Okay, I agree the label covers advertising. But what does this really mean? Unless the band already needs no introduction, they aren't getting TV or radio spots. (Let's not get into pay-for-play just yet)...
What are they getting? Posters? Unless you live in a major urban area, you're probally not going to see many of those, and even if you do they are probabally posted illegally. Besides I doubt you move many units based solely on what the album "looks like". Maybe the occasional impluse purchase, but I can't see many people buying the majority of their albums unheard.
That leaves airtime of some sort. The tradional channels are broadcast radio and cable TV. MTV and clones only play videos (or parts of them) and those are a whole 'nother expensive enterprise, which usually doesn't start until the band actually is successful. So we're back to broadcast radio. (We'll get to XM/Sirius in minute) The FM dial is pretty much a small set of genres with the same 12-24 songs in rotation around a slew of blather. Sure there are small indie stations, but those are dying off faster and faster. Probably because the labels would rather advertise the newest Britt Spears single on/to the local ClearChannel/Infinity franchise than spend the bucks to get some unknown played on a tiny little college station.
XM/Sirius is a little better, but you've still got a fairly small number of spots for a really huge number of potential songs/artists.
Where is left for the non-megastars? Pretty much concerts. Concerts get you something, but again, a narrow audience. I'm 29. I have a job, a wife, and a baby on the way. I live 30 minutes from the "hip Austin Music Scene". Even when bands I really like come to town I don't go see them. It's not something that fits into my lifestyle anymore.
Finally there is the internet. The last bands I've checked out where because someone's website said they liked them and I hunted around to find them. (Yes, usually on P2P of some sort). Once I did find them I downloaded a few tracks and looked at what else that user (the P2P one) was sharing and grabbed a handful of other stuff at random. Eventually I get around to listening to them, and delete 90% right off. Out of the remaining 10% I usually find a few tracks I like and then go out and buy the album. (This is pretty rare because I don't have the time to search and download...)
So, the way I see it, the "major labels" have two choices:
1) Not sell me anything, because I don't hear anything I like.
2) Accept the fact that P2P is a reality and produce a physical product which is inticing enough for people to bother buying it
When P2P was easy, I bought more CD's that I ever had at any other point in my life. As it got to be more of a hassle, I've bought less and less, and listen to the same old CDs again and again. I've brought close to 1000 CD's in my lifetime, but no more than 10-15 in the last two years. (And most of those were used)
"The Industry" is cutting itself out of the sweetest parts of the market (25 - 35 adults with 100K+ household income) in the hopes of locking in the 13-24 year old set.
You're forgetting the the truck's engines aren't 100% efficient. Not all the energy in the diesel is being extracted. By adding a little bit more hydrogen to the combustion process, the efficiency is boosted. (Not to 100%, but a small bit) So yes, the power to the wheels of the truck is marginally reduced by the additional load on the alternator, but that is more than offset by the increased energy extraction which results.
The hydrogen isn't so much a fuel, as a catalyst. A litte bit goes a long way, and more isn't necessarily better. Adding a small amount might improve efficency 10%, but adding twice as much doesn't result in a 20% boost, in fact, it would probably reduce the efficency.
I agree that tablets are still pretty niche market. The only time I've ever seen one being used by someone doing serious work is when I bought my house. Our house inspector had a tablet PC on some sort of sling attached to his body. When he wanted it, it swung around, and when he didn't, it just sorta hung out back by this hip. (The arrangement was a great deal more elegant than I am making it sound.)
The impressive part was that the guy took his notes on the tablet and documented issues with a digital camera as he went. When he was done, he emails out his report as a PDF containing photos accompaning all his remarks, as well as links to internet research about common problems. (For example, he might point out that something needed to be done, and link to a site which discussed the seriousness and pro vs. con aspects of various solutions)
I do see your point, and in some ways I agree. However, my point is that the average net user is just as identifiable with a IPv4 address as with an IPv6 one. Both addresses route to a physical endpoint which remains constant. Sure, you can run NAT and hide behing that endpoint, but that doesn't realistically buy you a whole lot.
For most home users NAT simply means the IP resolves to the "family members who use this address" as opposed to the individual computer using this address. Even this distinction isn't critical because you could have multiple users of the same computer, so even if I can track all requests from a single IP (or MAC) I don't know for sure they are the same person.
Where the pseudo-anonymity of IPv4 comes in is through DHCP. I don't believe (but am not positive) that the IPv6 spec prohibits DHCP, and in fact I would be surprised if it did. Most consumer ISPs want the easiest setup possible for their users. This usually means DHCP. Without it they have to educate their customers what to type in for IP, gateway, nameservers, etc... IPv6 reduces some of this, but doesn't make it go away. I suspect IPv6 ISPs will allow, and possibly encourage DHCP, and only provide static IPs at a premium.
The difference is that users who want to have multiple machines behind that top level IP won't have to buy a huge block of addresses.
IPv6 provides the power of a static IP to anyone who wants one, but if you don't want one, don't use one.
(I'm actually very pro-privacy, so it's odd for me to find myself arguing the other side.)
Re:Are we ready to surrender anonymity on the net?
on
IPv6 Still Hotly Debated
·
· Score: 3, Insightful
A couple of points:
1) With a static IP, especially if you have a DNS name to go along with it, you leave just as big of a footprint, if not more. (Since I've only got the one directly addressable IP, I might as well get a name to go with it, right? And then use something like DynDNS? Well, unless I register by proxy, I have to give my name, address, phone, etc...)
2) MAC address, while theoretically static, can easily be changed in most OSes and hardware. For example, my LinkSys router has an option to "clone MAC address" in the setup. The problem with changing your MAC address is that the prefixes indicate the vendor, and that might get you in trouble with someone who "owns" that prefix. (I doubt it though)
3) There is nothing preventing you from NAT'ing IPv6, and I suspect some people probably will simply for the quasi-deny-all-in firewall effect. Moreover, if you really want to be anonymous, IPv6 makes it much easier to implement things like "onion routing" because it's a lot easier for individuals to set up persistant servers.
The point is, you can control the "MAC" portion of the address, and the "public" portion is just as visible (or not) as with IPv4. Hell, you could change your MAC address every coupla minutes for a REALLY long time without ever repeating one if that's what you wanted. (Persistant connections be damned...)
I'm not bashing Java, but I don't completely agree with your statement that:
Cross-platform byte code enables you to deploy the same application on your PC or Mac workstation and have it function exactly the same as on a 64-processor Ultra server. It also means your application is "future-proof". Deploy it now on a 32-bits machine, later on a 64-bits machine without recompiling AND run it at speeds comparable to native code.
While this is theoretically true for byte code, in Java isn't not something you can depend on. I write Java code for a living, and our test team still has to test our product on every platform and OS we claim support for. Even the same version of the same vendor's JVM will have little OS-specific quirks which can add up to big problems if ignored.
More importantly, "write once, run anywhere" really should be qualified as "write once, run anywhere the exact same JVM is installed and a decent script/command/launcher exists to properly start your application with all the flags, classpaths, memory management, and other configuration needed for this platform"
This may seem sorta nit-picky but I spend a great deal of my time figuring out why a certain behavior only occurs on a certain platform. Usually it's because to do anything useful, you can only abstract the underlying operating system so far. For example, the way that Windows creates and manages a file is very different than the why most Unix systems do. An application which downloads a file from the network and overwrites an existing file often needs special handling on Windows because Windows sometimes will deny access to a file (if it thinks it is in use), or not synch the descriptors when you expect it to. This isn't either Java or Window's "fault" it's just the way things work.
However, I do agree that Java is a good language and makes a lot more "sense" from an API perspective than most other languages. Writing programs in C/C++ now seems very tedious to me. All the stuff I used to worry about without thinking about (buffer sizes, memory allocation, finding functions/libraries) in C/C++ which aren't issues in Java amazes me.
Luckily, I don't need (or want) Outlook in my life. I do however need Lotus Notes. I've been using the Codeweavers version for years to get this, and love it. The only major drawback was that the "new mail" notification popped up in its own window and stole focus. (Very annoying)
This latest version fixed it! It's right in the system tray where it belongs! Yay!
I can't thank the development team enough. Crossover Office is crucial for getting my work done everyday, and is the only software I have ever seen the need to pay money for. (Corporate site licenses cover the MS and Lotus stuff I run under Crossover, but Crossover I buy myself)
Note: I work for a company which develops software for this solution, but I do not speak for them in any official way. I'm also not going to plug my product by name, because that's not the point of this article. There aren't that many people doing this kind of work, and if you're really interested, you'll find us easily enough.
But this is pretty cool. The problem, of course, is how to decide whether someone is "secure" or not without running a scan on that computer. It isn't like infected computers are going to run around flagging routers of their infected status.
One of the principle components of the NAC architecture is something called the Cisco Trust Agent. The CTA is an authenticator framework which allows for multiple third-party vendor plugins. There are existing extensions for several major anti-virus technologies, and also much broader solutions. Basically, when you first show up on the network at a hardware level, the NAC-enabled device (router, switch, etc...) sends a challenge. The CTA then responds with information about the machine generated by CTA itself and any installed plugins. (This also can include authentication, which makes things that much more secure). The response is relayed to a Cisco Access Control Server (ACS), which then makes a decision about the state (posture) of the machine and pushes down an appropriate set of access controls.
What's even cooler, is that products like the one I work on, are capable of "closing the loop" and fixing violations. If your virus definitions are out of date, we can kick of a scan. If you're missing a Windows hotfix, we can install it. If you password length is too short, we can fix that too. All this is done by associating workflows with existing (and proven) configuration and provisioning management solutions.
I don't know about high-altitude booms, but at a low-level they are pretty amazing. As a child I was vacationing with my parents someplace on the North Carolina shore. There was military base somewhere nearby, and even though I was like 7 at time time, I still remember clearly the time we heard a "boom". It wasn't really a "boom" but more of a very long, rumbling thunder. At the time my dad described it as "someone rolling a huge log down the roof"
I suspect high-altitude booms are a much different thing, much like the difference between the noise a (sub-sonic) military jet at low altitude and high altitude.
Okay, this isn't a Microsoft bashing outright, but why the fsck does MS want to "kill" Google? Google makes a great product that has arguably is one of the most important and useful tools on the web. On top of this, they don't charge anything to use it.
Yes, Google makes money. Sure, money is good, and everyone want more. But for crying out loud, just because someone else has success in an area of business doesn't mean you have to squash them. Microsoft should focus on making Windows better. The reason Google is good is because they spend their effort trying to be the best search engine, not the only search engine.
I'm not saying companies shouldn't aggressively pursue their competitors, but this just reeks of jealousy. I know that Google has a lot of new services (and likely more on the way) which compete with Microsoft (GMail vs. Hotmail, Google Search vs. MSN Search, GTalk vs. Messenger, Google Earth vs. Terraserver) but still...
It would be different if I thought MS was going to build a product which would "kill" Google by simply being better, but I suspect the plan is more to cripple Google as much as possible, and bring everyone down to a "well, it could be better but this is good enough" level.
Best of luck. When I applied back in 99, I told my interviewers part of why I wanted to work for IBM was because I had had the oppertunity to take apart an RS/6000 machine once and was amazed at how clean and well laid-out everything was inside. I doubt that had anything to do with their hiring me, but it's funny that we think alike.
Oh, and it used to be (and may still be) that they made potential hires take this written test called the PAT. (Programming Aptitude Test). It's basically a bunch of thinly disguised logic puzzles and a couple of sections of "how fast can you do basic math".
I remember thinking I bombed it because I didn't seem to have enough time. I've asked around and everyone seems to have had that impression when they took it. My manager later told me I did really well on it, but I get the feeling it was more of a hold-over from when they were drafting coders off the assembly lines or something. (And yes, that happens. The guy in the next office over has been here for like 30+ years, and he started out assembling the carriages for typewriters and got drafted to try programming) Anyway, don't let the test rattle ya, cuz I suspect it's meaningless.
IBM is a cool place to work, and is actually pretty casual. The blue-suit days are gone. Good luck!
The video is very nice, and I'm told the DVI and S-Video out even work under Linux. (Haven't tried it yet) I've got the smaller screened version (1400 x 1040) instead of 1600x1200, but it still feels huge. This is the first time in a while I've found myself making fonts *bigger* to improve their readability.
I don't have the model with the fingerprint scanner, but that's used at a BIOS (or lower) level I beleive and I think it doesn't care what OS it is running.
I've also not tried the hard-drive parking yet, but there is preliminary support for it being developed which looks like it might even been in the -mm kernels soon. (HDAPS Project)
To be honest, I don't even know what "Rapid Restore" is, so I don't know if I'm benefiting from it. I've only had the laptop a week, and unfortunately work and home commitments have prevented me from playing with it as much as I would like.
The 9 cell battery seems to run a really long time, but I haven't pushed it hard.
I'm not sure that it means, but the T42p is current the standard laptop issued to developers in my division of IBM. I just got mine about a week ago, and so far, so good.
One thing to note is that even though it's easier to run Linux on that other computers there are still some warts. For example the built-in 802.11a/b/g card needs the MadWifi drivers, and to really make the display perform well you need the proprietary ATI drivers. Both of these taint the kernel.
On the plus side, a lot of stuff "just worked" for me out of the box or with little hassle. Most notably was ACPI suspend to RAM (S3). I can't tell you how much I've missed being able to just shut the lid and throw my laptop in mt briefcase for a day or two. (I should note you do have to add a line to the kernel boot options, and the lid-shut needs a script, but echo 3 >/proc/acpi/sleep works)
I'm really psyched that there is an actual ibm-acpi module in the standard kernel now...
Interesting ideas, but I'm not sure if any of them have both the political viabiliy to be accepted and the teeth to do what is supposedly being attempted by the new version of the GPL.
I'm not trying to be a jackass and tear you down, but I think there are flaws in all your suggestions which are fatal. Ultimately I don't think this problem is as solvable a DRM itself. You can put up elaborate barriers, and they might even work most the time, but ultimately, they aren't addressing the right tool for the "problem".
1. Holders of software patents cannot make use of the permissions granted by the GPL at all. I don't think it's possible to prevent them from using the software, but they could be prevented from creating derived works or from distributing copies.
I think this would pretty much kill the GPL in the corporate realm. Every major software player has some patent on something software related. So these people would be reduced to leechs at best.
2. Holders of software patents can make use of GPL-derived permissions, but only as long as they don't enforce their software patents.
If patent holders don't enforce their patents, then the patents are useless. (and effectively become public domain, similiar to trademarks). The amount of money big players like IBM and Sun make off licensing their patents is a signifigant chunk of their overall profit. By the same token, the money and resources it takes to patent even the simpliest of software ideas (which are frivolous at best) is non-trivial. Even if I interperate you to mean "only enforce defensively" there are problems. The major one is that the companies would go out of business overnight as their stock price tanked. Giving away the patent licensing fees would be seen as a VERY dumb move by the money-driven. The other one becomes defining "defensively". Defensive how? When someone sues them for infringement? What keeps me from using a suite over say, one-click shopping to hit my oppenent with everything I have in my portfolio? More importantly you can't selectively enforce your patent rights informally. You either explicitly grant rights or your retain them. The patent system is a "least permission" model. You start with no rights, and have to be explicitly given them.
3. Holders of software patents can use GPL-derived permissions, and can enforce their patents, but if they enforce their patents against *any* GPL project, they lose the right to use any GPL-derived permissions.
So basically, this means any patent can be used freely by anyone releasing under the GPL. But if I decide to sue, then the GPL people have to stop, and I have to stop using their stuff. There are multiple problems here as well. The major one is I don't think anyone making money off a patent is going to happily hand out a license to anyone using GPL. Even if they did, there is a more serious danger. A single party could easily kill large segments of the GPL world with a little planning. Let's say XYZ corp has the patent on some data format "X". X is a good format, and soon everyone is using it everywhere. Whole projects depend on it at a low level. Then one day XYZ corp says: "Nah, we don't want you to use X anymore, and we're cool with not using your stuff. In fact, we're going out of business" So, everyone who uses X in the GPL world now has to immediately stop their product AND get their users to stop. Then they get to rewrite from the ground up. Ouch. Of course the whole concept depends on selective patent enforcement, which I believe isn't valid anyway...
4. Holders of software patents can use GPL-derived permissions, and can enforce their patents, but if they claim patents that are infringed by a GPL project, they do not have the GPL-provided permissions for that project. They can gain the use of the GPL for that project by explicitly licensing the patent
for all GPL use.
This doesn't really change anything from idea 3. If I own the patent for "X" a
There is no way a company IBM's size has so many projects and patents there is no way they could comply with this. For example, most (all?) Thinkpads have a DRM chip in them. Sure, these are technically Lenovo now, but still. Ironically, the only support of the DRM functions I'm aware of is via a Linux kernel module with was/is made freely available by IBM and is open source. (Dunno about the license offhand)
And what about the customers of companies who insist (in some cases for good reason) on either DRM or closed source? People like the military like the idea of a hardware-validated OS very much, and have the bucks to make or break even the biggest players. And a lot of customers don't want the source of their apps made public. Companies can't afford to tell these customers: "sorry, we can't write the code you ask for"
I loathe and detest abuses of DRM and really don't think software should be patentable. And while the GPL is becoming increasingly powerful, it's got no where near the power to force a fundamental change in how the industry works. It's a nice dream, but it's just going to piss off the corporate-types which are just now starting to come around.
I'm surprised no one else has commented on this line from the article yet:
David Kaefer, Microsoft's director of intellectual property licensing, said it was open to letting other firms patent its innovations.
Maybe I'm missing something, but even in context, I keep parsing this sentence as meaning: "Microsoft is okay with other firms patenting Microsoft innovations". Sort of a whoever-thinks-it's-patentable-first thing.
I can't beleive MS would endorse this position. Did anyone else take it the same way?
I didn't find the point irrelavant at all. It's not about the majority of the use, it's about the minority. Almost all the material Google indexes is "legal" but I'm sure they also index pages with have say... serial numbers for software. Google might make a tiny portion of it's profit from these sorts of searches, but that's an unavoidable side-effect of their major business.
Smith and Wesson make a tiny profit from people who buy guns to use in crimes. Probably they make a slightly bigger (but still miniscule) profit from people who steal guns and then buy S&W ammo to use in crimes. There is no way for S&W to prevent this, just as there is no way for Google to prevent people from misusing information they search for.
Think of the song 867-5309 (Jenny). That was a real woman's phone number, and I believe she did sue and win over having it used in the song. This is the same situation.
Publishers actually have less free speech than the general public because they are speaking to more than one person. There is an expectation of relevance in journalism. Reporters are free to report things which aren't common knowledge, but they are expected to use their power responsibly.
CNet did what they did to provoke a response. They took a non-story and made it "interesting" by singling someone out. If they want to be childish, they have no grounds for complaining when Google is childish back.
I tried this (after a long pause thinking about running a random.exe) on my Fedora Core 3 machine under Crossover Office. No luck. In fact, it completely hung the keyboard on the display I invoked it on. I use a multiheaded setup under Xinerama, so that may have contributed to the problem. Fortunately I was able to kill it using a shell in another window.:(
Re:For those who don't want to RTFA, the top 10:
on
10 Technologies MIA
·
· Score: 2, Interesting
One thing to consider is that your old rotary phone has a speaker that is over an inch in diameter and has an actual honest-to-good magnet in it. Speakers in today's electronics tend to me cheap-ass piezo-electrics only slightly bigger than a pencil eraser. I've never heard one of those tiny speakers which sounded as good as a cheap paper-coned magnetic one.
Magnetic speakers are cheap and mass-produced as well, but they are also heavy, and can't be easily placed next to other circuitry without problems.
Okay, I admit it. People think I'm a security freak, but I still run 802.11b with WEP enabled at home. I've got strong keys, I filter MACs, I disable beaconing, and have put up other minor fortifications, but I still know I'm running pretty open.
So why haven't I improved things?
Simple. Even though I'm a pretty technical Linux user, I've been unable to really feel confident going out and buying 802.11g stuff with WPA, because the existing documentation on the net is pretty bad.
I'm waiting for the mythical "someone else" to set up a nice, straight-forward site that says "here are the cards you can buy at store X which support Linux and don't require binary drivers, patched kernels, and other crap" Sure, there are lists of chipsets, but the actual stores don't list the chipset in particular products often, and the vendors often have multiple versions of the same card with different chipsets.
I think a lot of the problem is the actual hardware industry itself. 802.11b wasn't hard to get Linux support for, but because of the software controlled radio in 802.11g chipsets, it's a bit tricker legally.
And don't get me started on Bluetooth. I got a new phone which has it, and I'd love to buy a little USB Bluetooth dongle so I can play with it, but right now the main Linux Bluetooth page has been asked to take down their list of devices known to work under Linux, because someone in the Bluetooth SIG complained the devices weren't technically qualified. (link) What a load of crap! So instead of getting a dongle which might not work, I'm just not going to get one at all. Everyone loses.
PCMCIA Firewire card is marginally easier, but again, trying to track down and actual card for sale which matches the user-reported specs and models is pretty damn hard. I spent conservatively 3 hours online and in Fry's reading before I got a card which works great until you eject it and panic the kernel.
I guess where I'm going with this rant is that wireless security (in the non-Windows world) would probably be better if the "standards" followed went a bit deeper and were more open to allowing outsiders to confidently buy products. All I'm asking for is a label or a sticker on the box telling me what chipset and version the device uses. It's not hard, and it shouldn't be a secret. Anyone technically savvy to make a purchasing decision based on chipset is technically savvy to figure out what chipset is in a device once they've bought it and spread the word.
Actually I did read not only TFA, but the other articles involved before posting. Sure, there is some valid concern about how much personal data Google might be tracking, especially given their privacy policies, but that's not why Google is pissed at cNet.
I notice your account has a URL listed, and that URL has a resume. It's perfectly legal for me to comb through all your pages, cross-reference it with other sources. I could even respond by posting every bit of minute trivia about you on this page. And then you could check your website logs, and figure out my IP, which would tell you who my employer is. And if you dug through my posting history, you could likely figure out where I live, who I am, and post the exact same type of information about me.
But neither one of us is going to do that. It might technically be legal, but it's a really jerk move, and pretty much defines "ad hominium" attacks. Generally talking about someone's wife, children, personal life, etc is a very inciting tactic to take in an argument. When I was growing up, you could always tell when an argument between two kids was getting out of hand when someone started talking trash (especially true trash) about someone's family members.
I'd take it very personally if someone posted my wife's name and contact information in response to something I said. Even more so if it was something my business did policy-wise. And beleive me, it would be very hard to resist the urge to strike back at the poster using any means I had at my disposal.
Personal information which may or may not be present in Google's non-public servers is a completely separate issue.
I have to say I side with Google on this one. Google simply indexes information. It's not possible (or desirable) for them to filter and screen the content of what they index.
Yes, you can use Google to track down a distressing amount of personal information about some people, but this is a function of the information being made available by third parties. Google just makes it easier to find all these sources quickly.
People that gripe about (or sue) Google based on their indexing "bad" things, need to step back and think of the Web as more of a library, with each page as a book. Google serves as a card catalog, helping you find the books that have the information you are interested in. If somebody goes to the library and looks up a bunch of personal information on you (which is possible, just slower) you don't get mad at the makers of the card catalog. Your anger should be directed first at the person who singled you out. Next, if the books contain something which shouldn't be public (unlike major stock sales, and other things from the article, which should be public) you ought to take it up with the author/publisher of the books.
cNet took a cheap shot at Google, and did it in a fairly childish way. The point they were trying to make is both obvious, and better made in a more mature fashion. That being said, I don't exactly think Google's response is "mature", but if they want to respond in kind, I don't blame them.
I'm not an amateur radio person, and have no interest in becoming one. At the same time, I like knowing that somewhere near me there is likely someone who can communicate with other people around the world under pretty much any conditions short of a massive EMP.
So, I have no personal stake in the requirements other than I'd like these people to be good enough to justify their access to the rather valuable section of the RF spectrum we've set aside for them. I'd like to think they might be able to put together a radio from components if needed.
Even more important, I like knowing that there is a group of people out their who can communicate over pretty much any channel, radio or not, using the simpliest possible code. There are times (albeit rare) when the only communication you have is banging on a pipe with a large wrench, or flashing a light. I highly doubt that "Radio Shack" has the components to build a decent long-range HAM setup these days, but pretty much anyone can figure out a way to transmit long and short pulses using stuff found anywhere.
Slashdot Mirror
Okay, I agree the label covers advertising. But what does this really mean? Unless the band already needs no introduction, they aren't getting TV or radio spots. (Let's not get into pay-for-play just yet)...
What are they getting? Posters? Unless you live in a major urban area, you're probally not going to see many of those, and even if you do they are probabally posted illegally. Besides I doubt you move many units based solely on what the album "looks like". Maybe the occasional impluse purchase, but I can't see many people buying the majority of their albums unheard.
That leaves airtime of some sort. The tradional channels are broadcast radio and cable TV. MTV and clones only play videos (or parts of them) and those are a whole 'nother expensive enterprise, which usually doesn't start until the band actually is successful. So we're back to broadcast radio. (We'll get to XM/Sirius in minute) The FM dial is pretty much a small set of genres with the same 12-24 songs in rotation around a slew of blather. Sure there are small indie stations, but those are dying off faster and faster. Probably because the labels would rather advertise the newest Britt Spears single on/to the local ClearChannel/Infinity franchise than spend the bucks to get some unknown played on a tiny little college station.
XM/Sirius is a little better, but you've still got a fairly small number of spots for a really huge number of potential songs/artists.
Where is left for the non-megastars? Pretty much concerts. Concerts get you something, but again, a narrow audience. I'm 29. I have a job, a wife, and a baby on the way. I live 30 minutes from the "hip Austin Music Scene". Even when bands I really like come to town I don't go see them. It's not something that fits into my lifestyle anymore.
Finally there is the internet. The last bands I've checked out where because someone's website said they liked them and I hunted around to find them. (Yes, usually on P2P of some sort). Once I did find them I downloaded a few tracks and looked at what else that user (the P2P one) was sharing and grabbed a handful of other stuff at random. Eventually I get around to listening to them, and delete 90% right off. Out of the remaining 10% I usually find a few tracks I like and then go out and buy the album. (This is pretty rare because I don't have the time to search and download...)
So, the way I see it, the "major labels" have two choices:
1) Not sell me anything, because I don't hear anything I like.
2) Accept the fact that P2P is a reality and produce a physical product which is inticing enough for people to bother buying it
When P2P was easy, I bought more CD's that I ever had at any other point in my life. As it got to be more of a hassle, I've bought less and less, and listen to the same old CDs again and again. I've brought close to 1000 CD's in my lifetime, but no more than 10-15 in the last two years. (And most of those were used)
"The Industry" is cutting itself out of the sweetest parts of the market (25 - 35 adults with 100K+ household income) in the hopes of locking in the 13-24 year old set.
The hydrogen isn't so much a fuel, as a catalyst. A litte bit goes a long way, and more isn't necessarily better. Adding a small amount might improve efficency 10%, but adding twice as much doesn't result in a 20% boost, in fact, it would probably reduce the efficency.
The impressive part was that the guy took his notes on the tablet and documented issues with a digital camera as he went. When he was done, he emails out his report as a PDF containing photos accompaning all his remarks, as well as links to internet research about common problems. (For example, he might point out that something needed to be done, and link to a site which discussed the seriousness and pro vs. con aspects of various solutions)
For most home users NAT simply means the IP resolves to the "family members who use this address" as opposed to the individual computer using this address. Even this distinction isn't critical because you could have multiple users of the same computer, so even if I can track all requests from a single IP (or MAC) I don't know for sure they are the same person.
Where the pseudo-anonymity of IPv4 comes in is through DHCP. I don't believe (but am not positive) that the IPv6 spec prohibits DHCP, and in fact I would be surprised if it did. Most consumer ISPs want the easiest setup possible for their users. This usually means DHCP. Without it they have to educate their customers what to type in for IP, gateway, nameservers, etc... IPv6 reduces some of this, but doesn't make it go away. I suspect IPv6 ISPs will allow, and possibly encourage DHCP, and only provide static IPs at a premium.
The difference is that users who want to have multiple machines behind that top level IP won't have to buy a huge block of addresses.
IPv6 provides the power of a static IP to anyone who wants one, but if you don't want one, don't use one.
(I'm actually very pro-privacy, so it's odd for me to find myself arguing the other side.)
1) With a static IP, especially if you have a DNS name to go along with it, you leave just as big of a footprint, if not more. (Since I've only got the one directly addressable IP, I might as well get a name to go with it, right? And then use something like DynDNS? Well, unless I register by proxy, I have to give my name, address, phone, etc...)
2) MAC address, while theoretically static, can easily be changed in most OSes and hardware. For example, my LinkSys router has an option to "clone MAC address" in the setup. The problem with changing your MAC address is that the prefixes indicate the vendor, and that might get you in trouble with someone who "owns" that prefix. (I doubt it though)
3) There is nothing preventing you from NAT'ing IPv6, and I suspect some people probably will simply for the quasi-deny-all-in firewall effect. Moreover, if you really want to be anonymous, IPv6 makes it much easier to implement things like "onion routing" because it's a lot easier for individuals to set up persistant servers.
The point is, you can control the "MAC" portion of the address, and the "public" portion is just as visible (or not) as with IPv4. Hell, you could change your MAC address every coupla minutes for a REALLY long time without ever repeating one if that's what you wanted. (Persistant connections be damned...)
Cross-platform byte code enables you to deploy the same application on your PC or Mac workstation and have it function exactly the same as on a 64-processor Ultra server. It also means your application is "future-proof". Deploy it now on a 32-bits machine, later on a 64-bits machine without recompiling AND run it at speeds comparable to native code.
While this is theoretically true for byte code, in Java isn't not something you can depend on. I write Java code for a living, and our test team still has to test our product on every platform and OS we claim support for. Even the same version of the same vendor's JVM will have little OS-specific quirks which can add up to big problems if ignored.
More importantly, "write once, run anywhere" really should be qualified as "write once, run anywhere the exact same JVM is installed and a decent script/command/launcher exists to properly start your application with all the flags, classpaths, memory management, and other configuration needed for this platform"
This may seem sorta nit-picky but I spend a great deal of my time figuring out why a certain behavior only occurs on a certain platform. Usually it's because to do anything useful, you can only abstract the underlying operating system so far. For example, the way that Windows creates and manages a file is very different than the why most Unix systems do. An application which downloads a file from the network and overwrites an existing file often needs special handling on Windows because Windows sometimes will deny access to a file (if it thinks it is in use), or not synch the descriptors when you expect it to. This isn't either Java or Window's "fault" it's just the way things work.
However, I do agree that Java is a good language and makes a lot more "sense" from an API perspective than most other languages. Writing programs in C/C++ now seems very tedious to me. All the stuff I used to worry about without thinking about (buffer sizes, memory allocation, finding functions/libraries) in C/C++ which aren't issues in Java amazes me.
This latest version fixed it! It's right in the system tray where it belongs! Yay!
I can't thank the development team enough. Crossover Office is crucial for getting my work done everyday, and is the only software I have ever seen the need to pay money for. (Corporate site licenses cover the MS and Lotus stuff I run under Crossover, but Crossover I buy myself)
Thanks again, and keep up the good work.
But this is pretty cool. The problem, of course, is how to decide whether someone is "secure" or not without running a scan on that computer. It isn't like infected computers are going to run around flagging routers of their infected status.
One of the principle components of the NAC architecture is something called the Cisco Trust Agent. The CTA is an authenticator framework which allows for multiple third-party vendor plugins. There are existing extensions for several major anti-virus technologies, and also much broader solutions. Basically, when you first show up on the network at a hardware level, the NAC-enabled device (router, switch, etc...) sends a challenge. The CTA then responds with information about the machine generated by CTA itself and any installed plugins. (This also can include authentication, which makes things that much more secure). The response is relayed to a Cisco Access Control Server (ACS), which then makes a decision about the state (posture) of the machine and pushes down an appropriate set of access controls.
What's even cooler, is that products like the one I work on, are capable of "closing the loop" and fixing violations. If your virus definitions are out of date, we can kick of a scan. If you're missing a Windows hotfix, we can install it. If you password length is too short, we can fix that too. All this is done by associating workflows with existing (and proven) configuration and provisioning management solutions.
I suspect high-altitude booms are a much different thing, much like the difference between the noise a (sub-sonic) military jet at low altitude and high altitude.
Yes, Google makes money. Sure, money is good, and everyone want more. But for crying out loud, just because someone else has success in an area of business doesn't mean you have to squash them. Microsoft should focus on making Windows better. The reason Google is good is because they spend their effort trying to be the best search engine, not the only search engine.
I'm not saying companies shouldn't aggressively pursue their competitors, but this just reeks of jealousy. I know that Google has a lot of new services (and likely more on the way) which compete with Microsoft (GMail vs. Hotmail, Google Search vs. MSN Search, GTalk vs. Messenger, Google Earth vs. Terraserver) but still...
It would be different if I thought MS was going to build a product which would "kill" Google by simply being better, but I suspect the plan is more to cripple Google as much as possible, and bring everyone down to a "well, it could be better but this is good enough" level.
Oh, and it used to be (and may still be) that they made potential hires take this written test called the PAT. (Programming Aptitude Test). It's basically a bunch of thinly disguised logic puzzles and a couple of sections of "how fast can you do basic math".
I remember thinking I bombed it because I didn't seem to have enough time. I've asked around and everyone seems to have had that impression when they took it. My manager later told me I did really well on it, but I get the feeling it was more of a hold-over from when they were drafting coders off the assembly lines or something. (And yes, that happens. The guy in the next office over has been here for like 30+ years, and he started out assembling the carriages for typewriters and got drafted to try programming) Anyway, don't let the test rattle ya, cuz I suspect it's meaningless.
IBM is a cool place to work, and is actually pretty casual. The blue-suit days are gone. Good luck!
I work for IBM. :)
I don't have the model with the fingerprint scanner, but that's used at a BIOS (or lower) level I beleive and I think it doesn't care what OS it is running.
I've also not tried the hard-drive parking yet, but there is preliminary support for it being developed which looks like it might even been in the -mm kernels soon. (HDAPS Project)
To be honest, I don't even know what "Rapid Restore" is, so I don't know if I'm benefiting from it. I've only had the laptop a week, and unfortunately work and home commitments have prevented me from playing with it as much as I would like.
The 9 cell battery seems to run a really long time, but I haven't pushed it hard.
One thing to note is that even though it's easier to run Linux on that other computers there are still some warts. For example the built-in 802.11a/b/g card needs the MadWifi drivers, and to really make the display perform well you need the proprietary ATI drivers. Both of these taint the kernel.
On the plus side, a lot of stuff "just worked" for me out of the box or with little hassle. Most notably was ACPI suspend to RAM (S3). I can't tell you how much I've missed being able to just shut the lid and throw my laptop in mt briefcase for a day or two. (I should note you do have to add a line to the kernel boot options, and the lid-shut needs a script, but echo 3 > /proc/acpi/sleep works)
I'm really psyched that there is an actual ibm-acpi module in the standard kernel now...
I'm not trying to be a jackass and tear you down, but I think there are flaws in all your suggestions which are fatal. Ultimately I don't think this problem is as solvable a DRM itself. You can put up elaborate barriers, and they might even work most the time, but ultimately, they aren't addressing the right tool for the "problem".
1. Holders of software patents cannot make use of the permissions granted by the GPL at all. I don't think it's possible to prevent them from using the software, but they could be prevented from creating derived works or from distributing copies.
I think this would pretty much kill the GPL in the corporate realm. Every major software player has some patent on something software related. So these people would be reduced to leechs at best.
2. Holders of software patents can make use of GPL-derived permissions, but only as long as they don't enforce their software patents.
If patent holders don't enforce their patents, then the patents are useless. (and effectively become public domain, similiar to trademarks). The amount of money big players like IBM and Sun make off licensing their patents is a signifigant chunk of their overall profit. By the same token, the money and resources it takes to patent even the simpliest of software ideas (which are frivolous at best) is non-trivial. Even if I interperate you to mean "only enforce defensively" there are problems. The major one is that the companies would go out of business overnight as their stock price tanked. Giving away the patent licensing fees would be seen as a VERY dumb move by the money-driven. The other one becomes defining "defensively". Defensive how? When someone sues them for infringement? What keeps me from using a suite over say, one-click shopping to hit my oppenent with everything I have in my portfolio? More importantly you can't selectively enforce your patent rights informally. You either explicitly grant rights or your retain them. The patent system is a "least permission" model. You start with no rights, and have to be explicitly given them.
3. Holders of software patents can use GPL-derived permissions, and can enforce their patents, but if they enforce their patents against *any* GPL project, they lose the right to use any GPL-derived permissions.
So basically, this means any patent can be used freely by anyone releasing under the GPL. But if I decide to sue, then the GPL people have to stop, and I have to stop using their stuff. There are multiple problems here as well. The major one is I don't think anyone making money off a patent is going to happily hand out a license to anyone using GPL. Even if they did, there is a more serious danger. A single party could easily kill large segments of the GPL world with a little planning. Let's say XYZ corp has the patent on some data format "X". X is a good format, and soon everyone is using it everywhere. Whole projects depend on it at a low level. Then one day XYZ corp says: "Nah, we don't want you to use X anymore, and we're cool with not using your stuff. In fact, we're going out of business" So, everyone who uses X in the GPL world now has to immediately stop their product AND get their users to stop. Then they get to rewrite from the ground up. Ouch. Of course the whole concept depends on selective patent enforcement, which I believe isn't valid anyway...
4. Holders of software patents can use GPL-derived permissions, and can enforce their patents, but if they claim patents that are infringed by a GPL project, they do not have the GPL-provided permissions for that project. They can gain the use of the GPL for that project by explicitly licensing the patent for all GPL use.
This doesn't really change anything from idea 3. If I own the patent for "X" a
And what about the customers of companies who insist (in some cases for good reason) on either DRM or closed source? People like the military like the idea of a hardware-validated OS very much, and have the bucks to make or break even the biggest players. And a lot of customers don't want the source of their apps made public. Companies can't afford to tell these customers: "sorry, we can't write the code you ask for"
I loathe and detest abuses of DRM and really don't think software should be patentable. And while the GPL is becoming increasingly powerful, it's got no where near the power to force a fundamental change in how the industry works. It's a nice dream, but it's just going to piss off the corporate-types which are just now starting to come around.
David Kaefer, Microsoft's director of intellectual property licensing, said it was open to letting other firms patent its innovations.
Maybe I'm missing something, but even in context, I keep parsing this sentence as meaning: "Microsoft is okay with other firms patenting Microsoft innovations". Sort of a whoever-thinks-it's-patentable-first thing.
I can't beleive MS would endorse this position. Did anyone else take it the same way?
Smith and Wesson make a tiny profit from people who buy guns to use in crimes. Probably they make a slightly bigger (but still miniscule) profit from people who steal guns and then buy S&W ammo to use in crimes. There is no way for S&W to prevent this, just as there is no way for Google to prevent people from misusing information they search for.
Think of the song 867-5309 (Jenny). That was a real woman's phone number, and I believe she did sue and win over having it used in the song. This is the same situation.
Publishers actually have less free speech than the general public because they are speaking to more than one person. There is an expectation of relevance in journalism. Reporters are free to report things which aren't common knowledge, but they are expected to use their power responsibly.
CNet did what they did to provoke a response. They took a non-story and made it "interesting" by singling someone out. If they want to be childish, they have no grounds for complaining when Google is childish back.
I tried this (after a long pause thinking about running a random .exe) on my Fedora Core 3 machine under Crossover Office. No luck. In fact, it completely hung the keyboard on the display I invoked it on. I use a multiheaded setup under Xinerama, so that may have contributed to the problem. Fortunately I was able to kill it using a shell in another window. :(
Magnetic speakers are cheap and mass-produced as well, but they are also heavy, and can't be easily placed next to other circuitry without problems.
So why haven't I improved things?
Simple. Even though I'm a pretty technical Linux user, I've been unable to really feel confident going out and buying 802.11g stuff with WPA, because the existing documentation on the net is pretty bad.
I'm waiting for the mythical "someone else" to set up a nice, straight-forward site that says "here are the cards you can buy at store X which support Linux and don't require binary drivers, patched kernels, and other crap" Sure, there are lists of chipsets, but the actual stores don't list the chipset in particular products often, and the vendors often have multiple versions of the same card with different chipsets.
I think a lot of the problem is the actual hardware industry itself. 802.11b wasn't hard to get Linux support for, but because of the software controlled radio in 802.11g chipsets, it's a bit tricker legally.
And don't get me started on Bluetooth. I got a new phone which has it, and I'd love to buy a little USB Bluetooth dongle so I can play with it, but right now the main Linux Bluetooth page has been asked to take down their list of devices known to work under Linux, because someone in the Bluetooth SIG complained the devices weren't technically qualified. (link) What a load of crap! So instead of getting a dongle which might not work, I'm just not going to get one at all. Everyone loses.
PCMCIA Firewire card is marginally easier, but again, trying to track down and actual card for sale which matches the user-reported specs and models is pretty damn hard. I spent conservatively 3 hours online and in Fry's reading before I got a card which works great until you eject it and panic the kernel.
I guess where I'm going with this rant is that wireless security (in the non-Windows world) would probably be better if the "standards" followed went a bit deeper and were more open to allowing outsiders to confidently buy products. All I'm asking for is a label or a sticker on the box telling me what chipset and version the device uses. It's not hard, and it shouldn't be a secret. Anyone technically savvy to make a purchasing decision based on chipset is technically savvy to figure out what chipset is in a device once they've bought it and spread the word.
Wow... my first rant. Sorry about that....
I notice your account has a URL listed, and that URL has a resume. It's perfectly legal for me to comb through all your pages, cross-reference it with other sources. I could even respond by posting every bit of minute trivia about you on this page. And then you could check your website logs, and figure out my IP, which would tell you who my employer is. And if you dug through my posting history, you could likely figure out where I live, who I am, and post the exact same type of information about me.
But neither one of us is going to do that. It might technically be legal, but it's a really jerk move, and pretty much defines "ad hominium" attacks. Generally talking about someone's wife, children, personal life, etc is a very inciting tactic to take in an argument. When I was growing up, you could always tell when an argument between two kids was getting out of hand when someone started talking trash (especially true trash) about someone's family members.
I'd take it very personally if someone posted my wife's name and contact information in response to something I said. Even more so if it was something my business did policy-wise. And beleive me, it would be very hard to resist the urge to strike back at the poster using any means I had at my disposal.
Personal information which may or may not be present in Google's non-public servers is a completely separate issue.
Yes, you can use Google to track down a distressing amount of personal information about some people, but this is a function of the information being made available by third parties. Google just makes it easier to find all these sources quickly.
People that gripe about (or sue) Google based on their indexing "bad" things, need to step back and think of the Web as more of a library, with each page as a book. Google serves as a card catalog, helping you find the books that have the information you are interested in. If somebody goes to the library and looks up a bunch of personal information on you (which is possible, just slower) you don't get mad at the makers of the card catalog. Your anger should be directed first at the person who singled you out. Next, if the books contain something which shouldn't be public (unlike major stock sales, and other things from the article, which should be public) you ought to take it up with the author/publisher of the books.
cNet took a cheap shot at Google, and did it in a fairly childish way. The point they were trying to make is both obvious, and better made in a more mature fashion. That being said, I don't exactly think Google's response is "mature", but if they want to respond in kind, I don't blame them.
So, I have no personal stake in the requirements other than I'd like these people to be good enough to justify their access to the rather valuable section of the RF spectrum we've set aside for them. I'd like to think they might be able to put together a radio from components if needed.
Even more important, I like knowing that there is a group of people out their who can communicate over pretty much any channel, radio or not, using the simpliest possible code. There are times (albeit rare) when the only communication you have is banging on a pipe with a large wrench, or flashing a light. I highly doubt that "Radio Shack" has the components to build a decent long-range HAM setup these days, but pretty much anyone can figure out a way to transmit long and short pulses using stuff found anywhere.