...but this directive was never intended to apply in the case of such a massive breach. The context doesn't change the necessity of following rules though. There are a few issues that make the military's decision ot block the press sites laughable.
First, just because there might be classified information available, you do not ban members from a resource or location in it's entirety. This would be like banning military members without a top secret clearance and a need to know from Fort Meade or Fort Huachuca. Just as there are justifiable things to do in these bases that do not involve reading elint, there are justifiable reasons to read the New York Times.
Another issue is the fact that this policy does nothing to solve the source leak. As a former COMSEC NCO, I have to question the logic. Just putting a sticker on the phone that says "Ivan's listening" isn't going to solve this. Assange is not the spy. His organization is the vehicle by which military members and other government employees have used to break their vows to the government and American people. These are the people who need to be found and punished, because these are the people who have broken the law.
Finally, if unclassified military controlled computers are the only machines on the planet which do not have access to our secrets, are they still secrets or something else?
The biggest issue with the request is that we don't really know what the customer really does. Sure, they're a small NGO, but do they do accounting, document storage, engineering, or classified work? All of these have specific requirements that change the loadout.
Second, how does the client interface with vendors and their clients? What data format do they need to output or accept? This, more than anything else, will determine software requirements. What proficiency are the users? If they are very skilled at a specific software title, your customer can spend money on that software to avoid retraining. If they are unskilled, a learning curve will occur regardless, so software selection can be made with capability and cost as primary factors.
From the answer to these and other questions, you can derive the desktop software loadout, data flow, and server requirements. From these in turn you can derive network architecture, desktop specifications, and server specifications. Note the process:
Regulations > Interface > Tasks > Users > Desktop OS and Software > Data Flow > Server OS and Software > Network > Server Spec. > Desktop Spec.
Now that we covered this, a little general info. To rack or not to rack. If you are installing new infrastructure and your budget allows, place all network and server infrastructure in a rack. Lay it out and lock it down. Track every wire, origin, destination, and termination. This will go a long way to saving time when you must look uber-proficient.
Desktop apps and operating system? Depends. If the client has never used a computer or is only marginally proficient, use Ubuntu 10.10 and save yourself an ungodly amount of labor. Just make sure that you nail down a service contract first. After training the users and some limited admin work, you will have little to do, if you rely on per hour support with a limited client base, you will go broke supporting Ubuntu, but your clients will be happy.
As far as the server goes, things are a little more flexible, unless there are some specific server side applications. Assuming there are not, I have has excellent results running Ubuntu Server. Even in a windows-centric organization using Exchange, I have run Ubuntu 10.10 server and Citadel Server replacing an aged exchange server, saved myself days and my client a truckload of money.
Network is fairly straightforward. For 20 users on a DSL backbone, a 10/100 switch is fine for office use unless there is heavy data access on the servers. If there is, or you are running VoIP, use gigabit. As a security solution, use Endian Community firewall on a dedicated machine, segmenting the system in a logical manner (guest access, trusted network, and VoIP for example).
The specifics of the server will depend on the client use. Some basics: I nearly always spec Dell or SuperMicro. Don't choose a 1U server unless you need the space. The vertical clearance is simply too tight, heat is an issue, and standard PC parts don't fit. This will increase the likelihood of a failure and increase the support costs. Exception to this is the Atom based servers from SuperMicro. If you don't need the processing horsepower of a Zeon (like a basic fileserver) and you are not running Windows Server, the Atom based 1U servers they make are an extremely cost effective solution.
Finally, what kind of desktop? User's choice. Whatever you do, do not deploy a laptop as a desktop and expect them to use the laptop keyboard. Between the ergonomics and construction, this will be a nightmare for all concerned. Use an external keyboard and mouse. The chief selling point of a laptop is the built in UPS you get for free (battery), but make sure the asset doesn't walk off. Another point, if you do deploy Linux on the desktop, use extreme caution and make sure the hardware is supported straight out of the box. I installed Ubuntu laptops as desktops at a client, using certified equipment (Dell Inspiron
Making software of any kind should be oriented on performing a task. Copying what others did, other than using familiar convention to reduce learning curves, just muddies things. Viewed objectively, a skilled user of Ubuntu 10.10 and Windows Seven are about on par with their weapons of choice, and only occasionally have a "ooh... that's neat, I wish mine did that" moment.
To move forward, we need to identify the tasks users perform and rank them by their frequency. Evaluate each in turn on how efficiently this task is handled. Make changes as appropriate. Develop new solutions as needed. Walk away from convention when legacy limits abilities. We can make these changes more rapidly than Microsoft or Apple, and become the preferred tool with only a little marketing.
You make a kick ass game platform year ahead of it's time, charge too much for it, and of course industry support takes a little while to warm up. By the time the game industry fully supports you (arguably), another niche finds you, the scientific community. Rather than get their parents to buy one, these folks buy 5 to 100 and require no tech support. Rather than embrace the paradigm shift which stands to make you more money than expected, you orphan them with a software update. Sorry Sony. Your products are reasonably good, but you have no clue what to do with them, especially when an unexpected application rises up.
...when are you going to pay Paramount royalties because the star destroyers and Millenium Falcon use FTL? Of course you may wish to sue medical laser manufacturers who use a similar form factor. George, you need to send your lawyers packing to the same deep dark hole that Sun sent their legal team to when they sued the Republic of Java for infringement. Just sayin'.
I am concerned for what we are becoming. The US is leading the way, but it is certainly not an exclusively American phenomenon. While our educational system teaches to the lowest common denominator, the public is lulled to sleep by the mediocrity of their existence. Where someone rises above the median, they are punished and ostracised by societies leaders and their hired thugs who change the popular opinions and even meanings of words to suit their interests. This has happened before, but we call those the dark ages. I guess we should be glad they aren't burning people at the stake yet.
Perhaps it is time to hide what we truly are and experiment in secret. Given that these events are cyclical in nature, we can simply sandbag our findings until the public is once again interested in taking steps forward. This too has been done before, though the media has sensationalized the history. The world's best and brightest working in secret with a society bent on ignorance at all costs, including the life and liberty of those great minds. Perhaps it is time to build an intellectual subculture, after all, most of it is already there.
If your target market shifts and you don't change with it, your business will fail. The fault is not the market change, but rather the business and its management for not adapting to the new paradigm. Likewise, the Internet did not break government, but rather any government which fails due to the introduction of the internet failed to adapt. The fault is the overpaid morons who were in control.
This has occurred before, but usually for more serious social reasons. Look up what caused the French civil war, American revolution, and American civil war. Each of these events can be traced to an inability or unwillingness to adapt by one or more parties. Unfortunately, it's generally not the stodgy government officials nor the vocal minority of rabble-rousers that pay the bill in blood, but rather the poor masses. Let's hope this revolution caused by information becomes a fight of heated words in a chat room.
Is it possible that our keepers want a complete lack of accountability? In that case, I want no traffic cameras and other municipal surveillance. Law enforcement officers are human, humans with power. As such there needs to be checks and balances. Without these we become a police state and the true power comes form those who wield it without accountability. If the police get their wish and can act as they see fit with no accountability, I might as well move to China... the food leaves a bit to be desired, but 20 years from now they will enjoy more freedoms than we will.
To save power about 4 years ago used an Asus Terminator C3 as a server to replace some aging HP netservers. The Via C3 chip is interesting to get set up due to the lack of floating point math, but it did the job and with little power. I selected components which drew less power. You don't need a 7200 rpm HD. After installation, remove the CD drive. It adds to the current draw, especially at boot.
Take the 5 and 12V lines from the PC power supply and bring them out the back of the computer and use them to run stuff that normally runs on a wall wart. The PC has a switch mode power supply which is much more efficient that the simple linear supply in the wall warts. Better still is the fact that switch mode supplies are at their most efficient at 70-90% of their rated capacity. The transformers the wall warts use draw a little power any time they're plugged in, not just when the thing they power is on.
I fabricated a mount for the Yoggie Soho firewall and Linksys SD-2008 switch and put them in a drive bay. My cablemodem won't fit, so it sits to the side, also powered by 12V from the power supply. Also set the BIOS to come on when you need it to, and put it to bed at night. You will save a lot more with the thing off. If you remember to shut it down, great, if not, oh well.
I will eventually switch over to a Intel Atom or ARM architecture, but for right now this works. My entire system, firewall, switch, cablemodem, and server pulls about 60 watts under use. It is doubtful I will be able to increase efficiency beyond this sufficiently to merit an upgrade. It will wait until I have a failure, and something tells me that could be a long wait, but I am thinking of using a netbook like my Samsung NC-10 when I finally do.
For a contract, we had to rent one of the high dollar TDRs, so we deceided to put some assumptions to the test. Many assumptions we had about cables went away that day. Now, with the exception of patch cables, we always make our own in house.
The first assumption to die was that a factory cable is superior to a field crimped one. Testing a name brand 100ft cable vs. a 100ft self made, we discovered that our cable was superior in FEXT and Skew, comparable in loss and NEXT.
The second is that expensive cable is better than cheap Home Depot stuff. Testing cables 50ft in length, we discovered no real difference in the cables electrically, but that the cheap stuff with the conformal riser cladding retained it's electrical characteristics better than the more expensive PVC or Plenum in situations where the cable moves, is zip or velcro tied, or is supported at points along a span.
The keys to good cables are to keep the twists unmolested as close to the RJ-45 or punch block as you can. Use good tools, not the $19.95 crimper. Test every connection (wether you crimped it or not) and document them.
Putting my money where my mouth is, We warranty our cables for 25 years, and have never had a valid claim.
Quite true, but after the engineering's done we have to make it so the user can drive. This requires feedback, education, and understanding the expectations of the user.
Once upon a time (about 1971), Ford and Mazda began to talk engines. Mazda built a rotary engine for Ford to eval. Ford tried it in a '66 mustang and after barely getting the Mustang to move, declared the result a failure. Oddly enough, Mazda rotaries were generally quicker than other vehicles in their class, but Ford never stopped to ask why, because at the time "cubes are king" (bigger engines are the answer to everything).
The reason it failed is that Ford didn't do their homework. The stock 289 engine that came with the Mustang would begin to develop torque around 1400 RPM and peaked at about 4000RPM. The rotary engine used begin to develop their torque around 3500 RPM, peaking at about 7000RPM. This meant that when the automatic transmission engaged, it was barely turning fast enough to keep itself running.
Had they put a stickshift on it and trained the driver to shift high, the rotary engine would have run the 289 some serious competition at a lighter weight and comparable gas mileage. Alternatively, they could have geared down the rotary engine before it got to the automatic transmission with similarly good results.
My point? Ford didn't know rotary engines, and Asus, Samsung, and HP really don't know Linux. Things do not simply randomly work... it's rather the other way around.
To solve the engine issue, Mazda techs could have simply educated the Ford techs about the idiosyncracies of their design. Likewise, Linux users would have valuable information to netbook manufacturers, enhancing their products' useability.
You are right on some counts, but wrong on others. You make some excellent points about the difference between the mindset of the programmers and those of the average user. This is what makes it so important to train both parties to deal with and respect the idiosyncracies of the other.
To do so techies MUST train the user, and the user must train the techies!
l337 73ch5 please note: In order for a idiot user to RTFM there has to be a FM! Get off your high horse and write something other than code or pay someone to do it. If you are a support geek and have that napoleonic complex to your users, retake the A+ training portion about professionalism. You obviously missed something!
Since techies don't think in quite the same way as non-techies, in order to help someone improve their understanding we have to learn to think like a user. In order to do this, the user has to communicate what the issue is, and since they may not have the ability to do so, we need to help them out with this. We do this by asking questions in a non- condecending way as you would expect a doctor to do when diagnosing what ails you.
Hostile toward non techies... There is an elitist attitude in tech on both sides of this issue. How does the Windows world solve it? Professional support personnel. Good documentation. Local knowledge base.
If you use similar resources (I am assuming internet forums) to solve a Windows issue, and you'll see trolls there too. This vocal minority has given on line support a bad name. Since this is one of the only ways to get support for Linux, the stink has rubbed off on the innocent.
Judging by what you say and don't in your post, I would hazard a guess you started out life as a VB programmer. Microsoft radically altered the way we program with this, and Linux does not work the same way.
Core Linux programming more closely resembles old school mainframe programming in many ways. This is not better or worse, just different. It is more efficient at the code level, but puts a very clear boundary between the presentation and processing layers. This makes it more challenging for Linux programmers to wrap their brains around both sides, and explains the lack of documentation and sometimes glacial improvements to the desktop, though this has improved.
I agree that we need to make apps that "Joe Random's retarded stepchild can use". At the same time, how do you create innovation, even a clean start, without a learning curve? One gal who works for me is a 50 year old grandmother with less than 5 years of actual seat time on computers. She has become so comfortable with Ubuntu and now Fedora10, I doubt I would survive the day if I moved her to XP or Vista.
My point is that everything is a challenge, and "Joe Random's retarded stepchild" probably needs to stick with bejeweled or Tux Type" if they are that challenged.
You are right that the issue of Netbook vs. Laptop vs. Desktop is irrelavent. The form gets chosen by use and social pressure, not by the IT weenie. The real issue is making the OS compatible with new and oddball hardware like the 1024 x 600 screens and chipsets on Netbooks.
As an example, loading XP on my Samsung NC10 is a challenge because of all the driver issues. Fedora10 just worked right down to the webcam, bluetooth, and wireless card.
The use of netbooks, Open Source Software, and Linux must be critically evaluated with clear expectations. Using the wrong tool of the job is like trying to use a Mazda Miata to move a house, or worse, fuelling a diesel truck with unleaded. Use the right tool for the job or you will get poor results regardless of the quality of the tool.
In the real world, there is a place for netbooks, and a place for desktop replacement laptops, just as there is a place for closed source software and open source software.
If you need a computer on the go, and can deal with a slightly smaller keyboard, get a netbook.
If you want a 17" screen and are basically looking for a desktop unit with a built in UPS, get a big laptop.
If you want to go to Best Buy and grab software, stick with Windows.
If you want good purpose built software for free download, try Linux.
If you want stable, top dollar, industry standard software that surprises you with a bug now and again where the manufacturer will spend a lot of time fingerpointing to the OS or drivers, use closed source software.
If you want decent free or low cost software capable of 90% of the functions of closed source software with the occasional bug where the fix is on the net ready to be googled or the programmer will work to fix it, but may take a while, use open source software.
Find where those break over points are. These will be unique to the organization and the user. We use Fedora10 on the desktop on both laptops and netbooks in our office. These were chosen carefully a good fits. Your mileage will vary.
...they do this in IT because they honestly see IT pros as kid geniuses you can pay off in twinkies and a pat on the head.
I see this all the time even though I am considerably older than the average age of my clients.
The other ones consider us part of overhead, since we aren't in sales, even though without us they really couldn't even spell CRM. Oh, and please don't mod this one +funny, because it's really not.
While reducing moving parts and heat generating components are important, as is upgrade path, the real issue is software longevity. The next version of your OS and specific application may not last 15 years thanks to the planned obsolescence of the industry.
My plan for something like this would include an Intel Atom processor integrated motherboard with 2GB of ram The idea is to prevent swapping to keep the HDs longer. Run a 200W DC power supply with an external AC adapter. Put the whole mess in a large steel case and block off the vents. The steel case will dissipate the heat from the HD and CPU, provided your office stays below 85 degrees.
As I said before, the OS and software are the real trick. While you could run Win95 / DOS, there are other options with clear benefits. The existing database is probably either an embedded flat file or a Lotus Smart Suite / DBase3 clone. The latter will have issues with the next part of this plan, so you may need to upgrade the software or modify it to handle a modern database like MySQL.
My solution? run Linux. You can use Linux to create a virtual DOS or Windows machine using VMWare or the Open Source equivalents. Since the database may run client-server, it's possible to run the database on the host machine, while running the DOS / Win95 image in non-persistent mode making it safe from modification.
Simply back up the virtual machines once, and your database on a regular cycle for disaster recovery. If it ever goes down or you want to upgrade the OS, simply reload Linux and restore the database and your virtual machines.
Another simpler option is to use wine or DOSemu to run the software as is, but these can have some compatibility issues if you upgrade the host OS at a later date.
This plan will allow the machines to run at a walk rather than a run, reduce moving parts to a minimum, present the least learning curve for your users, and keep you running no matter what fails.
Ummm... There is a small flaw with your argument. Nuclear reactors work off decay of fissionable material, a natural and predictable process with a set rate. This rate is modified somewhat using materials to buffer the reaction.
Bottom line is that a nuclear power plant produces the most power per fuel expended (efficiency) at a fairly high production rate. This "sweet spot" varies by reactor design.
If we had only nuclear reactors, you would be functionally correct, but in order to get the most "bang for the buck" (no pun intended, honest), nuke plants should be and generally are run at or close to that magic number, using the excess power generated to offset coal or oil burning plants in other areas.
Using more power generally doesn't create more reactors. It will delay the downgrading and decommissioning of the old oil and coal burning plants however, creating more problems with acid rains and global warming. You were right to be concerned, but the issue is worse than you thought.
...learn and understand the capabilities of seLinux and read the NSA Security guides. They don't do it for you (would you really want them to???) they tell you how, step by step.
Be careful you don't lock you or your employees out of their own system... it is a possibility if you really go aggressive with seLinux enforcement.
Quite true. Most teachers here spend about 10% of their day teaching. the other 90% is state and federally mandated paperwork, most of it relating to "no child left behind".
Nice idea, but if the teachers spend more time documenting than doing, there will be predictable results.
...besides, what if Johnny is a real moron? The classmates must trudge along, being bored and unchallenged. Admittedly, this is an oversimplification, but the results are clear.
I know a teacher that was going to put that on their car, but decided not to. What's the point if the majority of the graduating class can't read it within the time it takes for a light to turn green?
Here's scary: CO Workforce has an IT dept of less than 5 techs for the entire state.
They have open wireless routers half the time because the end user gets a Netgear or Linksys from Walmart rather than wait for their overworked IT folks to get around to it.
They often share hardware with other agencies without locking them down or performing an inspection prior to returning them to a State network.
Training and keeping up with advances? With this much of an overworked and underappreciated IT dept, the last training they got was when they were in school.
My point? It's not the IT weenie's fault. If you want to yell at someone, make sure it gives the IT department more funding for more positions. Consider:
If work remains constant, more bodies = problems solved quicker.
Problem solved quicker = more time.
More time = more time for training and learning.
More time for training and learning = less inaccurate statements and stupid decisions.
Honestly, I am really trying not to don the tin foil hat here... but the personal result of this is disgust. If we can make a voting system (my son and I did so in my living room), why can't a company like Diebold? Once a pattern emerges, the answers become obvious, though anecdotal.
1. Did Diebold and it's board make a substantial contribution to one party or the other? If so, one cannot expect neutrality.
2. Does Diebold use Open Source Software and techniques in the coding of their voting machines? If not, one cannot expect transparency and impartial peer review.
3. Would any system that is tested and shown to have a 0 variance be better than Diebold's? If so, we cannot expect fair elections from their machines.
Could Tech Savvy individuals develop a better system with more transparency and accountability? Given the correct answers to the first three questions, Yes.
Remember the 10 shortest words of greatest power: "If it is to be, it is up to me." If you are serious about this, the next step is to develop a competitive system and present it. Personally, I am working on a parlimentary system, and while the code is similar, it would not scale to a national election. The problem is authentication, and the preventing "vote early, vote often" while maintaining anonymity. Here are some basic standards / benchmarks that seem common sense to me:
Open Source must be used, and staff coders must become familiar with every line of code. It follows therefore that the OS and software must be compiled from scratch. Any improvements to the code must be handled in accordance with GPL
Accountability and error checking must occur at each point where data is converted, correlated, or handed off. Parity checks or checksums are insufficient. Use MD5 checksums or come up with an even better system.
Once the system begins testing, publish the code and issue cash awards for breaking it. Put the black hats to work for you, and for Christ's sake, LISTEN TO THEM!
Using Open Source means the source code itself will be free. This does not mean the package including approved hardware, expertise and training will be free. That's where the money is made. Odds are that it'll be cheaper than the Diebold solution.
Just remember to keep it simple at every level, and it must have an attractive GUI. Show, don't tell. The average Joe hates to read, and it's easier to go multilingual if text is at a minimum. Aside from the user experience, these must be set up, used, maintained, and read by relatively untrained people. KISS always applies.
Slashdot Mirror
...but this directive was never intended to apply in the case of such a massive breach. The context doesn't change the necessity of following rules though. There are a few issues that make the military's decision ot block the press sites laughable.
First, just because there might be classified information available, you do not ban members from a resource or location in it's entirety. This would be like banning military members without a top secret clearance and a need to know from Fort Meade or Fort Huachuca. Just as there are justifiable things to do in these bases that do not involve reading elint, there are justifiable reasons to read the New York Times.
Another issue is the fact that this policy does nothing to solve the source leak. As a former COMSEC NCO, I have to question the logic. Just putting a sticker on the phone that says "Ivan's listening" isn't going to solve this. Assange is not the spy. His organization is the vehicle by which military members and other government employees have used to break their vows to the government and American people. These are the people who need to be found and punished, because these are the people who have broken the law.
Finally, if unclassified military controlled computers are the only machines on the planet which do not have access to our secrets, are they still secrets or something else?
The biggest issue with the request is that we don't really know what the customer really does. Sure, they're a small NGO, but do they do accounting, document storage, engineering, or classified work? All of these have specific requirements that change the loadout.
Second, how does the client interface with vendors and their clients? What data format do they need to output or accept? This, more than anything else, will determine software requirements. What proficiency are the users? If they are very skilled at a specific software title, your customer can spend money on that software to avoid retraining. If they are unskilled, a learning curve will occur regardless, so software selection can be made with capability and cost as primary factors.
From the answer to these and other questions, you can derive the desktop software loadout, data flow, and server requirements. From these in turn you can derive network architecture, desktop specifications, and server specifications. Note the process:
Regulations > Interface > Tasks > Users > Desktop OS and Software > Data Flow > Server OS and Software > Network > Server Spec. > Desktop Spec.
Now that we covered this, a little general info. To rack or not to rack. If you are installing new infrastructure and your budget allows, place all network and server infrastructure in a rack. Lay it out and lock it down. Track every wire, origin, destination, and termination. This will go a long way to saving time when you must look uber-proficient.
Desktop apps and operating system? Depends. If the client has never used a computer or is only marginally proficient, use Ubuntu 10.10 and save yourself an ungodly amount of labor. Just make sure that you nail down a service contract first. After training the users and some limited admin work, you will have little to do, if you rely on per hour support with a limited client base, you will go broke supporting Ubuntu, but your clients will be happy.
As far as the server goes, things are a little more flexible, unless there are some specific server side applications. Assuming there are not, I have has excellent results running Ubuntu Server. Even in a windows-centric organization using Exchange, I have run Ubuntu 10.10 server and Citadel Server replacing an aged exchange server, saved myself days and my client a truckload of money.
Network is fairly straightforward. For 20 users on a DSL backbone, a 10/100 switch is fine for office use unless there is heavy data access on the servers. If there is, or you are running VoIP, use gigabit. As a security solution, use Endian Community firewall on a dedicated machine, segmenting the system in a logical manner (guest access, trusted network, and VoIP for example).
The specifics of the server will depend on the client use. Some basics: I nearly always spec Dell or SuperMicro. Don't choose a 1U server unless you need the space. The vertical clearance is simply too tight, heat is an issue, and standard PC parts don't fit. This will increase the likelihood of a failure and increase the support costs. Exception to this is the Atom based servers from SuperMicro. If you don't need the processing horsepower of a Zeon (like a basic fileserver) and you are not running Windows Server, the Atom based 1U servers they make are an extremely cost effective solution.
Finally, what kind of desktop? User's choice. Whatever you do, do not deploy a laptop as a desktop and expect them to use the laptop keyboard. Between the ergonomics and construction, this will be a nightmare for all concerned. Use an external keyboard and mouse. The chief selling point of a laptop is the built in UPS you get for free (battery), but make sure the asset doesn't walk off. Another point, if you do deploy Linux on the desktop, use extreme caution and make sure the hardware is supported straight out of the box. I installed Ubuntu laptops as desktops at a client, using certified equipment (Dell Inspiron
Making software of any kind should be oriented on performing a task. Copying what others did, other than using familiar convention to reduce learning curves, just muddies things. Viewed objectively, a skilled user of Ubuntu 10.10 and Windows Seven are about on par with their weapons of choice, and only occasionally have a "ooh... that's neat, I wish mine did that" moment. To move forward, we need to identify the tasks users perform and rank them by their frequency. Evaluate each in turn on how efficiently this task is handled. Make changes as appropriate. Develop new solutions as needed. Walk away from convention when legacy limits abilities. We can make these changes more rapidly than Microsoft or Apple, and become the preferred tool with only a little marketing.
You make a kick ass game platform year ahead of it's time, charge too much for it, and of course industry support takes a little while to warm up. By the time the game industry fully supports you (arguably), another niche finds you, the scientific community. Rather than get their parents to buy one, these folks buy 5 to 100 and require no tech support. Rather than embrace the paradigm shift which stands to make you more money than expected, you orphan them with a software update. Sorry Sony. Your products are reasonably good, but you have no clue what to do with them, especially when an unexpected application rises up.
...when are you going to pay Paramount royalties because the star destroyers and Millenium Falcon use FTL? Of course you may wish to sue medical laser manufacturers who use a similar form factor. George, you need to send your lawyers packing to the same deep dark hole that Sun sent their legal team to when they sued the Republic of Java for infringement. Just sayin'.
I am concerned for what we are becoming. The US is leading the way, but it is certainly not an exclusively American phenomenon. While our educational system teaches to the lowest common denominator, the public is lulled to sleep by the mediocrity of their existence. Where someone rises above the median, they are punished and ostracised by societies leaders and their hired thugs who change the popular opinions and even meanings of words to suit their interests. This has happened before, but we call those the dark ages. I guess we should be glad they aren't burning people at the stake yet.
Perhaps it is time to hide what we truly are and experiment in secret. Given that these events are cyclical in nature, we can simply sandbag our findings until the public is once again interested in taking steps forward. This too has been done before, though the media has sensationalized the history. The world's best and brightest working in secret with a society bent on ignorance at all costs, including the life and liberty of those great minds. Perhaps it is time to build an intellectual subculture, after all, most of it is already there.
If your target market shifts and you don't change with it, your business will fail. The fault is not the market change, but rather the business and its management for not adapting to the new paradigm. Likewise, the Internet did not break government, but rather any government which fails due to the introduction of the internet failed to adapt. The fault is the overpaid morons who were in control. This has occurred before, but usually for more serious social reasons. Look up what caused the French civil war, American revolution, and American civil war. Each of these events can be traced to an inability or unwillingness to adapt by one or more parties. Unfortunately, it's generally not the stodgy government officials nor the vocal minority of rabble-rousers that pay the bill in blood, but rather the poor masses. Let's hope this revolution caused by information becomes a fight of heated words in a chat room.
Is it possible that our keepers want a complete lack of accountability? In that case, I want no traffic cameras and other municipal surveillance. Law enforcement officers are human, humans with power. As such there needs to be checks and balances. Without these we become a police state and the true power comes form those who wield it without accountability. If the police get their wish and can act as they see fit with no accountability, I might as well move to China... the food leaves a bit to be desired, but 20 years from now they will enjoy more freedoms than we will.
To save power about 4 years ago used an Asus Terminator C3 as a server to replace some aging HP netservers. The Via C3 chip is interesting to get set up due to the lack of floating point math, but it did the job and with little power. I selected components which drew less power. You don't need a 7200 rpm HD. After installation, remove the CD drive. It adds to the current draw, especially at boot.
Take the 5 and 12V lines from the PC power supply and bring them out the back of the computer and use them to run stuff that normally runs on a wall wart. The PC has a switch mode power supply which is much more efficient that the simple linear supply in the wall warts. Better still is the fact that switch mode supplies are at their most efficient at 70-90% of their rated capacity. The transformers the wall warts use draw a little power any time they're plugged in, not just when the thing they power is on.
I fabricated a mount for the Yoggie Soho firewall and Linksys SD-2008 switch and put them in a drive bay. My cablemodem won't fit, so it sits to the side, also powered by 12V from the power supply. Also set the BIOS to come on when you need it to, and put it to bed at night. You will save a lot more with the thing off. If you remember to shut it down, great, if not, oh well.
I will eventually switch over to a Intel Atom or ARM architecture, but for right now this works. My entire system, firewall, switch, cablemodem, and server pulls about 60 watts under use. It is doubtful I will be able to increase efficiency beyond this sufficiently to merit an upgrade. It will wait until I have a failure, and something tells me that could be a long wait, but I am thinking of using a netbook like my Samsung NC-10 when I finally do.
For a contract, we had to rent one of the high dollar TDRs, so we deceided to put some assumptions to the test. Many assumptions we had about cables went away that day. Now, with the exception of patch cables, we always make our own in house.
The first assumption to die was that a factory cable is superior to a field crimped one. Testing a name brand 100ft cable vs. a 100ft self made, we discovered that our cable was superior in FEXT and Skew, comparable in loss and NEXT.
The second is that expensive cable is better than cheap Home Depot stuff. Testing cables 50ft in length, we discovered no real difference in the cables electrically, but that the cheap stuff with the conformal riser cladding retained it's electrical characteristics better than the more expensive PVC or Plenum in situations where the cable moves, is zip or velcro tied, or is supported at points along a span.
The keys to good cables are to keep the twists unmolested as close to the RJ-45 or punch block as you can. Use good tools, not the $19.95 crimper. Test every connection (wether you crimped it or not) and document them.
Putting my money where my mouth is, We warranty our cables for 25 years, and have never had a valid claim.
Quite true, but after the engineering's done we have to make it so the user can drive. This requires feedback, education, and understanding the expectations of the user.
Once upon a time (about 1971), Ford and Mazda began to talk engines. Mazda built a rotary engine for Ford to eval. Ford tried it in a '66 mustang and after barely getting the Mustang to move, declared the result a failure. Oddly enough, Mazda rotaries were generally quicker than other vehicles in their class, but Ford never stopped to ask why, because at the time "cubes are king" (bigger engines are the answer to everything).
The reason it failed is that Ford didn't do their homework. The stock 289 engine that came with the Mustang would begin to develop torque around 1400 RPM and peaked at about 4000RPM. The rotary engine used begin to develop their torque around 3500 RPM, peaking at about 7000RPM. This meant that when the automatic transmission engaged, it was barely turning fast enough to keep itself running.
Had they put a stickshift on it and trained the driver to shift high, the rotary engine would have run the 289 some serious competition at a lighter weight and comparable gas mileage. Alternatively, they could have geared down the rotary engine before it got to the automatic transmission with similarly good results.
My point? Ford didn't know rotary engines, and Asus, Samsung, and HP really don't know Linux. Things do not simply randomly work... it's rather the other way around.
To solve the engine issue, Mazda techs could have simply educated the Ford techs about the idiosyncracies of their design. Likewise, Linux users would have valuable information to netbook manufacturers, enhancing their products' useability.
Educate, don't bitch.
Billco,
You are right on some counts, but wrong on others. You make some excellent points about the difference between the mindset of the programmers and those of the average user. This is what makes it so important to train both parties to deal with and respect the idiosyncracies of the other.
To do so techies MUST train the user, and the user must train the techies!
l337 73ch5 please note: In order for a idiot user to RTFM there has to be a FM! Get off your high horse and write something other than code or pay someone to do it. If you are a support geek and have that napoleonic complex to your users, retake the A+ training portion about professionalism. You obviously missed something!
Since techies don't think in quite the same way as non-techies, in order to help someone improve their understanding we have to learn to think like a user. In order to do this, the user has to communicate what the issue is, and since they may not have the ability to do so, we need to help them out with this. We do this by asking questions in a non- condecending way as you would expect a doctor to do when diagnosing what ails you.
Hostile toward non techies... There is an elitist attitude in tech on both sides of this issue. How does the Windows world solve it? Professional support personnel. Good documentation. Local knowledge base.
If you use similar resources (I am assuming internet forums) to solve a Windows issue, and you'll see trolls there too. This vocal minority has given on line support a bad name. Since this is one of the only ways to get support for Linux, the stink has rubbed off on the innocent.
Judging by what you say and don't in your post, I would hazard a guess you started out life as a VB programmer. Microsoft radically altered the way we program with this, and Linux does not work the same way.
Core Linux programming more closely resembles old school mainframe programming in many ways. This is not better or worse, just different. It is more efficient at the code level, but puts a very clear boundary between the presentation and processing layers. This makes it more challenging for Linux programmers to wrap their brains around both sides, and explains the lack of documentation and sometimes glacial improvements to the desktop, though this has improved.
I agree that we need to make apps that "Joe Random's retarded stepchild can use". At the same time, how do you create innovation, even a clean start, without a learning curve? One gal who works for me is a 50 year old grandmother with less than 5 years of actual seat time on computers. She has become so comfortable with Ubuntu and now Fedora10, I doubt I would survive the day if I moved her to XP or Vista.
My point is that everything is a challenge, and "Joe Random's retarded stepchild" probably needs to stick with bejeweled or Tux Type" if they are that challenged.
You are right that the issue of Netbook vs. Laptop vs. Desktop is irrelavent. The form gets chosen by use and social pressure, not by the IT weenie. The real issue is making the OS compatible with new and oddball hardware like the 1024 x 600 screens and chipsets on Netbooks.
As an example, loading XP on my Samsung NC10 is a challenge because of all the driver issues. Fedora10 just worked right down to the webcam, bluetooth, and wireless card.
Cheers,
Ulf
The use of netbooks, Open Source Software, and Linux must be critically evaluated with clear expectations. Using the wrong tool of the job is like trying to use a Mazda Miata to move a house, or worse, fuelling a diesel truck with unleaded. Use the right tool for the job or you will get poor results regardless of the quality of the tool.
In the real world, there is a place for netbooks, and a place for desktop replacement laptops, just as there is a place for closed source software and open source software.
If you need a computer on the go, and can deal with a slightly smaller keyboard, get a netbook.
If you want a 17" screen and are basically looking for a desktop unit with a built in UPS, get a big laptop.
If you want to go to Best Buy and grab software, stick with Windows.
If you want good purpose built software for free download, try Linux.
If you want stable, top dollar, industry standard software that surprises you with a bug now and again where the manufacturer will spend a lot of time fingerpointing to the OS or drivers, use closed source software.
If you want decent free or low cost software capable of 90% of the functions of closed source software with the occasional bug where the fix is on the net ready to be googled or the programmer will work to fix it, but may take a while, use open source software.
Find where those break over points are. These will be unique to the organization and the user. We use Fedora10 on the desktop on both laptops and netbooks in our office. These were chosen carefully a good fits. Your mileage will vary.
...they do this in IT because they honestly see IT pros as kid geniuses you can pay off in twinkies and a pat on the head.
I see this all the time even though I am considerably older than the average age of my clients.
The other ones consider us part of overhead, since we aren't in sales, even though without us they really couldn't even spell CRM. Oh, and please don't mod this one +funny, because it's really not.
While reducing moving parts and heat generating components are important, as is upgrade path, the real issue is software longevity. The next version of your OS and specific application may not last 15 years thanks to the planned obsolescence of the industry.
My plan for something like this would include an Intel Atom processor integrated motherboard with 2GB of ram The idea is to prevent swapping to keep the HDs longer. Run a 200W DC power supply with an external AC adapter. Put the whole mess in a large steel case and block off the vents. The steel case will dissipate the heat from the HD and CPU, provided your office stays below 85 degrees.
As I said before, the OS and software are the real trick. While you could run Win95 / DOS, there are other options with clear benefits. The existing database is probably either an embedded flat file or a Lotus Smart Suite / DBase3 clone. The latter will have issues with the next part of this plan, so you may need to upgrade the software or modify it to handle a modern database like MySQL.
My solution? run Linux. You can use Linux to create a virtual DOS or Windows machine using VMWare or the Open Source equivalents. Since the database may run client-server, it's possible to run the database on the host machine, while running the DOS / Win95 image in non-persistent mode making it safe from modification.
Simply back up the virtual machines once, and your database on a regular cycle for disaster recovery. If it ever goes down or you want to upgrade the OS, simply reload Linux and restore the database and your virtual machines.
Another simpler option is to use wine or DOSemu to run the software as is, but these can have some compatibility issues if you upgrade the host OS at a later date.
This plan will allow the machines to run at a walk rather than a run, reduce moving parts to a minimum, present the least learning curve for your users, and keep you running no matter what fails.
Ummm... There is a small flaw with your argument. Nuclear reactors work off decay of fissionable material, a natural and predictable process with a set rate. This rate is modified somewhat using materials to buffer the reaction.
Bottom line is that a nuclear power plant produces the most power per fuel expended (efficiency) at a fairly high production rate. This "sweet spot" varies by reactor design.
If we had only nuclear reactors, you would be functionally correct, but in order to get the most "bang for the buck" (no pun intended, honest), nuke plants should be and generally are run at or close to that magic number, using the excess power generated to offset coal or oil burning plants in other areas.
Using more power generally doesn't create more reactors. It will delay the downgrading and decommissioning of the old oil and coal burning plants however, creating more problems with acid rains and global warming. You were right to be concerned, but the issue is worse than you thought.
...learn and understand the capabilities of seLinux and read the NSA Security guides. They don't do it for you (would you really want them to???) they tell you how, step by step.
Be careful you don't lock you or your employees out of their own system... it is a possibility if you really go aggressive with seLinux enforcement.
Considering we have 35MPH sustained winds, gusting to 50, I thought your username somewhat ironic.
Quite true. Most teachers here spend about 10% of their day teaching. the other 90% is state and federally mandated paperwork, most of it relating to "no child left behind".
...besides, what if Johnny is a real moron? The classmates must trudge along, being bored and unchallenged. Admittedly, this is an oversimplification, but the results are clear.
Nice idea, but if the teachers spend more time documenting than doing, there will be predictable results.
I know a teacher that was going to put that on their car, but decided not to. What's the point if the majority of the graduating class can't read it within the time it takes for a light to turn green?
Here's scary: CO Workforce has an IT dept of less than 5 techs for the entire state.
They have open wireless routers half the time because the end user gets a Netgear or Linksys from Walmart rather than wait for their overworked IT folks to get around to it.
They often share hardware with other agencies without locking them down or performing an inspection prior to returning them to a State network.
Training and keeping up with advances? With this much of an overworked and underappreciated IT dept, the last training they got was when they were in school.
My point? It's not the IT weenie's fault. If you want to yell at someone, make sure it gives the IT department more funding for more positions. Consider:
If work remains constant, more bodies = problems solved quicker.
Problem solved quicker = more time.
More time = more time for training and learning.
More time for training and learning = less inaccurate statements and stupid decisions.
Honestly, I am really trying not to don the tin foil hat here... but the personal result of this is disgust. If we can make a voting system (my son and I did so in my living room), why can't a company like Diebold? Once a pattern emerges, the answers become obvious, though anecdotal.
1. Did Diebold and it's board make a substantial contribution to one party or the other? If so, one cannot expect neutrality.
2. Does Diebold use Open Source Software and techniques in the coding of their voting machines? If not, one cannot expect transparency and impartial peer review.
3. Would any system that is tested and shown to have a 0 variance be better than Diebold's? If so, we cannot expect fair elections from their machines.
Could Tech Savvy individuals develop a better system with more transparency and accountability? Given the correct answers to the first three questions, Yes.
Remember the 10 shortest words of greatest power: "If it is to be, it is up to me." If you are serious about this, the next step is to develop a competitive system and present it. Personally, I am working on a parlimentary system, and while the code is similar, it would not scale to a national election. The problem is authentication, and the preventing "vote early, vote often" while maintaining anonymity. Here are some basic standards / benchmarks that seem common sense to me:
Open Source must be used, and staff coders must become familiar with every line of code. It follows therefore that the OS and software must be compiled from scratch. Any improvements to the code must be handled in accordance with GPL
Accountability and error checking must occur at each point where data is converted, correlated, or handed off. Parity checks or checksums are insufficient. Use MD5 checksums or come up with an even better system.
Once the system begins testing, publish the code and issue cash awards for breaking it. Put the black hats to work for you, and for Christ's sake, LISTEN TO THEM!
Using Open Source means the source code itself will be free. This does not mean the package including approved hardware, expertise and training will be free. That's where the money is made. Odds are that it'll be cheaper than the Diebold solution.
Just remember to keep it simple at every level, and it must have an attractive GUI. Show, don't tell. The average Joe hates to read, and it's easier to go multilingual if text is at a minimum. Aside from the user experience, these must be set up, used, maintained, and read by relatively untrained people. KISS always applies.