If you've ever looked at the patchwork of sales tax rules in this country you can quickly see this solves a major problem. There are literally 10's of thousands of sales tax jurisdictions in the US, pretty much every county at a minimum, and often each city or town inside of the county. It's not just different rates, but also different rules. Food is taxed in one place, taxed at a different rate in another, and not taxed in a third. And what is "food"? You don't really want to know the answer to that question, it's probably 10,000 pages long! Having one rate, or perhaps one national base rate and a per state "option"; but more importantly one set of classification rules would really solve a major hurdle for small online retailers, and make it practical for them to collect tax.
The largest problem this creates is who gets the revenue? Taxes generally pay for infrastructure (roads, bridges, fire departments, etc), so it makes sense for some of the revenue to go where the seller is located, and some where the buyer is located. In brick and mortar sales these tend to be the same place, but won't be for Internet sales. Plus, Internet sales depend on transportation. The goods are shipped by truck and rail, probably across many states in the middle. Some of the money needs to go to those states to build interstates and airports and rail yards to get the goods from seller to buyer.
There are some other small problems. For instance if the money is collected and distributed via the fed, can it be used as a stick to get the states to do other things? The tax may be regressive, depending on how it is implemented. Many localities exempt food for instance as a means of assisting the poor, squaring those rules into a national set of rules will be difficult.
Still, overall I think the country needs something like this to happen. The idea that we can collect no taxes on a significant fraction of the business activity is just crazy. Many other countries already have a VAT tax because of issues like this, so the US is really falling behind. No one likes taxes, but we all like the things taxes achieve (on some level), so let's find the simplest, least evil, and fairest way to collect them. Going from 10,000+ sets of rules and rates down to 1 would be a huge step.
If you live in a large city there will be one or more companies that do data destruction for corporations in your city that need to meet various standards for contracts they are working on, often for the government. Most of this will be bulk paper shredding and the like, but most will also do hard drives. Many will let you show up at the site and watch your hard drive go in the machine and get spit out in shreds, if you care. They are all certified (I forget the standard), they will bar code the drive, inventory it, and then shred it, and give you a certificate with the inventory numbers and serials confirming they were destroyed. The service is cheap if you don't make them come pick them up, I think around $10 a drive for single quantities.
However, I'm mostly with the other folks in this thread. If you write over with zeros in a pass or two pretty much anyone except the NSA won't stand a chance of recovering any data. Even someone like the NSA would only do it if there was no other way, due to the cost of getting that data. Basically, no one cares about you enough, unless you're like a major drug kingpin or something on the side.
I degauss mine, only because work has a machine (from years ago) so it's "free" and only takes a couple of minutes...but I also only do it on dead drives. If I have a drive I'm just not using right now I write zeros to it and store it to be used as a scratch drive. Only after it won't spin up anymore does it get degaussed.
The other DIY method mentioned is to drill a quarter inch hole in the platters with a drill press. Pro data recovery companies could get some data off for lots (10's of thousands) of money unless you drill like 20 holes it in, but it would keep any non-pro users from reading the drive.
Overwrite them and be happy. Shred them and know the stuff that comes out the back end at least gets recycled. It's all about your paranoia level.
I've long thought the best thing for health care in this country would be for the law to be struck down. Too many people in this down economy already like provisions of it (no pre-existing conditions, keeping kids on your insurance longer). Were it to be unconstitutional I think there would be a large swell of folks pushing for them to find some way to re-establish the law and make it constitutional.
Single payer becomes the obvious choice. It may be that the way to single payer is to lose in the Supreme Court.
The reason this post is stupid is that infrastructure is long term. When you go to the trouble of sending out a crew to dig up and put fiber in the ground your putting in an infrastructure asset that should have a 15-30 year lifespan. The fact that can average machine can't saturate it today means we're being forward thinking.
Like many tech-savvy home users I want a file server at home,but the reality is 95% of the time it is doing nothing. Be it noise (fan less), eco-friendly (use less electricity) or room friendly (make less heat) a low powered fan less system seems to be a great idea.
But it's damn hard to find. There seems to be an assumption that low powered and fan less means you want super-small. Pico, or mini-ITX. No one seems to think you would use it for a file server, so > 2 SATA cuts your choices by 50%, and > 4 almost doesn't exist.
I'd be ok with CPU's with speed step that could step _way_ down, like down to 5W, but they don't seem to exist. No one makes motherboards with Intel mobile processors for end users. Repurposing a laptop has other drawbacks.
I think it's an untapped market...perhaps a small one, but big enough more solutions should exist.
I generally agree with the poster, the pilot is at the controls, so I think taking away his water bottle and pocket knife are silly. The only thing that gives me some pause is the FedEx 705 hijacking. I find a lot of people are not aware of this event, but basically a FedEx pilot scheduled a jumpseat (i.e. he wasn't flying, just riding along) trip and proceeded to attack the air crew with a hammer. When you look at this incident, you wonder if he wasn't able to take a hammer on the plane how the outcome might have been different.
At the same time, the attempt was a failure. In effect, the system worked. Since that time, FedEx has all but banned jumpseat trips, which I think may be an overreaction.
I think the real problem here is that the traveling public is holding the airlines to an unobtainable standard. Cars kill thousands each year. Busses less, but still some people every year. Trains less, but still some. Planes, many years, kill no one. When they finally do have an incident the reaction is totally disproportionate for some reason. Why we expect planes to never be hijacked or crashed, but are ok with 40,000 people a year dying in auto accidents makes no sense to me. Even if removing the TSA added 100 deaths a year by plane (which I think is unlikely), it would still be the safest mode of transport, and I'd gladly take that trade off to not be groped each time I fly, and to be able to walk onto a plane 10 minutes before it leaves like used to be possible with the shuttles.
I read the headline, and I'm like "Flare", Sun never had a server called Flare. Fire, sure, did they typo that, or what?
Then I open it to see this is about The Sun Unleashing Most Powerful Solar Flare Since 2006. There, a much more useful headline. That wasn't hard, was it?
I came here to say this another way around, if all of your developers were Ken Thompson level folks, you wouldn't need formal code reviews.
Excellent developers know what they are doing, research alternatives, run tests, simulations, and performance analysis, and submit awesome code. Someone reviewing it is unlikely to make improvements by reading it, but rather would have to study the problem in depth to improve a solution.
If your reviewer can simply read the code and point out errors or problems you probably should fire the person who wrote it. If they aren't good enough, and can't do a complete enough job to pass a visual inspection you have no chance of turning out high quality software.
Code reviews should be less frequent, and more in depth. Have a programmer present on a section of code after he's figured out all the details. Describe the alternatives he investigated, and why this worked so well. After an hour presentation then his peers might be up to speed enough to make substantial comments on the code.
I wonder if this won't end up being another Airbus moment. I remember one of the early Airbus planes that failed at the air show (A300 prototype?). While the cause is too complicated to detail here, part of the issue was that Airbus believes the computers are highly accurate, and in many cases if there seems to be a discrepancy between what the pilot wants and what the computer wants, it has the computer win.
Now, if we assume the sensors stopped sending good data for any reason the computers had bad inputs. The pilots may have been trying to do the right thing and the computer may have overridden part of their inputs.
One thing I'm surprised hasn't been done yet is to simulate a frozen pitot tube. Fly a simulator in the same simulated conditions and make the pitot tube reading go bad. What do the computers do with that bad input?
I seriously hope this doesn't end up being a computer issue, but as programmers like to say Garbage In, Garbage Out. The fact that in some cases the computers can override pilot inputs is one of the few things that make me nervous about Airbus planes.
I love to talk with the managers/owners of restaurants, and have reviewed plenty on Yelp and Urbanspoon. Of course they are always very concerned about their bad reviews, and looking for ways to make the bad reviews go away. My advice to them is always the same. Leave those people alone, they are already unhappy. Rather, get your happy customers to leave good reviews and drown them out.
I've encouraged several businesses to pay the $5 for a Yelp or Urbanspoon sticker to put on their door. To claim their owner pages, and use them to post specials and send updates to regulators. To drop a reward on their "duke" on Yelp. You know what? In 2-3 months they easily amass tens or hundreds of positive reviews. Now the situation is 150 good reviews to the one bad one they were worried about, and their reputation is just fine.
The only reason these are such a big deal for most businesses is that people who feel they have been wronged are more likely to speak up. If you have only one review and it's bad, well, you look like a bad business. You will never satisfy 100% of your customers, so just get the 99% that you do to drown them out.
Easy, cheap, and builds loyalty with your regulars. Plus, you now have great reviews, so when people visit the area or move there and have nothing to go on but the reviews you'll be one of the first they try.
I use a good old pre-digital hanging lateral file. Hanging filing cabinets can be quite expensive, but a good one with ball bearing slides does the job quite well. A lateral 2-drawer is excellent because you can do letter sized in one drawer, and legal sized in the other.
But the real question is what to keep and why. You want to make it easy to get rid of what you don't need to keep anymore.
Bills. Best thing, make 12 folders, one for each month. At the start of the month shred the year old bills. You want a year of bills so you can check your yearly costs, seasonal trends, or in some markets you may have to detail electrical or gas costs to sell your hose quickly. More than a year isn't necessary.
Bank statements, 401k statements, pay stubs etc. 7 years. Basically financial stuff should be for 7 years. The IRS generally won't expect you to have older paperwork.
Tax returns. This should be the 7 year rule, but each year after I'm done I put that entire year in a folder and keep it forever. It's a small amount of space, and will provide a treasure trove of documents to shock your grand kids with when you are 80.
Home sales paperwork. I also keep this forever. At a minimum you need it as long as you have a mortgage, IMHO.
Warranties. As long as they last, which may be a long time for some things.
Lastly, things permanently installed in a house (new AC units, stoves, etc) should have all their stuff dropped in a drawer in the kitchen. Leave them for the next owner. They will love you.
Now, that you know what to keep you need to get rid of everything else, using a good shredder. Do yourself a favor and go ahead and spring for a quality "small business" unit in the $250-$400 range. The cheap home ones will choke when you try and put your junk mail in them and break after a few years. Your small business one will swallow those credit card offers that come in the mail without even opening, and easily eat your month of bills in one feed once a month.
I would not trust the government to be my only mail provider.
I do think it might be nice to have a walled garden government account. Every citizen gets an account, only government agencies can use it for official business. No spam. Secured. The government today won't do business for most things over e-mail because it goes over the big bad internet, and this could change it, and make a lot of things faster and more efficient. Since it's correspondence with the government in the first place I don't care if they can snoop on it.
Why, yes indeed. I worked in industry for many years, and I can tell you that no workers were more highly valued than those who were unable to do even the simplest things by themselves. "Let's collaborate!" they would say, and our hearts warmed instantly and we leapt into action, "helping" our valued coworkers, doing their work for them. In contrast, those with highly valuable skillsets, able to quickly solve difficult problems, those were as dirt to us. "Be off with you!" we'd cry, "and never dare to cross our path again!" Yes, as sweatyboatman says, nothing is more valuable in the real world than incompetence!
I see you understand the path to management, and more respect and money.
...can't they pipe real time sensor data into the internet, and let the gamers directly control the countermeasures? Why have a trained, staffed military when you can just start a new "battle" online that contains real data, rather than computer made up data for once? The gamers never even need to know!
...is the assumption that users want the code, and can make use of it.
Samba (by choosing GPLv3) wants to require Apple (in this case) to provide source code to Samba to anyone who uses OSX. The fundamental assumption in that paragraph is that a user can do something useful with the code.
My grandmother can't. Even if Apple shipped the full source on each CD it would make her no more free or less free, no more self sufficient or dependent. It would be about as useful as giving her a book written in Sanskrit.
But wait the GPL folks will argue, sure, most people won't be able to use it, but what about the programmers? Well, I'm a programmer. I want my Mac to read and write windows file shares. If it doesn't work, I'm going to open a ticket with Apple and make them fix it, even if the code is right there on my computer. The GPL folks fail to release that in many cases working and bug free trump the ability to modify; at least in the real world.
In a standard commercial world what the GPLv3 does is not create more freedom. It creates a burden on companies developing GPLv3 software to set up an entire distribution system for the benefit of a fraction of their user base. Like well under 1%. Probably under.1%. Maybe under.01% for a lot of products. When you have a million slaves it is hollow to tell them they are "more free" because you managed to free one of them.
But the real kicker here is that Samba is still free. Any OSX user can download the code, compile it on their OSX machine, and use it. If GPLv3 makes any sense it is on something like the kernel, or device drivers where various proprietary magic is required to make things go, and without the company distributing that information you really are screwed. To require a company to redistribute source to a user land program on a POSIX OS really is just lunacy. Anyone who wants more recent Samba than ships with OSX is already downloading it from Samba!
The BSD folks started to get it more right, and then backed off. The original 4 clause BSD license, with the so-called "advertising" clause was close to a good idea. The problem with it was that it really said advertising. Like if Apple made a TV commercial they would need to have some announcer read the names of every open source package included in OSX. The solution of simply removing it wasn't good either.
What was needed was a slightly different clause, requiring the user to be notified that they are using BSD licensed software, and where to find more information. One can imagine the MOTD on a brand new Unix system saying something like "This system contains BSD licensed software, for more information see/usr/share/bsd.txt. The file would contain not only the license, but the URL of each project involved, the name of the software, and version installed. The user is now aware of the open source project, and knows where to get the source code. That would have done more to promote open source software than either removing the various clauses (the BSD "solution"), or adding onerous provisions (the GPL "solution").
At the end of the day though, it's technical folks trying to solve a business problem they don't understand and don't think should exist. They want to oversimplify others businesses, and want to believe that every business should be run the way they want. Every company should just release all source, instructions on how to compile it yourself, and so on. That's straight up idealism with a healthy dose of ignorance. In going for a 100% solution, the open source world keeps turning down a 95% solution, and ending up with a 5% solution. A pretty stupid trade off, if you ask me.
There is one up side. I've never found Samba particularly stable, and so if Apple has to re-write it I have every faith they will actually do a better job, and give me one less reason to use Samba.
I have a number of applications where I want a low powered "desktop" form factor. That probably means Mini-ITX or something like that. The canonical example? A home file server. It's not in use 90% of the time, and I'd like my power bill to go down and the heat load to go down. A chip with a super low power standby mode would be nice.
Unfortunately Intel chips don't get that power sippy even when speed-stepped down. VIA makes some semi-interesting chips, but they seem to be integrated with a bunch of features I don't want. I understand the HTPC market is big, but I don't need HDMI out on my file server. What's more, even providing two SATA ports seems to be the exception, not the rule.
I think an ARM "server board" in a mini-ITX factor would sell well. 6xSata, GigaBit Ethernet (even if it can't quite fill it), lots of DIMM slots so you can buy lower density cheaper DIMMS, but still keep lots of cache in memory for things like file servers.
Many years ago Vocational Technical schools churned out welders, plumbers, electricians, and all sorts of other skilled trades by the boatload. Not everyone was cut out to be a white collar employee, and so if you didn't go to college you could choose these schools to learn a trade and get the skills necessary to get a good job.
These programs have fallen by the wayside along with America's manufacturing. We don't need as many of those workers, so we don't train them.
There is a new economy though, an information economy. Yesterdays Professional Engineers are today's MCSE's and CCIE's designing information systems. These high end jobs still require a college education, as much for the non-technical (e.g. communications) skills as for their technical parts.
For each one of the architects of the information age there are hundreds of technicians. Just like a P.E. may have designed building built by a crew of 1,000 skilled workers in the past, today an information architect designs a data center built by hundreds. These "for profit colleges" specialize in associates (2 year) degrees with the tech skills necessary to fill these jobs. They tech the technical bits, but go really light on the reading, writing, and math skills that would actually give people the fundamentals; just like VoTech schools of old. The welder of old didn't need to know at a 14" beam was required for the weight load and how to calculate it, just how to lay down a perfect bead. The information tech of today doesn't need to know why there's a three layer switching fabric, just how to run Cat5 cables and test them.
Where the "for profit colleges" mislead people is they want them to think they are getting the same education as a 4 year traditional college. They are not. Look at the curriculum online or talk to people who have attended one. These institutions teach you how to do, not how to think.
Somehow it became stigmatized to have not attended college. Never mind that I've seen plenty of 6 figure skilled tradesmen, and seen plenty of 4 year college graduates struggle to get a $40k job. If these schools marketed themselves as VoTech they would be more honest, but no one would go. They are forced into marketing themselves as something they are not, and then folks are surprised, and disappointed with the output.
I'm not very happy with the Democrats position of government should tell you what to do, nor do I think the Republican argument of let the market sort it out makes any sense. Lightbulbs are a prime case.
Consumers are resistant to change, even if it is good. They need to be pushed towards CFL's, as in many cases they do save energy. But they are not better in every case. Sometimes you want the heat, sometimes it's a decorative look, color temperature, size and shape, who knows. For instance, I use some 10W incandescent bulbs for some applications, and you just can't find CFL's that put out that few lumens.
So what's the middle ground? I would be ok with a modest tax on incandescent bulbs, particularly if the money could be channeled to efforts to recycle CFL's, develop better CFL's, and similar work. I think an outright ban though is wrong, and would argue it should be repealed. While I am 90% CFL's, I want to keep getting some incandescents for select applications.
A NATO stealth bomber strike to take them out with missiles intended for that (i understand some ordinance can fry electronics). Enough is enough. Fuck Gaddafi!
I am not generally in favor of the US getting involved in these sorts of disputes, but when a government starts slaughtering civilians perhaps it is time. It would absolutely be best to do this under NATO or UN sanction, but that may take too much time.
I'm sure we can figure out the locations of the transmitters, and rather than a stealth bomber this sounds like a perfect use of cruise missiles. No risking our own air crew, or the political down side of killed or captured military folks. Heck, with no communication we can deny we did it, the rebels must have blown up the sites.
The "ordinance that can fry electronics" is a Electromagnetic Pulse, and the best way to do it with a weapon is a nuke. I don't think nuking Libya would help the situation.
I think it is often useful to compare with companies outside the "tech stock distortion bubble". Give or take a billion or so in market cap, is Facebook worth as much as:
The Boeing Company ($53B) Dow Chemical ($44B) General Motors ($54B) Kraft Foods ($54B)
Or some big names worth a lot less:
Coke ($500M) Sears ($10B) Wendys/Arby's ($2B)
The real issue is that Facebook doesn't really create new revenue. What they have done is take time people used to spend doing other things and transferred it to Facebook. Where before advertisers might have had to buy ads on Google, or a beer cup at a event, or a billboard by the road they are now spending it for ads on Facebook. Most of Facebooks gains are going to come at the expense of these other forms of advertising.
Advertisers spent $6.4 billion on ads online in 2010. Even if Facebook tool 100% of them, you now know the upper limit on their revenue stream, or more less. Those betting on multiples more upside on the stock are, insane.
I agree that the wrong answers were more interesting. Another category with a similar outcome was the one where each answer had a double meaning as a computer keyboard key and something else; Watson clearly didn't "get" the concept of computer keys. I don't think it tried to answer any of them, or maybe only one of them, and it's confidence was really low on every question in that category.
The explanation makes more sense, that it was somewhat discounting the category names. I think that's an area where they need to do more work.
I don't think this is quite a fair assessment. I downloaded the app to my iPad, and will try it at least for the first two free weeks.
While I'll admit about 80% of it looks like a nice digitization of a print magazine, there are a number of aspects where they deviate. There are embedded movies and photos. No, not just still photos but 360 degree panoramas you can scroll around. The crossword and sudoku are both interactive, you can actually play them on the iPad. A couple of the advertisements have interactive elements. as well, allowing you to zoom in on areas of interest. Pages about sports teams embed their real time twitter feed, and interesting idea.
Revolutionary, no, not quite. Interesting, yes; it is a glimpse of how a magazine will evolve in the digital age. Could they do more? Sure, and I bet they will over time.
The drawbacks I see are that it makes my iPad feel sluggish. Most things on it are quite zippy, but several operations in The Daily are quite slow. The second is price, I don't really mind 99 cents for an issue, but I'm sort of annoyed the other option is $40 for an entire year. A year is a long time, and comes from slow print relationships of a magazine per month. I think perhaps $10 for a quarter would be a much better deal for the consumer, and not lock them in for what is an eternity in the digital space.
My bigger question is, why can't this sort of content just be delivered in Safari? The answer may be in the in-App charging and DRM, but if so that's a bit of a lame reason. Web sites for major magazines should look this good, on a standard browser.
Slashdot Mirror
If you've ever looked at the patchwork of sales tax rules in this country you can quickly see this solves a major problem. There are literally 10's of thousands of sales tax jurisdictions in the US, pretty much every county at a minimum, and often each city or town inside of the county. It's not just different rates, but also different rules. Food is taxed in one place, taxed at a different rate in another, and not taxed in a third. And what is "food"? You don't really want to know the answer to that question, it's probably 10,000 pages long! Having one rate, or perhaps one national base rate and a per state "option"; but more importantly one set of classification rules would really solve a major hurdle for small online retailers, and make it practical for them to collect tax.
The largest problem this creates is who gets the revenue? Taxes generally pay for infrastructure (roads, bridges, fire departments, etc), so it makes sense for some of the revenue to go where the seller is located, and some where the buyer is located. In brick and mortar sales these tend to be the same place, but won't be for Internet sales. Plus, Internet sales depend on transportation. The goods are shipped by truck and rail, probably across many states in the middle. Some of the money needs to go to those states to build interstates and airports and rail yards to get the goods from seller to buyer.
There are some other small problems. For instance if the money is collected and distributed via the fed, can it be used as a stick to get the states to do other things? The tax may be regressive, depending on how it is implemented. Many localities exempt food for instance as a means of assisting the poor, squaring those rules into a national set of rules will be difficult.
Still, overall I think the country needs something like this to happen. The idea that we can collect no taxes on a significant fraction of the business activity is just crazy. Many other countries already have a VAT tax because of issues like this, so the US is really falling behind. No one likes taxes, but we all like the things taxes achieve (on some level), so let's find the simplest, least evil, and fairest way to collect them. Going from 10,000+ sets of rules and rates down to 1 would be a huge step.
If you live in a large city there will be one or more companies that do data destruction for corporations in your city that need to meet various standards for contracts they are working on, often for the government. Most of this will be bulk paper shredding and the like, but most will also do hard drives. Many will let you show up at the site and watch your hard drive go in the machine and get spit out in shreds, if you care. They are all certified (I forget the standard), they will bar code the drive, inventory it, and then shred it, and give you a certificate with the inventory numbers and serials confirming they were destroyed. The service is cheap if you don't make them come pick them up, I think around $10 a drive for single quantities.
However, I'm mostly with the other folks in this thread. If you write over with zeros in a pass or two pretty much anyone except the NSA won't stand a chance of recovering any data. Even someone like the NSA would only do it if there was no other way, due to the cost of getting that data. Basically, no one cares about you enough, unless you're like a major drug kingpin or something on the side.
I degauss mine, only because work has a machine (from years ago) so it's "free" and only takes a couple of minutes...but I also only do it on dead drives. If I have a drive I'm just not using right now I write zeros to it and store it to be used as a scratch drive. Only after it won't spin up anymore does it get degaussed.
The other DIY method mentioned is to drill a quarter inch hole in the platters with a drill press. Pro data recovery companies could get some data off for lots (10's of thousands) of money unless you drill like 20 holes it in, but it would keep any non-pro users from reading the drive.
Overwrite them and be happy. Shred them and know the stuff that comes out the back end at least gets recycled. It's all about your paranoia level.
I've long thought the best thing for health care in this country would be for the law to be struck down. Too many people in this down economy already like provisions of it (no pre-existing conditions, keeping kids on your insurance longer). Were it to be unconstitutional I think there would be a large swell of folks pushing for them to find some way to re-establish the law and make it constitutional.
Single payer becomes the obvious choice. It may be that the way to single payer is to lose in the Supreme Court.
The reason this post is stupid is that infrastructure is long term. When you go to the trouble of sending out a crew to dig up and put fiber in the ground your putting in an infrastructure asset that should have a 15-30 year lifespan. The fact that can average machine can't saturate it today means we're being forward thinking.
Like many tech-savvy home users I want a file server at home,but the reality is 95% of the time it is doing nothing. Be it noise (fan less), eco-friendly (use less electricity) or room friendly (make less heat) a low powered fan less system seems to be a great idea.
But it's damn hard to find. There seems to be an assumption that low powered and fan less means you want super-small. Pico, or mini-ITX. No one seems to think you would use it for a file server, so > 2 SATA cuts your choices by 50%, and > 4 almost doesn't exist.
I'd be ok with CPU's with speed step that could step _way_ down, like down to 5W, but they don't seem to exist. No one makes motherboards with Intel mobile processors for end users. Repurposing a laptop has other drawbacks.
I think it's an untapped market...perhaps a small one, but big enough more solutions should exist.
HP has a software business? Besides bloatware on a new HP PC?
Seriously, name 5 software titles HP makes that a random computer user might know.
I generally agree with the poster, the pilot is at the controls, so I think taking away his water bottle and pocket knife are silly. The only thing that gives me some pause is the FedEx 705 hijacking. I find a lot of people are not aware of this event, but basically a FedEx pilot scheduled a jumpseat (i.e. he wasn't flying, just riding along) trip and proceeded to attack the air crew with a hammer. When you look at this incident, you wonder if he wasn't able to take a hammer on the plane how the outcome might have been different.
At the same time, the attempt was a failure. In effect, the system worked. Since that time, FedEx has all but banned jumpseat trips, which I think may be an overreaction.
I think the real problem here is that the traveling public is holding the airlines to an unobtainable standard. Cars kill thousands each year. Busses less, but still some people every year. Trains less, but still some. Planes, many years, kill no one. When they finally do have an incident the reaction is totally disproportionate for some reason. Why we expect planes to never be hijacked or crashed, but are ok with 40,000 people a year dying in auto accidents makes no sense to me. Even if removing the TSA added 100 deaths a year by plane (which I think is unlikely), it would still be the safest mode of transport, and I'd gladly take that trade off to not be groped each time I fly, and to be able to walk onto a plane 10 minutes before it leaves like used to be possible with the shuttles.
I read the headline, and I'm like "Flare", Sun never had a server called Flare. Fire, sure, did they typo that, or what?
Then I open it to see this is about The Sun Unleashing Most Powerful Solar Flare Since 2006. There, a much more useful headline. That wasn't hard, was it?
There's a reason why I read ./ less every day...
I came here to say this another way around, if all of your developers were Ken Thompson level folks, you wouldn't need formal code reviews.
Excellent developers know what they are doing, research alternatives, run tests, simulations, and performance analysis, and submit awesome code. Someone reviewing it is unlikely to make improvements by reading it, but rather would have to study the problem in depth to improve a solution.
If your reviewer can simply read the code and point out errors or problems you probably should fire the person who wrote it. If they aren't good enough, and can't do a complete enough job to pass a visual inspection you have no chance of turning out high quality software.
Code reviews should be less frequent, and more in depth. Have a programmer present on a section of code after he's figured out all the details. Describe the alternatives he investigated, and why this worked so well. After an hour presentation then his peers might be up to speed enough to make substantial comments on the code.
I wonder if this won't end up being another Airbus moment. I remember one of the early Airbus planes that failed at the air show (A300 prototype?). While the cause is too complicated to detail here, part of the issue was that Airbus believes the computers are highly accurate, and in many cases if there seems to be a discrepancy between what the pilot wants and what the computer wants, it has the computer win.
Now, if we assume the sensors stopped sending good data for any reason the computers had bad inputs. The pilots may have been trying to do the right thing and the computer may have overridden part of their inputs.
One thing I'm surprised hasn't been done yet is to simulate a frozen pitot tube. Fly a simulator in the same simulated conditions and make the pitot tube reading go bad. What do the computers do with that bad input?
I seriously hope this doesn't end up being a computer issue, but as programmers like to say Garbage In, Garbage Out. The fact that in some cases the computers can override pilot inputs is one of the few things that make me nervous about Airbus planes.
I love to talk with the managers/owners of restaurants, and have reviewed plenty on Yelp and Urbanspoon. Of course they are always very concerned about their bad reviews, and looking for ways to make the bad reviews go away. My advice to them is always the same. Leave those people alone, they are already unhappy. Rather, get your happy customers to leave good reviews and drown them out.
I've encouraged several businesses to pay the $5 for a Yelp or Urbanspoon sticker to put on their door. To claim their owner pages, and use them to post specials and send updates to regulators. To drop a reward on their "duke" on Yelp. You know what? In 2-3 months they easily amass tens or hundreds of positive reviews. Now the situation is 150 good reviews to the one bad one they were worried about, and their reputation is just fine.
The only reason these are such a big deal for most businesses is that people who feel they have been wronged are more likely to speak up. If you have only one review and it's bad, well, you look like a bad business. You will never satisfy 100% of your customers, so just get the 99% that you do to drown them out.
Easy, cheap, and builds loyalty with your regulars. Plus, you now have great reviews, so when people visit the area or move there and have nothing to go on but the reviews you'll be one of the first they try.
Use the technology, don't fight it.
I use a good old pre-digital hanging lateral file. Hanging filing cabinets can be quite expensive, but a good one with ball bearing slides does the job quite well. A lateral 2-drawer is excellent because you can do letter sized in one drawer, and legal sized in the other.
But the real question is what to keep and why. You want to make it easy to get rid of what you don't need to keep anymore.
Lastly, things permanently installed in a house (new AC units, stoves, etc) should have all their stuff dropped in a drawer in the kitchen. Leave them for the next owner. They will love you.
Now, that you know what to keep you need to get rid of everything else, using a good shredder. Do yourself a favor and go ahead and spring for a quality "small business" unit in the $250-$400 range. The cheap home ones will choke when you try and put your junk mail in them and break after a few years. Your small business one will swallow those credit card offers that come in the mail without even opening, and easily eat your month of bills in one feed once a month.
I would not trust the government to be my only mail provider.
I do think it might be nice to have a walled garden government account. Every citizen gets an account, only government agencies can use it for official business. No spam. Secured. The government today won't do business for most things over e-mail because it goes over the big bad internet, and this could change it, and make a lot of things faster and more efficient. Since it's correspondence with the government in the first place I don't care if they can snoop on it.
Why, yes indeed. I worked in industry for many years, and I can tell you that no workers were more highly valued than those who were unable to do even the simplest things by themselves. "Let's collaborate!" they would say, and our hearts warmed instantly and we leapt into action, "helping" our valued coworkers, doing their work for them. In contrast, those with highly valuable skillsets, able to quickly solve difficult problems, those were as dirt to us. "Be off with you!" we'd cry, "and never dare to cross our path again!" Yes, as sweatyboatman says, nothing is more valuable in the real world than incompetence!
I see you understand the path to management, and more respect and money.
...can't they pipe real time sensor data into the internet, and let the gamers directly control the countermeasures? Why have a trained, staffed military when you can just start a new "battle" online that contains real data, rather than computer made up data for once? The gamers never even need to know!
I mean, it's just a little wave. It's not like a full on nuclear disaster.
Samba (by choosing GPLv3) wants to require Apple (in this case) to provide source code to Samba to anyone who uses OSX. The fundamental assumption in that paragraph is that a user can do something useful with the code.
My grandmother can't. Even if Apple shipped the full source on each CD it would make her no more free or less free, no more self sufficient or dependent. It would be about as useful as giving her a book written in Sanskrit.
But wait the GPL folks will argue, sure, most people won't be able to use it, but what about the programmers? Well, I'm a programmer. I want my Mac to read and write windows file shares. If it doesn't work, I'm going to open a ticket with Apple and make them fix it, even if the code is right there on my computer. The GPL folks fail to release that in many cases working and bug free trump the ability to modify; at least in the real world.
In a standard commercial world what the GPLv3 does is not create more freedom. It creates a burden on companies developing GPLv3 software to set up an entire distribution system for the benefit of a fraction of their user base. Like well under 1%. Probably under .1%. Maybe under .01% for a lot of products. When you have a million slaves it is hollow to tell them they are "more free" because you managed to free one of them.
But the real kicker here is that Samba is still free. Any OSX user can download the code, compile it on their OSX machine, and use it. If GPLv3 makes any sense it is on something like the kernel, or device drivers where various proprietary magic is required to make things go, and without the company distributing that information you really are screwed. To require a company to redistribute source to a user land program on a POSIX OS really is just lunacy. Anyone who wants more recent Samba than ships with OSX is already downloading it from Samba!
The BSD folks started to get it more right, and then backed off. The original 4 clause BSD license, with the so-called "advertising" clause was close to a good idea. The problem with it was that it really said advertising. Like if Apple made a TV commercial they would need to have some announcer read the names of every open source package included in OSX. The solution of simply removing it wasn't good either.
What was needed was a slightly different clause, requiring the user to be notified that they are using BSD licensed software, and where to find more information. One can imagine the MOTD on a brand new Unix system saying something like "This system contains BSD licensed software, for more information see /usr/share/bsd.txt. The file would contain not only the license, but the URL of each project involved, the name of the software, and version installed. The user is now aware of the open source project, and knows where to get the source code. That would have done more to promote open source software than either
removing the various clauses (the BSD "solution"), or adding onerous provisions (the GPL "solution").
At the end of the day though, it's technical folks trying to solve a business problem they don't understand and don't think should exist. They want to oversimplify others businesses, and want to believe that every business should be run the way they want. Every company should just release all source, instructions on how to compile it yourself, and so on. That's straight up idealism with a healthy dose of ignorance. In going for a 100% solution, the open source world keeps turning down a 95% solution, and ending up with a 5% solution. A pretty stupid trade off, if you ask me.
There is one up side. I've never found Samba particularly stable, and so if Apple has to re-write it I have every faith they will actually do a better job, and give me one less reason to use Samba.
I have a number of applications where I want a low powered "desktop" form factor. That probably means Mini-ITX or something like that. The canonical example? A home file server. It's not in use 90% of the time, and I'd like my power bill to go down and the heat load to go down. A chip with a super low power standby mode would be nice.
Unfortunately Intel chips don't get that power sippy even when speed-stepped down. VIA makes some semi-interesting chips, but they seem to be integrated with a bunch of features I don't want. I understand the HTPC market is big, but I don't need HDMI out on my file server. What's more, even providing two SATA ports seems to be the exception, not the rule.
I think an ARM "server board" in a mini-ITX factor would sell well. 6xSata, GigaBit Ethernet (even if it can't quite fill it), lots of DIMM slots so you can buy lower density cheaper DIMMS, but still keep lots of cache in memory for things like file servers.
Many years ago Vocational Technical schools churned out welders, plumbers, electricians, and all sorts of other skilled trades by the boatload. Not everyone was cut out to be a white collar employee, and so if you didn't go to college you could choose these schools to learn a trade and get the skills necessary to get a good job.
These programs have fallen by the wayside along with America's manufacturing. We don't need as many of those workers, so we don't train them.
There is a new economy though, an information economy. Yesterdays Professional Engineers are today's MCSE's and CCIE's designing information systems. These high end jobs still require a college education, as much for the non-technical (e.g. communications) skills as for their technical parts.
For each one of the architects of the information age there are hundreds of technicians. Just like a P.E. may have designed building built by a crew of 1,000 skilled workers in the past, today an information architect designs a data center built by hundreds. These "for profit colleges" specialize in associates (2 year) degrees with the tech skills necessary to fill these jobs. They tech the technical bits, but go really light on the reading, writing, and math skills that would actually give people the fundamentals; just like VoTech schools of old. The welder of old didn't need to know at a 14" beam was required for the weight load and how to calculate it, just how to lay down a perfect bead. The information tech of today doesn't need to know why there's a three layer switching fabric, just how to run Cat5 cables and test them.
Where the "for profit colleges" mislead people is they want them to think they are getting the same education as a 4 year traditional college. They are not. Look at the curriculum online or talk to people who have attended one. These institutions teach you how to do, not how to think.
Somehow it became stigmatized to have not attended college. Never mind that I've seen plenty of 6 figure skilled tradesmen, and seen plenty of 4 year college graduates struggle to get a $40k job. If these schools marketed themselves as VoTech they would be more honest, but no one would go. They are forced into marketing themselves as something they are not, and then folks are surprised, and disappointed with the output.
I'm not very happy with the Democrats position of government should tell you what to do, nor do I think the Republican argument of let the market sort it out makes any sense. Lightbulbs are a prime case.
Consumers are resistant to change, even if it is good. They need to be pushed towards CFL's, as in many cases they do save energy. But they are not better in every case. Sometimes you want the heat, sometimes it's a decorative look, color temperature, size and shape, who knows. For instance, I use some 10W incandescent bulbs for some applications, and you just can't find CFL's that put out that few lumens.
So what's the middle ground? I would be ok with a modest tax on incandescent bulbs, particularly if the money could be channeled to efforts to recycle CFL's, develop better CFL's, and similar work. I think an outright ban though is wrong, and would argue it should be repealed. While I am 90% CFL's, I want to keep getting some incandescents for select applications.
A NATO stealth bomber strike to take them out with missiles intended for that (i understand some ordinance can fry electronics). Enough is enough. Fuck Gaddafi!
I am not generally in favor of the US getting involved in these sorts of disputes, but when a government starts slaughtering civilians perhaps it is time. It would absolutely be best to do this under NATO or UN sanction, but that may take too much time.
I'm sure we can figure out the locations of the transmitters, and rather than a stealth bomber this sounds like a perfect use of cruise missiles. No risking our own air crew, or the political down side of killed or captured military folks. Heck, with no communication we can deny we did it, the rebels must have blown up the sites.
The "ordinance that can fry electronics" is a Electromagnetic Pulse, and the best way to do it with a weapon is a nuke. I don't think nuking Libya would help the situation.
I think it is often useful to compare with companies outside the "tech stock distortion bubble". Give or take a billion or so in market cap, is Facebook worth as much as:
The Boeing Company ($53B)
Dow Chemical ($44B)
General Motors ($54B)
Kraft Foods ($54B)
Or some big names worth a lot less:
Coke ($500M)
Sears ($10B)
Wendys/Arby's ($2B)
The real issue is that Facebook doesn't really create new revenue. What they have done is take time people used to spend doing other things and transferred it to Facebook. Where before advertisers might have had to buy ads on Google, or a beer cup at a event, or a billboard by the road they are now spending it for ads on Facebook. Most of Facebooks gains are going to come at the expense of these other forms of advertising.
http://www.iab.net/about_the_iab/recent_press_releases/press_release_archive/press_release/pr-111710
Advertisers spent $6.4 billion on ads online in 2010. Even if Facebook tool 100% of them, you now know the upper limit on their revenue stream, or more less. Those betting on multiples more upside on the stock are, insane.
I agree that the wrong answers were more interesting. Another category with a similar outcome was the one where each answer had a double meaning as a computer keyboard key and something else; Watson clearly didn't "get" the concept of computer keys. I don't think it tried to answer any of them, or maybe only one of them, and it's confidence was really low on every question in that category.
The explanation makes more sense, that it was somewhat discounting the category names. I think that's an area where they need to do more work.
I don't think this is quite a fair assessment. I downloaded the app to my iPad, and will try it at least for the first two free weeks.
While I'll admit about 80% of it looks like a nice digitization of a print magazine, there are a number of aspects where they deviate. There are embedded movies and photos. No, not just still photos but 360 degree panoramas you can scroll around. The crossword and sudoku are both interactive, you can actually play them on the iPad. A couple of the advertisements have interactive elements. as well, allowing you to zoom in on areas of interest. Pages about sports teams embed their real time twitter feed, and interesting idea.
Revolutionary, no, not quite. Interesting, yes; it is a glimpse of how a magazine will evolve in the digital age. Could they do more? Sure, and I bet they will over time.
The drawbacks I see are that it makes my iPad feel sluggish. Most things on it are quite zippy, but several operations in The Daily are quite slow. The second is price, I don't really mind 99 cents for an issue, but I'm sort of annoyed the other option is $40 for an entire year. A year is a long time, and comes from slow print relationships of a magazine per month. I think perhaps $10 for a quarter would be a much better deal for the consumer, and not lock them in for what is an eternity in the digital space.
My bigger question is, why can't this sort of content just be delivered in Safari? The answer may be in the in-App charging and DRM, but if so that's a bit of a lame reason. Web sites for major magazines should look this good, on a standard browser.
In the old NANOG preso it was 0.078%, in the current referenced Yahoo! quotes its 0.05% and you've got it at 0.024%.
I sense a trend....