You don't need relativity. Newtonian physics (and specifically his work on describing gravity) is enough to discredit the notion that the Earth goes around the Sun. The motion of the Sun, like the motion of the Earth, is acted upon by the gravitational forces associated with every mass in the universe.
And I suppose you thing the Earth goes around the Sun?
(Both ideas are wrong in exactly the same way. It would be closer to say that the Earth and the Sun move together, but this is still an oversimplification. Every mass in the universe exerts gravitational pull on the Earth and on the Sun as well. As for the pull they exert on one another, it is equal: Earth pulls on the Sun with exactly the same force that the Sun pulls on Earth.)
I think Asia sort of was a country at one time (right next to Bythinia), but then it sort of got taken over by various other regional powers (the Persians, the Greeks, the Romans, the Caliphate,...) and ended up in the hands of the Turks.
> If you've never learned functional programming, JavaScript is a good language to learn in.
Most of the functional programming techniques I know, I learned in Perl.
It's interesting. I am sufficiently conversant in Emacs lisp to do significant amounts of useful stuff (including custom major modes for special editing situations), but I never understood lambda expressions in lisp. In Perl, though, anonymous subroutines are so straightforward, I immediately understood them and was able to start using them myself after reading barely a page of basic explanation. When I studies scheme I spent *hours* reading over and over and over again the chapter on closures and just couldn't figure the stupid things out. Then somebody on Perlmonks posted some Perl code that used lexical closures, and all became clear; I now use them myself without hesitation.
I've heard Perl6 is going to support continuations. I'm looking forward to it. Maybe if I see them in Perl code I'll understand those, too. I *sure* have not had an easy time trying to understand the explanations of them in the documentation for pure functional languages.
How about translation? I've seen the results of currently available translation software, and there is LOTS of room for improvement there. Can we design the new CAPTCHA tests so that breaking them will improve the state of the art of translation software?
Arial looks better than at least sixty percent of the Helvetica typefaces I've seen, and in any event Helvetica is not an available option for most Windows users, so being redundant with Helvetica does not make Arial non-useful to Windows users.
However, once Verdana was introduced, Arial ceased to be necessary. Verdana is better than Arial in pretty much ever respect (err, except Verdana runs large, but you can correct that easily by reducing the point size), so at this point I would call Arial "obsolete" and "redundant" as well as "mediocre".
But I wouldn't call it "bad".
Times New Roman, now that's bad. A crime against aesthetics.
> The reason to hate it is that it's the Universal "Specialty" font. If you don't want a > serif font, or a plain font like Arial, the first tool of choice is Comic Sans.
Ah. That's because it's widely available.
If the Core Fonts collection had left out the uglier-than-sin fonts (TNR and Courier New) and the redundant ones (Arial and Trebuchet aren't needed if you've got Verdana; Arial Black isn't needed if you've got Impact; TNR isn't needed ever, but especially not if you've got Georgia; Courier New is a crime against humanity in any case but an especially unnecessary one if you've got Andale Mono) and instead included a three or four more legitimately useful different-look fonts (like, say, a nice caligraphic decorative font, a decent script or brushscript font, a bubble-letters or stencil font, and maybe a grunge font), the total number of fonts would have been smaller, but the total usefulness would have been much greater.
But, you know, that's not what they did.
And when they redid the fonts for Vista (Calibri, Cambria, Candara, etc), they made the same mistake. Too much redundancy, not enough variety. Most people (who don't have a typesetting or graphic-arts background) aren't going to see the difference between Calibri and Candara even if you put the same text set in each of them side by side.
> how are a smudged O and C easier to tell apart than a smudged o and c
The part where they're different is a bit larger, so less likely to be completely obscured.
But I think the part about no descenders is more relevant. Early terminals had *small* numbers of pixels per character. The uppercase letters (well, the standard Latin ones with no diacritical marks that are used in English and included in ASCII) can, in a pinch, all be represented in in a 5x5 grid. It's ugly, but it works.
> written Hebrew doesn't actually supply any indication of what vowels should be inserted between the consonants.
It's more complicated than that. Masoretic writing does supply the vowels for most words (as diacritical marks, but it supplies them). However, it was taboo to *pronounce* God's name at the time, so the qithev/qere system (used to indicate prefered readings) indicated that the reader should pronounce the general word for a lord (usually romanized as Adonai) instead. Because of the way the qithev/qere system works, this means the consonants of the original text are preserved (because they lacked the nerve to actually change the letters of the text), but the vowels (being only diacritical marks, i.e., pronunciation guides, and therefore not sacred, or something) were changed to the ones for the replacement word. (The consonants for the replacement word are written in the margin. Why the replacement vowels couldn't be put in the margin too, and the original ones left in the main text, I don't know.)
> This means that the Abrahamic deity might actually have been called "Yahoo Wahoo"
That is not a particularly likely combination.
But yeah, we don't know the correct vowels for sure.
> > It is easily readable > Yes. Compared to, say, Wingdings.
Actually, it's one of only three or four fonts my sister (who teaches lower elementary school) is willing to use for classroom materials, because it's one of the only ones the kids can read, because it uses the simple letter forms they teach the kids in kindergarten. The biggest points of contention for most fonts are the lowercase letters a and g. A few other sans fonts use the simple-form g, but almost none of them use the simple-form lowercase a.
Now, one could argue that the schools *should* be teaching the normal lowercase forms that are ordinarily used in almost all print materials throughout the entire English-speaking world. But they *aren't*. (It may be partly because the more common forms are more complex and therefore require more coordination to write. A lot of kindergarten students struggle to get the stick on the right side of the circle for lowercase a, so asking them to write the Times form of the letter admittedly seems a bit much.)
Anyway, I would argue that Comic Sans is better than *several* of the other fonts from Microsoft's "Core Fonts for the Web" initiative.
Georgia and Verdana, of course, are clearly the best of the batch. They actually look good, and furthermore they look good together, which is a fairly big deal. You've got to have a basic serif and a basic sans font that look okay together, and this is a reasonably good pair. I've seen better pairs, but not *many* of them, and especially not ones that were available in 1996. Also, Georgia has a real actual honest-to-goodness italic face, which even manages to LOOK GOOD, which is a fairly rare quality. (I'm not a big fan of most italic faces, as a rule. If anybody knows of a freely-available sans-serif font that actually looks good in italic, I'd sure like to hear about it, because as yet I've not seen one.) Verdana runs a little on the large side, but you can fix that by decrementing the point size, so it's not exactly a deal-killer.
Impact and Andale Mono are acceptable for their intended purposes (wet paint signs and source code, respectively). Lucida Console is in some ways better than Andale Mono, but it's not freely redistributable. Bitstream Vera Sans Mono is alright, but it didn't come out until later.
But after that, really, the fifth-best one in the pack is Comic Sans. Bear with me...
Arial and Arial Black and Trebuchet aren't actively ugly, but they're mediocre, and more to the point they're also pretty redundant with other, better fonts from the same initiative (Verdana, Impact, and Verdana, respectively). Admittedly, Arial dates to Windows 3.x and thus is older than Verdana, but once Verdana was produce we no longer needed Arial for anything (because Verdana looks better), so why was it still included, why is it *still* included with Windows? Why? As for Trebuchet, I never understood why it was needed at any time. It was never *bad*, but it also never had anything to offer over other fonts that were already available.
Then we come to Times New Roman, which is uglier than a half-shaved mandrill, and Courier New, which is the most heinously hideous excuse for a font ever created in True Type format. (I've seen bitmapped screen fonts that were worse, but not many.)
I assume we're not going to try to compare Webdings with anything because, you know, it was never intended for the same basic purpose as the other core fonts (namely, typesetting actual words).
So, Comic Sans isn't the best font ever, but it's orders of magnitude nicer looking than TNR, to say nothing of Courier New, and furthermore it offers a significant stylistic difference from other available fonts, unlike Arial and Arial Black and Trebuchet. It's not as good as Verdana or Georgia, and its niche is (arguably) not as important as the ones for Impact and Andale Mono, but it still serves a useful purpose.
You don't like Comics Sans? Hey, fine, uninstall it from your computer, and websites that try to use it (and don't provide a fallback alternative) will render in whatever font you set as your browser's default. Voila.
But personally I don't see what's so bad about it.
Times New Roman is the one I wish was never created.
There are three basic problems with trains, all of which would have to be addressed.
First, trains can only take you where the tracks go, so unless you want to run as many tracks as there are roads, your possible destinations are severely limited.
Second, even if the tracks go to your destination, it's not economically viable to take a train there unless a good number of people *all* want to go there, at more or less the same time. This works out reasonably well in areas of high population density (e.g., Chicago), but it doesn't work out well everywhere.
Third, when you get to your destination, you don't have a car. This is the real killer in America. In Europe it works out, because once you get to your destination you can use public transportation to get around, but in North America too many destinations aren't populated at sufficient density to make that practical (see the second point). It's okay if you're going to the big city, because they all have busses and taxis and in many cases a subway or commuter rail or whatnot as well. But outside the big city this is a real problem. *Most* Americans live in smaller communities, where trains and buses aren't economically viable and taxi services only come out if you book them ahead of time. Supposing for the moment that we had rail service to every city in North America. I could then take the train from Galion to Pataskala (a place I go, oh, a couple times a year or so), but I'd arrive in Pataskala with no way to get around locally. I could take the train to Ontario (where we do most of our shopping), but then I wouldn't have a good way to get around over there. I could take the train to Bucyrus (the county seat), or Marion (another nearby county seat), or Hartville (where I have family), or Ashland (other side of the family), but in each case I wouldn't have a way to get around once I get there.
This doesn't mean passenger trains can't be useful. They can. Airplanes suffer from all these problems even more, and yet we have them and use them. But I don't see how trains could replace cars. Supplement, yes. Replace, no.
> Personally, I'd prefer if all the prices I saw were post-tax--as long as the > breakdown is shown on the receipts, or even on price tags on display shelves.
I'd vote for having both the pre-tax and post-tax price on the tag, in the same size font. Retailers would probably hate it, but as a consumer I'd consider it the best of both worlds: you know how much you're going to pay, but you also can easily see how much the tax is.
> Like gas. I'm pretty sure it was built into the pump price when I visited Florida a couple years ago,
Yes. Florida is in the US, so there is a federal excise tax on gasoline. Also cigarettes and alcohol (unless it's denatured or otherwise technically impotable, e.g., isopropanol is inherently exempt), and maybe a couple of other things.
> and it's like that in Canada--even though in both places, other taxable items > (e.g. groceries, computers, services, etc) list the pre-tax price.
That's state sales tax (and, incidentally, groceries are exempt in most states, but I don't know for sure about Florida). State sales tax is added at the cash register in most (possibly all?) US states. There are arguments for and against this, but the strongest argument (that I'm aware of) in favor (of listing the pre-tax price) is that it doesn't *hide* the tax. For historical reasons, Americans have traditionally been pretty touchy certain about tax issues, and hidden taxes are a real hot button. We put up with it on alcohol and cigarettes because of the general view in our culture that nobody *should* be buying those things anyway. (Yes, I know Europeans view alcohol very differently, but we're talking about US taxes here, so how Europeans look at it isn't particularly important.) And we put up with it on gas mainly because *theoretically* all of that money goes to maintain the highways, which are mainly used by people who buy gas, in more or less direct proportion to how much gas they buy, approximately.
It's possible that a general pre-price-tag tax would go over better these days, as a lot of Americans don't really care very much about their history any more, and convenience has become a much more important issue to us than it used to be in my grandparents' generation. But a hundred years ago you'd have had half the population at your doorstep with torches and pitchforks if they thought you were serious about enacting such a thing here. Even now, being the *first* politician to promote it might not exactly guarantee your reelection, if you get my drift.
> I always wondered why they call it a value "added" tax. Seems more like value subtracting to me.
You misunderstood how the modifiers apply. You're thinking "value-added" modifies tax, but it's more that "value" modifies "added tax". That is, it's an added tax based on value. The amount of tax that is added is calculated based on the value of the item, as determined by its price.
There is also a persistent rumor that Canada's GST stands for "General Sales Tax", but in fact it stands for "Gargantuan Sales Tax", a reference to the fact that it's several times larger than the sales tax in any normal country.
Excise taxes traditionally are only levied on specific goods (e.g., alcohol, cigarettes, gasoline).
There are reasonable arguments both for and against using excise tax instead of post-tag sales tax. On the one hand, it's more convenient for the consumer -- you don't have to stop and calculate whether you've got enough money in your pocket to buy the thing. On the other hand, it also makes the tax less noticeable, which *could* be construed as devious and a bad thing, on the grounds that you're not being up front with people.
It would also theoretically be possible to have it halfway in between, either by setting it up so that both prices are required to be on the tag in the same size of font, or by setting it up so that exactly half of the tax is included in the posted price. But such a measure is not exactly entirely certain to be enacted by next Tuesday, if you get my drift, because it not only ensures that people *realize* whenever their taxes go up (which a lot of lawmakers wouldn't necessarily be real enthusiastic about) but also ensures that people know how much they're really going to spend before they put the items in the cart (which the retailers might not be real excited about), so it's kind of a lose-lose situation for everyone but hoi polloi.
Give me the budget and a red pencil, and I'll make your head spin.
Seriously, it's not that tough. There's an enormous amount of completely gratuitous stuff in the US federal budget. If I ran my budget that way I'd be going deeper and deeper into debt, too, instead of putting money into savings every month. If GM ran their budget that way, they'd be in real... oh, wait, they ARE in real financial trouble, so nevermind, they're a bad example.
There are a small number of big places you could cut the budget. Times change, and we *add* new departments to keep up with the changing times, but we never remove or shrink old ones. The USDA, for instance, could reasonably be expected to do everything it actually NEEDS to do on 10% of its current budget. The Dept. of Education is scarcely even necessary *at all*. Education is constitutionally a reserved power, so, hey, just let the states worry about it. Not the federal government's problem. Not the federal government's *business*. (There are a few things the federal goverment has to be involved in, like approving student visas, but it's nowhere near enough to justify an entire department and cabinet position.)
But mostly you would trim a whole lot of little bits of waste out of many, *MANY* little obscure nooks and crannies of the budget.
(And no, you wouldn't do it all at once, because that would shock the economy in undesirable ways. Abruptness is harmful. So you trim a little this year, a little next year, a little more, a little more...)
We do power off our PCs at night where I work, and I could write you a short book on reasons NOT to do so. There are assorted technical reasons, and there are some fairly compelling security reasons (especially if you use Microsoft Windows, which is a fairly common scenario), but most of all there are social reasons.
Technical reasons? Where to begin? How about with "the shut-down interface changes slightly with every revision of Windows (in some cases even for service packs, e.g., XP SP2 changed it) *and* is different for Windows systems that are joined to a Windows domain versus ones that aren't, so there's more staff training than you really want to put into something as non-mission-related as turning the PCs off at night. If you have staff who are actually comfortable with computers and use them at home, you probably haven't even noticed this. I can hear you now, saying exactly what I would have said before I started working with end users: "What do you mean, it changed, it's been the same since 1996." But you're thinking in higher-level terms, like a person who *understands* the interface, so you probably think shutting down Windows is two steps (Start->Shut Down, and click the yeah-shut-down button). But end users don't think that way. Breaking it down the way they see the computer interface, performing this task is about eight steps, too many to easily remember. Some of them have it written down on a half-sheet of paper taped up somewhere near the PC, I kid you not. It goes something like this (depending on what version of Windows they have): To turn off the computer: 1. move the mouse to the lower-left corner where it says Start. 2. Click with the left button and let up. If nothing happens[note 1], click the left button again until it does. 3. When the thing comes up, move the mouse to the red button. 5. Click the left button again. If nothing happens, keep clicking the left button until something does happen. 6. Move the mouse until the arrow points to the OK button. 7. Click the left button again. If nothing happens, keep clicking the left button until something does happen. 8. Wait until the screen goes dark. 9. Turn off the power bar.
Now, if you carefully compare several different versions of Windows, you will discover that steps 3 and 6 on that list have changed with almost every version. And if you join the computer to a domain and then unjoin it, you'll discover that some of the steps appear, disappear, or change. We have computers on a Windows domain on one subnet, and other computers that aren't on one, and the same people are turning both kinds off at night, so users have to keep straight at least two shutdown styles, even if we *did* keep everything on the same version of Windows. Bonus.
Plus, of course, it's not economically feasible for an organization our size to keep everything on the same version. That would mean replacing all the hardware every time Microsoft decides to do a new release, and that would mean we'd have to spend about ten times our whole computer budget just on replacement workstation hardware. So we actually have three or four shutdown styles at any given time, and the number has only gotten that small recently since we've phased out all the legacy systems.
Don't look at me like that. I'm serious, we just got rid of the last Windows 98 system about a year ago, and the last MacOS 9 system about a year before that. The Linux systems all do shutdown -h on a cron job, though, so they don't have to learn the Gnome and KDE shutdown interfaces, which is a nice break. (Now, if somebody tells me about an easy way to make standard Windows workstations turn themselves off automatically without user intervention at set times each day of the week, I'll simultaneously want to jump for joy and also smack my head against the wall for not finding out sooner.)
Let's see, that's one technical reason I've discussed so far, right?
In the interest of brevity, I'll jump to chapter two, security reasons. Two words: Automat
> Overtaxed? Are you kidding me? If anything, Americans are extremely > UNDER taxed. Have you looked at your deficit recently?
The deficit isn't what it is because the tax rates are low. In fact, they are higher than they have ever been in our history, and a *great* deal higher than they would actually need to be if sane people were in charge of the budget.
The deficit is what it is because we spend preposterously obscene amounts of tax money on things the government has absolutely no business spending any money on. Give me the budget and a red pencil, and I'll show you how to cut taxes to less than half the current rate, in every income bracket, *and* operate in the black.
> My understanding is that taxing interstate commerce is prohibited by the constitution.
I'm not a lawyer either, but I don't remember seeing that in the constitution. Which article were you looking at?
The issue I see is that, traditionally, the federal government collects income tax, excise tax, estate tax, and a couple of other categories, but not sales or property tax. Those traditionally go to the state and local governments, exclusively.
The other thing is, some states already have rules for collecting sales tax on purchases that don't have sales tax withheld by the retailer. Ohio, for instance, calls this "use tax", and it goes on your state income tax return. It applies to any (non-exempt) purchase you make wherein sales tax was not already charged, irrespective of *why* it wasn't charged (unless it wasn't charged because it's exempt).
There are also a few weird wrinkles that you'd want to consult a lawyer or tax expert about if you were doing any significant quantity. One example I can think of is clothing purchases, which are tax-exempt in some states (e.g., PA), but not others. I don't happen to know whether the exemption in the state of purchase would extend to the use tax rules in Ohio (where clothing purchases are *not* per-se tax exempt, though most food purchases are). If you happen to be at the mall while visiting relatives in Pennsylvania and see a shirt you just have to have, I wouldn't worry overly much about this -- take your best guess as to whether it's exempt or not and let it go, it's a few cents anyhow, and believe me tax auditors have better things to do than hunt down people who underpaid by less than a dollar. But if you were making larger or more frequent purchases in this kind of category, then you'd want to explore the finer points of the rules in greater detail.
All of which is to say, if the states were to all get together and agree on a uniform set of rules, that would probably be a good thing on the whole, assuming the rules they agree on aren't inherently unreasonable or something. But I do think the sales tax money should still go to the states, not the federal government. I know *they* don't think so, but the federal government collects more than plenty already, and if they've got budget problems it's because they need to learn a little restraint.
> in several years of occasionally being hammered with similar attacks, > they have yet to even guess the one username that is allowed ssh access.
Actually, this consideration is why it is widely regarded as a Good Idea to disallow ssh logins directly as root (or other well-known usernames, but root in particular). I usually allow root ssh logins, for convenience, but I take this into account when I construct the root passwords.
> Yes, because often you can release code in binary form that you're not > allowed to release in source form. That happens, umm, never.
No, actually, in the proprietary world that does happen. When id released the Descent source code, it lacked sound support, because they didn't have source-release rights to a sound library they'd used. (The D1X folks then proceeded to get sound support working, of course. But it took a few months.) Very early versions of OpenOffice.org didn't have spellcheck, I think because Star Division hadn't developed their own, though I was never totally clear on the details of that one. Pegasus Mail can't be easily ported to Linux, at least partly because David Harris doesn't have the necessary rights to release or port the editing control widget that it uses pervasively.
> Anyone with passwords turned on is not secure IMHO
That entirely depends what your passwords are. Mine are generally more than fifteen characters long and include non-alphabetic characters, so a brute-force attack that goes minutes between attempts, like the ones in this article, is fairly unlikely to get anywhere before my hardware dies of old age.
Slashdot Mirror
> One of the fantastic perks of relativity
You don't need relativity. Newtonian physics (and specifically his work on describing gravity) is enough to discredit the notion that the Earth goes around the Sun. The motion of the Sun, like the motion of the Earth, is acted upon by the gravitational forces associated with every mass in the universe.
> The Sun doesn't go around the Earth either
And I suppose you thing the Earth goes around the Sun?
(Both ideas are wrong in exactly the same way. It would be closer to say that the Earth and the Sun move together, but this is still an oversimplification. Every mass in the universe exerts gravitational pull on the Earth and on the Sun as well. As for the pull they exert on one another, it is equal: Earth pulls on the Sun with exactly the same force that the Sun pulls on Earth.)
I think Asia sort of was a country at one time (right next to Bythinia), but then it sort of got taken over by various other regional powers (the Persians, the Greeks, the Romans, the Caliphate, ...) and ended up in the hands of the Turks.
> If you've never learned functional programming, JavaScript is a good language to learn in.
Most of the functional programming techniques I know, I learned in Perl.
It's interesting. I am sufficiently conversant in Emacs lisp to do significant amounts of useful stuff (including custom major modes for special editing situations), but I never understood lambda expressions in lisp. In Perl, though, anonymous subroutines are so straightforward, I immediately understood them and was able to start using them myself after reading barely a page of basic explanation. When I studies scheme I spent *hours* reading over and over and over again the chapter on closures and just couldn't figure the stupid things out. Then somebody on Perlmonks posted some Perl code that used lexical closures, and all became clear; I now use them myself without hesitation.
I've heard Perl6 is going to support continuations. I'm looking forward to it. Maybe if I see them in Perl code I'll understand those, too. I *sure* have not had an easy time trying to understand the explanations of them in the documentation for pure functional languages.
> > But the first OSX made Vista's problems look like first day of school jitters.
> As a user of both, I assure you: no it didn't.
That very much depends on exactly which third-party applications you were trying to use in each case.
> It's just a ridiculous myth that gets repeated on Slashdot and nowhere else.
That's not entirely true. It also gets repeated in comments on people's blogs.
How about translation? I've seen the results of currently available translation software, and there is LOTS of room for improvement there. Can we design the new CAPTCHA tests so that breaking them will improve the state of the art of translation software?
> But Arial is bad
Arial is merely mediocre.
> because it is a bad impersonation of Helvetica
Arial looks better than at least sixty percent of the Helvetica typefaces I've seen, and in any event Helvetica is not an available option for most Windows users, so being redundant with Helvetica does not make Arial non-useful to Windows users.
However, once Verdana was introduced, Arial ceased to be necessary. Verdana is better than Arial in pretty much ever respect (err, except Verdana runs large, but you can correct that easily by reducing the point size), so at this point I would call Arial "obsolete" and "redundant" as well as "mediocre".
But I wouldn't call it "bad".
Times New Roman, now that's bad. A crime against aesthetics.
> The reason to hate it is that it's the Universal "Specialty" font. If you don't want a
> serif font, or a plain font like Arial, the first tool of choice is Comic Sans.
Ah. That's because it's widely available.
If the Core Fonts collection had left out the uglier-than-sin fonts (TNR and Courier New) and the redundant ones (Arial and Trebuchet aren't needed if you've got Verdana; Arial Black isn't needed if you've got Impact; TNR isn't needed ever, but especially not if you've got Georgia; Courier New is a crime against humanity in any case but an especially unnecessary one if you've got Andale Mono) and instead included a three or four more legitimately useful different-look fonts (like, say, a nice caligraphic decorative font, a decent script or brushscript font, a bubble-letters or stencil font, and maybe a grunge font), the total number of fonts would have been smaller, but the total usefulness would have been much greater.
But, you know, that's not what they did.
And when they redid the fonts for Vista (Calibri, Cambria, Candara, etc), they made the same mistake. Too much redundancy, not enough variety. Most people (who don't have a typesetting or graphic-arts background) aren't going to see the difference between Calibri and Candara even if you put the same text set in each of them side by side.
> There are very few fonts where "Q" does not have a stroke below the baseline.
You're looking at modern fonts.
Look at some old bitmapped screen fonts that have 5-pixel-tall letters with a sixth pixel for underlining.
> how are a smudged O and C easier to tell apart than a smudged o and c
The part where they're different is a bit larger, so less likely to be completely obscured.
But I think the part about no descenders is more relevant. Early terminals had *small* numbers of pixels per character. The uppercase letters (well, the standard Latin ones with no diacritical marks that are used in English and included in ASCII) can, in a pinch, all be represented in in a 5x5 grid. It's ugly, but it works.
> written Hebrew doesn't actually supply any indication of what vowels should be inserted between the consonants.
It's more complicated than that. Masoretic writing does supply the vowels for most words (as diacritical marks, but it supplies them). However, it was taboo to *pronounce* God's name at the time, so the qithev/qere system (used to indicate prefered readings) indicated that the reader should pronounce the general word for a lord (usually romanized as Adonai) instead. Because of the way the qithev/qere system works, this means the consonants of the original text are preserved (because they lacked the nerve to actually change the letters of the text), but the vowels (being only diacritical marks, i.e., pronunciation guides, and therefore not sacred, or something) were changed to the ones for the replacement word. (The consonants for the replacement word are written in the margin. Why the replacement vowels couldn't be put in the margin too, and the original ones left in the main text, I don't know.)
> This means that the Abrahamic deity might actually have been called "Yahoo Wahoo"
That is not a particularly likely combination.
But yeah, we don't know the correct vowels for sure.
> > It is easily readable
> Yes. Compared to, say, Wingdings.
Actually, it's one of only three or four fonts my sister (who teaches lower elementary school) is willing to use for classroom materials, because it's one of the only ones the kids can read, because it uses the simple letter forms they teach the kids in kindergarten. The biggest points of contention for most fonts are the lowercase letters a and g. A few other sans fonts use the simple-form g, but almost none of them use the simple-form lowercase a.
Now, one could argue that the schools *should* be teaching the normal lowercase forms that are ordinarily used in almost all print materials throughout the entire English-speaking world. But they *aren't*. (It may be partly because the more common forms are more complex and therefore require more coordination to write. A lot of kindergarten students struggle to get the stick on the right side of the circle for lowercase a, so asking them to write the Times form of the letter admittedly seems a bit much.)
Anyway, I would argue that Comic Sans is better than *several* of the other fonts from Microsoft's "Core Fonts for the Web" initiative.
Georgia and Verdana, of course, are clearly the best of the batch. They actually look good, and furthermore they look good together, which is a fairly big deal. You've got to have a basic serif and a basic sans font that look okay together, and this is a reasonably good pair. I've seen better pairs, but not *many* of them, and especially not ones that were available in 1996. Also, Georgia has a real actual honest-to-goodness italic face, which even manages to LOOK GOOD, which is a fairly rare quality. (I'm not a big fan of most italic faces, as a rule. If anybody knows of a freely-available sans-serif font that actually looks good in italic, I'd sure like to hear about it, because as yet I've not seen one.) Verdana runs a little on the large side, but you can fix that by decrementing the point size, so it's not exactly a deal-killer.
Impact and Andale Mono are acceptable for their intended purposes (wet paint signs and source code, respectively). Lucida Console is in some ways better than Andale Mono, but it's not freely redistributable. Bitstream Vera Sans Mono is alright, but it didn't come out until later.
But after that, really, the fifth-best one in the pack is Comic Sans. Bear with me...
Arial and Arial Black and Trebuchet aren't actively ugly, but they're mediocre, and more to the point they're also pretty redundant with other, better fonts from the same initiative (Verdana, Impact, and Verdana, respectively). Admittedly, Arial dates to Windows 3.x and thus is older than Verdana, but once Verdana was produce we no longer needed Arial for anything (because Verdana looks better), so why was it still included, why is it *still* included with Windows? Why? As for Trebuchet, I never understood why it was needed at any time. It was never *bad*, but it also never had anything to offer over other fonts that were already available.
Then we come to Times New Roman, which is uglier than a half-shaved mandrill, and Courier New, which is the most heinously hideous excuse for a font ever created in True Type format. (I've seen bitmapped screen fonts that were worse, but not many.)
I assume we're not going to try to compare Webdings with anything because, you know, it was never intended for the same basic purpose as the other core fonts (namely, typesetting actual words).
So, Comic Sans isn't the best font ever, but it's orders of magnitude nicer looking than TNR, to say nothing of Courier New, and furthermore it offers a significant stylistic difference from other available fonts, unlike Arial and Arial Black and Trebuchet. It's not as good as Verdana or Georgia, and its niche is (arguably) not as important as the ones for Impact and Andale Mono, but it still serves a useful purpose.
You don't like Comics Sans? Hey, fine, uninstall it from your computer, and websites that try to use it (and don't provide a fallback alternative) will render in whatever font you set as your browser's default. Voila.
But personally I don't see what's so bad about it.
Times New Roman is the one I wish was never created.
Umm, maybe, but...
There are three basic problems with trains, all of which would have to be addressed.
First, trains can only take you where the tracks go, so unless you want to run as many tracks as there are roads, your possible destinations are severely limited.
Second, even if the tracks go to your destination, it's not economically viable to take a train there unless a good number of people *all* want to go there, at more or less the same time. This works out reasonably well in areas of high population density (e.g., Chicago), but it doesn't work out well everywhere.
Third, when you get to your destination, you don't have a car. This is the real killer in America. In Europe it works out, because once you get to your destination you can use public transportation to get around, but in North America too many destinations aren't populated at sufficient density to make that practical (see the second point). It's okay if you're going to the big city, because they all have busses and taxis and in many cases a subway or commuter rail or whatnot as well. But outside the big city this is a real problem. *Most* Americans live in smaller communities, where trains and buses aren't economically viable and taxi services only come out if you book them ahead of time. Supposing for the moment that we had rail service to every city in North America. I could then take the train from Galion to Pataskala (a place I go, oh, a couple times a year or so), but I'd arrive in Pataskala with no way to get around locally. I could take the train to Ontario (where we do most of our shopping), but then I wouldn't have a good way to get around over there. I could take the train to Bucyrus (the county seat), or Marion (another nearby county seat), or Hartville (where I have family), or Ashland (other side of the family), but in each case I wouldn't have a way to get around once I get there.
This doesn't mean passenger trains can't be useful. They can. Airplanes suffer from all these problems even more, and yet we have them and use them. But I don't see how trains could replace cars. Supplement, yes. Replace, no.
> Personally, I'd prefer if all the prices I saw were post-tax--as long as the
> breakdown is shown on the receipts, or even on price tags on display shelves.
I'd vote for having both the pre-tax and post-tax price on the tag, in the same size font. Retailers would probably hate it, but as a consumer I'd consider it the best of both worlds: you know how much you're going to pay, but you also can easily see how much the tax is.
> Like gas. I'm pretty sure it was built into the pump price when I visited Florida a couple years ago,
Yes. Florida is in the US, so there is a federal excise tax on gasoline. Also cigarettes and alcohol (unless it's denatured or otherwise technically impotable, e.g., isopropanol is inherently exempt), and maybe a couple of other things.
> and it's like that in Canada--even though in both places, other taxable items
> (e.g. groceries, computers, services, etc) list the pre-tax price.
That's state sales tax (and, incidentally, groceries are exempt in most states, but I don't know for sure about Florida). State sales tax is added at the cash register in most (possibly all?) US states. There are arguments for and against this, but the strongest argument (that I'm aware of) in favor (of listing the pre-tax price) is that it doesn't *hide* the tax. For historical reasons, Americans have traditionally been pretty touchy certain about tax issues, and hidden taxes are a real hot button. We put up with it on alcohol and cigarettes because of the general view in our culture that nobody *should* be buying those things anyway. (Yes, I know Europeans view alcohol very differently, but we're talking about US taxes here, so how Europeans look at it isn't particularly important.) And we put up with it on gas mainly because *theoretically* all of that money goes to maintain the highways, which are mainly used by people who buy gas, in more or less direct proportion to how much gas they buy, approximately.
It's possible that a general pre-price-tag tax would go over better these days, as a lot of Americans don't really care very much about their history any more, and convenience has become a much more important issue to us than it used to be in my grandparents' generation. But a hundred years ago you'd have had half the population at your doorstep with torches and pitchforks if they thought you were serious about enacting such a thing here. Even now, being the *first* politician to promote it might not exactly guarantee your reelection, if you get my drift.
> I always wondered why they call it a value "added" tax. Seems more like value subtracting to me.
You misunderstood how the modifiers apply. You're thinking "value-added" modifies tax, but it's more that "value" modifies "added tax". That is, it's an added tax based on value. The amount of tax that is added is calculated based on the value of the item, as determined by its price.
There is also a persistent rumor that Canada's GST stands for "General Sales Tax", but in fact it stands for "Gargantuan Sales Tax", a reference to the fact that it's several times larger than the sales tax in any normal country.
HTH.HAND.
Excise taxes traditionally are only levied on specific goods (e.g., alcohol, cigarettes, gasoline).
There are reasonable arguments both for and against using excise tax instead of post-tag sales tax. On the one hand, it's more convenient for the consumer -- you don't have to stop and calculate whether you've got enough money in your pocket to buy the thing. On the other hand, it also makes the tax less noticeable, which *could* be construed as devious and a bad thing, on the grounds that you're not being up front with people.
It would also theoretically be possible to have it halfway in between, either by setting it up so that both prices are required to be on the tag in the same size of font, or by setting it up so that exactly half of the tax is included in the posted price. But such a measure is not exactly entirely certain to be enacted by next Tuesday, if you get my drift, because it not only ensures that people *realize* whenever their taxes go up (which a lot of lawmakers wouldn't necessarily be real enthusiastic about) but also ensures that people know how much they're really going to spend before they put the items in the cart (which the retailers might not be real excited about), so it's kind of a lose-lose situation for everyone but hoi polloi.
> To an extent, you can. But what do you cut?
Give me the budget and a red pencil, and I'll make your head spin.
Seriously, it's not that tough. There's an enormous amount of completely gratuitous stuff in the US federal budget. If I ran my budget that way I'd be going deeper and deeper into debt, too, instead of putting money into savings every month. If GM ran their budget that way, they'd be in real... oh, wait, they ARE in real financial trouble, so nevermind, they're a bad example.
There are a small number of big places you could cut the budget. Times change, and we *add* new departments to keep up with the changing times, but we never remove or shrink old ones. The USDA, for instance, could reasonably be expected to do everything it actually NEEDS to do on 10% of its current budget. The Dept. of Education is scarcely even necessary *at all*. Education is constitutionally a reserved power, so, hey, just let the states worry about it. Not the federal government's problem. Not the federal government's *business*. (There are a few things the federal goverment has to be involved in, like approving student visas, but it's nowhere near enough to justify an entire department and cabinet position.)
But mostly you would trim a whole lot of little bits of waste out of many, *MANY* little obscure nooks and crannies of the budget.
(And no, you wouldn't do it all at once, because that would shock the economy in undesirable ways. Abruptness is harmful. So you trim a little this year, a little next year, a little more, a little more...)
We do power off our PCs at night where I work, and I could write you a short book on reasons NOT to do so. There are assorted technical reasons, and there are some fairly compelling security reasons (especially if you use Microsoft Windows, which is a fairly common scenario), but most of all there are social reasons.
Technical reasons? Where to begin? How about with "the shut-down interface changes slightly with every revision of Windows (in some cases even for service packs, e.g., XP SP2 changed it) *and* is different for Windows systems that are joined to a Windows domain versus ones that aren't, so there's more staff training than you really want to put into something as non-mission-related as turning the PCs off at night. If you have staff who are actually comfortable with computers and use them at home, you probably haven't even noticed this. I can hear you now, saying exactly what I would have said before I started working with end users: "What do you mean, it changed, it's been the same since 1996." But you're thinking in higher-level terms, like a person who *understands* the interface, so you probably think shutting down Windows is two steps (Start->Shut Down, and click the yeah-shut-down button). But end users don't think that way. Breaking it down the way they see the computer interface, performing this task is about eight steps, too many to easily remember. Some of them have it written down on a half-sheet of paper taped up somewhere near the PC, I kid you not. It goes something like this (depending on what version of Windows they have): To turn off the computer: 1. move the mouse to the lower-left corner where it says Start. 2. Click with the left button and let up. If nothing happens[note 1], click the left button again until it does. 3. When the thing comes up, move the mouse to the red button. 5. Click the left button again. If nothing happens, keep clicking the left button until something does happen. 6. Move the mouse until the arrow points to the OK button. 7. Click the left button again. If nothing happens, keep clicking the left button until something does happen. 8. Wait until the screen goes dark. 9. Turn off the power bar.
Now, if you carefully compare several different versions of Windows, you will discover that steps 3 and 6 on that list have changed with almost every version. And if you join the computer to a domain and then unjoin it, you'll discover that some of the steps appear, disappear, or change. We have computers on a Windows domain on one subnet, and other computers that aren't on one, and the same people are turning both kinds off at night, so users have to keep straight at least two shutdown styles, even if we *did* keep everything on the same version of Windows. Bonus.
Plus, of course, it's not economically feasible for an organization our size to keep everything on the same version. That would mean replacing all the hardware every time Microsoft decides to do a new release, and that would mean we'd have to spend about ten times our whole computer budget just on replacement workstation hardware. So we actually have three or four shutdown styles at any given time, and the number has only gotten that small recently since we've phased out all the legacy systems.
Don't look at me like that. I'm serious, we just got rid of the last Windows 98 system about a year ago, and the last MacOS 9 system about a year before that. The Linux systems all do shutdown -h on a cron job, though, so they don't have to learn the Gnome and KDE shutdown interfaces, which is a nice break. (Now, if somebody tells me about an easy way to make standard Windows workstations turn themselves off automatically without user intervention at set times each day of the week, I'll simultaneously want to jump for joy and also smack my head against the wall for not finding out sooner.)
Let's see, that's one technical reason I've discussed so far, right?
In the interest of brevity, I'll jump to chapter two, security reasons. Two words: Automat
> Overtaxed? Are you kidding me? If anything, Americans are extremely
> UNDER taxed. Have you looked at your deficit recently?
The deficit isn't what it is because the tax rates are low. In fact, they are higher than they have ever been in our history, and a *great* deal higher than they would actually need to be if sane people were in charge of the budget.
The deficit is what it is because we spend preposterously obscene amounts of tax money on things the government has absolutely no business spending any money on. Give me the budget and a red pencil, and I'll show you how to cut taxes to less than half the current rate, in every income bracket, *and* operate in the black.
> My understanding is that taxing interstate commerce is prohibited by the constitution.
I'm not a lawyer either, but I don't remember seeing that in the constitution. Which article were you looking at?
The issue I see is that, traditionally, the federal government collects income tax, excise tax, estate tax, and a couple of other categories, but not sales or property tax. Those traditionally go to the state and local governments, exclusively.
The other thing is, some states already have rules for collecting sales tax on purchases that don't have sales tax withheld by the retailer. Ohio, for instance, calls this "use tax", and it goes on your state income tax return. It applies to any (non-exempt) purchase you make wherein sales tax was not already charged, irrespective of *why* it wasn't charged (unless it wasn't charged because it's exempt).
There are also a few weird wrinkles that you'd want to consult a lawyer or tax expert about if you were doing any significant quantity. One example I can think of is clothing purchases, which are tax-exempt in some states (e.g., PA), but not others. I don't happen to know whether the exemption in the state of purchase would extend to the use tax rules in Ohio (where clothing purchases are *not* per-se tax exempt, though most food purchases are). If you happen to be at the mall while visiting relatives in Pennsylvania and see a shirt you just have to have, I wouldn't worry overly much about this -- take your best guess as to whether it's exempt or not and let it go, it's a few cents anyhow, and believe me tax auditors have better things to do than hunt down people who underpaid by less than a dollar. But if you were making larger or more frequent purchases in this kind of category, then you'd want to explore the finer points of the rules in greater detail.
All of which is to say, if the states were to all get together and agree on a uniform set of rules, that would probably be a good thing on the whole, assuming the rules they agree on aren't inherently unreasonable or something. But I do think the sales tax money should still go to the states, not the federal government. I know *they* don't think so, but the federal government collects more than plenty already, and if they've got budget problems it's because they need to learn a little restraint.
> in several years of occasionally being hammered with similar attacks,
> they have yet to even guess the one username that is allowed ssh access.
Actually, this consideration is why it is widely regarded as a Good Idea to disallow ssh logins directly as root (or other well-known usernames, but root in particular). I usually allow root ssh logins, for convenience, but I take this into account when I construct the root passwords.
> Yes, because often you can release code in binary form that you're not
> allowed to release in source form. That happens, umm, never.
No, actually, in the proprietary world that does happen. When id released the Descent source code, it lacked sound support, because they didn't have source-release rights to a sound library they'd used. (The D1X folks then proceeded to get sound support working, of course. But it took a few months.) Very early versions of OpenOffice.org didn't have spellcheck, I think because Star Division hadn't developed their own, though I was never totally clear on the details of that one. Pegasus Mail can't be easily ported to Linux, at least partly because David Harris doesn't have the necessary rights to release or port the editing control widget that it uses pervasively.
> Anyone with passwords turned on is not secure IMHO
That entirely depends what your passwords are. Mine are generally more than fifteen characters long and include non-alphabetic characters, so a brute-force attack that goes minutes between attempts, like the ones in this article, is fairly unlikely to get anywhere before my hardware dies of old age.