> Yet the countries with the advanced high-tech military hardware > still fell to the swarming hordes that out-produced them materially
Have you ever played freeciv?
Production is indeed a very big deal. Technology only matters if you're *WAY* imbalanced (like, howitzers attacking pikemen) or if your production capacities are very comparable. Given one side with superior technology and the other side with superior production, the superior production will win every time. (Unless it's controlled by the AI. The AI can't win against an intelligent human player, of course.)
> A lesson the US probably should keep in mind going into the 21st century.
Because there's another country with an economy ten or twenty times larger than ours, like our economy was ten or twenty times larger than Japan's in WWII? Which country would that be?
Obviously, the international balance of power always changes over time, so at some point in the future another nation will be much more powerful than the US. Duh. But going into the 21st century, this event appears to lie beyond the foreseeable future. I would not care to hazard a guess as to which nation it will be, or when.
> "being too late in the war" and "the enemy intelligence > was too good" are not likely to be unrelated facts
The lateness in the war is sure to have significantly exacerbated the intelligence imbalance, I would say.
But, frankly, once Germany started losing, there really isn't any technology that Japan could have developed that would have allowed them to win WWII. Prolong the war, yes. Win, no. The allied forces were backed by too much (economic and industrial) infrastructure, and the US was too geographically vast. With Italy out of the picture and Germany on the retreat, Japan was just plain outmatched.
The war was won (or lost, depending which side you're on) in Normandy.
> Obviously, these would go around the tip of S. America
This is really only obvious if you don't really grasp how *far* that is, through waters controlled by Allied forces almost the whole way. Frankly, it's the *third* most obvious way to go, with the second being under the arctic ice.
Having read the article, I do think the southern route is the one they intended, but it's not really a sane plan, with the submarine technology available at the time. In order to even consider it, you have to reject the central and northern routes as unworkable, and the sheer magnitude and audacity of the southern route is enough to make you go back and consider them again.
I mean, yes, guys like James Cook did junk like that in the age of exploration, but they weren't sailing through waters patrolled by hostile enemies the whole way, so it's really not the same. When you're at war, supply lines *matter*. Japan sending ships around Cape Horn and through the British-controlled south Atlantic to invade the US east coast would be like Napoleon sending troops through hostile Russia to invade China. (Napoleon, of course, wasn't *quite* that dumb. He had no intention of invading China. He was just invading Russia, which is like if Japan had tried to take the Falklands. It still ended badly for Napoleon's army, though.)
> Could they even carry enough fuel to make a trip like that?
Upon further reading (of the article), it appears that one of these subs could indeed carry enough fuel to go that far, and then some. (It was 400 feet long. For a WWII submarine, that's huge.)
My point about surfacing regularly for air while traveling thousands of miles through Allied waters stands, however. Especially that late in the war, when the Japanese navy didn't really even have the run of the Pacific. What made them think they could function in the *Atlantic*, thousands of miles from their supply lines, where Allied forces were significantly more concentrated and Allied supply lines much shorter? Crazy.
> I have doubts about this - with the Panama canal under > Allied control, getting to the east coast USA from > Japan would have been VERY far-ranging.
Indeed.
You can't sneak a sub surreptitiously through a canal lock, even with Jedi mind tricks. It takes *hours* to traverse the canal at Panama, and the entire time you'd be a sitting duck for land-based assault from the US-controlled canal zone.
That leaves going under the arctic ice, which to my knowledge was beyond the available submarine technology of that day even for Germany...
Or you could try to go clear around Cape Horn. What fun *that* would be in a ship of that era. Granted, a west-to-east passage around the horn is easier than going the other way, but we're still talking about thousands of miles, most of it through waters controlled by Allied forces, not least the British navy, and of course you'd be in the US Navy's front yard (where we had the better part of our navy, since we viewed Germany as the greater threat) for the last leg. On the entire trip there would have been *no* place where a Japanese ship could really be safe putting in to port, and furthermore you couldn't count on any Axis surface ships to assist with resupplying. (Germany could've tried to get a surface ship into the mid Atlantic to rendezvous, but it would be significantly trickier to pull off than just sending German subs to raid the US east coast.)
For a WWII-era submarine, the whole idea is just plain loony. Could they even carry enough fuel to make a trip like that? Fresh water? Certainly they'd have to surface dozens of times for air along the way, and risk being sighted by an allied plane every single time.
A late-cold-war nuclear sub could do it, sure. But that's different. Nuclear subs are much faster and can go days without surfacing. Heck, a nuclear sub could make the trip under the arctic ice. But I don't think a WWII-era sub could do that.
> If gnome (and linux in general) wants to escape > the geek-in-a-basement marketshare, it has to focus > on the average non-tech user.
Actually, focusing *exclusively* on the average non-tech user will put you where Apple was in 1995, before they got Jobs back.
There has to be a balance. A good rule of thumb is, the default setup should be geared toward the end user, but it's important to provide more flexible features and configurability for more advanced users.
> And no, pasting a link to a windows share is not what this user does.
Actually, users do (or try to do) an astonishingly wide variety of things, many of which do not make a heck of a lot of sense to a computer geek. Pasting a link to a CIFS fileshare into a web browser and expecting it to show them the contents of the fileshare is barely the tip of the iceberg. They print email that they don't want, even if they're being charged by the page. They type URLs into search boxes. They use the mouse upside down. They use spaces to center text on a line and blank lines to create double spaced copy. They type several pages of text into a six-line textarea on a web page, hit print, and expect to see all of the text they typed on the paper, and they get very upset when it's not there, because they *need* it to be there, to show the court. They set the zoom level in print setup to 300% and get upset when the ends of some of the lines get chopped off. They print white text and get upset if it's not legible. (I tell them the printer is all out of white ink.)
> Instead, this user is interested in finding "that god-damn file" > that he saved somewhere yesterday morning and has no idea where it is.
Yeah, recently used file lists are useful. Welcome to the state of the art of 1991.
> He also wants to easily find that openoffice window that got > lost in the 20 windows he opened and never closed in the last hour.
Actually, in all the considerable time I have spent working with, observing, and supporting end users, I have never yet seen an end user deliberately open more than one window at a time. I've seen them get several browser windows because client-side script links open windows without their knowledge; this, of course, confuses the user, who typically doesn't understand why the back button won't work. But I've never EVER seen an end user open two word processing windows at once.
I'm sure there *are* users who do this. As I said, users do an astonishingly wide variety of things. But opening lots of windows is not at all common among the computer illiterate, I can tell you that for free.
Now, power users open lots of windows. Amazingly large numbers of windows. But power users also like for things to be configurable and are willing (usually eager) to learn how to do new things. That's a whole different category of users.
> So what if their default configuration has just the > one panel? Thats how I configure it anyway.
I used to try to fit everything on one panel, but it made the task list (or window list, or whatever they call it now) too cramped. I tend to have numerous windows open at once...
I ended up settling on a panel across the bottom with very little on it, mostly just the window list, so that it gets pretty much the whole panel and thus has plenty of space. Then I put my launchers and panel apps and stuff on a side panel on the left. There's a screenshot of my setup (taken for a different purpose, but it shows the whole desktop) here: http://mistersanity.blogspot.com/2009/10/screenshot.html
I know, most people don't have that many panel launchers because they use desktop icons. But I don't like that setup because it causes you to constantly have to minimize everything, which gets old fast. And yes, I know you can minimize everything in one fell swoop, but I don't want to do that, because I don't want everything minimized. I want my windows to stay where I put them, thank you very much. I want to be able to glance down and see the bottom line of the xxms2-status readout showing what's currently playing. I want to be able to see that big dclock from across the room. And I want to be able to open a new window or launch a new application without minimizing what I'm currently working on, because I want to be able to alt-tab back to it in one quick motion, without going through other less-recent windows first. So I don't minimize everything, and thus I go *weeks* without seeing my wallpaper.
So I use panel launchers instead of desktop shortcuts. Hence, the second panel. Plus, that gives me a handy place to put panel applets, which is useful. The UIM applet, for instance, has become indispensable as I'm studying a foreign language. In a single-panel setup, I'd end up begrudging it the space, because I'd need the room for more entries on my window list.
YMMV. Configurability is good.
And as much as I complain about the reduced and ever shrinking configurability in Gnome 2.x, it still beats the everliving pants off Windows in that department.
Indeed, but I have a simple solution for that: chmod ugo-x `which nautilus` Sorted.
I use a similarly heavy-handed approach to getting rid of the stupid window manager that comes with Gnome, making it a symlink to sawfish.
Unfortunately it's not so easy to undo the numerous undesirable changes that have been made to the panel over the years (starting with 2.0).
> Just boot XP and clone Windows Explorer, mkay...?
Ugh, no. In the name of all that is sane, no. Even the file manager that came with Windows 3.1 (winfile.exe, I believe it was called) was better than the horrible monstrosity that is Windows Explorer.
Although, in fairness, I'm not really part of the core target market for graphical file managers. I've been doing file management mostly from the command line since PC-DOS 3.3, and once I discovered tab completion in the nineties that was pretty much all she wrote for graphical file management.
Still, Windows Explorer? Nautilus, granted, isn't really better, but comparing a file manager to Nautilus is kind of like comparing a mailreader to Outlook Express. Something could be a whole lot better than that and still be completely wretched, so the comparison isn't really useful.
> Maybe they're trying to innovate and do something new and different.
Innovation would be okay if we could turn off the "innovations" we dislike. But the general pattern with Gnome (starting from version 2.0) is that such changes, especially the most undesirable ones, are usually mandatory, i.e., it is impossible to configure things back to the way we want them, impossible to get back basic functionality that we had in version 1.4.
> Microsoft has now started to do the same thing, with the > inability to turn off the $#@! "improvements" and get back > the normal "classic" behavior in Seven like you could do in > previous versions. Thus, they have started down the Gnome path.
Come to think of it, the Firefox team has now started down this path as well. So I guess it's an emerging trend now, and the Gnome people were on the leading edge of it seven years ago. What I want to know is, who's going to lead the way in swinging the pendulum back the other direction?
> Gnome3 looks unusable anyways, delay it forever. > Go through the early tour and tell me that is more > usable. I've no idea wtf they were thinking.
"How can we make this more *different* from what long-time users want and expect, so that they have a harder time configuring it back toward normal behavior and an even harder time adjusting to the idiosyncrasies that can't be configured away?"
This is what the Gnome developers have been thinking ever since they started work on version 2.0. It has become their modus operandi. (Microsoft has now started to do the same thing, with the inability to turn off the $#@! "improvements" and get back the normal "classic" behavior in Seven like you could do in previous versions. Thus, they have started down the Gnome path.)
You want to see a useful Gnome desktop environment, that can be easily configured to behave in the desirable fashion? Use version 1.4 sometime. The downside is, it's basically impossible to use modern applications with it, because of library version dependency conflicts.
> Um... taking time doesn't necessarily mean it gets done right.
Indeed. To date, the best Gnome release was version 1.4, which came out just eleven months after the previous release and was a significant improvement in a number of ways. Gnome 2, in contrast, seems to actively get worse with each passing release. Well, except for gnome-terminal. gnome-terminal is actually better. Everything else is worse.
Where I come from, the police have a much more positive public image than Visa. Not as positive as the fire department or the public library, but pretty positive.
> A lot of the criticism of comics and comic books come from people > who think that kids are just looking at the pictures and not putting > them together with the words,' says Tilley. 'But you could easily > make some of the same criticisms of picture books - that kids are > just looking at pictures, and not at the words.'
Umm, kids read picture books up through about Kindergarten, and yeah, they *are* basically just looking at the pictures and, hopefully, listening -- mom and dad are supposed to be reading to them at that point, obviously.
Kids who are old enough to actually *read*, however, tend to read books that are mostly or entirely text. Charlotte's Web, Adventures of Tom Sawyer, The Hobbit, Voyage of the Dawn Treader, that sort of thing. You know, actual books. Did they compare the comic books to those?
Sure, in most jurisdictions it's illegal for a dead man to marry, but maybe the laws in Scotland are different. I mean, the men over there wear dresses, so you know they're a little weird, right?
> God is a libertarian that wants people to voluntarily be socialists:)
Not exactly. Socialists are generous with other people's stuff. God would prefer you be generous with your own stuff (and, really, even that is a fairly minor point).
> Meanwhile, it's much easier to shout about the > sins of homosexuals, since, if they exist at all, > they exist as a minority in the congregation.
You know, I hear this a lot, but I've never yet found a church that actually spends more time talking about homosexuality than about heterosexual adultery and other sins. Quite the contrary, churches that talk about sin at all (which, granted, leaves out a fair few these days) usually focus on the ones that people seem to have a problem with, for the most part.
If you think churches rant about homosexuality all the time and ignore every other sin, you obviously watch too much television. Try turning the TV off and actually going to church sometime.
> The funny part is a lot of people argue, strongly, that > self-signed certs aren't any less secure than CA-signed certs.
It's not significantly harder for the bad guys to get a CA-signed cert than it is for the good guys. Ipso facto, CA signing does not add any significant security. It only *seems* to do so, if you ignore the question of who can get a certificate.
What does add significant security is if the client software flashes big red glowing alarm bells if the certificate changes compared to last time. OpenSSH does this. Mail clients don't and definitely should. Web browsers don't and perhaps should (although there would need to be some provision for legitimately transitioning to a new certificate without setting off said alarms, which in order to be secure would require nontrivial design changes to the way https certs are handled on the server as well as the client).
The lesson here is that application of encryption often results in an overall system that is much weaker than you might think based on the strength of the encryption itself. When you are looking at designing a secure system, selecting strong encryption can be useful, but only if you apply it in a secure way. It is my considered opinion that the use of SSL on the web adds very little security -- not due to any inherent flaw in the design of SSL, but due to the way in which https uses it.
Attacks based on using the MAC (or anything else related to the data link layer) require the attacker to already be on the local network segment. An upstream middle-man attacker can't do that stuff, because routers (well, sane routers anyway) don't forward anything below the network layer.
That doesn't mean such attacks have no relevance at all. The principle of defense in depth dictates that you *do* want to defend, as much as possible, against attacks originating on the LAN. But these attacks are not as prevalent or dangerous as ones that can be perpetrated from a remote or upstream system, such as the MITM vulnerability discussed in the article.
Trying to make strong AI in the image of a human is an extreme case of getting the cart before the horse, like if Henry Ford had tried to design a transdimensional intergalactic spacecraft instead of just a touring car. When AI researchers can create a computer system with greater overall intelligence than a mosquito (not just better in one small area like vision, but overall better across the board), then maybe they can start thinking about going for the intelligence level of smarter insects and possibly crustaceans...
> Rather, the expectations that we have about someone talking > strongly influence what sounds we perceive in the first place. > Hardly any of these [properties] have been duplicated in algorithms > yet, and it's not even certain that that would be a good idea.
It's not clear to me that whether it's a good idea will ever even be an important question to ask, because I don't think anyone has any idea *how* it could be done. Even if AI researchers wanted to try that approach, they wouldn't know how to even get started. Anyone who says otherwise is going to give you yet more approaches that are NOTHING like the way humans do it. For instance, they might propose keeping an enormous database of which words and phrases are statistically most likely to "come next" after a certain thing is said. But humans don't do that. If they expect the speaker to say a certain thing, it's not because it statistically most often comes after what they just said, or any other straightforwardly computable reason, but rather because the listener understands the overall flow of the speaker's line of thinking.
That doesn't mean computers can't do a lot of useful stuff with their capabilities. On the contrary, if computers processed information the way humans do, they'd be a good deal *less* useful. We don't need machines that think like people. We already have people. But computers can do things that humans are remarkably *bad* at, like searching through huge amounts of information quickly. They don't *duplicate* our abilities, they *complement* them. That's much more useful.
> Only idiots, marketers, businessmen and outsiders ever thought we would > be completely replaced by artificially intelligent machines.
You're understating the primitiveness of existing AI. Completely replacing humans? That's *hundreds* of orders of magnitude beyond what AI researchers are even *trying* to do at this point.
Last I checked we are approximately 0% of the way to being able to create a computer system with even arthropod-level "intelligence". The best minds in AI are still working working on *individual* basic problems, like vision. They believe they're making progress on these individual problems, but the idea of putting it all together and creating anything with a total intelligence approaching that of a cockroach is, for the time being at least, science fiction.
People think AI is making progress because computers can be programmed to do a lot of things that *seem* smart, but that's all computational tricks (like exhausting all the possible future moves in a game of chess and picking the branch that has the highest percentage of positive outcomes) and pre-programmed details (like an endgame library). It's the programmers who are learning, not the software. The software is as dumb as a box of rocks.
In fact, I will lay it on the line: software is no more intelligent today than it was in the sixties. Once we moved past if/then and conditional jump constructs to abstract high-level data structures containing both code and data references (e.g., lookup tables with callbacks in them), no significant additional progress has been made beyond that. We're stuck, and we don't know how to do any better. We can add more computational power and other resources, but all that does is speed things up. The results aren't qualitatively better, though we certainly can get them faster.
> lets not try to claim that ANY of their releases are > anything other than bleeding edge beta quality releases
Ubuntu is what it is because of the circumstances of its birth.
At the time, Debian stable had not been meaningfully updated in, approximately, forever. The cool kids were trying out Linux 2.6, and meanwhile Debian stable offered you the choice of the "new and experimental" Linux 2.2, or the tried and true Linux 2.0. What? Linux 2.4? We can't put that in stable, it's only six years old!
Ubuntu, or at least a large part of its popularity, was born out of frustration with this situation. The official Debian line at the time was that "stable" means "doesn't change often", but people were starting to think a new version would not come out *ever*. A lot of people started playing around with testing and/or unstable, but those are really a bit *too* bleeding-edge for most purposes.
Something intermediate was needed, something safer and saner than running off the testing repository (an actual *release*, in other words), but built out of software released in the current decade. Warty was built out of that would eventually become Sarge, but it was built as an actual *release*. This was sorely needed at the time, so it instantly became popular, and the rest is history.
So if you think Ubuntu is less stable than Debian stable,... that's kind of the point.
Of course, Debian releases have been coming out a little more frequently since sarge. Etch for instance came out practically overnight, by Debian standards. So the disadvantages of using Debian stable are somewhat less now. But Ubuntu remains an intermediate distribution, more current than Debian stable but an actual stable release unlike Debian testing. That's its niche. That's its role. And the next time a Debian release takes half as long to come out as sarge did, I'll be very glad Ubuntu is around as an option.
Slashdot Mirror
> Yet the countries with the advanced high-tech military hardware
> still fell to the swarming hordes that out-produced them materially
Have you ever played freeciv?
Production is indeed a very big deal. Technology only matters if you're *WAY* imbalanced (like, howitzers attacking pikemen) or if your production capacities are very comparable. Given one side with superior technology and the other side with superior production, the superior production will win every time. (Unless it's controlled by the AI. The AI can't win against an intelligent human player, of course.)
> A lesson the US probably should keep in mind going into the 21st century.
Because there's another country with an economy ten or twenty times larger than ours, like our economy was ten or twenty times larger than Japan's in WWII? Which country would that be?
Obviously, the international balance of power always changes over time, so at some point in the future another nation will be much more powerful than the US. Duh. But going into the 21st century, this event appears to lie beyond the foreseeable future. I would not care to hazard a guess as to which nation it will be, or when.
> "being too late in the war" and "the enemy intelligence
> was too good" are not likely to be unrelated facts
The lateness in the war is sure to have significantly exacerbated the intelligence imbalance, I would say.
But, frankly, once Germany started losing, there really isn't any technology that Japan could have developed that would have allowed them to win WWII. Prolong the war, yes. Win, no. The allied forces were backed by too much (economic and industrial) infrastructure, and the US was too geographically vast. With Italy out of the picture and Germany on the retreat, Japan was just plain outmatched.
The war was won (or lost, depending which side you're on) in Normandy.
> Obviously, these would go around the tip of S. America
This is really only obvious if you don't really grasp how *far* that is, through waters controlled by Allied forces almost the whole way. Frankly, it's the *third* most obvious way to go, with the second being under the arctic ice.
Having read the article, I do think the southern route is the one they intended, but it's not really a sane plan, with the submarine technology available at the time. In order to even consider it, you have to reject the central and northern routes as unworkable, and the sheer magnitude and audacity of the southern route is enough to make you go back and consider them again.
I mean, yes, guys like James Cook did junk like that in the age of exploration, but they weren't sailing through waters patrolled by hostile enemies the whole way, so it's really not the same. When you're at war, supply lines *matter*. Japan sending ships around Cape Horn and through the British-controlled south Atlantic to invade the US east coast would be like Napoleon sending troops through hostile Russia to invade China. (Napoleon, of course, wasn't *quite* that dumb. He had no intention of invading China. He was just invading Russia, which is like if Japan had tried to take the Falklands. It still ended badly for Napoleon's army, though.)
> Could they even carry enough fuel to make a trip like that?
Upon further reading (of the article), it appears that one of these subs could indeed carry enough fuel to go that far, and then some. (It was 400 feet long. For a WWII submarine, that's huge.)
My point about surfacing regularly for air while traveling thousands of miles through Allied waters stands, however. Especially that late in the war, when the Japanese navy didn't really even have the run of the Pacific. What made them think they could function in the *Atlantic*, thousands of miles from their supply lines, where Allied forces were significantly more concentrated and Allied supply lines much shorter? Crazy.
> I have doubts about this - with the Panama canal under
> Allied control, getting to the east coast USA from
> Japan would have been VERY far-ranging.
Indeed.
You can't sneak a sub surreptitiously through a canal lock, even with Jedi mind tricks. It takes *hours* to traverse the canal at Panama, and the entire time you'd be a sitting duck for land-based assault from the US-controlled canal zone.
That leaves going under the arctic ice, which to my knowledge was beyond the available submarine technology of that day even for Germany...
Or you could try to go clear around Cape Horn. What fun *that* would be in a ship of that era. Granted, a west-to-east passage around the horn is easier than going the other way, but we're still talking about thousands of miles, most of it through waters controlled by Allied forces, not least the British navy, and of course you'd be in the US Navy's front yard (where we had the better part of our navy, since we viewed Germany as the greater threat) for the last leg. On the entire trip there would have been *no* place where a Japanese ship could really be safe putting in to port, and furthermore you couldn't count on any Axis surface ships to assist with resupplying. (Germany could've tried to get a surface ship into the mid Atlantic to rendezvous, but it would be significantly trickier to pull off than just sending German subs to raid the US east coast.)
For a WWII-era submarine, the whole idea is just plain loony. Could they even carry enough fuel to make a trip like that? Fresh water? Certainly they'd have to surface dozens of times for air along the way, and risk being sighted by an allied plane every single time.
A late-cold-war nuclear sub could do it, sure. But that's different. Nuclear subs are much faster and can go days without surfacing. Heck, a nuclear sub could make the trip under the arctic ice. But I don't think a WWII-era sub could do that.
> If gnome (and linux in general) wants to escape
> the geek-in-a-basement marketshare, it has to focus
> on the average non-tech user.
Actually, focusing *exclusively* on the average non-tech user will put you where Apple was in 1995, before they got Jobs back.
There has to be a balance. A good rule of thumb is, the default setup should be geared toward the end user, but it's important to provide more flexible features and configurability for more advanced users.
> And no, pasting a link to a windows share is not what this user does.
Actually, users do (or try to do) an astonishingly wide variety of things, many of which do not make a heck of a lot of sense to a computer geek. Pasting a link to a CIFS fileshare into a web browser and expecting it to show them the contents of the fileshare is barely the tip of the iceberg. They print email that they don't want, even if they're being charged by the page. They type URLs into search boxes. They use the mouse upside down. They use spaces to center text on a line and blank lines to create double spaced copy. They type several pages of text into a six-line textarea on a web page, hit print, and expect to see all of the text they typed on the paper, and they get very upset when it's not there, because they *need* it to be there, to show the court. They set the zoom level in print setup to 300% and get upset when the ends of some of the lines get chopped off. They print white text and get upset if it's not legible. (I tell them the printer is all out of white ink.)
> Instead, this user is interested in finding "that god-damn file"
> that he saved somewhere yesterday morning and has no idea where it is.
Yeah, recently used file lists are useful. Welcome to the state of the art of 1991.
> He also wants to easily find that openoffice window that got
> lost in the 20 windows he opened and never closed in the last hour.
Actually, in all the considerable time I have spent working with, observing, and supporting end users, I have never yet seen an end user deliberately open more than one window at a time. I've seen them get several browser windows because client-side script links open windows without their knowledge; this, of course, confuses the user, who typically doesn't understand why the back button won't work. But I've never EVER seen an end user open two word processing windows at once.
I'm sure there *are* users who do this. As I said, users do an astonishingly wide variety of things. But opening lots of windows is not at all common among the computer illiterate, I can tell you that for free.
Now, power users open lots of windows. Amazingly large numbers of windows. But power users also like for things to be configurable and are willing (usually eager) to learn how to do new things. That's a whole different category of users.
> So what if their default configuration has just the
> one panel? Thats how I configure it anyway.
I used to try to fit everything on one panel, but it made the task list (or window list, or whatever they call it now) too cramped. I tend to have numerous windows open at once...
I ended up settling on a panel across the bottom with very little on it, mostly just the window list, so that it gets pretty much the whole panel and thus has plenty of space. Then I put my launchers and panel apps and stuff on a side panel on the left. There's a screenshot of my setup (taken for a different purpose, but it shows the whole desktop) here:
http://mistersanity.blogspot.com/2009/10/screenshot.html
I know, most people don't have that many panel launchers because they use desktop icons. But I don't like that setup because it causes you to constantly have to minimize everything, which gets old fast. And yes, I know you can minimize everything in one fell swoop, but I don't want to do that, because I don't want everything minimized. I want my windows to stay where I put them, thank you very much. I want to be able to glance down and see the bottom line of the xxms2-status readout showing what's currently playing. I want to be able to see that big dclock from across the room. And I want to be able to open a new window or launch a new application without minimizing what I'm currently working on, because I want to be able to alt-tab back to it in one quick motion, without going through other less-recent windows first. So I don't minimize everything, and thus I go *weeks* without seeing my wallpaper.
So I use panel launchers instead of desktop shortcuts. Hence, the second panel. Plus, that gives me a handy place to put panel applets, which is useful. The UIM applet, for instance, has become indispensable as I'm studying a foreign language. In a single-panel setup, I'd end up begrudging it the space, because I'd need the room for more entries on my window list.
YMMV. Configurability is good.
And as much as I complain about the reduced and ever shrinking configurability in Gnome 2.x, it still beats the everliving pants off Windows in that department.
> Gnome is still plagued by Nautilus
Indeed, but I have a simple solution for that:
chmod ugo-x `which nautilus`
Sorted.
I use a similarly heavy-handed approach to getting rid of the stupid window manager that comes with Gnome, making it a symlink to sawfish.
Unfortunately it's not so easy to undo the numerous undesirable changes that have been made to the panel over the years (starting with 2.0).
> Just boot XP and clone Windows Explorer, mkay...?
Ugh, no. In the name of all that is sane, no. Even the file manager that came with Windows 3.1 (winfile.exe, I believe it was called) was better than the horrible monstrosity that is Windows Explorer.
Although, in fairness, I'm not really part of the core target market for graphical file managers. I've been doing file management mostly from the command line since PC-DOS 3.3, and once I discovered tab completion in the nineties that was pretty much all she wrote for graphical file management.
Still, Windows Explorer? Nautilus, granted, isn't really better, but comparing a file manager to Nautilus is kind of like comparing a mailreader to Outlook Express. Something could be a whole lot better than that and still be completely wretched, so the comparison isn't really useful.
> Maybe they're trying to innovate and do something new and different.
Innovation would be okay if we could turn off the "innovations" we dislike. But the general pattern with Gnome (starting from version 2.0) is that such changes, especially the most undesirable ones, are usually mandatory, i.e., it is impossible to configure things back to the way we want them, impossible to get back basic functionality that we had in version 1.4.
> Microsoft has now started to do the same thing, with the
> inability to turn off the $#@! "improvements" and get back
> the normal "classic" behavior in Seven like you could do in
> previous versions. Thus, they have started down the Gnome path.
Come to think of it, the Firefox team has now started down this path as well. So I guess it's an emerging trend now, and the Gnome people were on the leading edge of it seven years ago. What I want to know is, who's going to lead the way in swinging the pendulum back the other direction?
> Gnome3 looks unusable anyways, delay it forever.
> Go through the early tour and tell me that is more
> usable. I've no idea wtf they were thinking.
"How can we make this more *different* from what long-time users want and expect, so that they have a harder time configuring it back toward normal behavior and an even harder time adjusting to the idiosyncrasies that can't be configured away?"
This is what the Gnome developers have been thinking ever since they started work on version 2.0. It has become their modus operandi. (Microsoft has now started to do the same thing, with the inability to turn off the $#@! "improvements" and get back the normal "classic" behavior in Seven like you could do in previous versions. Thus, they have started down the Gnome path.)
You want to see a useful Gnome desktop environment, that can be easily configured to behave in the desirable fashion? Use version 1.4 sometime. The downside is, it's basically impossible to use modern applications with it, because of library version dependency conflicts.
> Um... taking time doesn't necessarily mean it gets done right.
Indeed. To date, the best Gnome release was version 1.4, which came out just eleven months after the previous release and was a significant improvement in a number of ways. Gnome 2, in contrast, seems to actively get worse with each passing release. Well, except for gnome-terminal. gnome-terminal is actually better. Everything else is worse.
Where I come from, the police have a much more positive public image than Visa. Not as positive as the fire department or the public library, but pretty positive.
> Which browsers actually do this? Is Mozilla actually participating in that nonsense?
Yeah, I was wondering that as well. Personally, I would not willingly use a browser that I believe does this.
> A lot of the criticism of comics and comic books come from people
> who think that kids are just looking at the pictures and not putting
> them together with the words,' says Tilley. 'But you could easily
> make some of the same criticisms of picture books - that kids are
> just looking at pictures, and not at the words.'
Umm, kids read picture books up through about Kindergarten, and yeah, they *are* basically just looking at the pictures and, hopefully, listening -- mom and dad are supposed to be reading to them at that point, obviously.
Kids who are old enough to actually *read*, however, tend to read books that are mostly or entirely text. Charlotte's Web, Adventures of Tom Sawyer, The Hobbit, Voyage of the Dawn Treader, that sort of thing. You know, actual books. Did they compare the comic books to those?
Sure, in most jurisdictions it's illegal for a dead man to marry, but maybe the laws in Scotland are different. I mean, the men over there wear dresses, so you know they're a little weird, right?
> God is a libertarian that wants people to voluntarily be socialists :)
Not exactly. Socialists are generous with other people's stuff. God would prefer you be generous with your own stuff (and, really, even that is a fairly minor point).
> Meanwhile, it's much easier to shout about the
> sins of homosexuals, since, if they exist at all,
> they exist as a minority in the congregation.
You know, I hear this a lot, but I've never yet found a church that actually spends more time talking about homosexuality than about heterosexual adultery and other sins. Quite the contrary, churches that talk about sin at all (which, granted, leaves out a fair few these days) usually focus on the ones that people seem to have a problem with, for the most part.
If you think churches rant about homosexuality all the time and ignore every other sin, you obviously watch too much television. Try turning the TV off and actually going to church sometime.
> The funny part is a lot of people argue, strongly, that
> self-signed certs aren't any less secure than CA-signed certs.
It's not significantly harder for the bad guys to get a CA-signed cert than it is for the good guys. Ipso facto, CA signing does not add any significant security. It only *seems* to do so, if you ignore the question of who can get a certificate.
What does add significant security is if the client software flashes big red glowing alarm bells if the certificate changes compared to last time. OpenSSH does this. Mail clients don't and definitely should. Web browsers don't and perhaps should (although there would need to be some provision for legitimately transitioning to a new certificate without setting off said alarms, which in order to be secure would require nontrivial design changes to the way https certs are handled on the server as well as the client).
The lesson here is that application of encryption often results in an overall system that is much weaker than you might think based on the strength of the encryption itself. When you are looking at designing a secure system, selecting strong encryption can be useful, but only if you apply it in a secure way. It is my considered opinion that the use of SSL on the web adds very little security -- not due to any inherent flaw in the design of SSL, but due to the way in which https uses it.
Attacks based on using the MAC (or anything else related to the data link layer) require the attacker to already be on the local network segment. An upstream middle-man attacker can't do that stuff, because routers (well, sane routers anyway) don't forward anything below the network layer.
That doesn't mean such attacks have no relevance at all. The principle of defense in depth dictates that you *do* want to defend, as much as possible, against attacks originating on the LAN. But these attacks are not as prevalent or dangerous as ones that can be perpetrated from a remote or upstream system, such as the MITM vulnerability discussed in the article.
Trying to make strong AI in the image of a human is an extreme case of getting the cart before the horse, like if Henry Ford had tried to design a transdimensional intergalactic spacecraft instead of just a touring car. When AI researchers can create a computer system with greater overall intelligence than a mosquito (not just better in one small area like vision, but overall better across the board), then maybe they can start thinking about going for the intelligence level of smarter insects and possibly crustaceans...
> Rather, the expectations that we have about someone talking
> strongly influence what sounds we perceive in the first place.
> Hardly any of these [properties] have been duplicated in algorithms
> yet, and it's not even certain that that would be a good idea.
It's not clear to me that whether it's a good idea will ever even be an important question to ask, because I don't think anyone has any idea *how* it could be done. Even if AI researchers wanted to try that approach, they wouldn't know how to even get started. Anyone who says otherwise is going to give you yet more approaches that are NOTHING like the way humans do it. For instance, they might propose keeping an enormous database of which words and phrases are statistically most likely to "come next" after a certain thing is said. But humans don't do that. If they expect the speaker to say a certain thing, it's not because it statistically most often comes after what they just said, or any other straightforwardly computable reason, but rather because the listener understands the overall flow of the speaker's line of thinking.
That doesn't mean computers can't do a lot of useful stuff with their capabilities. On the contrary, if computers processed information the way humans do, they'd be a good deal *less* useful. We don't need machines that think like people. We already have people. But computers can do things that humans are remarkably *bad* at, like searching through huge amounts of information quickly. They don't *duplicate* our abilities, they *complement* them. That's much more useful.
> Only idiots, marketers, businessmen and outsiders ever thought we would
> be completely replaced by artificially intelligent machines.
You're understating the primitiveness of existing AI. Completely replacing humans? That's *hundreds* of orders of magnitude beyond what AI researchers are even *trying* to do at this point.
Last I checked we are approximately 0% of the way to being able to create a computer system with even arthropod-level "intelligence". The best minds in AI are still working working on *individual* basic problems, like vision. They believe they're making progress on these individual problems, but the idea of putting it all together and creating anything with a total intelligence approaching that of a cockroach is, for the time being at least, science fiction.
People think AI is making progress because computers can be programmed to do a lot of things that *seem* smart, but that's all computational tricks (like exhausting all the possible future moves in a game of chess and picking the branch that has the highest percentage of positive outcomes) and pre-programmed details (like an endgame library). It's the programmers who are learning, not the software. The software is as dumb as a box of rocks.
In fact, I will lay it on the line: software is no more intelligent today than it was in the sixties. Once we moved past if/then and conditional jump constructs to abstract high-level data structures containing both code and data references (e.g., lookup tables with callbacks in them), no significant additional progress has been made beyond that. We're stuck, and we don't know how to do any better. We can add more computational power and other resources, but all that does is speed things up. The results aren't qualitatively better, though we certainly can get them faster.
> lets not try to claim that ANY of their releases are
... that's kind of the point.
> anything other than bleeding edge beta quality releases
Ubuntu is what it is because of the circumstances of its birth.
At the time, Debian stable had not been meaningfully updated in, approximately, forever. The cool kids were trying out Linux 2.6, and meanwhile Debian stable offered you the choice of the "new and experimental" Linux 2.2, or the tried and true Linux 2.0. What? Linux 2.4? We can't put that in stable, it's only six years old!
Ubuntu, or at least a large part of its popularity, was born out of frustration with this situation. The official Debian line at the time was that "stable" means "doesn't change often", but people were starting to think a new version would not come out *ever*. A lot of people started playing around with testing and/or unstable, but those are really a bit *too* bleeding-edge for most purposes.
Something intermediate was needed, something safer and saner than running off the testing repository (an actual *release*, in other words), but built out of software released in the current decade. Warty was built out of that would eventually become Sarge, but it was built as an actual *release*. This was sorely needed at the time, so it instantly became popular, and the rest is history.
So if you think Ubuntu is less stable than Debian stable,
Of course, Debian releases have been coming out a little more frequently since sarge. Etch for instance came out practically overnight, by Debian standards. So the disadvantages of using Debian stable are somewhat less now. But Ubuntu remains an intermediate distribution, more current than Debian stable but an actual stable release unlike Debian testing. That's its niche. That's its role. And the next time a Debian release takes half as long to come out as sarge did, I'll be very glad Ubuntu is around as an option.
Which part of "North Dakota" didn't you understand?