Which is only a band-aid, really. Misspell support, you don't even need AOL.
The way to deal with people who fall for phishing schemes is not to coddle them, but to let them get stung and hope that teaches them their lesson. Or we could simply start taking the warning labels off of things, and let natural selection take its course -- "Do not stop blade with hands or genitals."
I imagine "AOLamers", who can't even spell "you" properly, wouldn't notice that, either. And besides, what's stopping people from doing that already with email?
You are right about one thing, though: It's about control. If it was a fully open Jabber server, people might actually start switching off of AOL's servers, even if the majority of their buddy list is still on AOL -- which means eventually, fewer people that AOL can log conversations from, or filter key words out of -- and also probably kills them being able to charge so much for IM on cell phones.
Fun fact: I once got spammed on my phone by AOL (or was it Yahoo?) offering an IM service. They sent it over SMS (which is basically the same thing, fucktards), but what's worse is, my SMS doesn't charge me for outgoing messages, only incoming ones. So, whoever it was was sending me spam on my dime. They only did it once, but that is one reason I refuse to pay for that kind of service. (But I don't pay for my phone, so at least there, it's moot.)
I never liked tabbed windows much, really. The place where they make sense is, for instance, Firefox, where I'm constantly opening new tabs and closing them, and it's nice to have them reasonably well organized. But in general, I'd much rather just be able to rotate through similar windows, so I can see them all at once -- for instance, on OS X, I usually opened four Terminal windows and rotated through them with command+left/right, but they were 80x24 terminals, so they were all visible at once.
I believe this can be done in Beryl, but I haven't gotten around to trying. The tricky part is getting everything to group properly.
Except the advantage of being interoperable with every other IM service out there that decides to use it.
Is there actually a compelling technical reason to use their AIM protocol instead of Jabber? Because I can think of a couple of compelling reasons to use Jabber instead of AIM.
When are they going to reimplement AIM via Jabber, so that AIM users can easily talk to Google Talk users and everyone else?
That would leave only Yahoo and MSN...
But really, it seems obvious to me that they are not implementing OpenID because they like open standards. Otherwise, why aren't they actually using open standards elsewhere?
There is a solution: Authenticate your OpenID once, manually. You could even do it with a browser extension. Then, whatever they spoof, they won't be able to authenticate as you to anywhere else, only to the site you're trying to login to.
To put it in really simple terms, they'll get your username, but not your password.
By the way, we already have this problem. If someone steals your identity (social security number, etc), they can use that to gain access to most things you have, including your bank. The trick is to use single sign on to also reduce the number of places you can be compromised -- you can always pick an uber-secure OpenID provider, or roll your own.
And if you really want, you can use different IDs for different sites.
I bring this up because even if your solution works, it takes control out of our hands.
I don't mind Google, because if they ever start being obnoxious in their search ads, I can easily switch to some other search engine, or even attempt to build one myself.
What you're talking about implies a lot of industry cooperation, which also implies that there'd be a monopoly on this service. Which means it would be overpriced and under-featured. They'd arbitrarily move normal content to "premium", and you wouldn't be able to do anything other than cancel and slowly try to save up and re-buy the stuff. They'd be able to set prices wherever they want, with the same result. They'd advertise just as much as Cable TV -- have you seen those fucking things? Can't even let you enjoy the 5-10 minutes of the show you get between ads without sliding in some little ad that takes up a quarter of the screen, animates, and makes an occasional sound or two. Except that with Cable and Satellite, if I get sick of it, I can cancel my subscription and go buy a DVD, which won't have ads...
Which brings up another thing: DVDs can have unskippable ads. You can skip them in VLC, but only because VLC cracks the DRM.
So, the only way I would ever subscribe to something like this is if they gave everything to me DRM-free. If they could manage a distribution system which is faster and better than the existing networks (think BitTorrent), and if they would actually just give me the DVD in, say, a matroska file, I'd subscribe and stay subscribed. Yes, of course this means I could just share the file with all my friends, but I can do that anyway -- have music execs even looked on peer-to-peer networks lately? DRM ISN'T WORKING! It also means I could just subscribe and download as much as I could in a month, then unsubscribe -- which is, after all, what they deserve; they should be making enough new content to keep me interested -- I would subscribe to cable or satellite TV to watch a show I like, so what makes them think I wouldn't do the same over the Internet? MythTV already makes it ludicrously easy for me to share that show of cable or satellite, why do they think the Internet will make it any easier?
And if you really have zero affiliation to any company, why are you posting as Anonymous, you Coward?!
If I could implant all my media devices with a unique-to-me identifier and then transfer any content I have paid for *from any source* to any of my devices then I'd be happy with such DRM.
So would I.
Trouble is, one of those devices is my Linux desktop, and I currently play all my media with mplayer and other similar programs. I would accept DRM that these could handle, as long as you realize that this automatically means that I can decrypt it out of the DRM anyway.
And no, no proprietary forks of mplayer. I can and do play with the source code on all parts of my system. I can do things that the content providers may not have thought of, but if we're using DRM everywhere, pretty much by definition, I can only do things that content providers have thought of. As a simple example, when ripping a DVD, I can either leave it as a DVD image (and play it off my hard drive as if I had the physical DVD there), or save some space by re-encoding to h.264 in an MKV, and using optical character recognition to convert subtitles. With all subtitles in a text format, aside from simply being able to use whatever font I want and put them wherever on my screen is convenient, I can then do a full-text search of DVDs!
Other neat side effect: You now have 3 clients instead of 300.
I would say use PGP internally and enforce it, and include it in your spam rules. That way, clients who send encrypted/signed messages can be sure they get through, but clients are not required to use encryption.
Content filtering is the only workable solution I've found. If done right, it doesn't slow anything down at all (my own email is instantaneous), and is more accurate than anything else -- I NEVER lose mail, the worst that ever happens is it goes in "unsure". I get maybe 100 spams a day, maybe 200, and less than 10 gets to "unsure", less than 1 a week gets through to my actual inbox.
Spam cannot be solved with law enforcement. Even assuming we had a 100% reliable definition of spam, and it was illegal everywhere -- neither of which is true or can ever be true -- who are you going to sue? The person who didn't keep their Windows box free of spyware? Well, actually, I wouldn't mind that -- but you still have the problem that spammers are hard to track, and even if you do, they aren't necessarily anywhere that has antispam laws, and antispam laws are every bit as unreliable as traditional content-based filtering -- you know, the kind that specifically looks for "viagra", and requires just as many updates as antivirus software, and is just as ineffective? That's hard enough when done in Internet Time, and I seriously doubt our legal system -- or any legal system -- can keep up.
Antivirus/antispyware -- have to agree with you there. Look instead to the open source community -- things like dspam and bogofilter, which have no vested interest in fooling you into a solution that doesn't work.
Relay blacklisting also doesn't work -- too many false positives, too much politics involved. It also does nothing to stop zombies. I'm tempted to run some statistics on where my spam is coming from, but offhand, I'd guess at least 20% and maybe as much as 60 or 80% of it is zombies.
And it really is pretty impossible to stop spam completely, that'd be like a "war on drugs". More than once, I've been tempted by the dark side. I think to myself that I could setup the mother of all botnets, send spam, extort gambling sites, collect credit card numbers, make myself a few million, and retire, without ever really getting caught. And I'm 20.
I've never done that, but that's not because it's impossible, or even particularly hard.
That doesn't mean we shouldn't try legal action, but just like security, spam is something that is actually much easier to simply do yourself. Make your network a fortress, and then worry about whether you want to actually try to punish the spammers and crackers, or just let them bounce off.
As for me, my own solution is a simple bogofilter setup. I will modify it when I get the time (it's a personal email server) to throttle IPs and netblocks which repeatedly send large amounts of spam, but aside from the bandwidth waste, it works flawlessly. Basically, I get maybe 100-200 spams a day, which get sorted into a "spam" folder, which I glance at and delete -- there hasn't been anything legit in there for about six months. I get another 20 spams or so which go into "unsure", which occasionally has a false positive. And I get maybe 1-2 spams a week which actually make it through to my inbox.
So it works flawlessly for me as a user, so all I need to do now is performance hacks -- and throttling looks like the right answer here.
Actually, ranton did a better job than I could of refuting your points. But let's suppose you're right on one count:
God forbade the fruits of the Tree of Knowledge because it would disturb the balance of Paradise by introducing the knowledge of Evil.
I would like to think that, as human beings, good and evil goes deeper than just the knowledge of them. That is, I would like to think that merely "knowledge of Evil" is not enough to make us actually become evil.
Because if that is the case, it implies that we have no free will -- that we don't make decisions, our knowledge and experience makes them for us. And I must reject that, because I have to believe the opposite: Knowledge is power.
In fact, the demand that we consciously choose to do Good over Evil is in fact a validation of the worth of independent thought.
But again: God did not punish Adam and Eve for lying to him, or for clothing themselves. He punished them for eating the fruit -- for making themselves capable of independent thought. They weren't even punished for the independent thought, just for making themselves capable of it.
It always amazes me, reading the Old Testament, that a being supposedly all-powerful is also the most power-hungry being I've ever heard of.
Could be. But in my experience, the kids more frequently have their act together than the school.
Second, I believe that when school kids are on school property using school equipment, the school should get to decide what they're allowed to do.
The big problem for me here isn't so much that the school's deciding what they get to do, but that a very flawed program is deciding what they get to do. My employer sure has the right to block my internet use, but they don't, because unrestricted Web access makes me much more effective at my job, and they can always glance over my shoulder or read proxy logs if they want to find out what I'm doing.
Sorry, school time is time that students should be using for, I dunno, learning.
Which is why in high school, at least one of my teachers taught everyone to use proxies. Basically, the school paid their ISP to filter our HTTP, but they didn't even try to implement proxies or outbound firewalls locally, so all you had to do was find a web proxy (we used megaproxy) and type https instead of http.
Later, we replaced this with proxies like Squid and Polipo, run at people's houses, but the point I'm making is that the teacher was the first to teach us to use proxies, because too often, the filter got in the way of learning. We'd be trying to do a research paper, and half the research out there would be arbitrarily blocked, for no good reason. I used Slashdot, even, for current events and technology and such, and Slashdot is blocked because it has comments, and someone could *gasp!* swear in the comments, or even post ASCII PR0N OMG ITS THE END OF THE WORLD!!!
MySpace and MSN don't qualify, if this is really what they're looking to get to.
I never use MySpace, but I frequently use IM, even at work. I probably learn, and teach, as much through IM as I do hands-on.
I suppose next you'll be wanting to block speaking in schools, period, because if you speak, you could be gossiping and wasting time? Never mind that speech -- and all other forms of communication -- are essential tools for actual education...
For that matter, let's get rid of anything that's not always and immediately used for learning. No chairs, no desks, you sit on the hard floor, chairs are luxuries, you can learn there on the floor... No clothes, don't want you staring at each other's earrings and such and thinking about fashion while you should be learning...
I hate to invoke slippery slope, but believe me when I say that the more open a learning environment is, the better. If you're worried the kids aren't focused, give them assignments, and mark them down when they don't complete them -- or better yet, actually supervise them, and don't let them use MySpace if it's not relevant -- or ask them to explain if it is relevant.
hollared at as censorship or some other poorly-fitting term
In what way is this not censorship? Or would you rather we make up a happy politically-correct bullshit term for it -- I don't know, "network restrictions"?
Censorship fits. Burden is on you to show why it doesn't.
"We think kids should be allowed to do whatever they want, but we also think you should make them learn material they don't want to at the same time."
Wrong. We think kids should be allowed to do whatever they want, so long as it isn't illegal and doesn't prevent other kids from learning the material.
And that really isn't mutually exclusive with learning the material. Believe it or not, some kids do want to learn, and some kids do want the grades. But if none of the kids want to learn what you're teaching, perhaps it's time to rethink what you're teaching and how you're teaching it?
There was a physics teacher at my high school who was an absolute genius. He could probably
I mean, if you're going to teach the Bible, at least teach what was actually in it, but I can't help but wonder...
The original sin was eating from the tree of knowledge of good and evil. So, actually learning -- particularly learning about ethics -- is what damned us all. Curiosity is a bad thing.
I realize they were specifically ordered by God not to do that. But is that really different than, say, being ordered by the Chinese Government not to search Google.com for Tiananmen Square? What makes God so special compared to an oppressive government or human dictator?
I actually don't mean this as a direct attack on the idea of God, and I can save that for another debate. I'm just pointing out that the result of both of these is about the same -- either puritanical fear of sex or Luddite-like fear of knowledge. Think about it -- if God said "Don't kill anything" and Adam kills Eve, I could understand that as Original Sin. If he said "You belong together forever", and Adam had a fight with Eve and went off to screw sheep, I could understand. I mean, if they raped, killed, tortured, maimed, stole, or any number of things, I could understand... But I cannot accept curiosity and independent thought as Original Sin.
But that IS what the Bible tends to teach, so it's not surprising to me that we see people ignorant of Original Sin, when wanting to not be ignorant was the Original Sin anyway.
I think it's great that Gnome doesn't make everything retardedly point-n-click(tm) to configure.
There's a difference between wanting things to be configurable with a text file and configurable by patching and recompiling. Besides, the two are not mutually exclusive.
I like text files because they're searchable and concise, two attributes that can not be put on some dumb GUI.
You know, I like text files too, but GUIs can be both searchable and concise. Try OS X's System Preferences, for instance -- you can actually search for a config option, and you'll get visual results.
Side note: Why does Linus have to use Metacity with Gnome?
I suppose he doesn't, but would it still be Gnome if he didn't? And that's just one example. This started with discussion about gimp-print, which I don't think can be replaced so easily.
It's simple to do, but he might have to open the Gconf to do it... which means he won't because there isn't a Giant Pretty Blue Button to automatically do it.
Remember who you're talking about. You think the Linux kernel has Giant Pretty Blue Buttons... well, anywhere? What about gcc, or git?
Your work computer can be managed by the company you work at; they can even revoke root if they're concerned about security. There are actually a few existing distributed filesystems for Linux, though most of them suck, and the few I've seen with the potential not to suck either cost money or are a long way from being stable on Linux. Haven't seen ANY of these on Windows.
Someone mentioned backup, which isn't a big deal. Ever heard of RAID? Yeah, it could be something like that.
Although if it's a desktop PC, the 40 gigs probably isn't worth the power required to keep your computer on, and they're probably better with insanely aggressive local stores to speed up disk access. But again, these kind of suck for Linux.
The closest thing to a solid, clean design that I can find for Linux network filesystems is NFS, and that doesn't have any of the features we're talking about -- not without combining it with one of these other kludges...
"Lucy doesn't want her children to screw up the computer such that double-clicking anything fails to work."
I have to agree with Linus that this kind of thinking is not just wrong, but damaging.
If Lucy lets her children use a computer that she depends on, she really should give them their own accounts so they can't screw up anything. And they will find a way -- trying to protect the user from their own stupidity is like trying to make bits not copyable, or water not wet, or RMS shave his beard. I'm sorry, but no system out there protects her completely, and they shouldn't. On Win2K, her kind could simply drag her toolbar down -- hiding the Start menu -- and she'd be useless. On XP, they just have to unlock it before they autohide it -- few more right-clicks, I guess.
As it is, it seems he will have to fork something. This discussion was originally about gimp-print. He can run KDE, but if he actually, say, runs the Gimp, he'll have to use the GNOME print dialog -- which is one reason I hate GNOME and KDE; they've gotten too big as discrete projects. They are not the Unix Way. The Print dialog should be a very small library that everyone uses, which depends on nothing irrelevant... but then, KDE uses QT and GNOME uses GTK+, making this difficult... You get the idea.
By the way, I would much rather have Linus keep working on Linux than go off on a crusade forking GNOME stuff. There are better people for that job, and better things for Linus to do. Ultimately it's his choice, though, so take what we say here with a grain of salt...
While they integrate well with KDE or GNOME, I have Beryl working well without any kind of desktop environment running.
Not much slower than Fluxbox to start, and the only place performance suffers is games -- and then only in windowed mode (once you disable redirection for fullscreen windows). In a few years even that won't matter much -- my UT2004 framerate dropped by half or so when running windowed, so in a few years, that'll be down to maybe a 5-10% hit, if that. CPU usage is less than 10% -- more like 5% -- when dragging around a wobbling, translucent, animating terminal (running top). This is running at 1600x1200.
It's actually faster in at least a few places. For instance, as counter-intuitive as this seems, dragging around a large-ish Cedega window -- with full wobbly-windows in play -- was MUCH faster and smoother in Beryl than in Fluxbox, probably because Fluxbox can't really do any acceleration on that kind of movement.
I haven't found anything I could configure in Fluxbox that I can't configure or replace in Beryl. But then, I used Fluxbox pretty minimally. The tabbed window feature just isn't very useful when you can't see multiple tabs of a window at once, so I've learned to like multiple desktops more. I didn't use a menu to launch apps, I used a run command, and in fact, I've bound Fluxbox's run to a key in Beryl.
And by the way, speaking of configurability, YOU CAN TURN OFF AERO IN VISTA! DUH!
If Linux ever becomes popular, it would be conceivable for a user to use someone else's machine, or expect instructions in an introductory book to work.
Both of these are actually obscenely easy to deal with. If I let other people use my machine, I give them their own account and put them on GNOME. (I run straight Beryl with elements of Fluxbox.) And if you're reading an introductory book, you'll be dealing with defaults anyway.
Sane defaults, but configurability, is the way to go. And by the way, this is true of more than just Linux. I've tried to use someone else's Mac, and couldn't find the program I wanted easily because his desktop was absolutely fucking PILED with documents, something like 10 deep on top of the "hard drive" icon, making it kind of difficult to get to "Applications". Someone else's Windows, and you find they've got the status bar auto-hiding at the top of the screen. And for that matter, I use the dvorak layout, so...
I mean, I understand the point of that. That is why, for instance, game consoles are designed the way they are -- you can toss a controller to anyone and have them join the party.
But configurability can be done in such a way that it doesn't hurt usability. And, in fact, it has to be done that way, because if you nix configurability, you kill usability.
Except the way things are going, the GNOME modem configuration wizard will have one gigantic button which says "Intarweb" and connects you to AOL. If you don't like that, you can always recompile or use KDE...
And yes, OSS does do that. Usually it's done by forking small projects and testing development versions, then merging those changes back. Sometimes it's done by keeping alternatives of very small systems available -- like Cron, for instance -- often based on forks of the same code, too, so you can merge stuff back.
KDE and GNOME are simply too big for me to accept them as reasonable alternatives. If it was just metacity vs kwin, that'd be alright -- after all, I can replace both of them with Beryl, at both ends. But if you replace one, you end up replacing a whole set of things -- they each have different theming systems, configuration interfaces, file browsing interfaces, virtual filesystem interfaces...
It's hard to measure it, but I think you should do a reasonably objective comparison: Open the two of them side by side. Ask yourself: Is KDE's taskbar significantly different than GNOME's panel, in terms of functionality? What about the virtual desktop pager -- they really look identical. Why do we need both of them?
Too often, the best answer anyone can give is that the underlying tech is too different. In other words: The Gnome Panel will use GTK+, probably Bonobo, maybe gconf, etc. Not sure what KDE uses instead of gconf, but the KDE panel would use QT and Kparts. And there's dozens of other libraries that they each have, which essentially duplicate the same functionality, but the whole systems are so big that they really have to have different libraries for a lot of it.
I mean, what are you going to do, implement a new toolbar in pure X? Because neither project wants to depend on the other. You won't get GNOME to depend on QT, and you won't get KDE to depend on GTK+. And this kind of thing is pretty pervasive. It's a fairly massive project -- probably being done anyway, but still pretty massive -- to try to absorb the common, maturing features of each into common mechanisms like gstreamer and dbus.
And there will ALWAYS be feature overlap here.
What we really need is to refactor things a bit -- which is what they're doing. Find functionality common to both which can be isolated into a tiny library that depends on neither. Rinse and repeat until the things that make KDE different than GNOME are handled in something smaller than Fluxbox.
Unless that happens, I'm going to consider it a truly massive waste, both in terms of manpower and my computer resources -- every day I have to run something depending on each project, meaning I need both QT and GTK+, even though they both do exactly the same thing.
At least, I think both KDE and GNOME are GPL'd... right? I hope? Because licensing issues can make a bad situation worse... Cedega forked wine back when it was under a more generous license, and promised to contribute back to the WineHQ project -- but it doesn't look like they've given back any code, so Wine is now under LPGL, meaning Cedega could contribute code to Wine (by relicensing it), but they can no longer take code from Wine -- effectively a permanent fork. Compiz and Beryl have similar problems -- I believe Compiz is under an MIT-like license, while Beryl is GPL'd, which means Compiz improvements can go into Beryl, but not the other way around.
People don't want configurability, they want something that works out of the box.
In what world are these mutually exclusive?
That's why despite being free, people will pay 200$ for a copy of Windows.
Windows is free?
Come on, with a post that short, you can afford to read it aloud to yourself and see if it makes sense.
They don't want to compile things, they want it to work out of the box.
Funny, that seems like what Linus is doing here. He wants to be able to change his right-click without recompiling! In what way is that wrong?
Oh, by the way, Windows allows programs change options in your right-click menu without rebooting, much less recompiling.
They don't want to edit config files,
Which is why good Linux UIs make this configurable through a nice GUI. Point and click your way to what you want -- just like on Windows.
they want it to self discover out of the box.
You know, your post reads like you think this has something to do with Linux not detecting hardware (except it does; it's not 1999 anymore), but that's not the issue at all. It's about UI preferences, and for fuck's sake, how is my computer supposed to know what kind of UI I want?
Oh, right, there's Clippy, which tries to guess what I'm doing ("It looks like you're writing a letter..."), and Vista's UAC, which asks me before it does anything ("Are you sure you want to use the Internet? That could be dangerous!")... I suppose that's my computer "discovering" what I want out of the box. But you know what, every single person I've talked to -- not just on Slashdot; Microsoft zealots included -- hates Clippy and UAC with a passion, because the defaults are fine for most people, and for the people who care, they'd rather hunt for the settings they care about than be bombarded about absolutely everything.
Lopez did seem like a squabbling child. However, the Schaller and Torvalds quotes you mention seem decent.
Yes, Linus is pissed, but he's not wrong, and there's actually very good signal-to-noise ratio there. If the GNOME team can hold their blood pressure steady as they read it, there could actually be some useful dialog there, carefully disguised as a flamewar.
Slashdot Mirror
Which is only a band-aid, really. Misspell support, you don't even need AOL.
The way to deal with people who fall for phishing schemes is not to coddle them, but to let them get stung and hope that teaches them their lesson. Or we could simply start taking the warning labels off of things, and let natural selection take its course -- "Do not stop blade with hands or genitals."
So I can't currently register AOLCustumerService?
Get it? Look at that nick closely...
I imagine "AOLamers", who can't even spell "you" properly, wouldn't notice that, either. And besides, what's stopping people from doing that already with email?
You are right about one thing, though: It's about control. If it was a fully open Jabber server, people might actually start switching off of AOL's servers, even if the majority of their buddy list is still on AOL -- which means eventually, fewer people that AOL can log conversations from, or filter key words out of -- and also probably kills them being able to charge so much for IM on cell phones.
Fun fact: I once got spammed on my phone by AOL (or was it Yahoo?) offering an IM service. They sent it over SMS (which is basically the same thing, fucktards), but what's worse is, my SMS doesn't charge me for outgoing messages, only incoming ones. So, whoever it was was sending me spam on my dime. They only did it once, but that is one reason I refuse to pay for that kind of service. (But I don't pay for my phone, so at least there, it's moot.)
I never liked tabbed windows much, really. The place where they make sense is, for instance, Firefox, where I'm constantly opening new tabs and closing them, and it's nice to have them reasonably well organized. But in general, I'd much rather just be able to rotate through similar windows, so I can see them all at once -- for instance, on OS X, I usually opened four Terminal windows and rotated through them with command+left/right, but they were 80x24 terminals, so they were all visible at once.
I believe this can be done in Beryl, but I haven't gotten around to trying. The tricky part is getting everything to group properly.
Except the advantage of being interoperable with every other IM service out there that decides to use it.
Is there actually a compelling technical reason to use their AIM protocol instead of Jabber? Because I can think of a couple of compelling reasons to use Jabber instead of AIM.
When are they going to reimplement AIM via Jabber, so that AIM users can easily talk to Google Talk users and everyone else?
That would leave only Yahoo and MSN...
But really, it seems obvious to me that they are not implementing OpenID because they like open standards. Otherwise, why aren't they actually using open standards elsewhere?
There is a solution: Authenticate your OpenID once, manually. You could even do it with a browser extension. Then, whatever they spoof, they won't be able to authenticate as you to anywhere else, only to the site you're trying to login to.
To put it in really simple terms, they'll get your username, but not your password.
By the way, we already have this problem. If someone steals your identity (social security number, etc), they can use that to gain access to most things you have, including your bank. The trick is to use single sign on to also reduce the number of places you can be compromised -- you can always pick an uber-secure OpenID provider, or roll your own.
And if you really want, you can use different IDs for different sites.
By this I mean the functionality that I had in Fluxbox. At the end of the day, the only real difference in how I work is the eye candy.
I bring this up because even if your solution works, it takes control out of our hands.
I don't mind Google, because if they ever start being obnoxious in their search ads, I can easily switch to some other search engine, or even attempt to build one myself.
What you're talking about implies a lot of industry cooperation, which also implies that there'd be a monopoly on this service. Which means it would be overpriced and under-featured. They'd arbitrarily move normal content to "premium", and you wouldn't be able to do anything other than cancel and slowly try to save up and re-buy the stuff. They'd be able to set prices wherever they want, with the same result. They'd advertise just as much as Cable TV -- have you seen those fucking things? Can't even let you enjoy the 5-10 minutes of the show you get between ads without sliding in some little ad that takes up a quarter of the screen, animates, and makes an occasional sound or two. Except that with Cable and Satellite, if I get sick of it, I can cancel my subscription and go buy a DVD, which won't have ads...
Which brings up another thing: DVDs can have unskippable ads. You can skip them in VLC, but only because VLC cracks the DRM.
So, the only way I would ever subscribe to something like this is if they gave everything to me DRM-free. If they could manage a distribution system which is faster and better than the existing networks (think BitTorrent), and if they would actually just give me the DVD in, say, a matroska file, I'd subscribe and stay subscribed. Yes, of course this means I could just share the file with all my friends, but I can do that anyway -- have music execs even looked on peer-to-peer networks lately? DRM ISN'T WORKING! It also means I could just subscribe and download as much as I could in a month, then unsubscribe -- which is, after all, what they deserve; they should be making enough new content to keep me interested -- I would subscribe to cable or satellite TV to watch a show I like, so what makes them think I wouldn't do the same over the Internet? MythTV already makes it ludicrously easy for me to share that show of cable or satellite, why do they think the Internet will make it any easier?
And if you really have zero affiliation to any company, why are you posting as Anonymous, you Coward?!
So would I.
Trouble is, one of those devices is my Linux desktop, and I currently play all my media with mplayer and other similar programs. I would accept DRM that these could handle, as long as you realize that this automatically means that I can decrypt it out of the DRM anyway.
And no, no proprietary forks of mplayer. I can and do play with the source code on all parts of my system. I can do things that the content providers may not have thought of, but if we're using DRM everywhere, pretty much by definition, I can only do things that content providers have thought of. As a simple example, when ripping a DVD, I can either leave it as a DVD image (and play it off my hard drive as if I had the physical DVD there), or save some space by re-encoding to h.264 in an MKV, and using optical character recognition to convert subtitles. With all subtitles in a text format, aside from simply being able to use whatever font I want and put them wherever on my screen is convenient, I can then do a full-text search of DVDs!
Other neat side effect: You now have 3 clients instead of 300.
I would say use PGP internally and enforce it, and include it in your spam rules. That way, clients who send encrypted/signed messages can be sure they get through, but clients are not required to use encryption.
Content filtering is the only workable solution I've found. If done right, it doesn't slow anything down at all (my own email is instantaneous), and is more accurate than anything else -- I NEVER lose mail, the worst that ever happens is it goes in "unsure". I get maybe 100 spams a day, maybe 200, and less than 10 gets to "unsure", less than 1 a week gets through to my actual inbox.
Spam cannot be solved with law enforcement. Even assuming we had a 100% reliable definition of spam, and it was illegal everywhere -- neither of which is true or can ever be true -- who are you going to sue? The person who didn't keep their Windows box free of spyware? Well, actually, I wouldn't mind that -- but you still have the problem that spammers are hard to track, and even if you do, they aren't necessarily anywhere that has antispam laws, and antispam laws are every bit as unreliable as traditional content-based filtering -- you know, the kind that specifically looks for "viagra", and requires just as many updates as antivirus software, and is just as ineffective? That's hard enough when done in Internet Time, and I seriously doubt our legal system -- or any legal system -- can keep up.
Antivirus/antispyware -- have to agree with you there. Look instead to the open source community -- things like dspam and bogofilter, which have no vested interest in fooling you into a solution that doesn't work.
Relay blacklisting also doesn't work -- too many false positives, too much politics involved. It also does nothing to stop zombies. I'm tempted to run some statistics on where my spam is coming from, but offhand, I'd guess at least 20% and maybe as much as 60 or 80% of it is zombies.
And it really is pretty impossible to stop spam completely, that'd be like a "war on drugs". More than once, I've been tempted by the dark side. I think to myself that I could setup the mother of all botnets, send spam, extort gambling sites, collect credit card numbers, make myself a few million, and retire, without ever really getting caught. And I'm 20.
I've never done that, but that's not because it's impossible, or even particularly hard.
That doesn't mean we shouldn't try legal action, but just like security, spam is something that is actually much easier to simply do yourself. Make your network a fortress, and then worry about whether you want to actually try to punish the spammers and crackers, or just let them bounce off.
As for me, my own solution is a simple bogofilter setup. I will modify it when I get the time (it's a personal email server) to throttle IPs and netblocks which repeatedly send large amounts of spam, but aside from the bandwidth waste, it works flawlessly. Basically, I get maybe 100-200 spams a day, which get sorted into a "spam" folder, which I glance at and delete -- there hasn't been anything legit in there for about six months. I get another 20 spams or so which go into "unsure", which occasionally has a false positive. And I get maybe 1-2 spams a week which actually make it through to my inbox.
So it works flawlessly for me as a user, so all I need to do now is performance hacks -- and throttling looks like the right answer here.
Problem: What happens if the spammers discover you doing this, and send new spam with a link to your website?
Even if people do it manually, this is going to sting legitimate people who have nothing to do with the spam.
Actually, ranton did a better job than I could of refuting your points. But let's suppose you're right on one count:
I would like to think that, as human beings, good and evil goes deeper than just the knowledge of them. That is, I would like to think that merely "knowledge of Evil" is not enough to make us actually become evil.
Because if that is the case, it implies that we have no free will -- that we don't make decisions, our knowledge and experience makes them for us. And I must reject that, because I have to believe the opposite: Knowledge is power.
But again: God did not punish Adam and Eve for lying to him, or for clothing themselves. He punished them for eating the fruit -- for making themselves capable of independent thought. They weren't even punished for the independent thought, just for making themselves capable of it.
It always amazes me, reading the Old Testament, that a being supposedly all-powerful is also the most power-hungry being I've ever heard of.
Or we could go for Gangsta Rap OS! Don't even have to rhyme, just RESPEKT, BITCHEZ!!!
Could be. But in my experience, the kids more frequently have their act together than the school.
The big problem for me here isn't so much that the school's deciding what they get to do, but that a very flawed program is deciding what they get to do. My employer sure has the right to block my internet use, but they don't, because unrestricted Web access makes me much more effective at my job, and they can always glance over my shoulder or read proxy logs if they want to find out what I'm doing.
Which is why in high school, at least one of my teachers taught everyone to use proxies. Basically, the school paid their ISP to filter our HTTP, but they didn't even try to implement proxies or outbound firewalls locally, so all you had to do was find a web proxy (we used megaproxy) and type https instead of http.
Later, we replaced this with proxies like Squid and Polipo, run at people's houses, but the point I'm making is that the teacher was the first to teach us to use proxies, because too often, the filter got in the way of learning. We'd be trying to do a research paper, and half the research out there would be arbitrarily blocked, for no good reason. I used Slashdot, even, for current events and technology and such, and Slashdot is blocked because it has comments, and someone could *gasp!* swear in the comments, or even post ASCII PR0N OMG ITS THE END OF THE WORLD!!!
I never use MySpace, but I frequently use IM, even at work. I probably learn, and teach, as much through IM as I do hands-on.
I suppose next you'll be wanting to block speaking in schools, period, because if you speak, you could be gossiping and wasting time? Never mind that speech -- and all other forms of communication -- are essential tools for actual education...
For that matter, let's get rid of anything that's not always and immediately used for learning. No chairs, no desks, you sit on the hard floor, chairs are luxuries, you can learn there on the floor... No clothes, don't want you staring at each other's earrings and such and thinking about fashion while you should be learning...
I hate to invoke slippery slope, but believe me when I say that the more open a learning environment is, the better. If you're worried the kids aren't focused, give them assignments, and mark them down when they don't complete them -- or better yet, actually supervise them, and don't let them use MySpace if it's not relevant -- or ask them to explain if it is relevant.
In what way is this not censorship? Or would you rather we make up a happy politically-correct bullshit term for it -- I don't know, "network restrictions"?
Censorship fits. Burden is on you to show why it doesn't.
Wrong. We think kids should be allowed to do whatever they want, so long as it isn't illegal and doesn't prevent other kids from learning the material.
And that really isn't mutually exclusive with learning the material. Believe it or not, some kids do want to learn, and some kids do want the grades. But if none of the kids want to learn what you're teaching, perhaps it's time to rethink what you're teaching and how you're teaching it?
There was a physics teacher at my high school who was an absolute genius. He could probably
I mean, if you're going to teach the Bible, at least teach what was actually in it, but I can't help but wonder...
The original sin was eating from the tree of knowledge of good and evil. So, actually learning -- particularly learning about ethics -- is what damned us all. Curiosity is a bad thing.
I realize they were specifically ordered by God not to do that. But is that really different than, say, being ordered by the Chinese Government not to search Google.com for Tiananmen Square? What makes God so special compared to an oppressive government or human dictator?
I actually don't mean this as a direct attack on the idea of God, and I can save that for another debate. I'm just pointing out that the result of both of these is about the same -- either puritanical fear of sex or Luddite-like fear of knowledge. Think about it -- if God said "Don't kill anything" and Adam kills Eve, I could understand that as Original Sin. If he said "You belong together forever", and Adam had a fight with Eve and went off to screw sheep, I could understand. I mean, if they raped, killed, tortured, maimed, stole, or any number of things, I could understand... But I cannot accept curiosity and independent thought as Original Sin.
But that IS what the Bible tends to teach, so it's not surprising to me that we see people ignorant of Original Sin, when wanting to not be ignorant was the Original Sin anyway.
There's a difference between wanting things to be configurable with a text file and configurable by patching and recompiling. Besides, the two are not mutually exclusive.
You know, I like text files too, but GUIs can be both searchable and concise. Try OS X's System Preferences, for instance -- you can actually search for a config option, and you'll get visual results.
I suppose he doesn't, but would it still be Gnome if he didn't? And that's just one example. This started with discussion about gimp-print, which I don't think can be replaced so easily.
Remember who you're talking about. You think the Linux kernel has Giant Pretty Blue Buttons... well, anywhere? What about gcc, or git?
Your work computer can be managed by the company you work at; they can even revoke root if they're concerned about security. There are actually a few existing distributed filesystems for Linux, though most of them suck, and the few I've seen with the potential not to suck either cost money or are a long way from being stable on Linux. Haven't seen ANY of these on Windows.
Someone mentioned backup, which isn't a big deal. Ever heard of RAID? Yeah, it could be something like that.
Although if it's a desktop PC, the 40 gigs probably isn't worth the power required to keep your computer on, and they're probably better with insanely aggressive local stores to speed up disk access. But again, these kind of suck for Linux.
The closest thing to a solid, clean design that I can find for Linux network filesystems is NFS, and that doesn't have any of the features we're talking about -- not without combining it with one of these other kludges...
I have to agree with Linus that this kind of thinking is not just wrong, but damaging.
If Lucy lets her children use a computer that she depends on, she really should give them their own accounts so they can't screw up anything. And they will find a way -- trying to protect the user from their own stupidity is like trying to make bits not copyable, or water not wet, or RMS shave his beard. I'm sorry, but no system out there protects her completely, and they shouldn't. On Win2K, her kind could simply drag her toolbar down -- hiding the Start menu -- and she'd be useless. On XP, they just have to unlock it before they autohide it -- few more right-clicks, I guess.
As it is, it seems he will have to fork something. This discussion was originally about gimp-print. He can run KDE, but if he actually, say, runs the Gimp, he'll have to use the GNOME print dialog -- which is one reason I hate GNOME and KDE; they've gotten too big as discrete projects. They are not the Unix Way. The Print dialog should be a very small library that everyone uses, which depends on nothing irrelevant... but then, KDE uses QT and GNOME uses GTK+, making this difficult... You get the idea.
By the way, I would much rather have Linus keep working on Linux than go off on a crusade forking GNOME stuff. There are better people for that job, and better things for Linus to do. Ultimately it's his choice, though, so take what we say here with a grain of salt...
While they integrate well with KDE or GNOME, I have Beryl working well without any kind of desktop environment running.
Not much slower than Fluxbox to start, and the only place performance suffers is games -- and then only in windowed mode (once you disable redirection for fullscreen windows). In a few years even that won't matter much -- my UT2004 framerate dropped by half or so when running windowed, so in a few years, that'll be down to maybe a 5-10% hit, if that. CPU usage is less than 10% -- more like 5% -- when dragging around a wobbling, translucent, animating terminal (running top). This is running at 1600x1200.
It's actually faster in at least a few places. For instance, as counter-intuitive as this seems, dragging around a large-ish Cedega window -- with full wobbly-windows in play -- was MUCH faster and smoother in Beryl than in Fluxbox, probably because Fluxbox can't really do any acceleration on that kind of movement.
I haven't found anything I could configure in Fluxbox that I can't configure or replace in Beryl. But then, I used Fluxbox pretty minimally. The tabbed window feature just isn't very useful when you can't see multiple tabs of a window at once, so I've learned to like multiple desktops more. I didn't use a menu to launch apps, I used a run command, and in fact, I've bound Fluxbox's run to a key in Beryl.
And by the way, speaking of configurability, YOU CAN TURN OFF AERO IN VISTA! DUH!
Both of these are actually obscenely easy to deal with. If I let other people use my machine, I give them their own account and put them on GNOME. (I run straight Beryl with elements of Fluxbox.) And if you're reading an introductory book, you'll be dealing with defaults anyway.
Sane defaults, but configurability, is the way to go. And by the way, this is true of more than just Linux. I've tried to use someone else's Mac, and couldn't find the program I wanted easily because his desktop was absolutely fucking PILED with documents, something like 10 deep on top of the "hard drive" icon, making it kind of difficult to get to "Applications". Someone else's Windows, and you find they've got the status bar auto-hiding at the top of the screen. And for that matter, I use the dvorak layout, so...
I mean, I understand the point of that. That is why, for instance, game consoles are designed the way they are -- you can toss a controller to anyone and have them join the party.
But configurability can be done in such a way that it doesn't hurt usability. And, in fact, it has to be done that way, because if you nix configurability, you kill usability.
Except the way things are going, the GNOME modem configuration wizard will have one gigantic button which says "Intarweb" and connects you to AOL. If you don't like that, you can always recompile or use KDE...
Alternatives are good, code duplication is bad.
And yes, OSS does do that. Usually it's done by forking small projects and testing development versions, then merging those changes back. Sometimes it's done by keeping alternatives of very small systems available -- like Cron, for instance -- often based on forks of the same code, too, so you can merge stuff back.
KDE and GNOME are simply too big for me to accept them as reasonable alternatives. If it was just metacity vs kwin, that'd be alright -- after all, I can replace both of them with Beryl, at both ends. But if you replace one, you end up replacing a whole set of things -- they each have different theming systems, configuration interfaces, file browsing interfaces, virtual filesystem interfaces...
It's hard to measure it, but I think you should do a reasonably objective comparison: Open the two of them side by side. Ask yourself: Is KDE's taskbar significantly different than GNOME's panel, in terms of functionality? What about the virtual desktop pager -- they really look identical. Why do we need both of them?
Too often, the best answer anyone can give is that the underlying tech is too different. In other words: The Gnome Panel will use GTK+, probably Bonobo, maybe gconf, etc. Not sure what KDE uses instead of gconf, but the KDE panel would use QT and Kparts. And there's dozens of other libraries that they each have, which essentially duplicate the same functionality, but the whole systems are so big that they really have to have different libraries for a lot of it.
I mean, what are you going to do, implement a new toolbar in pure X? Because neither project wants to depend on the other. You won't get GNOME to depend on QT, and you won't get KDE to depend on GTK+. And this kind of thing is pretty pervasive. It's a fairly massive project -- probably being done anyway, but still pretty massive -- to try to absorb the common, maturing features of each into common mechanisms like gstreamer and dbus.
And there will ALWAYS be feature overlap here.
What we really need is to refactor things a bit -- which is what they're doing. Find functionality common to both which can be isolated into a tiny library that depends on neither. Rinse and repeat until the things that make KDE different than GNOME are handled in something smaller than Fluxbox.
Unless that happens, I'm going to consider it a truly massive waste, both in terms of manpower and my computer resources -- every day I have to run something depending on each project, meaning I need both QT and GTK+, even though they both do exactly the same thing.
At least, I think both KDE and GNOME are GPL'd... right? I hope? Because licensing issues can make a bad situation worse... Cedega forked wine back when it was under a more generous license, and promised to contribute back to the WineHQ project -- but it doesn't look like they've given back any code, so Wine is now under LPGL, meaning Cedega could contribute code to Wine (by relicensing it), but they can no longer take code from Wine -- effectively a permanent fork. Compiz and Beryl have similar problems -- I believe Compiz is under an MIT-like license, while Beryl is GPL'd, which means Compiz improvements can go into Beryl, but not the other way around.
In what world are these mutually exclusive?
Windows is free?
Come on, with a post that short, you can afford to read it aloud to yourself and see if it makes sense.
Funny, that seems like what Linus is doing here. He wants to be able to change his right-click without recompiling! In what way is that wrong?
Oh, by the way, Windows allows programs change options in your right-click menu without rebooting, much less recompiling.
Which is why good Linux UIs make this configurable through a nice GUI. Point and click your way to what you want -- just like on Windows.
You know, your post reads like you think this has something to do with Linux not detecting hardware (except it does; it's not 1999 anymore), but that's not the issue at all. It's about UI preferences, and for fuck's sake, how is my computer supposed to know what kind of UI I want?
Oh, right, there's Clippy, which tries to guess what I'm doing ("It looks like you're writing a letter..."), and Vista's UAC, which asks me before it does anything ("Are you sure you want to use the Internet? That could be dangerous!")... I suppose that's my computer "discovering" what I want out of the box. But you know what, every single person I've talked to -- not just on Slashdot; Microsoft zealots included -- hates Clippy and UAC with a passion, because the defaults are fine for most people, and for the people who care, they'd rather hunt for the settings they care about than be bombarded about absolutely everything.
Lopez did seem like a squabbling child. However, the Schaller and Torvalds quotes you mention seem decent.
Yes, Linus is pissed, but he's not wrong, and there's actually very good signal-to-noise ratio there. If the GNOME team can hold their blood pressure steady as they read it, there could actually be some useful dialog there, carefully disguised as a flamewar.