The South Korean method sounds like a build-up or additive process, to create a small graphene sheet. The Manchester transistor sounds like a subtractive process, to cut electron channels out of the graphene mesh. So can't they do each in order, and start making prototype atom-scale CPU's now? I guess they haven't figured out how to make the subtractive process targeted and repeatable yet, but this new larger sheet size sounds like a good place to start cutting!
I strongly discourage building your network around a single server.
I couldn't agree more. The strongest reason to use *any* AD or LDAP implementation is support for live network replication and load-balanced clustering.
Keep your two old servers but install the new OS, make the new server the master node, and replicate to the two old servers as active cluster slaves. Spend the money on more hardware for cluster reliability, not on one big monster server. You can even hot-fail-over IPs, DNS names, and shared storage, for any services that require a consistent single network source.
Clusters are always better than behemoths, when networking.
I agree with parent, but felt the need to plug Novell eDirectory a bit more.
I run IT at a small business (just over 50 workstations now) and we now run Novell SLES+EOS+eDirectory, and it is well worth the low price, especially compared to equivalent Microsoft server+client license costs. Unlike Microsoft licenses, with Novell some support is included. I assume their Academic License prices are even lower than their Small Business License.
We paid Novell for a 5 server pack and that's it -- no per-client licenses required. We also use their GINA instead of Windows logins, and iFolder/WebDAV is far superior to SMB, so we replaced every part of what we used to use Samba/AD for with Novell products. We considered using pGINA, but support for that has waned recently. You can use WebDrive on Windows clients to letter-mount WebDAV folder shares, as a cheaper alternative to iFolder licenses. We still have a couple of Samba 3.x server installations around for legacy Windows hardware and laptops, but their use is declining. iFolder security controls are much tighter than any SMB mounts. iFolder and WebDAV are also much easier to serve over all forms of VPN tunnels.
Some of Novell's products are proprietary, but their standards support behaves well with other standards-based OSS projects. Almost all their products have direct OSS equivalents, but it sounds like paying a little bit more for Novell support might be cost effective for your environment.
Barring that, Fedora Directory Server is probably the next best choice. I've tried standard distributions of Samba 3.x and OpenLDAP, with some success in the past. OpenLDAP's default BDB back-end always manages to fail eventually, and frequently restoring is a pain. Getting both Apache and Samba to work with the same LDAP authentication groups can also take some tweaking, but PAM helps there.
the only appropriate response is to refuse to pay your taxes
How do you suggest I keep my company from taking it out of my paychecks, and stop submitting W2 or 1099 forms? Tax evasion is a choice limited to the extremely wealthy, or to people who derive all of their income from a black market. There's not a lot of options for us legal citizen laborers.
Except for all of the air-to-air missiles, which are UAVs of a sort. -jcr
I like this concept a lot. I'm imagining a hybrid approach, where a manned aircraft acts as a sort of command-and-control station for a cloud of UAV units, flying in formation, or deployed from bays as needed. UAV units could be AI or remote controlled, either from the airship, or relayed from ground. AI goals/actions could be guided by targeting interfaces in the manned craft.
We stop thinking of them as missiles and think of them as kamikaze-AI explosive nodes, evasive-AI chaff nodes, target-AI gun nodes, swarm-AI laser/radio jammer nodes, advanced-AI ground attack nodes, diver-AI bomb nodes, scanning-AI remote sensing nodes...
I can't get that Protoss battleship from Starcraft out of my head now.
I agree with the parent and almost all preceding comments above +3, but I feel the need to clarify a little bit.
Currently, the merits of higher education in the job field are:
1. Job listings often specify a particular degree as a MINIMUM requirement. So you got that one.
2. Completing a degree proves that: a. You can fool someone into accepting you into their program based on High School merits, which are universally flimsy.
b. You can make about 80% of your contrived deadlines, or more.
c. You took a test on relevant subject matter at some point, so interviewing you shouldn't be a complete waste of time.
d. Despite dealing with all the complex life questions that come about when leaving home for more than a couple of months, you managed to get something done, or at least fool your professors to their satisfaction.
e. You got over fending for yourself. Your new employers don't have to act as if they're your new parents.
3. Job experience you managed to pick up while paying your expenses. Hopefully by now your new employers don't have to show you how to fill out an employment eligibility form, or handle a checkbook.
4. You meet new school friends, many of whom are smarter than you, who can help you out.
Anything else your education establishment claims is B.S. Sometimes, really friendly professors that like your work can arrange contacts, especially if they decide to be your mentor. This is rare, because professors can only handle becoming a role model for a small fraction of their students, and usually that's because they are trying to push more undergraduates into higher degree programs, or underpaid academic work in general.
So your best bet, in general, is to concentrate on item 3. You can also compete on low price/hour.
If you managed to get all the way through a degree without ANY relevant work experience, that usually means you were a spoiled trust fund baby brat, or at least your parents are rich enough to pay off your major expenses. Such brats usually spend most of their degree program partying, cribbing off their smarter frat/sorority friends, or paying for cash-strapped smart people to do their work for them. Tests can be crammed sufficiently otherwise. In this case, you will be dependent on your Greek friends to arrange for jobs for you, or on nepotism of some form. This latter option also precludes needing to post a question on Slashdot.
Our desktop setups actually do sound rather similar. I have a dual-monitor setup at home, and I tend to treat each monitor like your dedicated "virtual desktop" scheme. I keep active work on the bigger monitor to the left, usually with something like Visual Studio in "full screen", and communication apps sort of cascaded on the right.
Let me ask you just one thing: after your preferred web browser gained tabs (this was a while back for me since I use Firefox almost exclusively, though I have dabbled with Chrome), did you stick to having any separate browser windows, or did you go all tab-per-page mode? The reason I ask this is, it's an existing proof of my point. Given a more intuitive method to browse between a list of full(er) size panes, as opposed to spending time on window arrangement (such as manual paneling or cascading), I suspect most people would rather have full size panes, or at the very least a finite pre-arrangement of panes (which is arguably the point of most frame UI).
In part because of screen real estate constraints, mobile UI developers are hitting those "intuition with full pane" spots much more quickly. I think it's inevitable that these will reach back to preferred desktop settings, so much so that I'm willing to jump the gun and call something like Android a viable desktop UI, not just a mobile UI. I'll also jump the gun in hoping some of the cool push-pull interface ideas in Fennec will come back to desktop Firefox at some point.
I also think touchscreen/stylus tablet use on both desktops and mobiles will become more common within the next 1-3 years, so perhaps that introduces some bias.
Multi-window operation. Again, the target is applications where the resolution, screen size, and interface methods do not lend themselves well for multiple windows.
This is where I have to begin to disagree with you. I know it's a hard habit to give up, but I think the multi-window desktop UI paradigm we've been force-fed since the 80's is vastly overrated. I don't know how you do your tasks, but I do mine one at a time, and the really important windows are kept full-screen. The rare exception to that might be filesystem browsing, but a multi-frame browser is a simpler solution there. Recent studies show that humans are not natively multi-tasking, and asking them to do so tends to slow them down, where single-tasking serially is faster.
Full-screen apps with some sort of switching/tabbing interface, and some form of communication layer/functions between apps (minimally, cut'n'paste, you hear me APPLE??) is sufficient to coordinate any 2 apps. Android's inclusion of things like the contact manager in multi-app communication makes this even better. Even COM hasn't caught up with that level of cross-app communication.
Sony has a good example of a nice persistent interface with per-task near-full-screen or full-screen interfaces, in their XMB for the PS3 and PSP. The PS3 XMB exceeds the kind of resolutions you're talking about. I prefer Android's "shelf" metaphor/UI to standard desktop shortcut based UI. Desktop widgets forming a "home" UI are also becoming fairly universal, and Android already supports that better than X11. With the existence of remote desktops like VNC, X11 remote windowing is also outdated and just confusing more than anything.
Ask anyone who has tried to develop games under X -- the mouse handling is the worst. Android is made for touch screens, and has been proven to support multi-touch, so its cursor/mouse manager is probably already better than X.
The world needs an alternative to the old mis-guided "everything in a window pane" UI. Android may be just the thing.
I still think what Linux needs is an iTunes/iTunes Store like system.
Lindows/Linspire Click'n'Run was pretty close to this. Too bad they didn't do well, but maybe Xandros will do something with it.
This may seem like an unrelated idea at first, but just hear me out: I think this is one area where Google's Android and DalvikVM can become of use.
I just read that someone recently ported Android completely to an Asus Eee netbook. Further, I think a DalvikVM app on an OS-independent stack, and related browser plug-in, could be developed. Such a plug-in could go head-to-head with Flash and Silverlight. Android/DalvikVM already supports OpenGL ES 2.0, SGL, a stable Media Framework, Audio, etc. It would be the perfect environment for a platform-independent game makers, for both browser apps and local installs. Port the Android Market to any system with the DalvikVM app, and there you have your market place, regardless of OS or distribution.
The other issue is the hardcore gamer market, who we either have to cater to, or convince them somehow to go through the work of setting up multi-boot systems. This is one rare area where VM managers haven't caught up, especially in graphic driver support. I think the main pressure there needs to be put on the graphic and audio card makers, for FOSS or at least re-compilable drivers, and so they stop hamstringing all but the Win32 drivers (even Win64 still has problems on that front).
I think you're wrong that banning things never work
I don't mean to imply that's globally true -- just that we have good evidence on the banning of drugs.
I really appreciate your market-centric perspective on this. I think the other commenters have a problem with your use of the word "ban", because they see it as a binary (all illegal or completely legal) proposition. The truth is that "ban" is just one end of the government regulation spectrum, with the opposite being "totally untaxed and unregulated."
I don't think any of the legalization effort has anything to do with moving ANY drugs to an unregulated market. History shows that it makes much more sense to put ALL drugs into the same kind of regulated market, such as under the FDA. Then drugs with certain society-harming qualities, such as second-hand smoke or impaired motor control, get some more regulations than others. Specialized drugs even require approval by a doctor and/or a pharmacy.
No one here is talking about deregulating drugs, which I find very promising, given the numbers of libertarians at Slashdot (with 'l's both Big and small). All we're talking about here is changing the nature of regulation for Type IV substances, from a "ban" that just results in a larger black market, to a regulatory environment which allows for a limited (controlled) market. The controlled market then becomes taxable, which can then fund any health problems and black market suppression which may result, via market-specific taxation.
Right now the black market suppression and health effects, for all Type IV classified substances, are being paid for by ALL tax payers. Wouldn't it be great if those taxes were only paid for by the participants in the specific Type IV substance market causing the problems? I think the right regulatory model is somewhere between cigarettes, alcohol, and over-the-counter medications -- with doctors and pharmacies getting involved at higher dosage levels. Existing regulatory structures are sufficient for Type IV substances. There's never been any need for this costly "war".
Are you happy about having your pocket picked to rehabilitate those who've turned themselves into potted plants of the sort that they smoke?
In California, judges usually assign rehab instead of jail time. The numbers now show that the rehab sentences both lower recidivism rate, and are much cheaper to the State than prison time. Therefore, rehab is costing me less in taxes than prison costs, so yes, I would rather pay for rehab. Of course, if you legalize it, and regulate proper doses, they can pay for their own rehab via special sales taxes. If you're worried about taxes, you should be pushing for legalization ASAP, regardless of unrelated health insurance issues.
Previous lists of winners vs. losers in the drug war here seem to ignore the one most important sentence from TFA:
Anti-prohibition voters "saw what most Americans still fail to see today: That a failed drug prohibition can cause greater harm than the drug it was intended to banish."
It is impossible, at this point in time, to judge the true harm of most illegal drugs, because you can't even be sure of their true composition, nor the true list of upstream market profiteers. Unregulated drug markets, including all existing drug black-markets, are free to put anything they want in their products, and get those ingredients from anywhere, and still label them "pure" or "home grown". Who is going to call them out for false advertising? We can't even catch China putting poison in our toothpaste and dog food! How are we going to catch the black market poison profiteers?
How many cases of drug related impairment, addiction, or deaths can be blamed on the common practice of "cutting", or "diluting" pure/concentrated (transport friendly) drugs with dangerous chemicals -- even known poisons? There is also the problem of dose-fixing, where the first dose is always intended to be addictive, and subsequent doses are dealt carefully, to maintain that addiction. What ad-hoc drug experiment case study here can claim that their study materials were not tainted, and that the dose was measured correctly, with 100% certainty? I know there are some pre-med/bio-chem students here who might raise their hands, but the vast majority of Americans know nothing of what they ingest, including in those rare cases when the FDA enforces proper labeling.
The only valid way for government to ever deal with a moral question is to study it, then regulate it, and tax the hell out of it. All other moral legislation is unethical -- government can only ever properly deal in ethical questions, and can never ethically deal in moral interpretations.
This is part of the reason for the Constitutional guarantee of the separation between Church and State. My Church has a different definition of "religious experience" and "prayer aid" than yours. Keep your damned moralizing legislation out of my personal religion.
Whatever you may think of government, they are far more accountable to tax payers than the monopoly corporations are. Their extortion "rents" are yet another form of corrupt industry, creating hidden taxes on everyone. Yes, their lobby power also corrupts government, but that form of corruption is already much more transparent than internal corporate corruption, and it can be corrected with the right amount of populist legislation, journalism, and public protest. You can protest AT&T directly all you want -- they only pay attention to their almighty "profit motive". At this point, only government has the power to correct their estimate of "profit", and to break their monopoly stranglehold.
The original rally cry of the American Revolution was "no taxation without representation." I don't have any representation at AT&T, so why should I be forced to pay their monopoly extortion taxes?
This doesn't fix the main problem: the local loop monopolies will still extort as much as possible, for the necessary connection between the "community fiber" and the "real Internet", at the metro level fiber. Stating that the "last mile" is the main cost driver, and that house tails will fix that, is complete monopoly-driven fantasy B.S. The primary cost will always be the local loop private monopoly extortion taxes.
Here's what we need: part of the future Network Neutrality Bill should provide the possibility for people to use their own homes, without restriction, as network hubs, and form peering agreements at will. No ISP that connects to them can restrict allowing them peering agreements with anyone else.
Adjoining lot neighbors then dig a few feet of pipe between their homes, on their own property, but meeting at an agreed point on their fence line. They peer with each other, and own equipment that will automatically load balance their bandwidth between a shared pool of ISP uplinks, based on cost of bandwidth, QoS, etc. Once a big enough neighborhood of adjacent >=100Mbps peers is formed, new ISP players will have less to worry about in the last-mile build-outs -- they will just have to find the nearest neighbor in the group with a manageable adjacent CO, or find one neighbor in the group willing to host their equipment onsite within range of an existing management point.
I still don't get why the streets/highway/interstate analogies don't get more penetration when talking about Internet governance. Please try to imagine how awful driving would be, if different private monopolies each owned a piece of your commute. That's exactly the situation we have with Internet connections today. Unfortunately, only government can really fix that. Private at-home peering with neighbors is just a grassroots way of starting in that direction. The bigger changes need to happen at the Interstate network level, through forced competition, neutrality fair-play rules, and other government policy acts.
Bigotry and ignorance are connected. Therefore, I would say 100% of the 23% ignorant Texans who actually believe the "secret Muslim" BS out loud are also bigots. And the 23% number only represents the subset of bigots that are also stupid enough to say such a thing out loud. I'm sure the actual percentage of bigots there is much higher, and probably constitutes a super-majority, based on past polling. They are probably so bigoted that the fact that Barack Obama is half black (for some, maybe "only half" black), is the real problem for them. Calling Obama a "secret Muslim" is just a more "politically correct" excuse for hatred in their closet-racist/open-religious environment.
Can the parent be modded up any higher? How about a new category, like "Perfect Post". I'm really glad I scrolled down far enough to see someone else who's actually read enough about Barack Obama, including his own words, to form a valid opinion here. That seems to be severely lacking here in America, even here in Slashdot, as evidenced by other posts in this same thread.
One of the biggest problems with American democracy is the perpetuation of the fallacy, that one can form a valid opinion on any subject, without any research into that subject. Most of the time, what you have is NOT a valid opinion -- it is merely an agreement with someone else whose opinion you may trust, good or bad. Research references are one thing, but wholesale agreement with a random "expert", whose background and research you are not familiar with, is just ignorance with a better label.
I can say one thing in the favor of quoting random "experts". Consistently agreeing with ignorant knee-jerk bigots, like certain Texas pastors, or Fox News, clearly identifies you as a bigot. I am familiar with a variety of bigot media, unfortunately because my own racist family members quote it to me all the time. So, when you quote the bigot media, even without proper attribution, I can easily identify you. Thanks for the self-labeling!
If you haven't done your own research, please don't vote. The other fallacy of American democracy is that every vote, no matter how backwards or ignorant the voter, should count equally. In other words, if you can fool most of the people, just long enough for a single voting period, you should win. You can clearly see the results of this fallacy in all elections involving George Herbert Walker Bush. Drink beer with him as he watches Dick, or now Palin, push the button. I'm sure you'll have a great time, with the last beer ever.
I've considered SquirrelMail myself, but it does lack a lot of the UI features that the Gmail UI provides us now.
I would say the #1 feature that is un-beaten in ANY other webmail service (including Zimbra, which is also a very nice AJAXy UI) is the whole folder=label instead of folder=FS-folder concept. I can not even attempt to count (without extensive search help) how many times I've set up filters to apply multiple labels to the same items, or just multi-labeled manually before Archiving (removing "Inbox" label). It may seem like a small thing, but just knowing that multi-organizing my e-mail is not possibly costing me any disk quota is quite liberating. More IMAP servers need to adopt this storage approach, and more e-mail clients need to make it as easy to use.
The mail search is also un-beatable right now, but I fully expect to be able to add Google search to any application (Thunderbird plug-in, desktop search, appliance-based extranet indexing, etc.) anytime I need to in the future, so maybe that's for another topic thread.
If you go with Google, make sure their proposal has phone support for administrative accounts.
From all my reading before going with the Google Apps free version, I'm pretty sure both the paid and academic versions include direct support access like what you're talking about -- probably limited phone but unlimited e-mail. I'm fairly certain they would reimburse any phone charges (their charges, not your cell cost) if e-mail were ever completely down. I've never run across that case with them, using their free version (more than I can say for several other ISPs I've paid).
Having been on the free (Groups and e-mail based) support for about a year and a half myself, I can say that it's not all that bad. After surpassing their "Beta Apps" stage, their e-mail turn around time is about on par with M$ Windows (paid via box license) e-mail support options, and most general not-just-me problems were answered very quickly through Google Groups. Their FAQs pretty clearly answer any non-temporary-failure related questions, which is much better than matching M$ support FAQs.
Google least lock-in? No way - they'll own your calendars, your email accounts, your social networking, your website if you let them.
I think the point here is that Google has the very least lock-in, because:
1. They use well documented open standards, which makes all your data far more portable than something proprietary, with poorly documented interfaces, like Exchange.
2. They don't stop you from using your own servers, either as gateways to the Google servers, or as alternative client-facing services.
I would say Yahoo! has a chance at having equally low lock-in, but they should really be considered an unknown right now, due to all the acquisition and partnership possibilities floating around, and related license changes.
Set up your own MX gateways as redundant access points, in case Google goes down (and your other servers don't) for some reason. Gmail servers will even act as a basic storage/forwarding server for free, where you can setup a distinct forwarding rule for every account.
I've been using Google Apps free editions for quite a while, with my own DNS servers. I can change servers any time, and just use some OSS tool or another to leech all the data via the IMAP, POP3, and/or iCal standard interfaces, any time I want. Nothing is really stopping me from closing down the Google Apps account, deleting all its data, and formally requesting that they scrub related account history.
I've been considering setting up a server to provide an MX gateway host, that saves all incoming mail, and forwards to Google servers. Then I might set up a Zimbra site or some other webmail service, to provide mail access if Gmail is ever down, even if only temporarily. I haven't felt the need yet, but I like having the option.
In a manner of speaking, the Google Apps solution allows so many storage, forwarding, and redirection options, that the option left out of this list is "Use all 3!" Go crazy and add an OSS-Exchange cross-over solution into the mix, like Binari. I'm not saying that is cost effective (especially with proprietary M$ in the mix), but it is an existing option.
I agree with making Math Teaching more financially rewarding, but it must also be personally rewarding. That means the basic curriculum, especially its order and progression, must be changed.
In grade school, after about second grade, I just "knew" math. I was "good" at it, even though I never found it very interesting, because the minor logical progression made "sense" to me. But it could never be exciting, because it was still a form of logic force-fed by rote. No matter what my teachers said, even about fun things about the relationship between statistics and gambling, they couldn't help me do anything but churn through the homework, and study only to the tests. A's were easy but boring.
I can't ever remember being *excited* about math at all, until my first 400 level college course. We went over Euclid's Elements, going through all the axioms, proofs, and corralaries in order. I "knew" the trigonometric and geometric principals involved before this class, but I didn't fully GROK it all. Math never felt like a process of discovery, before that class. Before it was all force-fed logic, and now it is a mountain to be conquered.
If ALL Math was taught the same way, in the same order it was discovered, from the beginning line, it would be a whole different subject. Sure, it might take a little longer to get to the "basics" like Algebra and Calculus, but it would be a lot more fun.
Nerds learn by rote. Adventurers learn by discovery.
Because "average" people don't want their leaders to treat them like serfs or proles or subjects or children.
Exactly! We would much rather be fooled by rich trust fund babies (especially those with a nice dynastic last name), or husbands of rich heiresses, into thinking that they're "just like us." We would never want to elect an academic "elitist" who attended public schools, while being raised with the help of his/her grandparents. We would never want to elect someone who actually had to EARN her/his way up, via hard work and scholarships. We can't elect anyone who reached to the TOP of their class, by their own hard work, values, and determination. That's just un-American!
Not only is the interest rate over 11%, but the Feds took a 79.9% equity stake in AIG. The US Taxpayer now owns 80 percent of AIG.
So they got to keep 20%, for only 11% down? The last time I checked, if a bank foreclosed on a mortgage, that meant they got 100% of the house. It sounds like AIG still got off easy, to me.
Slashdot Mirror
Isn't 1x1 cm already big enough to form an entire CPU chip when each transistor is only 1x10 atoms?
http://www.sciencedaily.com/releases/2008/04/080417142452.htm
The South Korean method sounds like a build-up or additive process, to create a small graphene sheet. The Manchester transistor sounds like a subtractive process, to cut electron channels out of the graphene mesh. So can't they do each in order, and start making prototype atom-scale CPU's now? I guess they haven't figured out how to make the subtractive process targeted and repeatable yet, but this new larger sheet size sounds like a good place to start cutting!
I strongly discourage building your network around a single server.
I couldn't agree more. The strongest reason to use *any* AD or LDAP implementation is support for live network replication and load-balanced clustering.
Keep your two old servers but install the new OS, make the new server the master node, and replicate to the two old servers as active cluster slaves. Spend the money on more hardware for cluster reliability, not on one big monster server. You can even hot-fail-over IPs, DNS names, and shared storage, for any services that require a consistent single network source.
Clusters are always better than behemoths, when networking.
I agree with parent, but felt the need to plug Novell eDirectory a bit more.
I run IT at a small business (just over 50 workstations now) and we now run Novell SLES+EOS+eDirectory, and it is well worth the low price, especially compared to equivalent Microsoft server+client license costs. Unlike Microsoft licenses, with Novell some support is included. I assume their Academic License prices are even lower than their Small Business License.
We paid Novell for a 5 server pack and that's it -- no per-client licenses required. We also use their GINA instead of Windows logins, and iFolder/WebDAV is far superior to SMB, so we replaced every part of what we used to use Samba/AD for with Novell products. We considered using pGINA, but support for that has waned recently. You can use WebDrive on Windows clients to letter-mount WebDAV folder shares, as a cheaper alternative to iFolder licenses. We still have a couple of Samba 3.x server installations around for legacy Windows hardware and laptops, but their use is declining. iFolder security controls are much tighter than any SMB mounts. iFolder and WebDAV are also much easier to serve over all forms of VPN tunnels.
Some of Novell's products are proprietary, but their standards support behaves well with other standards-based OSS projects. Almost all their products have direct OSS equivalents, but it sounds like paying a little bit more for Novell support might be cost effective for your environment.
Barring that, Fedora Directory Server is probably the next best choice. I've tried standard distributions of Samba 3.x and OpenLDAP, with some success in the past. OpenLDAP's default BDB back-end always manages to fail eventually, and frequently restoring is a pain. Getting both Apache and Samba to work with the same LDAP authentication groups can also take some tweaking, but PAM helps there.
the only appropriate response is to refuse to pay your taxes
How do you suggest I keep my company from taking it out of my paychecks, and stop submitting W2 or 1099 forms? Tax evasion is a choice limited to the extremely wealthy, or to people who derive all of their income from a black market. There's not a lot of options for us legal citizen laborers.
(If it's not legal, expect them to try to pretend it's legal.)
You forgot the primary option they usually use: if it's not legal yet, spend lobbyist resources on making it legal, and retroactive if possible.
'WARNING: Excessive exposure to news about inept politicians who blame media for all their problems, and corresponding news that they can't do something as simple as balance a budget, has been linked to aggressive behavior.'
Agressive thoughts, anyway...
Except for all of the air-to-air missiles, which are UAVs of a sort. -jcr
I like this concept a lot. I'm imagining a hybrid approach, where a manned aircraft acts as a sort of command-and-control station for a cloud of UAV units, flying in formation, or deployed from bays as needed. UAV units could be AI or remote controlled, either from the airship, or relayed from ground. AI goals/actions could be guided by targeting interfaces in the manned craft.
We stop thinking of them as missiles and think of them as kamikaze-AI explosive nodes, evasive-AI chaff nodes, target-AI gun nodes, swarm-AI laser/radio jammer nodes, advanced-AI ground attack nodes, diver-AI bomb nodes, scanning-AI remote sensing nodes...
I can't get that Protoss battleship from Starcraft out of my head now.
I agree with the parent and almost all preceding comments above +3, but I feel the need to clarify a little bit.
Currently, the merits of higher education in the job field are:
1. Job listings often specify a particular degree as a MINIMUM requirement. So you got that one.
2. Completing a degree proves that:
a. You can fool someone into accepting you into their program based on High School merits, which are universally flimsy.
b. You can make about 80% of your contrived deadlines, or more.
c. You took a test on relevant subject matter at some point, so interviewing you shouldn't be a complete waste of time.
d. Despite dealing with all the complex life questions that come about when leaving home for more than a couple of months, you managed to get something done, or at least fool your professors to their satisfaction.
e. You got over fending for yourself. Your new employers don't have to act as if they're your new parents.
3. Job experience you managed to pick up while paying your expenses. Hopefully by now your new employers don't have to show you how to fill out an employment eligibility form, or handle a checkbook.
4. You meet new school friends, many of whom are smarter than you, who can help you out.
Anything else your education establishment claims is B.S. Sometimes, really friendly professors that like your work can arrange contacts, especially if they decide to be your mentor. This is rare, because professors can only handle becoming a role model for a small fraction of their students, and usually that's because they are trying to push more undergraduates into higher degree programs, or underpaid academic work in general.
So your best bet, in general, is to concentrate on item 3. You can also compete on low price/hour.
If you managed to get all the way through a degree without ANY relevant work experience, that usually means you were a spoiled trust fund baby brat, or at least your parents are rich enough to pay off your major expenses. Such brats usually spend most of their degree program partying, cribbing off their smarter frat/sorority friends, or paying for cash-strapped smart people to do their work for them. Tests can be crammed sufficiently otherwise. In this case, you will be dependent on your Greek friends to arrange for jobs for you, or on nepotism of some form. This latter option also precludes needing to post a question on Slashdot.
Targeted blue lights may correlate with lowered suicide rates:
http://www.physorg.com/news148153021.html
However, streetlights in general have not been proven to prevent any crime:
http://www.delscope.demon.co.uk/information/lightpollution.htm#security
http://amper.ped.muni.cz/light/crime/lp040_1h.html
Our desktop setups actually do sound rather similar. I have a dual-monitor setup at home, and I tend to treat each monitor like your dedicated "virtual desktop" scheme. I keep active work on the bigger monitor to the left, usually with something like Visual Studio in "full screen", and communication apps sort of cascaded on the right.
Let me ask you just one thing: after your preferred web browser gained tabs (this was a while back for me since I use Firefox almost exclusively, though I have dabbled with Chrome), did you stick to having any separate browser windows, or did you go all tab-per-page mode? The reason I ask this is, it's an existing proof of my point. Given a more intuitive method to browse between a list of full(er) size panes, as opposed to spending time on window arrangement (such as manual paneling or cascading), I suspect most people would rather have full size panes, or at the very least a finite pre-arrangement of panes (which is arguably the point of most frame UI).
In part because of screen real estate constraints, mobile UI developers are hitting those "intuition with full pane" spots much more quickly. I think it's inevitable that these will reach back to preferred desktop settings, so much so that I'm willing to jump the gun and call something like Android a viable desktop UI, not just a mobile UI. I'll also jump the gun in hoping some of the cool push-pull interface ideas in Fennec will come back to desktop Firefox at some point.
I also think touchscreen/stylus tablet use on both desktops and mobiles will become more common within the next 1-3 years, so perhaps that introduces some bias.
Multi-window operation. Again, the target is applications where the resolution, screen size, and interface methods do not lend themselves well for multiple windows.
This is where I have to begin to disagree with you. I know it's a hard habit to give up, but I think the multi-window desktop UI paradigm we've been force-fed since the 80's is vastly overrated. I don't know how you do your tasks, but I do mine one at a time, and the really important windows are kept full-screen. The rare exception to that might be filesystem browsing, but a multi-frame browser is a simpler solution there. Recent studies show that humans are not natively multi-tasking, and asking them to do so tends to slow them down, where single-tasking serially is faster.
Full-screen apps with some sort of switching/tabbing interface, and some form of communication layer/functions between apps (minimally, cut'n'paste, you hear me APPLE??) is sufficient to coordinate any 2 apps. Android's inclusion of things like the contact manager in multi-app communication makes this even better. Even COM hasn't caught up with that level of cross-app communication.
Sony has a good example of a nice persistent interface with per-task near-full-screen or full-screen interfaces, in their XMB for the PS3 and PSP. The PS3 XMB exceeds the kind of resolutions you're talking about. I prefer Android's "shelf" metaphor/UI to standard desktop shortcut based UI. Desktop widgets forming a "home" UI are also becoming fairly universal, and Android already supports that better than X11. With the existence of remote desktops like VNC, X11 remote windowing is also outdated and just confusing more than anything.
Ask anyone who has tried to develop games under X -- the mouse handling is the worst. Android is made for touch screens, and has been proven to support multi-touch, so its cursor/mouse manager is probably already better than X.
The world needs an alternative to the old mis-guided "everything in a window pane" UI. Android may be just the thing.
I still think what Linux needs is an iTunes/iTunes Store like system.
Lindows/Linspire Click'n'Run was pretty close to this. Too bad they didn't do well, but maybe Xandros will do something with it.
This may seem like an unrelated idea at first, but just hear me out: I think this is one area where Google's Android and DalvikVM can become of use.
I just read that someone recently ported Android completely to an Asus Eee netbook. Further, I think a DalvikVM app on an OS-independent stack, and related browser plug-in, could be developed. Such a plug-in could go head-to-head with Flash and Silverlight. Android/DalvikVM already supports OpenGL ES 2.0, SGL, a stable Media Framework, Audio, etc. It would be the perfect environment for a platform-independent game makers, for both browser apps and local installs. Port the Android Market to any system with the DalvikVM app, and there you have your market place, regardless of OS or distribution.
The other issue is the hardcore gamer market, who we either have to cater to, or convince them somehow to go through the work of setting up multi-boot systems. This is one rare area where VM managers haven't caught up, especially in graphic driver support. I think the main pressure there needs to be put on the graphic and audio card makers, for FOSS or at least re-compilable drivers, and so they stop hamstringing all but the Win32 drivers (even Win64 still has problems on that front).
I think you're wrong that banning things never work
I don't mean to imply that's globally true -- just that we have good evidence on the banning of drugs.
I really appreciate your market-centric perspective on this. I think the other commenters have a problem with your use of the word "ban", because they see it as a binary (all illegal or completely legal) proposition. The truth is that "ban" is just one end of the government regulation spectrum, with the opposite being "totally untaxed and unregulated."
I don't think any of the legalization effort has anything to do with moving ANY drugs to an unregulated market. History shows that it makes much more sense to put ALL drugs into the same kind of regulated market, such as under the FDA. Then drugs with certain society-harming qualities, such as second-hand smoke or impaired motor control, get some more regulations than others. Specialized drugs even require approval by a doctor and/or a pharmacy.
No one here is talking about deregulating drugs, which I find very promising, given the numbers of libertarians at Slashdot (with 'l's both Big and small). All we're talking about here is changing the nature of regulation for Type IV substances, from a "ban" that just results in a larger black market, to a regulatory environment which allows for a limited (controlled) market. The controlled market then becomes taxable, which can then fund any health problems and black market suppression which may result, via market-specific taxation.
Right now the black market suppression and health effects, for all Type IV classified substances, are being paid for by ALL tax payers. Wouldn't it be great if those taxes were only paid for by the participants in the specific Type IV substance market causing the problems? I think the right regulatory model is somewhere between cigarettes, alcohol, and over-the-counter medications -- with doctors and pharmacies getting involved at higher dosage levels. Existing regulatory structures are sufficient for Type IV substances. There's never been any need for this costly "war".
Are you happy about having your pocket picked to rehabilitate those who've turned themselves into potted plants of the sort that they smoke?
In California, judges usually assign rehab instead of jail time. The numbers now show that the rehab sentences both lower recidivism rate, and are much cheaper to the State than prison time. Therefore, rehab is costing me less in taxes than prison costs, so yes, I would rather pay for rehab. Of course, if you legalize it, and regulate proper doses, they can pay for their own rehab via special sales taxes. If you're worried about taxes, you should be pushing for legalization ASAP, regardless of unrelated health insurance issues.
Previous lists of winners vs. losers in the drug war here seem to ignore the one most important sentence from TFA:
Anti-prohibition voters "saw what most Americans still fail to see today: That a failed drug prohibition can cause greater harm than the drug it was intended to banish."
It is impossible, at this point in time, to judge the true harm of most illegal drugs, because you can't even be sure of their true composition, nor the true list of upstream market profiteers. Unregulated drug markets, including all existing drug black-markets, are free to put anything they want in their products, and get those ingredients from anywhere, and still label them "pure" or "home grown". Who is going to call them out for false advertising? We can't even catch China putting poison in our toothpaste and dog food! How are we going to catch the black market poison profiteers?
How many cases of drug related impairment, addiction, or deaths can be blamed on the common practice of "cutting", or "diluting" pure/concentrated (transport friendly) drugs with dangerous chemicals -- even known poisons? There is also the problem of dose-fixing, where the first dose is always intended to be addictive, and subsequent doses are dealt carefully, to maintain that addiction. What ad-hoc drug experiment case study here can claim that their study materials were not tainted, and that the dose was measured correctly, with 100% certainty? I know there are some pre-med/bio-chem students here who might raise their hands, but the vast majority of Americans know nothing of what they ingest, including in those rare cases when the FDA enforces proper labeling.
The only valid way for government to ever deal with a moral question is to study it, then regulate it, and tax the hell out of it. All other moral legislation is unethical -- government can only ever properly deal in ethical questions, and can never ethically deal in moral interpretations.
This is part of the reason for the Constitutional guarantee of the separation between Church and State. My Church has a different definition of "religious experience" and "prayer aid" than yours. Keep your damned moralizing legislation out of my personal religion.
Whatever you may think of government, they are far more accountable to tax payers than the monopoly corporations are. Their extortion "rents" are yet another form of corrupt industry, creating hidden taxes on everyone. Yes, their lobby power also corrupts government, but that form of corruption is already much more transparent than internal corporate corruption, and it can be corrected with the right amount of populist legislation, journalism, and public protest. You can protest AT&T directly all you want -- they only pay attention to their almighty "profit motive". At this point, only government has the power to correct their estimate of "profit", and to break their monopoly stranglehold.
The original rally cry of the American Revolution was "no taxation without representation." I don't have any representation at AT&T, so why should I be forced to pay their monopoly extortion taxes?
This doesn't fix the main problem: the local loop monopolies will still extort as much as possible, for the necessary connection between the "community fiber" and the "real Internet", at the metro level fiber. Stating that the "last mile" is the main cost driver, and that house tails will fix that, is complete monopoly-driven fantasy B.S. The primary cost will always be the local loop private monopoly extortion taxes.
Here's what we need: part of the future Network Neutrality Bill should provide the possibility for people to use their own homes, without restriction, as network hubs, and form peering agreements at will. No ISP that connects to them can restrict allowing them peering agreements with anyone else.
Adjoining lot neighbors then dig a few feet of pipe between their homes, on their own property, but meeting at an agreed point on their fence line. They peer with each other, and own equipment that will automatically load balance their bandwidth between a shared pool of ISP uplinks, based on cost of bandwidth, QoS, etc. Once a big enough neighborhood of adjacent >=100Mbps peers is formed, new ISP players will have less to worry about in the last-mile build-outs -- they will just have to find the nearest neighbor in the group with a manageable adjacent CO, or find one neighbor in the group willing to host their equipment onsite within range of an existing management point.
I still don't get why the streets/highway/interstate analogies don't get more penetration when talking about Internet governance. Please try to imagine how awful driving would be, if different private monopolies each owned a piece of your commute. That's exactly the situation we have with Internet connections today. Unfortunately, only government can really fix that. Private at-home peering with neighbors is just a grassroots way of starting in that direction. The bigger changes need to happen at the Interstate network level, through forced competition, neutrality fair-play rules, and other government policy acts.
Bigotry and ignorance are connected. Therefore, I would say 100% of the 23% ignorant Texans who actually believe the "secret Muslim" BS out loud are also bigots. And the 23% number only represents the subset of bigots that are also stupid enough to say such a thing out loud. I'm sure the actual percentage of bigots there is much higher, and probably constitutes a super-majority, based on past polling. They are probably so bigoted that the fact that Barack Obama is half black (for some, maybe "only half" black), is the real problem for them. Calling Obama a "secret Muslim" is just a more "politically correct" excuse for hatred in their closet-racist/open-religious environment.
Can the parent be modded up any higher? How about a new category, like "Perfect Post". I'm really glad I scrolled down far enough to see someone else who's actually read enough about Barack Obama, including his own words, to form a valid opinion here. That seems to be severely lacking here in America, even here in Slashdot, as evidenced by other posts in this same thread.
One of the biggest problems with American democracy is the perpetuation of the fallacy, that one can form a valid opinion on any subject, without any research into that subject. Most of the time, what you have is NOT a valid opinion -- it is merely an agreement with someone else whose opinion you may trust, good or bad. Research references are one thing, but wholesale agreement with a random "expert", whose background and research you are not familiar with, is just ignorance with a better label.
I can say one thing in the favor of quoting random "experts". Consistently agreeing with ignorant knee-jerk bigots, like certain Texas pastors, or Fox News, clearly identifies you as a bigot. I am familiar with a variety of bigot media, unfortunately because my own racist family members quote it to me all the time. So, when you quote the bigot media, even without proper attribution, I can easily identify you. Thanks for the self-labeling!
If you haven't done your own research, please don't vote. The other fallacy of American democracy is that every vote, no matter how backwards or ignorant the voter, should count equally. In other words, if you can fool most of the people, just long enough for a single voting period, you should win. You can clearly see the results of this fallacy in all elections involving George Herbert Walker Bush. Drink beer with him as he watches Dick, or now Palin, push the button. I'm sure you'll have a great time, with the last beer ever.
I've considered SquirrelMail myself, but it does lack a lot of the UI features that the Gmail UI provides us now.
I would say the #1 feature that is un-beaten in ANY other webmail service (including Zimbra, which is also a very nice AJAXy UI) is the whole folder=label instead of folder=FS-folder concept. I can not even attempt to count (without extensive search help) how many times I've set up filters to apply multiple labels to the same items, or just multi-labeled manually before Archiving (removing "Inbox" label). It may seem like a small thing, but just knowing that multi-organizing my e-mail is not possibly costing me any disk quota is quite liberating. More IMAP servers need to adopt this storage approach, and more e-mail clients need to make it as easy to use.
The mail search is also un-beatable right now, but I fully expect to be able to add Google search to any application (Thunderbird plug-in, desktop search, appliance-based extranet indexing, etc.) anytime I need to in the future, so maybe that's for another topic thread.
If you go with Google, make sure their proposal has phone support for administrative accounts.
From all my reading before going with the Google Apps free version, I'm pretty sure both the paid and academic versions include direct support access like what you're talking about -- probably limited phone but unlimited e-mail. I'm fairly certain they would reimburse any phone charges (their charges, not your cell cost) if e-mail were ever completely down. I've never run across that case with them, using their free version (more than I can say for several other ISPs I've paid).
Having been on the free (Groups and e-mail based) support for about a year and a half myself, I can say that it's not all that bad. After surpassing their "Beta Apps" stage, their e-mail turn around time is about on par with M$ Windows (paid via box license) e-mail support options, and most general not-just-me problems were answered very quickly through Google Groups. Their FAQs pretty clearly answer any non-temporary-failure related questions, which is much better than matching M$ support FAQs.
Google least lock-in? No way - they'll own your calendars, your email accounts, your social networking, your website if you let them.
I think the point here is that Google has the very least lock-in, because:
1. They use well documented open standards, which makes all your data far more portable than something proprietary, with poorly documented interfaces, like Exchange.
2. They don't stop you from using your own servers, either as gateways to the Google servers, or as alternative client-facing services.
I would say Yahoo! has a chance at having equally low lock-in, but they should really be considered an unknown right now, due to all the acquisition and partnership possibilities floating around, and related license changes.
Set up your own MX gateways as redundant access points, in case Google goes down (and your other servers don't) for some reason. Gmail servers will even act as a basic storage/forwarding server for free, where you can setup a distinct forwarding rule for every account.
I've been using Google Apps free editions for quite a while, with my own DNS servers. I can change servers any time, and just use some OSS tool or another to leech all the data via the IMAP, POP3, and/or iCal standard interfaces, any time I want. Nothing is really stopping me from closing down the Google Apps account, deleting all its data, and formally requesting that they scrub related account history.
I've been considering setting up a server to provide an MX gateway host, that saves all incoming mail, and forwards to Google servers. Then I might set up a Zimbra site or some other webmail service, to provide mail access if Gmail is ever down, even if only temporarily. I haven't felt the need yet, but I like having the option.
In a manner of speaking, the Google Apps solution allows so many storage, forwarding, and redirection options, that the option left out of this list is "Use all 3!" Go crazy and add an OSS-Exchange cross-over solution into the mix, like Binari. I'm not saying that is cost effective (especially with proprietary M$ in the mix), but it is an existing option.
I agree with making Math Teaching more financially rewarding, but it must also be personally rewarding. That means the basic curriculum, especially its order and progression, must be changed.
In grade school, after about second grade, I just "knew" math. I was "good" at it, even though I never found it very interesting, because the minor logical progression made "sense" to me. But it could never be exciting, because it was still a form of logic force-fed by rote. No matter what my teachers said, even about fun things about the relationship between statistics and gambling, they couldn't help me do anything but churn through the homework, and study only to the tests. A's were easy but boring.
I can't ever remember being *excited* about math at all, until my first 400 level college course. We went over Euclid's Elements, going through all the axioms, proofs, and corralaries in order. I "knew" the trigonometric and geometric principals involved before this class, but I didn't fully GROK it all. Math never felt like a process of discovery, before that class. Before it was all force-fed logic, and now it is a mountain to be conquered.
If ALL Math was taught the same way, in the same order it was discovered, from the beginning line, it would be a whole different subject. Sure, it might take a little longer to get to the "basics" like Algebra and Calculus, but it would be a lot more fun.
Nerds learn by rote. Adventurers learn by discovery.
Because "average" people don't want their leaders to treat them like serfs or proles or subjects or children.
Exactly! We would much rather be fooled by rich trust fund babies (especially those with a nice dynastic last name), or husbands of rich heiresses, into thinking that they're "just like us." We would never want to elect an academic "elitist" who attended public schools, while being raised with the help of his/her grandparents. We would never want to elect someone who actually had to EARN her/his way up, via hard work and scholarships. We can't elect anyone who reached to the TOP of their class, by their own hard work, values, and determination. That's just un-American!
Not only is the interest rate over 11%, but the Feds took a 79.9% equity stake in AIG. The US Taxpayer now owns 80 percent of AIG.
So they got to keep 20%, for only 11% down? The last time I checked, if a bank foreclosed on a mortgage, that meant they got 100% of the house. It sounds like AIG still got off easy, to me.