the most obscure, dated formats you can imagine? Wow, either you just got into computers in the last decade, or you have no imagination whatsoever. Haven't you ever had a student hand you a 720k Mac-formatted floppy with a Claris Works document on it? I would have thought that would actually *happen*. Things that I would not expect to happen today, but that might have just a few years ago, include a 5.25" DSDD floppy with a PC Write document on it, or maybe a WordWriter file, or perhaps an Apple-formatted 5.25" SSSD floppy with a/// EZ Pieces document, or possibly something from the Tandy or Commodore world. If you'd asked me to imagine an extreme case, I'd have said something about an early-model Brother dedicated electronic word processor.
Microsoft Works? That's practically mainstream. I think I even still have an old copy of version 4 sitting around someplace (not that it would likely run on a modern OS).
Your story says as much about the school's math program as it does about the EE program. There is a large variance in the quality of math programs. A poor math program teaches the students how to do some math. A *good* math program teaches the students how to think, and by two years into the program they can do well in practically any subject -- not just EE, but even stuff that may seem, to the casual observer, to be completely unrelated, such as the language arts.
It also may say something about the specific math majors in question. Note that these particular math majors were deliberately taking a non-required course, which did not have a reputation for being an easy course, outside their scope of their major. When you see a student doing that sort of thing, it's usually a good sign. It's like a certain former math undergrad I knew who deliberately signed up to take a drawing class intended for art majors as well as a course in ancient Greek grammar, as electives, the same semester of his junior year (having already completed the majority of his major requirements in his first two years). He aced both, of course.
> But we must really ask ourselves what kind of educational system we have that makes > kids WANT to cheat
You don't have to *make* people want to cheat; people just naturally want to cheat. I've even caught myself thinking up (sometimes elaborate) ways to cheat, even though I'm one of those guys the other kids always despised for breaking curves. (I'm not much smarter than average (honest, I'm not), but I perform well under pressure (don't get nervous, and when I have a mental block on a question I just mark it in the margin and come back later), and I tend to study rather more than average.) And it wasn't because I cared about the grades either; I was never motivated by grades. I studied because I wanted to understand the material.
I thought up (albeit never used) ways to cheat for no good reason, just, uhm, *because*. It's part of human nature.
> I never memorised physics or calculus formulae - I derived the formula needed for each > question from first principles when I reached a question needing that particular formula.
I memorized a small handful of ones that would save me a lot of steps of deriving, but in general I did as you did -- derived the formula on the fly. I first did this in High School Geometry. My proofs would have twice or three times as many steps as the canonical answer, but they were valid and sound and reached the expected conclusion, and I always finished the test early because I didn't have to sit and wrack my brain. Later I did the same thing in high school algebra, trig, physics, and calc, and then in various college math courses.
For other courses, like art history or whatnot, I made up cheat sheets for myself, then studied from the cheat sheets; a couple of hours before the exam I quizzed myself to see which items on the cheat sheets I still didn't know, put those on a shorter cheat sheet, and crammed it into short-term memory at the last minute.
So I never did anything that would technically be considered cheating in any sort of indictable way, but I didn't actually really *learn* every single minor little detail I was supposed to, although I did learn the major concepts well.
Slashdot Mirror
> a smaller chance of long-term survival?
Apparently you do not know your history. I'm not an Opera user myself (except for occasional testing purposes), but its long-term servival is pretty much settled. Of the dozen or so browsers with anything approaching significant market share, Opera is the second-oldest, after Netscape. In that time, Opera has never lost any significant percentage of what market share it has. Granted, it has also never had a very large market share, compared to the top two players. OTOH, which other two browsers are the top two players, and in what order, has changed three or four times while Opera has stayed consistently in the top four. Opera's long-term survival, assuming we define "long-term" in a way that makes sense in the context of the web browser market, isn't really in question.
Whether it will ever climb into the top two is another matter.
> Geeze, here it is 2006 and people still think that the return address in unsigned
> email means ANYTHING.
Well, yeah. I had to explain to two coworkers just last week that the scary messages they were getting weren't really from eBay, and they were quite surprised. (So I told them that if they were concerned that they might need to check their eBay accounts, to use the bookmarks they usually use to go there, because they would know that those really go to eBay. The link in this message only says it goes to eBay, and really it goes someplace else, to another site. Such gasps of outrage and astonishment as I then heard, you'd have thought I was telling them that their husbands lied about the business trip and were really with in Las Vegas with girlfriends.)
This is at least partly because of the way mailreaders present the data. Instead of showing the headers as part of the message (which is, essentially, how they're transmitted), most mail readers parse the headers and present certain pieces of data from them (the From address, for instance) separately from the message, as metadata. Well, yeah, it *is* metadata in a sense, but the way it's presented makes it appear, to the casual user, as if it's something the mailreader knows about the message, rather than something the message claims about itself. Other critical headers, such as Receives:, are not shown at all (unless the user specifically goes looking for them in a "Show All Headers" or "View Message Source" option or somesuch.
There are, of course, good solid usability reasons why these things are the way they are, but it doesn't take a doctor of psychology to tell you what people are going to think as a result.
Personally I'd like to see the information parsed out of the headers, especially the sender information, labelled just a little differently, e.g., "Claims to be From:". I'm not sure that would entirely solve the problem, but it might help a little. I'm also deeply annoyed that our ISP's mail server accepts HTML messages for delivery (if we had our own mail server in house it sure wouldn't), and that all the decent, deployable, user-friendly mail clients I can find happily render and display HTML mail. Even recent versions of Pegasus cannot, as near as I can determine, be configured to show the source or treat the HTML as an attachment.
> > As for Notepad, have you ever tried to edit a 6GB text file?
> vi can open files of any size because it caches to disk and only has part in memory
Notepad, as I'm sure you're aware, is not vi.
> And you can get vi for windows
So, to clarify, are you saying that because it is possible to obtain *other* software for Windows, that therefore Windows should not ship with a text editor that can edit large files? Perhaps there is no reason for Notepad to be included at all?
> I do this sort of thing all of the time with (32-bit) Safari on a computer with 1 GB
> of memory. It uses virtual memory to page out parts of memory to disk. I'd guess IE
> does the same thing.
Now I think you may be (if I understand what you're saying) confusing physical memory with address space. I don't *think* you're implying that each application should handle virtual memory management itself (rather than having the OS do this), although it sort of sounds that way in the above-quoted paragraph. Either way, though, whether the app does the virtual memory management or whether, as is more usual, the OS handles that, the address size still limits how much memory, real or virtual, can be addressed. If the hardware and kernel are 64-bit, therefore, it makes sense for applications such as web browsers also to be 64-bit.
> As for semantics, every word processor I know supports it and encourages its use.
> The 'right' way to use Word is to use Styles for formatting.
The word processors support that in *theory*, but in practice the interface doesn't work very well for that, and so vanishingly close to everyone ends up using word processors the ''wrong'' way, i.e., in a standard WYSIWYG semantics-are-irrelevant fashion.
> The font issue is not unique to web pages: If you e-mail a word processing document
> to someone who lacks the right fonts, it will look different on their computer.
Word processing is designed for printing on paper. What happens when you send it by email is about as relevant as what happens when you try to display it on a dumb terminal. (Yes, I know there are people who send word processing documents by email. There are also people who send cellphone text messages while driving. People *do* all sorts of inane things.)
> Actually hand coding HTML is just not realistic for most users.
Perhaps you would like to explain why it is unrealistic?
> I'd say it's probably coming from underground.
Interesting possibility. When I first read the story I of course immediately thought, like most everyone else here I suspect, "Wow, there must be some kind of interdimensional portal, and it's all coming through that from an alien planet!" However, now that you mention it, I suppose after all it *could* just be coming from underground.
> This is one thing that has always bugged me about (most) religions. On the one hand
> it is said that God (s/God/Deity of your choice) is infallible, and that the Bible
> (s/Bible/Religous text of your choice) is the word of God. But on the other hand,
> people say that their religion is flexible and adaptive.
Two different groups of people. "Flexible and adaptive" is the sort of thing the liberals and eccumenists say; conservatives (usually called "fundamentalists" if you're liberal) are the ones who believe in things like infallibility and inerrancy. There is very little overlap between the two groups.
Where it gets confusing is when you start trying to categorize denominations. In any given denomination, you usually have both sorts of people (although one or the other may dominate), so you hear both kinds of comments coming out of them.
> BTX form factor? didn't it just come out?
ISTR that this is not the first year it "just" came out and also not the first year that was going to be the year BTX completely took over. I suspect 2005 will not be the last year that BTX fails to gain any significant traction in the market. Fundamentally, the considerations are like this. On the one hand, there's a major tech company or two that likes BTX and wants it to be the industry standard. On the other hand, the benefits are minor, and nobody else cares, least of all consumers, and most of the big OEMs would just as soon stick with the existing industry standard. I don't know if this reminds *you* of anything, but it makes *me* think of the MicroChannel expansion bus which, year after year, relentlessly, continued to *not* take over the expansion bus market.
> Notepad? IE? How could those ever need more than 4 GB of memory?
... well, okay, so let's imagine that IE got improved to the point where you could actually use it as your regular web browser. In that case, you could have half a dozen IE windows each with 20-30 tabs open, and you could leave some of those windows open for weeks on end and, yeah, I could see it potentially reaching 4GB of memory then, especially with cacheing and prefetching.
As for Notepad, have you ever tried to edit a 6GB text file? (Granted, Notepad is *not* what I would use for any serious editing job in any case...) As for IE,
> is this a Windows XP flaw or is it just an Internet Explorer/Outlook flaw?
.jpg and .wmf among others. Most of the current exploits are abusing the IE/OE automatic-display feature get the image displayed (and thus, the exploit code executed), but if you use the Win32 version of Lynx as your browser, save one of the malicious .wmf image in the My Documents folder, and try to open it in the picture viewer that you find under All Programs->Accessories (IIRC), you'll get infected.
It is not an IE/OE flaw, although the flaw is significantly exacerbated by IE/OE's default behavior of automatically opening everything without prompting the user. However, the vulnerability is in the picture-viewer application that comes with Windows and is the default way to view several common image types, including
> So is it IE or Windows that is home to the vulnerability?
.wmf images.
We covered that the other day in the original article on the vulnerability. (This article is about how more sites are now exploiting it...) The vulnerability is in the picture viewer app/library that is included with Windows out of the box and is the default way to view certain kinds of images, including
Fill out a 1040. You know, with a pencil. Takes about ten minutes, plus another five or so for the Ohio IT-1040. I pick up the forms a the public library. Unless you've got something complicated going on, like self-employment or real-estate holdings, I just don't see the point in spending an extra $20 or more on special-purpose software that isn't useful for anything else and will have to be replaced with a new version next year. Maybe if the IRS distributed their own tax software, for free, I'd consider it, but then again frankly if it would probably have such an icky interface that I'd end up going with the paper forms anyhow.
Depends greatly on your target audience. If you're creating a site of AJAX resources for webmasters, it needs to handle resolutions from 1280x1024 up to 3200x2450 or so. Game tips and hits site? More like 800x600 to 1600x1200 or thereabouts.
If your site is of _general_ interest, however, you need to handle, at minimum, everything from 640x480 (lower if you care about cellphone users) up to at least 2048x1500 or so. At the lower end of that range, it's acceptable if the sidebar goes off the edge, but be very careful that the user doesn't have to scroll to read each line of any given column. At 800x600, try to make the entire layout fit within the width of the window. At the higher end, you'll be more concerned with how sparse the site looks and how the whitespace distributes.
Be aware also that higher resolutions are driving users to set minimum font sizes (e.g., nothing less than 16pt) and/or automatic text scaling (e.g., all font sizes 150% of what the page author specifies).
Images are the big problem. It would *really* help layouts if we could specify an image width in % or em units, but if you try that with a png (or any other bitmap format), it looks, in a word, bad. Webmasters at this point are desparately in need of widespread SVG support, but I suspect that it is still 5-10 years from being widespread enough that we can expect to work for enough users that we can use it as a replacement for almost all bitmapped graphics.
> When you say "usual moron who has never had kids" does that include morons that choose
> not to have kids, but still has a firm grip on how to deal with a child?
It includes everyone. Parents who have set out to raise their children in a certain way and failed to have it come together in the way that they originally intended frequently think they know more about child-rearing than anyone else. (This is probably *why* they have failed to work things out the way they wanted.) They want to blame the laws of nature for the problems that they've had, rather than looking for what mistakes they may have made. They couldn't have made any mistakes, of course. They did everything perfect. The fact that their little angels are hellians just proves that parenting is a hopeless task, and that anyone who says otherwise has obviously never attempted to raise children. Any examples to the contrary are deceptive anomolies, or simply untrue.
It isn't worth arguing with these people. By the time they reach the point of calling anyone who hasn't raised children a moron, they've already finished with attempting to raise their own children, so it's pointless (and possibly cruel) to try to help them see their mistakes.
One thing I know about raising children: it's difficult difficult (and possibly impossible) to know that you've succeeded before the children are fully grown, but many people fail before the children can walk or talk, and those of us who have seen children raised well (e.g., younger siblings) can spot it immediately. When you see the three-year-old dictating to the parents what they will fix for him to eat (just for example), you know that these are parents who will grow up to make comments like the grandparent post.
You're thinking of Euler's Relation, one of the links in the article. And you have it slightly off; the way you've written it, it'd be -1 instead of 1. To have the positive one you have to move it over to the left side, leaving zero standing by itself on the right.
> What humans do when you ask them to make something random is an interesting subject.
:: is a syntactically valid name for a subroutine, you may still be able to determine that A is syntactically valid and B is not, leaving you with B and E as possible answers, for (at least) a 50% chance of answering correctly (more if you _think_ you might actually know about C but just aren't quite sure), which IMO is fair, although AFAIK nobody ever accused his tests of being *easy*.
Indeed.
> Doctor Plaster's strategy sounds like a good one. There's still some nonzero
> random chance score
Yes, but it's 1/8 instead of the 1/2 you get with a regular true/false test, much deeper into flunking range. Nonetheless, *educated* guessing is useful. For instance, in my fake example, if you aren't certain whether
Then there's the other extreme, my eighth-grade current events teacher, who in the late 80s liked to put true/false questions on his test like "Nobody has been killed in the West Bank or Gaza Strip since Christmas".
> The obvious question is why test prep folks have been telling people bogus information
> for years? Either they are just bad at statistics (a scary thought, given the business
> they're in), or there's some psychological benefit to convincing people that picking
> the same letter will improve their score. Perhaps it's marginally faster during the
> exam? Hard to believe that's significant.
There are several reasons, all having to do with human psychology. The first is that, as you noted, most tests are prepared by humans; thus, the answers are neither random nor uniformly distributed.
The second reason is nervousness on the part of the test-taker. Most people who attend test-prep sessions or buy test-prep materials are nervous about how they are going to perform on the test, and they tend to be the sort of people who fall apart under the pressure in an intense test situation. If somebody like me, who doesn't get nervous and performs well under pressure, wants to prepare for a test, he just studies the material that the test covers (or, if it's general like the SAT, goes to bed an hour early the night before). When somebody's nervous, concrete advice like "mark B" is easier to follow than something abstract like "pick an answer at random"; the latter is more likely to augment the nervousness.
Another reason is because the test-prep people sound more authoritative and reliable handing out concrete advice; saying "If you don't know, it doesn't matter what you mark" sounds like punting, even if it happens to be true. People would think, "Then why do we pay you?"
> Assuming test answer key is truly random and uniformly distributed
::($b) <=> ::($a) } @quux;
Umm, if it were truly random, it would almost certainly not be uniformly distributed. In particular, it's not possible to guarantee that a truly random selection will be uniformly distributed, and it usually won't be. Going the other way, uaranteeing uniform distribution absolutely precludes true randomness.
> So, either strategy works equally well.
Either strategy works equally well if the test is either uniformly distributed *or* random. However, in practice, most tests are *intended* to be either uniformly distributed or random, but because they are generally constructed by humans it seldom works out that way. Most teachers have a subconscious bias for or against certain letters, and it's also not unusual for a teacher to have a bias for or against certain sequences. Notably, the sequence "abc" is just as likely as any other sequence to occur at random, but many teachers will deliberately avoid it under the mistaken impression that they are somehow "improving" the randomness. Also, it is well documented that if there are five choices, a through e, b and d are statistically better bets than a, c, or e (though this of course can vary with specific teachers).
I still like Doctor Plaster's method for preventing statistical approaches from being of any value in taking his multiple choice tests. Every question was constructed something like this (although the following question per se would never appear because he doesn't teach computer stuff):
Which of the following lines of Perl code contains a syntax error under perl 5.8.7?
A: print for grep { $_ lt $: } %foo;
B: return map { sprintf "%05d" $_ } @bar;
C: push @baz, sort {
D: A and B
E: B and C
F: A and C
G: All of the above
H: None of the above
Basically, it's three closely related true-and-false questions rolled together in an all-or-nothing fashion, so that random guessing is not useful. (The answer, incidentally, in this case is B, since both A and C are syntactically valid, albeit not the most elegant lines ever written. The error in B is the omission of the comma after the first argument to sprintf.)
It's possible to miss questions and still score 800. The test is deliberately built with more discriminatory capacity than is required to create the scores, to allow for adjustments based on comparison with past editions of the tests. Also, the basic SAT includes an entire _section_ that does not count toward your score, that is used for the express purpose of statistically comparing performances across editions; I am not sure whether the more specific tests such as Math II have this entire-extra-section feature, though.
> This coming term, I will have the choice of studying
.NET for the course, which is not something you'll likely be tempted to use after you finish the course. Yeah, there's mono, but it's not so similar to VS.NET as the somewhat-cross-platform JDK would be. So you won't be focused on the tools. Second, using C# for the course probably means a lot of the assignments will delve into the Win32 API, which has two benefits. On the one hand, as a Linux user, it gives you some insight into the other side of things, which is always advantageous; on the other hand, it means, again, that you'll be less tempted to continue using the stuff after the course. So you won't be focused on the nitty-gritty details of the language and the API. Not focused on the tools, not focused on the language or the API, means you'll be focused on the *concepts*, which is the whole point of the course anyhow.
> either Java or C# for my Object Oriented Programming
> class. Now I'm a diehard Linux user, so I'm slightly
> conflicted here. Which should I take?"
Assuming neither of these is a language that you really want to end up using after you're done with the course, take the C#. First, unless I've missed my guess, it means you'll be using MS Visual Studio
This is all assuming you don't want to build a long-term relationship with either language. If you *do* want to do so, then by all means, go with Java.
Either way, you should, on the side, also expose yourself to another object-oriented language, one with a really good object model. My personal recommendation would be Inform (get the Inform Designer's Manual; it's excellent), but Smalltalk also has its proponents, and there are other options.
> I'm often asked for a simple WYSIWYG html editor
Just say no.
HTML was not designed to be created with a WYSIWYG editor, and there are fundamental reasons why it should not be.
Among other things, the semantics get lost with a WYSIWYG editor, as the WYSIWYG editor has no way of knowing, from the fact that you made the font such-and-such, whether the content of that section is a header, just another part of the document that happens to be emphasized, or what. You made this phrase italic, does that mean it's a title of a work that should be placed in a cite element, or are you using the italics just for emphasis, or perhaps because you like the way Georgia looks in italic? The software doesn't know. Did you indent this because it's a definition, or because it's a block quotation, or just because you wanted it layed out that way? The editor can't tell.
Furthermore, WYSIWYG implies things that the web is not designed to guarantee, things that in practice will break the instant somebody looks at the page on a different computer. Will the typeface really look like that? (Does the user even have a typeface *called* Arial? If so, is it from the same foundry? Is the hinting the same?) Will the line really wrap between these two words? (Haha.) Is what you see *really* what you get?
Now, an HTML editor with a *preview* feature would be okay. But true WYSIWYG, like in word processing, is fundamentally incompatible with the web.
> Lotus SmartSuite and a user interface that is quite enjoyable to me
Reportedly, it also gets rave reviews from a high percentage of its other users, possibly even from both of them.
I'm not saying that it shouldn't be looked at; plenty of obscure things have ideas in them that are worth looking at. BeOS, for instance, has some interesting features that other OSes *still* would do well to look at. ITS had (optional) file versioning built in at the filesystem level, a feature I would really like to have in Linux. TOPS-20 had a very innovative integrated help system that nothing since has really equalled. And so on.
However, while I'm sure Lotus has some nifty ideas in it, I think you also need to understand that if the developers of some other software don't implement these ideas, it's not because they have some kind of personal vendetta or blind spot regarding Lotus. It's just because Lotus is, in a word, obscure (except for 123 for DOS, of course, which is not so much obscure as just dated). And if you think that because they don't implement the ideas that you plead and beg for each release if means they have a personal vendetta or blind spot against you, then you obviously have no idea how many thousands of users plead and beg for features each release. You and the other three people pleading for features from Lotus are getting drowned out by tens of thousands of users pleading and begging for features found in relatively more common software, such as Emacs, to say nothing of the hundreds of thousands of braindead and inspecific comments to the effect of "Please, for the love of all that is sane, make it more like Word!" Then there's the "You simply must stop adding any more features until you reduce the memory footprint so it will run on my Pentium 60 with 3MB of RAM" crowd. Et cetera, ad infinitum, ad nauseam, a veritable deluge of pleading and begging. A vaccuum? Please. The developers are closer to operating in a wind tunnel than a vaccuum.
You can stop checking each release to see if it magically transformed into Lotus. It didn't. It won't, unless you and the other Lotus ObscureSuite user write the code. I'm not saying this is the ideal scenario, but it is a reality you must accept, unless you are willing to put in the work required to change it (which means writing the code).
> I have students that turn things in in the most obscure, dated
/// EZ Pieces document, or possibly something from the Tandy or Commodore world. If you'd asked me to imagine an extreme case, I'd have said something about an early-model Brother dedicated electronic word processor.
> formats imaginable [...] Yep the Works files are the problem.
Wait, MS Works formats are the most obscure, dated formats you can imagine? Wow, either you just got into computers in the last decade, or you have no imagination whatsoever. Haven't you ever had a student hand you a 720k Mac-formatted floppy with a Claris Works document on it? I would have thought that would actually *happen*. Things that I would not expect to happen today, but that might have just a few years ago, include a 5.25" DSDD floppy with a PC Write document on it, or maybe a WordWriter file, or perhaps an Apple-formatted 5.25" SSSD floppy with a
Microsoft Works? That's practically mainstream. I think I even still have an old copy of version 4 sitting around someplace (not that it would likely run on a modern OS).
Your story says as much about the school's math program as it does about the EE program. There is a large variance in the quality of math programs. A poor math program teaches the students how to do some math. A *good* math program teaches the students how to think, and by two years into the program they can do well in practically any subject -- not just EE, but even stuff that may seem, to the casual observer, to be completely unrelated, such as the language arts.
It also may say something about the specific math majors in question. Note that these particular math majors were deliberately taking a non-required course, which did not have a reputation for being an easy course, outside their scope of their major. When you see a student doing that sort of thing, it's usually a good sign. It's like a certain former math undergrad I knew who deliberately signed up to take a drawing class intended for art majors as well as a course in ancient Greek grammar, as electives, the same semester of his junior year (having already completed the majority of his major requirements in his first two years). He aced both, of course.
> But we must really ask ourselves what kind of educational system we have that makes
> kids WANT to cheat
You don't have to *make* people want to cheat; people just naturally want to cheat. I've even caught myself thinking up (sometimes elaborate) ways to cheat, even though I'm one of those guys the other kids always despised for breaking curves. (I'm not much smarter than average (honest, I'm not), but I perform well under pressure (don't get nervous, and when I have a mental block on a question I just mark it in the margin and come back later), and I tend to study rather more than average.) And it wasn't because I cared about the grades either; I was never motivated by grades. I studied because I wanted to understand the material.
I thought up (albeit never used) ways to cheat for no good reason, just, uhm, *because*. It's part of human nature.
> I never memorised physics or calculus formulae - I derived the formula needed for each
> question from first principles when I reached a question needing that particular formula.
I memorized a small handful of ones that would save me a lot of steps of deriving, but in general I did as you did -- derived the formula on the fly. I first did this in High School Geometry. My proofs would have twice or three times as many steps as the canonical answer, but they were valid and sound and reached the expected conclusion, and I always finished the test early because I didn't have to sit and wrack my brain. Later I did the same thing in high school algebra, trig, physics, and calc, and then in various college math courses.
For other courses, like art history or whatnot, I made up cheat sheets for myself, then studied from the cheat sheets; a couple of hours before the exam I quizzed myself to see which items on the cheat sheets I still didn't know, put those on a shorter cheat sheet, and crammed it into short-term memory at the last minute.
So I never did anything that would technically be considered cheating in any sort of indictable way, but I didn't actually really *learn* every single minor little detail I was supposed to, although I did learn the major concepts well.