You think those are major features, I think that's unimportant entry level stuff.
The number one feature for me in choosing Gaim was a drop-in no-brainpower-required encryption solution. This feature is absent from all of the officially supported and third party IM clients I tried. (I also appreciate Gaim's cross-platform nature now that I've switched to a Linux desktop.)
You call the previously posted feature list "unimportant entry level stuff," and I partly agree: built-in client-based easy-to-use encryption is a basic feature that any IM client that purports to be a good IM client should support. What other clients offer this feature? (I'm seriously interested; I'm not opposed to switching if I can find something that works better than Gaim. Support for Yahoo videoconferencing on Linux would be a big plus.)
Re:gaim works for me, but loses ground from here
on
Linux Instant Messengers
·
· Score: 2, Interesting
What about the rest of you out there in the working world, do you IM that much?
My employer started using IM pretty regularly at the beginning of the year. They bought everyone webcams, and we use Yahoo Messenger to videoconference during meetings (some participants are geographically remote.) So whenever I have a meeting, I need to boot up the windows laptop (Gaim on Linux doesn't support Yahoo videoconferencing yet.) They seem unconcerned about the idea of sending unencrypted trade secrets out into the wild woolly internet, but such is life. Maybe Yahoo doesn't data mine conversations.
I IM with personal accounts periodically, but I generally don't log on during the workday (email's distracting enough) and I often forget to log in at home.
It's just like every other "realistic based-on-fact" show: it's close enough to look cool to the general public, but it makes experts laugh or cry. This is true across many domains; ask a cop what they think about the typical cop show, or a lawyer about the typical courtroom drama show, or a survivalist about "Survivor".
It does, sort of. My understanding is that there are algorithms for adding information to an image that guess based on what nearby pixels look like. Depending on the accuracy of the guess, the result might or might not resemble the original. If *I* had to guess, I'd say that the accuracy of the refinement drops with each successive iteration over the image, because on each pass, more of your input data is based on guesses that might be wrong.
Maybe if we're lucky a math geek will happen by and explain it all to us.
Note that the cost comparison isn't quite that cut and dried, because lots of people already have computers (and sometimes TV capture cards too.) If you're setting up an HD PVR, a computer based solution can be quite cost competitive, especially if your only HD-capable display is your computer monitor. (Whether you're able and inclined to deal with the other issues is another question.)
Inherently, a language with access to pointers is always going to be less secure than a language that does not have them, and uses automatic memory management. None of the security enhancements to the compiler can change code that assigns a pointer to the wrong object and overwrites the memory for it. None of them can prevent a bad cast from screwing things up.
I see this error a lot, and it's slightly annoying. You're treating pointers as if they all behave like C-style pointers, which isn't correct.
Java does have pointers. It doesn't let you orphan memory in the same way that C does, and it restricts the operations that you can perform on pointers, and it doesn't even call them pointers, but they are, in fact, pointers (variables that store a memory location.)
Furthermore, it is possible for a language with pointers to prevent a bad cast from screwing things up: don't allow casting! Or limit it to non-bad cases (which is a pretty big limitation.)
I don't know if they said the same thing when Netflix came out, but if they did, they'd be right. Not only are "borderline" movies cheaper to see, but when I'm finished, they don't embarrass me by sitting around on my DVD shelf.
South Vietnam was the side that the US was allied with
No shit. Of the two sides though, the successive South Vietnamese regimes far surpassed the communists in brutality and attrocities.
Regardless of the truth of this, it has nothing to do with whether or not the US abandoned South Vietnam, and it doesn't change the fact that the communists were an oppressive regime. The US abandoned the South Vietnamese, its supposed allies, to be conquered by an oppressive regime. Those facts are not in dispute.
A private monopolist will often endeavor to charge the highest price that the market will bear, while a public monopolist theoretically will not. (That's a big part of why private monopolies are a bad idea; really almost all monopolies are a bad idea, but in the case of utilities it often can't be helped.) Measuring efficiency in private vs public monopolies requires a much closer examination than simply comparing price.
What's the very worst that can happen if the Internet goes down?
Somebody somewhere panics and shoots off a bunch of nuclear missiles. Billions die, but I survive.
Now it's a race to see if I die from radiation or starvation. Good thing I have all this extra body fat. Once that's gone, I'll have the corpses of my family and my pets to eat.
Dang, radiation sickness really sucks. The other survivors all have it too, so I'm able to successfully fight them off for my share of the remnants of society.
Finally, after six months, I am too weak to defend myself, and am devoured alive by rodents.
I wake up. What? It's October 10th again? Look, it's just like the movie Groundhog Day, but lasting six months. Six terrible, terrible months. Forever.
That first one would truly be wonderful. I long for the day when government becomes less wasteful than, well, anything.
I'm not so sure about the second one. One of the goals of such a system would surely be to help those who slip through the cracks of society, and such people are often unable to work.
Perhaps care would be limited by economic collapse: treat everybody, until the government runs out of money (i.e. can't tax more or borrow more.)
Or perhaps care would be limited by civil war: treat everybody, until government mismanagement causes the citizenry to storm Congress, pitchforks and torches in hand.
Or perhaps care would be limited by race/culture: those who are or are descended from persecuted minorities get first dibs. (As a descendent of French Canadians, I'm all for this.) Note that this would only work until everybody intermarries or suddenly discovers non-white non-Europeans in their family history.
Or perhaps care would be limited by political persuasion: members of "the party" (whichever party or coalition is currently in power) get first dibs (while members of "the other party", if it exists, are stuck at the end of the line.) Rank within the party would need to be taken into account once everybody tries to join the party.
I've never heard of a software system that even approaches the complexity of a significant bridge. Never.
Forgive me if I find your post unconvincing, but your description of a bridge just doesn't sound that complicated to me. Do you know of any objective way to measure the complexity of problems in different domains, by which they could be compared? Comparing the complexity of a bridge (or house, or skyscraper, or airplane, or whatever) to the complexity of any given computer program does sound like an interesting pursuit. Maybe decision points, or variables... I don't know if there's an equivalent to a logical branch in bridge design, but surely there are equivalents to things like coupling and cohesion. Like, how many other widgits does this widgit touch? Through how many layers will a failure in this widgit cascade?
My intuition tells me that the complexity per developer of a piece of software is much higher than the complexity per... engineer? of a bridge. Computers make it very easy to program past your level of competence. But, I'd welcome a way to objectively verify my intuition.
I'm afraid I don't really share your faith in proofs of correctness for large systems. Apart from the problems scaling up these approaches they assume that you can easily mathematically describe how the thing is supposed to behave.
I shared your skepticism, but apparently nobody informed the formal methods community that their goals are impossible, and they've come up with a number of different approaches for proving the "correctness" (really, the logical consistency) of a program's design. This is a huge step forward from current industry practice.
It's true that mathematical depictions of a program's design and behavior don't eliminate bugs. But they can potentially eliminate certain kinds of bugs that can be difficult or time-consuming to fix after the fact, and if coupled with code generation tools, the overhead of formally describing a program's structure can be somewhat mitigated.
That being said, I would be interested in seeing studies that back up some of these claims with actual numbers.
I like your i18n example. Being able to react well to that kind of change is a good thing for a design methodology to be able to do.
That last phase, assembly, is where the stuff that is not fun enough for volunteers to typically do gets done. If a particular instance of Desktop Linux is not up to par, the assembly folks are the ones who are to blame. One success story in this area is Mac OS X, where the assembly folks took a bunch of open source programs, assembled them and some other programs into a cohesive whole and created a desktop that successfully meets the needs of most users.
However, the key need of 99.9999% of users that is met by Windows, is that it just works out of the box. Maybe not optimally and definitely not securely, but it does simply work.
Desktop Linux meets that standard today. It's not a very high standard.
I hereby submit that at least the GPL is part of the reason, as it "prevents" certain/many hardware companies from providing drivers.
I hear this claim periodically, but I'd like to see some actual evidence of it. My own employer provides hardware drivers for Linux (closed source; they buy into the "it'll reveal our trade secrets and that would be bad!" fallacy) because it's what customers want.
In any case, 100% coverage of all existing hardware or devices is not necessary to achieve the standard of "works out of the box, maybe not optimally or securely." Current Linux hardware/device support is enough to cover all of the hardware that exists on a wide range of actual machines.
...the video game industry has had ample time to step up to the plate on this. This has been an issue for over five years at this point...
What exactly is the issue? Why is it an issue?
This is exactly the kind of made-up problem that politicians waste their time on to try to gain a reputation for "doing something". It's a lot easier to pass a stupid law like this than to solve important problems.
That iFilm link takes me to an all-black page with text at the top that says "This Free IFILM Content Brought to You by Our Sponsor: Skip Ad". Clicking "Skip Ad" (or waiting a few seconds) causes the page to reload. That's a great way to increase your page load count, I guess, but it doesn't display content too well.
Unfortunately everyone except the douches around here realize that anyone who was around listening to music when that song was popular is also aware of the fact that the song doesn't speak of irony.
No, South Vietnam was the side that the US was allied with. When the US left, the North conquered the South and quickly established "reeducation" camps. For the US, it was a shameful episode, but for the South Vietnamese, it was much worse.
A properly formed e-mail from a reputable company nearly completely eliminates all possible intercepts.
Even organizations that should know better sometimes fail to do this. I once received an email message from an address at openvenue.com claiming to be from the ACM and asking me to go to confirmit.com to fill out a survey. Imagine my surprise when it turned out to actually be from the ACM. (To add further insult, when I emailed the ACM about it, the two line response was followed by two copies of a ten line signature, without delimiter. Sigh.)
It seems like any university that claims to give a well-rounded liberal arts education should include a course that covers issues of computer-related common sense and etiquette, such as "don't give your account details to strangers" and "don't use a signature 10 times as long as the body of your email."
It's true that abandoning our Vietnamese friends to be conquered by an oppressive regime was a huge screwup. Unfortunately, at this point there's no realistic way to atone for that mistake; the best we can do is allow our allies and their families to immigrate. We can also hope to learn from that mistake by not abandoning any more friends in their time of need.
Slashdot Mirror
Thanks (and to the other posters who suggested different programs too.) Looks like it's time to download the KDE 3.5 beta.
The number one feature for me in choosing Gaim was a drop-in no-brainpower-required encryption solution. This feature is absent from all of the officially supported and third party IM clients I tried. (I also appreciate Gaim's cross-platform nature now that I've switched to a Linux desktop.)
You call the previously posted feature list "unimportant entry level stuff," and I partly agree: built-in client-based easy-to-use encryption is a basic feature that any IM client that purports to be a good IM client should support. What other clients offer this feature? (I'm seriously interested; I'm not opposed to switching if I can find something that works better than Gaim. Support for Yahoo videoconferencing on Linux would be a big plus.)
My employer started using IM pretty regularly at the beginning of the year. They bought everyone webcams, and we use Yahoo Messenger to videoconference during meetings (some participants are geographically remote.) So whenever I have a meeting, I need to boot up the windows laptop (Gaim on Linux doesn't support Yahoo videoconferencing yet.) They seem unconcerned about the idea of sending unencrypted trade secrets out into the wild woolly internet, but such is life. Maybe Yahoo doesn't data mine conversations.
I IM with personal accounts periodically, but I generally don't log on during the workday (email's distracting enough) and I often forget to log in at home.
Right, "CommieOverlord." I'm sure they loved working together to build their workers' paradise.
It's just like every other "realistic based-on-fact" show: it's close enough to look cool to the general public, but it makes experts laugh or cry. This is true across many domains; ask a cop what they think about the typical cop show, or a lawyer about the typical courtroom drama show, or a survivalist about "Survivor".
Maybe if we're lucky a math geek will happen by and explain it all to us.
Note that the cost comparison isn't quite that cut and dried, because lots of people already have computers (and sometimes TV capture cards too.) If you're setting up an HD PVR, a computer based solution can be quite cost competitive, especially if your only HD-capable display is your computer monitor. (Whether you're able and inclined to deal with the other issues is another question.)
I see this error a lot, and it's slightly annoying. You're treating pointers as if they all behave like C-style pointers, which isn't correct.
Java does have pointers. It doesn't let you orphan memory in the same way that C does, and it restricts the operations that you can perform on pointers, and it doesn't even call them pointers, but they are, in fact, pointers (variables that store a memory location.)
Furthermore, it is possible for a language with pointers to prevent a bad cast from screwing things up: don't allow casting! Or limit it to non-bad cases (which is a pretty big limitation.)
I don't know if they said the same thing when Netflix came out, but if they did, they'd be right. Not only are "borderline" movies cheaper to see, but when I'm finished, they don't embarrass me by sitting around on my DVD shelf.
Regardless of the truth of this, it has nothing to do with whether or not the US abandoned South Vietnam, and it doesn't change the fact that the communists were an oppressive regime. The US abandoned the South Vietnamese, its supposed allies, to be conquered by an oppressive regime. Those facts are not in dispute.
Typo? Perhaps not.
In fact, that is how it is handled, more or less.
A private monopolist will often endeavor to charge the highest price that the market will bear, while a public monopolist theoretically will not. (That's a big part of why private monopolies are a bad idea; really almost all monopolies are a bad idea, but in the case of utilities it often can't be helped.) Measuring efficiency in private vs public monopolies requires a much closer examination than simply comparing price.
Somebody somewhere panics and shoots off a bunch of nuclear missiles. Billions die, but I survive.
Now it's a race to see if I die from radiation or starvation. Good thing I have all this extra body fat. Once that's gone, I'll have the corpses of my family and my pets to eat.
Dang, radiation sickness really sucks. The other survivors all have it too, so I'm able to successfully fight them off for my share of the remnants of society.
Finally, after six months, I am too weak to defend myself, and am devoured alive by rodents.
I wake up. What? It's October 10th again? Look, it's just like the movie Groundhog Day, but lasting six months. Six terrible, terrible months. Forever.
I'm not so sure about the second one. One of the goals of such a system would surely be to help those who slip through the cracks of society, and such people are often unable to work.
Or perhaps care would be limited by civil war: treat everybody, until government mismanagement causes the citizenry to storm Congress, pitchforks and torches in hand.
Or perhaps care would be limited by race/culture: those who are or are descended from persecuted minorities get first dibs. (As a descendent of French Canadians, I'm all for this.) Note that this would only work until everybody intermarries or suddenly discovers non-white non-Europeans in their family history.
Or perhaps care would be limited by political persuasion: members of "the party" (whichever party or coalition is currently in power) get first dibs (while members of "the other party", if it exists, are stuck at the end of the line.) Rank within the party would need to be taken into account once everybody tries to join the party.
Forgive me if I find your post unconvincing, but your description of a bridge just doesn't sound that complicated to me. Do you know of any objective way to measure the complexity of problems in different domains, by which they could be compared? Comparing the complexity of a bridge (or house, or skyscraper, or airplane, or whatever) to the complexity of any given computer program does sound like an interesting pursuit. Maybe decision points, or variables... I don't know if there's an equivalent to a logical branch in bridge design, but surely there are equivalents to things like coupling and cohesion. Like, how many other widgits does this widgit touch? Through how many layers will a failure in this widgit cascade?
My intuition tells me that the complexity per developer of a piece of software is much higher than the complexity per... engineer? of a bridge. Computers make it very easy to program past your level of competence. But, I'd welcome a way to objectively verify my intuition.
I shared your skepticism, but apparently nobody informed the formal methods community that their goals are impossible, and they've come up with a number of different approaches for proving the "correctness" (really, the logical consistency) of a program's design. This is a huge step forward from current industry practice.
It's true that mathematical depictions of a program's design and behavior don't eliminate bugs. But they can potentially eliminate certain kinds of bugs that can be difficult or time-consuming to fix after the fact, and if coupled with code generation tools, the overhead of formally describing a program's structure can be somewhat mitigated.
That being said, I would be interested in seeing studies that back up some of these claims with actual numbers.
I like your i18n example. Being able to react well to that kind of change is a good thing for a design methodology to be able to do.
And insurance. "Honey, this ginormous HDTV monitor will never replace you in my heart. But it's gonna try."
That last phase, assembly, is where the stuff that is not fun enough for volunteers to typically do gets done. If a particular instance of Desktop Linux is not up to par, the assembly folks are the ones who are to blame. One success story in this area is Mac OS X, where the assembly folks took a bunch of open source programs, assembled them and some other programs into a cohesive whole and created a desktop that successfully meets the needs of most users.
Desktop Linux meets that standard today. It's not a very high standard.
I hear this claim periodically, but I'd like to see some actual evidence of it. My own employer provides hardware drivers for Linux (closed source; they buy into the "it'll reveal our trade secrets and that would be bad!" fallacy) because it's what customers want.
In any case, 100% coverage of all existing hardware or devices is not necessary to achieve the standard of "works out of the box, maybe not optimally or securely." Current Linux hardware/device support is enough to cover all of the hardware that exists on a wide range of actual machines.
What exactly is the issue? Why is it an issue?
This is exactly the kind of made-up problem that politicians waste their time on to try to gain a reputation for "doing something". It's a lot easier to pass a stupid law like this than to solve important problems.
That iFilm link takes me to an all-black page with text at the top that says "This Free IFILM Content Brought to You by Our Sponsor: Skip Ad". Clicking "Skip Ad" (or waiting a few seconds) causes the page to reload. That's a great way to increase your page load count, I guess, but it doesn't display content too well.
Optimist.
No, South Vietnam was the side that the US was allied with. When the US left, the North conquered the South and quickly established "reeducation" camps. For the US, it was a shameful episode, but for the South Vietnamese, it was much worse.
Even organizations that should know better sometimes fail to do this. I once received an email message from an address at openvenue.com claiming to be from the ACM and asking me to go to confirmit.com to fill out a survey. Imagine my surprise when it turned out to actually be from the ACM. (To add further insult, when I emailed the ACM about it, the two line response was followed by two copies of a ten line signature, without delimiter. Sigh.)
It seems like any university that claims to give a well-rounded liberal arts education should include a course that covers issues of computer-related common sense and etiquette, such as "don't give your account details to strangers" and "don't use a signature 10 times as long as the body of your email."
It's true that abandoning our Vietnamese friends to be conquered by an oppressive regime was a huge screwup. Unfortunately, at this point there's no realistic way to atone for that mistake; the best we can do is allow our allies and their families to immigrate. We can also hope to learn from that mistake by not abandoning any more friends in their time of need.