> Vi and emacs are to an IDE what a Ford model T is to a 2006 Lexus.
I won't talk about vi, but this is a terrible analogy for Emacs. If an IDE is a 2006 Lexus, then Emacs is the car from the movie Batman Begins. It's much more intimidating and harder to learn how to use, but under that hard exterior there is all sorts of functionality that you previously never realized was possible.
> being forced to write things by hand often does produce a better assignment - > you are forced to redraft (at least once). When you redraft by hand, you have > to write every word again, which gives you time to think carefully about every > individual sentence each time you write it out. If you use a Word Processor, > then if you redraft at all, you're likely just to skim
I would have said pretty much exactly the opposite. When I had to write every draft by hand, I seldom revised anything more than two or at most three times. Frequently, I would hand in my second draft as final, even for English writing assignments, simply because the prospect of rewriting the thing by hand was so painful.
With the advent of word processing I found myself able to revise much more often, because it didn't take nearly as long. Frequently I would go through half a dozen drafts or more even for minor assignments in non-language courses.
For English composition teachers, the best piece of advice I could give would be to build repeated revision into the cirruculum -- assign the students to pick out some of their best work from earlier in the semester and improve it further. Good writing comes from good writers. Great writing comes from mediocre writers who tenaciously keep revising and improving their work for years or, in extreme cases, decades. Tolkein is an excellent example here: if you read the Book of Lost Tales, it becomes immediately apparent that *much* revision was needed, but Tolkein realized this and spent most of his life revising his material, eventually ending up with something really great.
Computer programming teachers would do well to keep that same principle in mind. However, I'm not convinced that the question of whether to use an IDE or not has anything to do with that.
Some languages should be taught with an IDE, some with a text editor (in most cases, letting the students each select any text editor they like, _even_ if they pick Notepad, because it just doesn't matter for purposes of learning the language, although you should mention in passing at least once that there are much better editors available), and for some languages you can really go either way, depending on what your goals are for the course. Every student should at some point have to take languages taught with an IDE and languages taught with a text editor.
C++ is an example of a language that should be taught with an IDE. Among other things, the documentation features of an IDE are really necessary. Even more obviously, trying to teach Visual Basic without the IDE would be ridiculous and pointless.
Perl is an example of a language that should be taught using text editors, letting the students pick whatever text editors they want. Teaching Perl with an IDE would be like teaching algebra with pottery wheels -- it would just get in the way.
Lisp is a special case and should be taught with Emacs specifically, both because there are features you just have to have (not least, grouping-symbol matching) and because Emacs is customized using a dialect of lisp. Also, if you're teaching lisp, it's too good an opportunity to pass up to slip Emacs in there, so that the students are exposed to that.
I can't tell you which category Java belongs in, because I never studied Java, but I can tell you that you should set aside the advice of anyone who bases his answer on experience learning or teaching other languages with or without an IDE, and focus on the answers given by people whose experience is with Java, because this is a question that has to be answered on a per-language basis.
> Look at it from the criminals' perspective. What is the most desirable gun > with which to commit a crime?
Depends on the crime.
> Is the AK-47,
Actually, in some parts of the world the AK-47 is extremely popular with criminals. These tend to be areas of the world with approximately zero real law enforcement, where the criminals can openly parade around with their weapons in plain view, sieze control of entire cities, and so forth -- i.e., the third world mostly, and parts of the second world.
> the M-16,
From a criminal perspective, the M-16 is in pretty much all ways inferior to the M-8 Carbine. However, neither is very popular with criminals and here's why: it's harder to obtain than the AK-47, especially in the second and third world (where assault rifles make sense). In the first world, where there is reasonably decent law enforcment, the dynamics are such that range is pretty much a non-issue, because everything has to be done up close. Under those circumstances, a submachine gun is preferable to an assault rifle, so the criminals will go with an Uzi (or a Thompson or MP5 or whatnot) when they need bulk firepower, or a handgun otherwise.
The M-8 or M-16 might be somewhat popular with criminals in Africa and the Middle East, if it were as easy to obtain there as the AK-47 is, although Kalashnikov's idiot-friendly maintenance is also a factor.
> All of these rifles are bulky, difficult to conceal, tough to wield in > tight circumstances,
The tight circumstances are key. If you know you'll NEVER have to fire on a target more than about a block away, you're going to go with the submachine gun over the assault rifle every time, and possibly even the handgun. The choice of the handgun versus the submachine gun is going to come down to whether you expect to be in firefights (gang warfare or somesuch -- get the submachine gun) or not (more "ordinary" crimes such as armed robbery -- you wave a handgun and don't wait around afterwards for the police to show up).
> Remarkably few crimes have been commited with.50 calibre bolt action rifles,
Assasinations. But yeah, non-automatic weapons are not mostly used for crime.
> or AK-47s,
Depends where you live. Seriously, in some parts of the world, the AK-47 is *THE* weapon of choice for criminals of all sorts.
> But how can we know the planets are indeed distributed as they are in our Solar > System, with a rocky planet with the right elements located in zone around the > star that can support liquid water for billions of years?
Actually, if you read the article, it becomes instantly apparent that these planets are *not* inhabitable by earthlife as they stand, and probably not even with significant terraforming.
The astroscientists are excited because they're now able to use stellar wobble to locate planets the size of Neptune, which is a significant improvement over years past when they could only find Jupiter-sized planets this way. The actual planets they found are not the main point here.
Foremost, the number one feature I miss is grouping-symbol matching. Emacs has had this since RAM was measured in kilobytes, and it's one of the Most Useful Features Ever. (I also miss it in textarea fields in my web browser.) I think MS Office is also missing this important feature.
There are a handful of edge cases in Writer (e.g., right before or after a table) where it's not always easy to insert or delete a paragraph. More than once I've ended up editing the XML by hand, and that should NOT be necessary (although it's a nice ability to have and creates a lot of flexibility for automatic document generation).
When you've got a chart in a Calc, it can be a pain to remember how many times to click, double-click, or right-click, and in what order, to get to a specific option (e.g., to modify the data range (select, right-click), or to switch to a logarithmic scale (select, double-click, then right-click)). Shouldn't it be possible to just right-click once, choose Properties from a context menu, and get a tabbed dialog box with *all* the available options that can be modified for the chart? The capabilities I need are available, but the UI needs work.
OOo Base is quite rough around the edges still. That's to be expected, as it's a relatively new addition, but it's very noticeable. Getting it configured to talk to a local RDBMS, for instance, is non-trivial and more than a little confusing. It _should_ be a small matter of selecting the type of database from a list and plugging in host, username, password, and database name. At least, it should be that easy for any of the major database systems (Postgres, MySQL, Oracle, MS SQL Server,...). Setting up a database-in-a-file (using the built-in database engine) isn't very easy either.
But most of all, and I do mean most of all, I want grouping-symbol matching. Please.
Alliteration. Ruby on Rails: it just _sounds_ good.
What it actually _is_, I don't know. I know approximately what Ruby is, although not intimately, but I have no idea what the rails signifiy in technical terms, although they presumably connote rapidity of some sort. If I were a Ruby programmer, I'd certainly want to know what this Rails thing is all about.
Whether it sounds good enough to make me (as a Ruby outsider) want to learn the language is another question. Frankly if I were to learn Ruby it would probably be more due to the good things I have heard about it from sources like perl6-language, than due to Rails, because whatever other tools you use the quality of the core language is critical to their usefulness, so as a programmer it's the style of the language I want to know about first. (This is part of why I never picked up Java: the GUI libraries it comes with are supposed to be really great, but nobody with similar tastes in languages to mine ever has many attractive things to say about the design of the language itself; Ruby, OTOH, does regularly get positive comments from people whose opinions on language design I respect greatly.) Nonetheless, if I *did* take the time to learn Ruby, and if I found that I liked it as a language, Rails would definitely go on my to-learn list then, *despite* that I don't actually know what it does. Because, you know, it just sounds good. Probably looks good on a resume too.
> Copper works, its easy to work with, and even when it fails, you can still shut > it off by squeezing the pipe shut at the break with a pair of pliers. It also > makes a great electrical ground, and is easily recycled. And there's no issue > with plasticizers leaching into the water you drink.
Yeah, I know all that. I wasn't saying anything against copper pipe. I was just pointing out that a lot of older homes contain other kinds of pipe. Usually the hot water pipes are copper, but frequently some of the cold water pipers are steel (or, in really old homes, lead, and the homeowner often doesn't know it).
> It is not like DDT is the only available pesticide
It's cheap. You get a lot of bang for your buck with DDT. Since African governments tend to want to limit their spending for public services (such as public health, highways, and education) to approximately nothing, DDT tends to be attractive to them. If you want them to use something else, the western world probably has to pick up the bill.
Black tea without sugar is fine, but green tea really needs the sugar as far as I'm concerned and as for coffee... I don't think any amount of sugar could make that stuff taste good.
> Also, if you've ever done any home plumbing, you'd know that even 50-year-old > copper pipe is in decent shape inside, after decades of attack by chlorine, > ozone, and good old H2O.
Where do *you* live? Around here, copper pipe only started to come into widespread use in the last fifty years or so; most older homes still have pipe made from other materials (frequently steel, the regular kind, not stainless), and even if your home is new, if you live in a municipality, even a small one, the pipes coming to your house are certainly not copper.
With that said, I prefer tapwater to bottled water, and as for micro-organisms, that's what the trace amounts of chlorine are for.
> You never drank water before did you? There IS taste.
It's not the water itself that has taste, but the minerals in it. Pure water isn't very nice to drink, largely because it's so hypotonic, but it doesn't really taste like anything. Tapwater does, because of the various trace ingredients.
> > Tap, preferably > See, you don't have taste (or perhaps good tap water)
I prefer most tapwater I've encountered to the bottled waters I've tried. It does vary, though...
> I think for me personally, it's the ammount of calcium that gives a good taste > (the more the better!)
Calcium, eh?
I know the tastes of iron (which I like), lime (which is not bad too), sulfur (which is overwhelming if there's much of it but can be okay in quite small amounts), chlorine (which is good in small amounts but unpleasant if overdone), and sodium (which I don't like in my water, even in quite small amounts)...
Say, doesn't lime have a lot of calcium in it? There's a lot of lime in the water where I went to college, and I liked the water there okay. (It also had noticeable iron, though, which doubtless helped the taste, as I tend to like iron in my water, preferably enough to make orange stains in the sinks after a few years.)
> In fact I can't really enjoy sweet drinks these days, they just taste too "cheap".
Around here, nothing's cheaper than tapwater, but I like it anyway. I don't mind sweet drinks (sometimes I put sugar in my tea, even quite a bit of sugar (although, sometimes I don't), and I drink Kool-Aid, especially in the summer, and from time to time pop), but I also like water.
> It's a lot harder to make low or sugarfree drinks without gas that are quite good.
Cut up a ginger root and boil it for a few minutes, then throw out the pieces and drink the water. (If it's too strong, dilute to taste with water.) This is good hot or iced, with or without sugar. I'm not sure I'd want it bottled from a factory, though.
> Seriously, I have never understood the thing that IT people have for sucking > down caffeine all day long
For some people, caffeine is extremely addictive. I'm not sure why some people are so much more affected by this than others, but while some of us (I'm one of the lucky ones, I guess) can drink a lot of caffeine one day and then suddenly switch to water or milk and feel fine, others drink a single Coke and a few hours later get very grouchy if they don't have another.
As for me, I can take caffeine or leave it. It doesn't do much for me or to me, one way or the other. I can drink a quart of strongly-brewed black tea and half an hour later go to bed and sleep normally; I can drink a 32-oz Mt Dew and then not have any caffeine for days, and I don't crave it. Not everyone is so lucky. My sister can't drink anything with caffeine in it after about 2pm, or she won't sleep that night. Problem is, if she has something with caffeine in it in the morning, she wants more at noontime, and more in the afternoon, and more later... and if she lets herself have the amounts of it she wants, she turns into an obsessive taskmaster, working (and driving others, if possible) at breakneck pace, intent on getting six days' worth of stuff done in the next six hours. It's something to behold. Once.
My guess is that the people who suck down caffeine all day long are the ones more like my sister than like me.
> The "I don't like the taste" argument against diet soft drinks is nothing more than > a bullshit excuse. If it isn't sweet enough for you, that means you're drinking so > much high-fructose corn syrup that your taste buds are desensitized to sweetness.
It's not the absense of sugar. I drink unsweetened things all the time: black tea, milk (okay, so that has natural sugar, but not high-fructose corn syrup like you were talking about), water (lots of this, preferably tapwater, preferably with enough iron in it to turn the sinks orange after a few years), and so on. (I also sometimes drink black tea _with_ sugar, even quite a bit of sugar. The flavour is different, but I actually like it either way. Green tea I only drink with sugar, though.) It's not the lack of sugar. I like sugar, but I don't mind drinking things without it.
The thing that makes diet softdrinks so impotable is that they taste like artificial sweeteners, i.e., positively incredibly *nasty*. Artificial sweeteners taste outlandishly terrible, on par with the old kind of envelope glue they used to use in the seventies before the nicer, more expensive, gentler-tasting kind of envelope glue was introduced. Aspartame is absolutely the worst, but the others aren't much better. I don't have any problem drinking unsweetened beverages, but I will NOT drink artificially-sweetened ones.
They don't make unsweetened pop, so if I'm going to drink pop, it's going to be the kind with sugar. (That said, I don't drink pop that often. Mostly water, milk, Kool-Aid (yeah, okay, so that's got just about as much sugar as pop), and tea. Pop I drink only occasionally.)
> any reason that isn't explicitly prohibited by law. In most jurisdictions, this > is limited to race, gender, religion, non-disqualifying handicap, age, and perhaps > a handful of other characteristics.
There are, in the US at least, a number of other specific things they dursn't fire you for, besides the discrimination-oriented ones. Refusing to falsify records, refusing to violate the law, attempting to unionize, taking time off for certain things as permitted under the FMLA, and so on and so forth.
And that's all just in terms of criminal law. There's also whatever the employee can sue you for under civil law, which is what really scares most employers.
You don't have to use a microkernel design to have userland drivers. Heck, Microsoft has started talking (so far just talking, as near as I can tell, but it's a start) about userland drivers, and you *know* they're not going to a microkernel. Moving drivers out of the kernel doesn't have to mean breaking the rest of the kernel into pieces as well.
> Also, given that some microkernels are only about 3500-6000 lines of code > (as opposed to Linux's million or so) it's relatively easy to make certain > that the code is bug free
Microkernels aren't only 3500-6000 lines of code if you count the whole kernel's worth of functionality. At least, not if they're written in a low-level language like C, which generally seems like the way to go for kernels.
I don't think it's necessary to start Linux over from scratch just yet. Sure, it will *eventually* be necessary to replace Linux, as it is with all software sooner or later, but most *nix software these days is sufficiently general that it will run on other kernels besides Linux, so when the time comes that shouldn't be a huge problem. There are, after all, other kernels around.
> Don't forget that this year will bring Linux on the desktop
Linux maybe, but not Debian. The whole "does not change often" thing may work for some kinds of servers, but on the desktop it's a nightmare. Oh, you want to install the security-critical bug-fix update to your web browser? I'm sorry, that requires a recent version of the widget toolkit it uses, which in turn requires recent versions of three-quarters of the stuff on your system, none of which is available for any distro more than about six months old. HTH.HAND.
Debian, whatever else it may be, is not a desktop-oriented distribution.
(And anyway, the days of Linux on the desktop are drawing to a close. We've been there, and we've done that, and it was okay, but now we're coming to the days of BSD on the desktop. Nietzsche is dead already, and NetCraft is up next.)
> Give me a good reason to upgrade [from potato] being this is my web/mail/dns/ftp box.
One thing that springs to mind off the top of my head is that IP Tables (the firewall system in Linux 2.4 and higher) is much nicer to work with than IP Chains (Linux 2.2, IIRC).
Even if you don't update the whole distro, I sure hope you've updated the services that are exposed to the internet from the versions in potato. Especially the mail server. (If you haven't, I'd seriously consider replacing the system altogether and decommissioning it, as it's probably too infested to consider repairable.)
It is reasonable to stay with gcc 2.95 for this type of system. It wouldn't be practical for a desktop, but for a server that does not change often it should be fine.
If you don't want to move to Linux 2.4, you might consider a later release of 2.2. I really do recommend 2.4 though, for IP Tables if nothing else. Very nice kernel, and quite mature at this point. For the web server, stay with Apache 1.x, but get a more recent Apache 1.x than the one that shipped with potato. For the mail server, you don't want anything to do with the one in potato. For dns, it depends on your needs -- do you really need all of bind's features, or could you be using something smaller and safer, along the lines of djbdns or whatnot? For ftp, it depends on whether you need to offer anonymous ftp on port 21. If you do, then you really want to pay attention to security issues for sure, which probably means staying on the latest release most of the time. (Which ftp daemon are you using, anyway?)
Do have a look at IP Tables. Really it's quite a nifty system.
Seriously. You don't need a white-hot performance-optimizing programmer, because nothing you're doing is going to tax the system resources of a Pentium II, even with _fairly_ inefficient code (as long as none of the algorithms it uses are worse than about O(n log n).) What you do need is somebody who understand the importance of robustness, and that's not nearly as hard to hire, or as expensive. Your existing solution is so archaic and so bad, any decent computer geek will be able to put together something you can live with in a matter of a couple of months, and then you'll have various improvement requests over the next year or so after that.
Just make sure, before you let the programmer start writing any code, that you explain to him all the important features of your existing software, how they work, and why you need them.
Vacation? You get vacation? Why, back in my day we were expected to write code 365 days a year. One year they decided to be nice and give us twenty minutes off for Christmas and an extra crust of bread. We might've gotten that every year, but the young whippersnappers had the nerve to *complain*. Ingrates. Spoiled it for all of us.
> Even for plain printed text. This falls squarely in the quality department for me.
It depends where you live. Around here, the only place you _ever_ see a ligature is in a stylized brand name (where its purpose is to distinguish the brand name from the normal English word it's based on), and it is normal even in heavily academic contexts for foreign words and phrases to appear in print sans any diacritical marks that they would have had in their original language: facade, jalapeno, resume, je ne sais quoi, and so on and so forth. (For that matter, Greek, Russian, Hebrew, and Chinese words are almost always written in Latin characters, even in heavily academic contexts, not because people don't know about other writing systems, but because the Latin alphabet is the standard and more convenient one.) Anyway, Word certainly *can* do oddball characters (albeit, it's a bit of a pain if you don't have a special keyboard), but most people _don't_ do them.
Quite aside from that, you're trying to compare the quality of a wine press versus a blender. They can both make juice, sure, but trying to compare their quality as juice-makers would be bizarre. Arguing the quality of MS Word against TeX is a bit like that, or, to use a more technical example, like arguing the quality of Gimp against Adobe Illustrator. Yeah, they both do graphics, but they represent completely different ways of thinking about graphics and as such they're not directly comparable. Gimp should be compared with Photoshop, not Illustrator. (If you want to compare an open-source project with Illustrator, you could reach for Inkscape, although that's not really a mature product yet. (I use it, and I like it, but it's... not done.)) MS Office should be compared with OpenOffice, not with TeX. (Microsoft does not, as far as I am aware, really have a product that attempts to directly compete with TeX, but if they did, and if they bundled it in with Office, it still wouldn't be used for most of the stuff people do with MS Word. (Whether it would gain any significant traction against TeX is another matter (and really beside the point (and yes, I've done a little lisp programming on occasion (why do you ask?))).))
Not yet. At the time of this writing, it's still supposed to include the Aero Glass interface, which allows window titlebars to be transparent (if you have a good video card).
> not to worry, at this rate it WILL be perfectly timed for the holiday > period... xmas 2007.
Don't hold your breath.
Projected release dates are meaningless until they're *this* quarter. Arguing about whether the date is going to get pushed back from the quarter after next to the quarter after that is just idle speculation. It's preposterously hard to predict release dates with any accuracy that far into the future for a product like Windows from a company like Microsoft. Vista will ship when it's ready. Any announced release dates prior to that are just marketing hype intended to keep people excited.
When they start saying they're going to release in the *current* quarter, then you figure they're actually close to having it ready to ship. Until then, who knows.
> Actually less when you consider Vista will run on 800mhz machines with 512mb > of RAM quite well.
Define "run". Windows XP, assuming you want to run more than two applications at once, creeps like a paraplegic on tranquilizers if you only give it 512MB of RAM, so I have a hard time imagining Vista will run in an acceptable fashion on that much.
(Not that modern Unix-like systems are terribly happy with only 512MB either. Really I think 1GB of RAM is pretty much a minimum these days irrespective of OS, if you want to run several apps at once. But your claim that Vista will be run in a reasonable fashion with only 512 MB of RAM is not believable.)
OTOH, I would not consider anything up to 2GB of RAM to be "insane specs", so you have a point as far as that is concerned. More than 2GB would be problematic at this time cheifly because far too many motherboards can only handle up to that amount. Good motherboards can take 4GB (and a few take more, but those tend to be for 64-bit systems), but most boards at this point still max out at 2GB.
Slashdot Mirror
> Vi and emacs are to an IDE what a Ford model T is to a 2006 Lexus.
I won't talk about vi, but this is a terrible analogy for Emacs. If an IDE is a 2006 Lexus, then Emacs is the car from the movie Batman Begins. It's much more intimidating and harder to learn how to use, but under that hard exterior there is all sorts of functionality that you previously never realized was possible.
> being forced to write things by hand often does produce a better assignment -
> you are forced to redraft (at least once). When you redraft by hand, you have
> to write every word again, which gives you time to think carefully about every
> individual sentence each time you write it out. If you use a Word Processor,
> then if you redraft at all, you're likely just to skim
I would have said pretty much exactly the opposite. When I had to write every draft by hand, I seldom revised anything more than two or at most three times. Frequently, I would hand in my second draft as final, even for English writing assignments, simply because the prospect of rewriting the thing by hand was so painful.
With the advent of word processing I found myself able to revise much more often, because it didn't take nearly as long. Frequently I would go through half a dozen drafts or more even for minor assignments in non-language courses.
For English composition teachers, the best piece of advice I could give would be to build repeated revision into the cirruculum -- assign the students to pick out some of their best work from earlier in the semester and improve it further. Good writing comes from good writers. Great writing comes from mediocre writers who tenaciously keep revising and improving their work for years or, in extreme cases, decades. Tolkein is an excellent example here: if you read the Book of Lost Tales, it becomes immediately apparent that *much* revision was needed, but Tolkein realized this and spent most of his life revising his material, eventually ending up with something really great.
Computer programming teachers would do well to keep that same principle in mind. However, I'm not convinced that the question of whether to use an IDE or not has anything to do with that.
Some languages should be taught with an IDE, some with a text editor (in most cases, letting the students each select any text editor they like, _even_ if they pick Notepad, because it just doesn't matter for purposes of learning the language, although you should mention in passing at least once that there are much better editors available), and for some languages you can really go either way, depending on what your goals are for the course. Every student should at some point have to take languages taught with an IDE and languages taught with a text editor.
C++ is an example of a language that should be taught with an IDE. Among other things, the documentation features of an IDE are really necessary. Even more obviously, trying to teach Visual Basic without the IDE would be ridiculous and pointless.
Perl is an example of a language that should be taught using text editors, letting the students pick whatever text editors they want. Teaching Perl with an IDE would be like teaching algebra with pottery wheels -- it would just get in the way.
Lisp is a special case and should be taught with Emacs specifically, both because there are features you just have to have (not least, grouping-symbol matching) and because Emacs is customized using a dialect of lisp. Also, if you're teaching lisp, it's too good an opportunity to pass up to slip Emacs in there, so that the students are exposed to that.
I can't tell you which category Java belongs in, because I never studied Java, but I can tell you that you should set aside the advice of anyone who bases his answer on experience learning or teaching other languages with or without an IDE, and focus on the answers given by people whose experience is with Java, because this is a question that has to be answered on a per-language basis.
> Look at it from the criminals' perspective. What is the most desirable gun
.50 calibre bolt action rifles,
> with which to commit a crime?
Depends on the crime.
> Is the AK-47,
Actually, in some parts of the world the AK-47 is extremely popular with criminals. These tend to be areas of the world with approximately zero real law enforcement, where the criminals can openly parade around with their weapons in plain view, sieze control of entire cities, and so forth -- i.e., the third world mostly, and parts of the second world.
> the M-16,
From a criminal perspective, the M-16 is in pretty much all ways inferior to the M-8 Carbine. However, neither is very popular with criminals and here's why: it's harder to obtain than the AK-47, especially in the second and third world (where assault rifles make sense). In the first world, where there is reasonably decent law enforcment, the dynamics are such that range is pretty much a non-issue, because everything has to be done up close. Under those circumstances, a submachine gun is preferable to an assault rifle, so the criminals will go with an Uzi (or a Thompson or MP5 or whatnot) when they need bulk firepower, or a handgun otherwise.
The M-8 or M-16 might be somewhat popular with criminals in Africa and the Middle East, if it were as easy to obtain there as the AK-47 is, although Kalashnikov's idiot-friendly maintenance is also a factor.
> All of these rifles are bulky, difficult to conceal, tough to wield in
> tight circumstances,
The tight circumstances are key. If you know you'll NEVER have to fire on a target more than about a block away, you're going to go with the submachine gun over the assault rifle every time, and possibly even the handgun. The choice of the handgun versus the submachine gun is going to come down to whether you expect to be in firefights (gang warfare or somesuch -- get the submachine gun) or not (more "ordinary" crimes such as armed robbery -- you wave a handgun and don't wait around afterwards for the police to show up).
> Remarkably few crimes have been commited with
Assasinations. But yeah, non-automatic weapons are not mostly used for crime.
> or AK-47s,
Depends where you live. Seriously, in some parts of the world, the AK-47 is *THE* weapon of choice for criminals of all sorts.
> But how can we know the planets are indeed distributed as they are in our Solar
> System, with a rocky planet with the right elements located in zone around the
> star that can support liquid water for billions of years?
Actually, if you read the article, it becomes instantly apparent that these planets are *not* inhabitable by earthlife as they stand, and probably not even with significant terraforming.
The astroscientists are excited because they're now able to use stellar wobble to locate planets the size of Neptune, which is a significant improvement over years past when they could only find Jupiter-sized planets this way. The actual planets they found are not the main point here.
Just grab your towell, and you're all set to go.
Foremost, the number one feature I miss is grouping-symbol matching. Emacs has had this since RAM was measured in kilobytes, and it's one of the Most Useful Features Ever. (I also miss it in textarea fields in my web browser.) I think MS Office is also missing this important feature.
...). Setting up a database-in-a-file (using the built-in database engine) isn't very easy either.
There are a handful of edge cases in Writer (e.g., right before or after a table) where it's not always easy to insert or delete a paragraph. More than once I've ended up editing the XML by hand, and that should NOT be necessary (although it's a nice ability to have and creates a lot of flexibility for automatic document generation).
When you've got a chart in a Calc, it can be a pain to remember how many times to click, double-click, or right-click, and in what order, to get to a specific option (e.g., to modify the data range (select, right-click), or to switch to a logarithmic scale (select, double-click, then right-click)). Shouldn't it be possible to just right-click once, choose Properties from a context menu, and get a tabbed dialog box with *all* the available options that can be modified for the chart? The capabilities I need are available, but the UI needs work.
OOo Base is quite rough around the edges still. That's to be expected, as it's a relatively new addition, but it's very noticeable. Getting it configured to talk to a local RDBMS, for instance, is non-trivial and more than a little confusing. It _should_ be a small matter of selecting the type of database from a list and plugging in host, username, password, and database name. At least, it should be that easy for any of the major database systems (Postgres, MySQL, Oracle, MS SQL Server,
But most of all, and I do mean most of all, I want grouping-symbol matching. Please.
Alliteration. Ruby on Rails: it just _sounds_ good.
What it actually _is_, I don't know. I know approximately what Ruby is, although not intimately, but I have no idea what the rails signifiy in technical terms, although they presumably connote rapidity of some sort. If I were a Ruby programmer, I'd certainly want to know what this Rails thing is all about.
Whether it sounds good enough to make me (as a Ruby outsider) want to learn the language is another question. Frankly if I were to learn Ruby it would probably be more due to the good things I have heard about it from sources like perl6-language, than due to Rails, because whatever other tools you use the quality of the core language is critical to their usefulness, so as a programmer it's the style of the language I want to know about first. (This is part of why I never picked up Java: the GUI libraries it comes with are supposed to be really great, but nobody with similar tastes in languages to mine ever has many attractive things to say about the design of the language itself; Ruby, OTOH, does regularly get positive comments from people whose opinions on language design I respect greatly.) Nonetheless, if I *did* take the time to learn Ruby, and if I found that I liked it as a language, Rails would definitely go on my to-learn list then, *despite* that I don't actually know what it does. Because, you know, it just sounds good. Probably looks good on a resume too.
> Copper works, its easy to work with, and even when it fails, you can still shut
> it off by squeezing the pipe shut at the break with a pair of pliers. It also
> makes a great electrical ground, and is easily recycled. And there's no issue
> with plasticizers leaching into the water you drink.
Yeah, I know all that. I wasn't saying anything against copper pipe. I was just pointing out that a lot of older homes contain other kinds of pipe. Usually the hot water pipes are copper, but frequently some of the cold water pipers are steel (or, in really old homes, lead, and the homeowner often doesn't know it).
> It is not like DDT is the only available pesticide
It's cheap. You get a lot of bang for your buck with DDT. Since African governments tend to want to limit their spending for public services (such as public health, highways, and education) to approximately nothing, DDT tends to be attractive to them. If you want them to use something else, the western world probably has to pick up the bill.
Black tea without sugar is fine, but green tea really needs the sugar as far as I'm concerned and as for coffee... I don't think any amount of sugar could make that stuff taste good.
> Also, if you've ever done any home plumbing, you'd know that even 50-year-old
> copper pipe is in decent shape inside, after decades of attack by chlorine,
> ozone, and good old H2O.
Where do *you* live? Around here, copper pipe only started to come into widespread use in the last fifty years or so; most older homes still have pipe made from other materials (frequently steel, the regular kind, not stainless), and even if your home is new, if you live in a municipality, even a small one, the pipes coming to your house are certainly not copper.
With that said, I prefer tapwater to bottled water, and as for micro-organisms, that's what the trace amounts of chlorine are for.
> You never drank water before did you? There IS taste.
It's not the water itself that has taste, but the minerals in it. Pure water isn't very nice to drink, largely because it's so hypotonic, but it doesn't really taste like anything. Tapwater does, because of the various trace ingredients.
> > Tap, preferably
> See, you don't have taste (or perhaps good tap water)
I prefer most tapwater I've encountered to the bottled waters I've tried. It does vary, though...
> I think for me personally, it's the ammount of calcium that gives a good taste
> (the more the better!)
Calcium, eh?
I know the tastes of iron (which I like), lime (which is not bad too), sulfur (which is overwhelming if there's much of it but can be okay in quite small amounts), chlorine (which is good in small amounts but unpleasant if overdone), and sodium (which I don't like in my water, even in quite small amounts)...
Say, doesn't lime have a lot of calcium in it? There's a lot of lime in the water where I went to college, and I liked the water there okay. (It also had noticeable iron, though, which doubtless helped the taste, as I tend to like iron in my water, preferably enough to make orange stains in the sinks after a few years.)
> In fact I can't really enjoy sweet drinks these days, they just taste too "cheap".
Around here, nothing's cheaper than tapwater, but I like it anyway. I don't mind sweet drinks (sometimes I put sugar in my tea, even quite a bit of sugar (although, sometimes I don't), and I drink Kool-Aid, especially in the summer, and from time to time pop), but I also like water.
> It's a lot harder to make low or sugarfree drinks without gas that are quite good.
Cut up a ginger root and boil it for a few minutes, then throw out the pieces and drink the water. (If it's too strong, dilute to taste with water.) This is good hot or iced, with or without sugar. I'm not sure I'd want it bottled from a factory, though.
> Seriously, I have never understood the thing that IT people have for sucking
> down caffeine all day long
For some people, caffeine is extremely addictive. I'm not sure why some people are so much more affected by this than others, but while some of us (I'm one of the lucky ones, I guess) can drink a lot of caffeine one day and then suddenly switch to water or milk and feel fine, others drink a single Coke and a few hours later get very grouchy if they don't have another.
As for me, I can take caffeine or leave it. It doesn't do much for me or to me, one way or the other. I can drink a quart of strongly-brewed black tea and half an hour later go to bed and sleep normally; I can drink a 32-oz Mt Dew and then not have any caffeine for days, and I don't crave it. Not everyone is so lucky. My sister can't drink anything with caffeine in it after about 2pm, or she won't sleep that night. Problem is, if she has something with caffeine in it in the morning, she wants more at noontime, and more in the afternoon, and more later... and if she lets herself have the amounts of it she wants, she turns into an obsessive taskmaster, working (and driving others, if possible) at breakneck pace, intent on getting six days' worth of stuff done in the next six hours. It's something to behold. Once.
My guess is that the people who suck down caffeine all day long are the ones more like my sister than like me.
> The "I don't like the taste" argument against diet soft drinks is nothing more than
> a bullshit excuse. If it isn't sweet enough for you, that means you're drinking so
> much high-fructose corn syrup that your taste buds are desensitized to sweetness.
It's not the absense of sugar. I drink unsweetened things all the time: black tea, milk (okay, so that has natural sugar, but not high-fructose corn syrup like you were talking about), water (lots of this, preferably tapwater, preferably with enough iron in it to turn the sinks orange after a few years), and so on. (I also sometimes drink black tea _with_ sugar, even quite a bit of sugar. The flavour is different, but I actually like it either way. Green tea I only drink with sugar, though.) It's not the lack of sugar. I like sugar, but I don't mind drinking things without it.
The thing that makes diet softdrinks so impotable is that they taste like artificial sweeteners, i.e., positively incredibly *nasty*. Artificial sweeteners taste outlandishly terrible, on par with the old kind of envelope glue they used to use in the seventies before the nicer, more expensive, gentler-tasting kind of envelope glue was introduced. Aspartame is absolutely the worst, but the others aren't much better. I don't have any problem drinking unsweetened beverages, but I will NOT drink artificially-sweetened ones.
They don't make unsweetened pop, so if I'm going to drink pop, it's going to be the kind with sugar. (That said, I don't drink pop that often. Mostly water, milk, Kool-Aid (yeah, okay, so that's got just about as much sugar as pop), and tea. Pop I drink only occasionally.)
> any reason that isn't explicitly prohibited by law. In most jurisdictions, this
> is limited to race, gender, religion, non-disqualifying handicap, age, and perhaps
> a handful of other characteristics.
There are, in the US at least, a number of other specific things they dursn't fire you for, besides the discrimination-oriented ones. Refusing to falsify records, refusing to violate the law, attempting to unionize, taking time off for certain things as permitted under the FMLA, and so on and so forth.
And that's all just in terms of criminal law. There's also whatever the employee can sue you for under civil law, which is what really scares most employers.
You don't have to use a microkernel design to have userland drivers. Heck, Microsoft has started talking (so far just talking, as near as I can tell, but it's a start) about userland drivers, and you *know* they're not going to a microkernel. Moving drivers out of the kernel doesn't have to mean breaking the rest of the kernel into pieces as well.
> Also, given that some microkernels are only about 3500-6000 lines of code
> (as opposed to Linux's million or so) it's relatively easy to make certain
> that the code is bug free
Microkernels aren't only 3500-6000 lines of code if you count the whole kernel's worth of functionality. At least, not if they're written in a low-level language like C, which generally seems like the way to go for kernels.
I don't think it's necessary to start Linux over from scratch just yet. Sure, it will *eventually* be necessary to replace Linux, as it is with all software sooner or later, but most *nix software these days is sufficiently general that it will run on other kernels besides Linux, so when the time comes that shouldn't be a huge problem. There are, after all, other kernels around.
> Don't forget that this year will bring Linux on the desktop
Linux maybe, but not Debian. The whole "does not change often" thing may work for some kinds of servers, but on the desktop it's a nightmare. Oh, you want to install the security-critical bug-fix update to your web browser? I'm sorry, that requires a recent version of the widget toolkit it uses, which in turn requires recent versions of three-quarters of the stuff on your system, none of which is available for any distro more than about six months old. HTH.HAND.
Debian, whatever else it may be, is not a desktop-oriented distribution.
(And anyway, the days of Linux on the desktop are drawing to a close. We've been there, and we've done that, and it was okay, but now we're coming to the days of BSD on the desktop. Nietzsche is dead already, and NetCraft is up next.)
> Give me a good reason to upgrade [from potato] being this is my web/mail/dns/ftp box.
One thing that springs to mind off the top of my head is that IP Tables (the firewall system in Linux 2.4 and higher) is much nicer to work with than IP Chains (Linux 2.2, IIRC).
Even if you don't update the whole distro, I sure hope you've updated the services that are exposed to the internet from the versions in potato. Especially the mail server. (If you haven't, I'd seriously consider replacing the system altogether and decommissioning it, as it's probably too infested to consider repairable.)
It is reasonable to stay with gcc 2.95 for this type of system. It wouldn't be practical for a desktop, but for a server that does not change often it should be fine.
If you don't want to move to Linux 2.4, you might consider a later release of 2.2. I really do recommend 2.4 though, for IP Tables if nothing else. Very nice kernel, and quite mature at this point. For the web server, stay with Apache 1.x, but get a more recent Apache 1.x than the one that shipped with potato. For the mail server, you don't want anything to do with the one in potato. For dns, it depends on your needs -- do you really need all of bind's features, or could you be using something smaller and safer, along the lines of djbdns or whatnot? For ftp, it depends on whether you need to offer anonymous ftp on port 21. If you do, then you really want to pay attention to security issues for sure, which probably means staying on the latest release most of the time. (Which ftp daemon are you using, anyway?)
Do have a look at IP Tables. Really it's quite a nifty system.
Seriously. You don't need a white-hot performance-optimizing programmer, because nothing you're doing is going to tax the system resources of a Pentium II, even with _fairly_ inefficient code (as long as none of the algorithms it uses are worse than about O(n log n).) What you do need is somebody who understand the importance of robustness, and that's not nearly as hard to hire, or as expensive. Your existing solution is so archaic and so bad, any decent computer geek will be able to put together something you can live with in a matter of a couple of months, and then you'll have various improvement requests over the next year or so after that.
Just make sure, before you let the programmer start writing any code, that you explain to him all the important features of your existing software, how they work, and why you need them.
Vacation? You get vacation? Why, back in my day we were expected to write code 365 days a year. One year they decided to be nice and give us twenty minutes off for Christmas and an extra crust of bread. We might've gotten that every year, but the young whippersnappers had the nerve to *complain*. Ingrates. Spoiled it for all of us.
> Even for plain printed text. This falls squarely in the quality department for me.
It depends where you live. Around here, the only place you _ever_ see a ligature is in a stylized brand name (where its purpose is to distinguish the brand name from the normal English word it's based on), and it is normal even in heavily academic contexts for foreign words and phrases to appear in print sans any diacritical marks that they would have had in their original language: facade, jalapeno, resume, je ne sais quoi, and so on and so forth. (For that matter, Greek, Russian, Hebrew, and Chinese words are almost always written in Latin characters, even in heavily academic contexts, not because people don't know about other writing systems, but because the Latin alphabet is the standard and more convenient one.) Anyway, Word certainly *can* do oddball characters (albeit, it's a bit of a pain if you don't have a special keyboard), but most people _don't_ do them.
Quite aside from that, you're trying to compare the quality of a wine press versus a blender. They can both make juice, sure, but trying to compare their quality as juice-makers would be bizarre. Arguing the quality of MS Word against TeX is a bit like that, or, to use a more technical example, like arguing the quality of Gimp against Adobe Illustrator. Yeah, they both do graphics, but they represent completely different ways of thinking about graphics and as such they're not directly comparable. Gimp should be compared with Photoshop, not Illustrator. (If you want to compare an open-source project with Illustrator, you could reach for Inkscape, although that's not really a mature product yet. (I use it, and I like it, but it's... not done.)) MS Office should be compared with OpenOffice, not with TeX. (Microsoft does not, as far as I am aware, really have a product that attempts to directly compete with TeX, but if they did, and if they bundled it in with Office, it still wouldn't be used for most of the stuff people do with MS Word. (Whether it would gain any significant traction against TeX is another matter (and really beside the point (and yes, I've done a little lisp programming on occasion (why do you ask?))).))
Not yet. At the time of this writing, it's still supposed to include the Aero Glass interface, which allows window titlebars to be transparent (if you have a good video card).
> not to worry, at this rate it WILL be perfectly timed for the holiday
> period... xmas 2007.
Don't hold your breath.
Projected release dates are meaningless until they're *this* quarter. Arguing about whether the date is going to get pushed back from the quarter after next to the quarter after that is just idle speculation. It's preposterously hard to predict release dates with any accuracy that far into the future for a product like Windows from a company like Microsoft. Vista will ship when it's ready. Any announced release dates prior to that are just marketing hype intended to keep people excited.
When they start saying they're going to release in the *current* quarter, then you figure they're actually close to having it ready to ship. Until then, who knows.
> Actually less when you consider Vista will run on 800mhz machines with 512mb
> of RAM quite well.
Define "run". Windows XP, assuming you want to run more than two applications at once, creeps like a paraplegic on tranquilizers if you only give it 512MB of RAM, so I have a hard time imagining Vista will run in an acceptable fashion on that much.
(Not that modern Unix-like systems are terribly happy with only 512MB either. Really I think 1GB of RAM is pretty much a minimum these days irrespective of OS, if you want to run several apps at once. But your claim that Vista will be run in a reasonable fashion with only 512 MB of RAM is not believable.)
OTOH, I would not consider anything up to 2GB of RAM to be "insane specs", so you have a point as far as that is concerned. More than 2GB would be problematic at this time cheifly because far too many motherboards can only handle up to that amount. Good motherboards can take 4GB (and a few take more, but those tend to be for 64-bit systems), but most boards at this point still max out at 2GB.