my sister-in-law teaches, and has college students turning in papers (electronically) written on "MS Write" and comparable because they can't afford MSOffice. So she tells them to download Open Office (sometimes gives them a quick demo) and her students love it. They can't believe what OO can do.
Thin client is the way to go. See LTSP.org It's built into Ubuntu.com (couple of easy commands to install). I've used it in home environment and start up manufacturing business. Perfect for Libraries and call-centers. IRC group on freenode.net(I think) at group "#ltsp"
Basically, server with P4 or better and 1-2GB ram is more than enough server for class of 30. Either buy thin clients or re-use former thick-clients (strip out all drives). Thin clients then only need 300-400Mhz, 128-256MB Ram, and maybe USB port for student's flash drive data.
Centrally manage server software. Auto software update is handled via Ubuntu already across the network (OS plus Open Office etc) - popup icon "do you want to update?" even.
Sometimes companies find that if they *gasp* remove half the workforce they move at four-times the speed!
That's what turnaround companies do - and can be successful.
Such as: Often one or two key people can design up a robust process to get things done in a few days..while a team of ten will take months to define that same project.
Or: How many emails get logged between a group of 3 people vs a department of 300? Is the customer experience suddenly better?
As organizations get larger, there are "naturally" more "approval levels" and "communication" and "revisions" and "new requests" and so on. The beast feeds itself, and that is before politics get started and expand the workload yet more.
That's why Entrepreneurs and Small Companies actually get started and have a chance to thrive and allow turnaround companies to exist.
I do a bit of consulting in this area - take a look at Lean Manufacturing. Most companies think it's only for the shop floor, but it really applies to the "Paperwork" processes that mostly hinge on the IT infrastructure. In the beginning code tries to duplicate paper forms and work flow but soon the code begins to define the work process.
Cleaning up the electronic work forms (and reducing the data required to fill out forms - simple as having to put in both State _&_ Zip Code to complex like requiring filling out 15 data fields when only 2 ever get used by the database miners) saves the company lots of wasted time, effort, and cost. Also issues with a lack of error proofing feedback causes waste - users don't know what isn't right and why the system throws them an error.
Big companies have ability to absorb waste in work processes for quite a while, until the economy tips over and then they shed people by the thousands.
It's easy to fix - but the steps need to be taken.
You've either hired right or wrong.
Then you've either managed them right or wrong.
You could graph this into four quadrants. And quickly see that only one quadrant will get any work done from home.
And that sub-set are more valuable than the other 75% you've got. So don't piss them off or they'll leave, telecommuting or not.
A good manager will find that self-motivated quadrant will likely work more than 8 hours - but it will be at all times of the day (you'll see emails timestamped from both midnight and 7am).
So watch the ones you need to and unleash as many as you can.
The purpose of the Patent system isn't to fund lawyers or IPO's but to provide short term protection to encourage inventors to share their new ideas. These ideas then end up in a public forum where other inventors can see the idea and either incorporate it or build upon it - all to advance the success of the country.
Frivolous lawsuits are a result of problems in the court system - "get out of my court with your hot coffee spill claim, buy iced tea next time" - you won't fix that here.
You can always seek a "defensive publication" where the invention is disclosed via the patent office but no patent is sought. There is no real expense to the company doing this (maybe a nominal insertion fee compared to the huge expense of a regular patent + required lawyer billable hours to put it together). That way no one else is able to patent your idea and lock your company out (hence the "defensive" name).
The IPO guys love patents and other IP and value it highly, but often they don't know how to be accurate on this valuation (they are typically just bankers or MBA's without Ccientific/Engineering technical degrees and only real training in finance calculating time-value-of-money).
What can you calculate as the possible long term value of your patent? That will decide more what to do with it.
Due to chip production processes, it's possible to get most of this functionality as "computer on a chip". However, aim a bit higher. There also might be the problem of getting chips that are this underpowered... Manufacturing plants will be setup to produce a range of products but tricky getting old stuff, like there is only one manufacturer for replacement radio vacuum tubes - and many tube types you just can't get.
How many will use this computer if their cell phone is more powerful? Build a platform off a cell phone.
Which brings up... many cell phones run on Linux - so there are light options.
People will want to get on the internet and getting information is important.
A basic paper-back sized box with several USB ports and a TV-out and a low speed Ethernet is all that's needed. Ports are expensive so only have the bare minimum. Then use USB keyboard and mouse like suggested. USB flash drive has full OS + user storage. They can use any machine by just plugging in the flash drive and booting up. Something like Damn-Small-Linux that installs on a 64MB flash drive or larger for more user space.
"...nice for a printer"
My first computer (Apple IIe!) was much more useful after getting a printer hooked to it. Before then it was just a cool toy.
Today the USB flash drives are quite inexpensive - so make sure there are USB ports, and a printer port (at least USB).
This here new fangled Internet, now accessing that would make this computer fancy. But cause greater need of processor speed and ram to display the web.
The desktop is pretty much there as a Killer.
Firefox, Open Office, and Thunderbird
Open Office outputs to.pdf formats. I know people who installed Open Office just to get the.pdf writer.
Also, those programs install in Windows, so most people will already have "test driven" most of what they will use under Linux. Wubi, dual booting, and liveCDs only make it easy to test drive the other programs.
How many cars and car parts do you think get tested, smashed, and redesigned before sold to people? It's called development.
How many computers are stress-tested before being sold at a local Big Box store? Don't want those things burning down the house. It's called development.
Development (and thus failures) are a natural step in the evolution of good things.
Build something. Test it until it breaks. Go back to the lab and improve the design. Repeat.
The problem that SpaceX has is the test models are very expensive - so where a regular business would test 100 door latches produced for homes to see if they can open/shut 100k times each without breaking, they have one or two parts that cost more than an entire subdivision of homes and you get to experiment once.
Meanwhile everyone is watching every "test" - not like finding out if a batch of turn-signal light bulbs fail after three blinks in the bulb-makers development lab resulting in a decision to test with larger filaments - no shocking news story but real situations that go on with everything people use. When you have a rocket problem then the news feeds pick it up.
SpaceX issue here still seems like progress.
Software boxes showing "minimum requirements" for Windows, Mac, Linux would go a long way toward expanding this.
Open Office and Firefox/Thunderbird were the key bits of software to get real movement to Linux. The user community of Ubuntu is fantastic and really the key for Ubuntu to continue expanding.
The next wave of people are those with these types of specialized programs (CAD, graphics, etc). In the end, people just want the OS to be out of their way so they can use the programs they are familiar with.
A risk to the Linux OS community is the number of flavors of Linux (choice is good once you're "inside", but not to convince others to "enter").
So Ubuntu should really add in their name "Ubuntu Linux" so the brand consistency is maintained. Then when software product boxes state "minimum requirements" they can keep one or two major listings (maybe Red Hat, Debian (Ubuntu), etc).
It will be funny a few years from now... Product X has a handy chart of the minimum recommended hardware requirements: Pentium "10+" for Vista, Pentium "8+" for Mac, or a Pentium "3+" for Linux!!
Is a BMW, Mercedes, Lexus, Cadillac, etc easier to use than a Toyota, Honda, GM, or Ford product?
Price is a discriminator of status. A lot of people like to awe their friends with the fact they plunked down $2,500 for a trendy item. Other people like to awe their friends in being frugal ("look at the deal I got for free!").
The two camps will never socialize.
Like automotive quality, ease of use is a given and you only lose customers if they find it is difficult, like Vista not working with a lot of hardware and having to manually track down drivers (if there are any)... Except there are some hard-core Linux users who purposefully avoid Ubuntu because they like the exclusivity of "rolling their own" or using Slax, just like others like the elitism of high/low price, this group likes the exclusivity of hard effort like mountain climbing or extreme sports.
By the way, I've actually found the latest 8.04 HH install is the easiest and hassle free yet of the Ubuntu series (I'm using Xubuntu mostly). Continuous improvement every six months is great and needs to continue. I look forward to each new launch.
Microsoft will fail if it doesn't embrace Open Source. The original Microsoft model was to offer low-cost alternatives to the competition and become the standard. Now Open Source is providing equally good (or frequently better) products for free.
Legions of programmers are coding for fun what Microsoft employees do for their mind-numbing work while having to deal with middle-management political games and all that other day-to-day dribble that makes working only valuable for a paycheck. Those Open Source programmers actually have a reason and find meaning in what they are doing!
How can Microsoft turn things around? By getting involved with Open Source. I blogged about the 10 major details they need to consider a while back:
Discretionary Thoughts
The most successful Open Source Linux version right now is Ubuntu (especially when combined with its related forks like Kubuntu, Xubuntu, Mint, and others). A major reason is Shuttleworth has kept the project on a six month schedule. It's easy to forecast and incorporate consumer feature requests with that kind of upgrade cycle (and it helps focus the programmers and beta testers). Can Microsoft launch a new Windows every Six Months? And continue doing so for over three+ years?
pfSense seems to have failover and is easy to setup on a headless pc (only need a Pentium-II with 128MB ram and strip out all non-essential drives, multiple user boot options from cd/floppy to USB thumb-drive choices will determine which stay).
For the backup WAN line... look for some pringles "cantenna" discussion on google search to create a wave-guide antenna (can be paired with a discarded satellite dish for more signal strength). then aim it at a friend's house several miles away with a duplicate receiver antenna and second pfSense box.
You're more likely to get a backup signal outside of your local dsl/cable spigot. Both you and your friend can share each other's broadband redundancy features for no "extra" cost.
For those trying to figure out "why do you want to do this"... The phone cable in my neighborhood has a bad connection on a few sets of wires... twice a year, every year for the ten years I've been here, when it's foggy and wet (fall/spring) the phones go staticy and neither dial-up nor DSL work until they switch your line. Then a few days later the neighbor calls to complain their line is no good (it's your old wires from two days prior since everyone was just swapped) and they switch them to another set. Comical, this goes on for days until the cable/junction box dries out and everyone is happy (I now know that when I return from work or errands and the phone truck is somewhere on the street that I had better plan on being out). The problem is Murphy is involved that you've got business to take care of - home based businesses for some, consulting clients for others - especially as fuel prices rise and more pressure to telecommute.
If you're stuck in a big company job driving you nuts with boredom... look for a small company that needs help. The small company relies on each person to do the work of twenty - often twenty different types of work (sales, engineering, quality, shipping). Big companies have enough people to redundantly slice up any one job across twenty people. So it's very boring and very quickly and fraught with lots of politics (boredom begets politics - no other way to spice up the day).
Or start a business in your field in your spare time (at home or while surfing at work, it seems). Then you're the one person creating the whole business. When you're successful you can then hire a few people to do the drudge work. With a bit more success you are suddenly a CEO getting 10-100x the pay for not doing anything. But you created a business that employs hundreds of people. (different issues if it's an inherited CEO position by a "professional manager").
Rather than staying bored, look for alternatives. Downsize your lifestyle so you're not a slave to the credit card or mortgage and take a few more chances. Your largest liability is being bored in a job where you are not growing professionally, nor caring to do differently.
Dream up a money saving project that can turn around your company, or double sales, and try implementing it (you'll need to work through all the other bored workers and their "nope won't work here"/"tried that two years ago"/"Joe didn't like that the last time someone suggested it"/etc.
Work is only as boring and lifeless as you wish to make it. "Just whistle while you work"..those darn seven dwarves.
Dangerous to store with HDDs. They are more likely to be incompatible with future interfaces (even USB) than you'd expect. I had a very reputable manufacturer-built computer, an IBM 386 from 1991, that I couldn't reuse the HDD in my 1998 Dell machine because the interface had changed already (maybe there was a converter...or not).
That's without even going into the mechanical and electrical issues with HDD.
Optical will be "better" but they have their own issues.
The only thing known to last for thousands of years are pyramids and sculptures in the desert.
Difference in interpretation, Consumerreports is on what you should realistically consider for yourself. The previous post is the historic mortgage lender/banker model. It's how people get upsized and in trouble but tend to muddle through in non-bubble markets.
The hours stem from the underdog mission. You've got to work hard at all hours to beat the evil empire(s). Anything less and you're showing your not loyal enough to the cause.
Money is just the baseline for a job - if it's ridiculously low then people leave, if it's reasonably close then they stay based on other factors (a lot of research, Maslow and others).
Probably the main reason people stay is the "sense of passion" for the product - they are proud to be working on them which leads into... They still have the sense of a mission or struggling underdog to fight their way beyond 5% market share against the evil empire(s).
Get enough people that want to join such a mission and supply/demand kicks in and you can hire a good engineer for next to nothing. There are a dozen other equally talented engineers willing to work for nothing to be part of something.
Then remember, it's not a computer company anymore... When Wall Street marks up the value of Apple shares it's because of the number of iPods projected to sell in the next quarter, not computers.
Cost of cameras and video/sound editing (especially Linux & Open Source) have really dropped. The internet is a definite distribution medium. Get some people together and make short and long movies. Start a web site. Look for fan sites and fan films (star wars to star trek) for passion and methods. Go to youtube. With costs falling - it's very possible to launch your own brand.
I saw where George Lukas was said to have modeled his rebellion against the movie empire. He was in the business and wanted a hit to get out to make his own stuff. (why o why jarr-jarr?). Have some fun like the guys doing "Chad Vader".
So make your own and bootstrap up.
One of the problems with many of the light distros is you load up a program and there isn't a quick nor obvious way to add the program to the "start menu"... Some do automatically, but most of the lighter ones you have to go searching.../bin..nope../sbin...nope...nope...not there... etc.
I did a Xubuntu install recently and then loaded up many of the lighter weight window managers via apt-get to try them out ( I was going to do another install so I wasn't worried about borking the system). Enlightenment, fluxbox, flwm, icewm, jwm, openbox, and saphire. Plan was to do a server install and an alternate WM.
A couple of them I liked, but I ended up back at Xfce on standard Xubuntu because it still had the best (of the lighter WMs) program finding menu without a bunch of internet searches to find the "tricks". Though Xfce is not as good for menus as KDE (so when recommending someone upgrade their old PC to try out Linux I have them start with Kubuntu).
Go with Ubuntu "server-install" - better updates to future-protect yourself. Hardcore-types might fiddle with Debian, FreeBSD, Gentoo, etc. But there is a growing userbase of the X/K/Ubuntu's that provide many hints and fixes and forums to get and keep you going.
Most P1's don't have the "energy star" ratings... and can be problematic to idle HDDs or go into standby. The PIIs and newer will generally be better about this (I swapped out a P1 motherboard for a PII because the NAS software at the time couldn't spin down the HDD's with the P1).
However, if you build a NAS (like FreeNAS.org) or router (pfsense/monowall) on a stripped out computer it will use something like 32watts or less running (I just finished a pfsense AP on a PII-300Mhz and this was a real reading using a Kill-a-watt unit), and with idle I've had thin clients (LTSP.org) down to 15watts in sleep mode. A dedicated newly purchased router might only be 12-15watts (but that assumes you "recycled" the old unit and spent how much energy manufacturing and shipping the new router?).
The new pc builders also demand 400-500Watt power supplies as "you have to have that much for peak performance" - then they laugh at anyone trying to use a 200Watt ps with modern power hungry CPU's, but 200watt is what was in the P1's and PII's.
Most of the discussion on pre-installed systems here seems to be relative to Dell,... last I saw both Lenovo and HP were also providing Linux desktop (Ubuntu) systems.
,br>
For even more savings _and_ send the Linux message at the same time.. look for the Eee laptop pc... a second slashdot article today shows Asus is releasing a desktop version too.
And if you're really into saving... find someone who is upgrading from XP to Vista and can't use their old computer any more "too slow!" - or go to Ebay. Reformat the hard drive with Ubuntu.com 8.04 and you're on your way. This way you'll save the environment from making a new machine (one statistic is 2 tons of raw materials are needed to make a new pc...not including the energy to transform it from materials into something).
I use a P4-2.4Ghz for email, surfing, office documents for my consulting business (running Kubuntu). I run a refurbished P4-3Ghz machine for Finite Element Analysis (on PCLinuxOS) - heavy number crunching. There's a 1.7Ghz machine running Xubuntu with a better graphics card to play FPS's. All machines were rescued from neighbors replacing/upgrading their windows machines. The only purchases were that I ebay'd a faster processor and additional RAM for the 3Ghz machine.
Slashdot Mirror
my sister-in-law teaches, and has college students turning in papers (electronically) written on "MS Write" and comparable because they can't afford MSOffice. So she tells them to download Open Office (sometimes gives them a quick demo) and her students love it. They can't believe what OO can do.
Thin client is the way to go. See LTSP.org It's built into Ubuntu.com (couple of easy commands to install). I've used it in home environment and start up manufacturing business. Perfect for Libraries and call-centers. IRC group on freenode.net(I think) at group "#ltsp"
Basically, server with P4 or better and 1-2GB ram is more than enough server for class of 30. Either buy thin clients or re-use former thick-clients (strip out all drives). Thin clients then only need 300-400Mhz, 128-256MB Ram, and maybe USB port for student's flash drive data.
Centrally manage server software. Auto software update is handled via Ubuntu already across the network (OS plus Open Office etc) - popup icon "do you want to update?" even.
bliss.
Sometimes companies find that if they *gasp* remove half the workforce they move at four-times the speed!
That's what turnaround companies do - and can be successful.
Such as: Often one or two key people can design up a robust process to get things done in a few days..while a team of ten will take months to define that same project.
Or: How many emails get logged between a group of 3 people vs a department of 300? Is the customer experience suddenly better?
As organizations get larger, there are "naturally" more "approval levels" and "communication" and "revisions" and "new requests" and so on. The beast feeds itself, and that is before politics get started and expand the workload yet more.
That's why Entrepreneurs and Small Companies actually get started and have a chance to thrive and allow turnaround companies to exist.
Discretionary Thoughts
I do a bit of consulting in this area - take a look at Lean Manufacturing. Most companies think it's only for the shop floor, but it really applies to the "Paperwork" processes that mostly hinge on the IT infrastructure. In the beginning code tries to duplicate paper forms and work flow but soon the code begins to define the work process.
Cleaning up the electronic work forms (and reducing the data required to fill out forms - simple as having to put in both State _&_ Zip Code to complex like requiring filling out 15 data fields when only 2 ever get used by the database miners) saves the company lots of wasted time, effort, and cost. Also issues with a lack of error proofing feedback causes waste - users don't know what isn't right and why the system throws them an error.
Big companies have ability to absorb waste in work processes for quite a while, until the economy tips over and then they shed people by the thousands.
It's easy to fix - but the steps need to be taken.
John
Discretionary Thoughts
You've either hired right or wrong. Then you've either managed them right or wrong.
You could graph this into four quadrants. And quickly see that only one quadrant will get any work done from home.
And that sub-set are more valuable than the other 75% you've got. So don't piss them off or they'll leave, telecommuting or not.
A good manager will find that self-motivated quadrant will likely work more than 8 hours - but it will be at all times of the day (you'll see emails timestamped from both midnight and 7am).
So watch the ones you need to and unleash as many as you can.
The purpose of the Patent system isn't to fund lawyers or IPO's but to provide short term protection to encourage inventors to share their new ideas. These ideas then end up in a public forum where other inventors can see the idea and either incorporate it or build upon it - all to advance the success of the country.
Frivolous lawsuits are a result of problems in the court system - "get out of my court with your hot coffee spill claim, buy iced tea next time" - you won't fix that here.
You can always seek a "defensive publication" where the invention is disclosed via the patent office but no patent is sought. There is no real expense to the company doing this (maybe a nominal insertion fee compared to the huge expense of a regular patent + required lawyer billable hours to put it together). That way no one else is able to patent your idea and lock your company out (hence the "defensive" name).
The IPO guys love patents and other IP and value it highly, but often they don't know how to be accurate on this valuation (they are typically just bankers or MBA's without Ccientific/Engineering technical degrees and only real training in finance calculating time-value-of-money).
What can you calculate as the possible long term value of your patent? That will decide more what to do with it.
Due to chip production processes, it's possible to get most of this functionality as "computer on a chip". However, aim a bit higher. There also might be the problem of getting chips that are this underpowered... Manufacturing plants will be setup to produce a range of products but tricky getting old stuff, like there is only one manufacturer for replacement radio vacuum tubes - and many tube types you just can't get.
How many will use this computer if their cell phone is more powerful? Build a platform off a cell phone.
Which brings up... many cell phones run on Linux - so there are light options.
People will want to get on the internet and getting information is important.
A basic paper-back sized box with several USB ports and a TV-out and a low speed Ethernet is all that's needed. Ports are expensive so only have the bare minimum. Then use USB keyboard and mouse like suggested. USB flash drive has full OS + user storage. They can use any machine by just plugging in the flash drive and booting up. Something like Damn-Small-Linux that installs on a 64MB flash drive or larger for more user space.
"...nice for a printer" My first computer (Apple IIe!) was much more useful after getting a printer hooked to it. Before then it was just a cool toy. Today the USB flash drives are quite inexpensive - so make sure there are USB ports, and a printer port (at least USB). This here new fangled Internet, now accessing that would make this computer fancy. But cause greater need of processor speed and ram to display the web.
The desktop is pretty much there as a Killer. Firefox, Open Office, and Thunderbird
.pdf formats. I know people who installed Open Office just to get the .pdf writer.
Open Office outputs to
Also, those programs install in Windows, so most people will already have "test driven" most of what they will use under Linux. Wubi, dual booting, and liveCDs only make it easy to test drive the other programs.
How many cars and car parts do you think get tested, smashed, and redesigned before sold to people? It's called development. How many computers are stress-tested before being sold at a local Big Box store? Don't want those things burning down the house. It's called development. Development (and thus failures) are a natural step in the evolution of good things. Build something. Test it until it breaks. Go back to the lab and improve the design. Repeat. The problem that SpaceX has is the test models are very expensive - so where a regular business would test 100 door latches produced for homes to see if they can open/shut 100k times each without breaking, they have one or two parts that cost more than an entire subdivision of homes and you get to experiment once. Meanwhile everyone is watching every "test" - not like finding out if a batch of turn-signal light bulbs fail after three blinks in the bulb-makers development lab resulting in a decision to test with larger filaments - no shocking news story but real situations that go on with everything people use. When you have a rocket problem then the news feeds pick it up. SpaceX issue here still seems like progress.
Software boxes showing "minimum requirements" for Windows, Mac, Linux would go a long way toward expanding this.
Open Office and Firefox/Thunderbird were the key bits of software to get real movement to Linux. The user community of Ubuntu is fantastic and really the key for Ubuntu to continue expanding.
The next wave of people are those with these types of specialized programs (CAD, graphics, etc). In the end, people just want the OS to be out of their way so they can use the programs they are familiar with.
A risk to the Linux OS community is the number of flavors of Linux (choice is good once you're "inside", but not to convince others to "enter").
So Ubuntu should really add in their name "Ubuntu Linux" so the brand consistency is maintained. Then when software product boxes state "minimum requirements" they can keep one or two major listings (maybe Red Hat, Debian (Ubuntu), etc).
It will be funny a few years from now... Product X has a handy chart of the minimum recommended hardware requirements: Pentium "10+" for Vista, Pentium "8+" for Mac, or a Pentium "3+" for Linux!!
It's not about ease of use, alone.
... Except there are some hard-core Linux users who purposefully avoid Ubuntu because they like the exclusivity of "rolling their own" or using Slax, just like others like the elitism of high/low price, this group likes the exclusivity of hard effort like mountain climbing or extreme sports.
It's about marketing and status.
Is a BMW, Mercedes, Lexus, Cadillac, etc easier to use than a Toyota, Honda, GM, or Ford product?
Price is a discriminator of status. A lot of people like to awe their friends with the fact they plunked down $2,500 for a trendy item. Other people like to awe their friends in being frugal ("look at the deal I got for free!").
The two camps will never socialize.
Like automotive quality, ease of use is a given and you only lose customers if they find it is difficult, like Vista not working with a lot of hardware and having to manually track down drivers (if there are any)
By the way, I've actually found the latest 8.04 HH install is the easiest and hassle free yet of the Ubuntu series (I'm using Xubuntu mostly). Continuous improvement every six months is great and needs to continue. I look forward to each new launch.
Microsoft will fail if it doesn't embrace Open Source. The original Microsoft model was to offer low-cost alternatives to the competition and become the standard. Now Open Source is providing equally good (or frequently better) products for free.
Legions of programmers are coding for fun what Microsoft employees do for their mind-numbing work while having to deal with middle-management political games and all that other day-to-day dribble that makes working only valuable for a paycheck. Those Open Source programmers actually have a reason and find meaning in what they are doing!
How can Microsoft turn things around? By getting involved with Open Source. I blogged about the 10 major details they need to consider a while back: Discretionary Thoughts
For some additional insight on getting Microsoft to help itself, here is another article. Microsoft needs to embrace Lean Manufacturing to get projects completed earlier and ultimately with more commercial success: Microsoft Ignores the Lean Manufacturing Model & how Vista became a Problem
The most successful Open Source Linux version right now is Ubuntu (especially when combined with its related forks like Kubuntu, Xubuntu, Mint, and others). A major reason is Shuttleworth has kept the project on a six month schedule. It's easy to forecast and incorporate consumer feature requests with that kind of upgrade cycle (and it helps focus the programmers and beta testers). Can Microsoft launch a new Windows every Six Months? And continue doing so for over three+ years?
Challenging... But not impossible.
pfSense seems to have failover and is easy to setup on a headless pc (only need a Pentium-II with 128MB ram and strip out all non-essential drives, multiple user boot options from cd/floppy to USB thumb-drive choices will determine which stay).
For the backup WAN line... look for some pringles "cantenna" discussion on google search to create a wave-guide antenna (can be paired with a discarded satellite dish for more signal strength). then aim it at a friend's house several miles away with a duplicate receiver antenna and second pfSense box.
You're more likely to get a backup signal outside of your local dsl/cable spigot. Both you and your friend can share each other's broadband redundancy features for no "extra" cost.
For those trying to figure out "why do you want to do this"... The phone cable in my neighborhood has a bad connection on a few sets of wires... twice a year, every year for the ten years I've been here, when it's foggy and wet (fall/spring) the phones go staticy and neither dial-up nor DSL work until they switch your line. Then a few days later the neighbor calls to complain their line is no good (it's your old wires from two days prior since everyone was just swapped) and they switch them to another set. Comical, this goes on for days until the cable/junction box dries out and everyone is happy (I now know that when I return from work or errands and the phone truck is somewhere on the street that I had better plan on being out). The problem is Murphy is involved that you've got business to take care of - home based businesses for some, consulting clients for others - especially as fuel prices rise and more pressure to telecommute.
If you're stuck in a big company job driving you nuts with boredom... look for a small company that needs help. The small company relies on each person to do the work of twenty - often twenty different types of work (sales, engineering, quality, shipping). Big companies have enough people to redundantly slice up any one job across twenty people. So it's very boring and very quickly and fraught with lots of politics (boredom begets politics - no other way to spice up the day).
Or start a business in your field in your spare time (at home or while surfing at work, it seems). Then you're the one person creating the whole business. When you're successful you can then hire a few people to do the drudge work. With a bit more success you are suddenly a CEO getting 10-100x the pay for not doing anything. But you created a business that employs hundreds of people. (different issues if it's an inherited CEO position by a "professional manager").
Rather than staying bored, look for alternatives. Downsize your lifestyle so you're not a slave to the credit card or mortgage and take a few more chances. Your largest liability is being bored in a job where you are not growing professionally, nor caring to do differently.
Dream up a money saving project that can turn around your company, or double sales, and try implementing it (you'll need to work through all the other bored workers and their "nope won't work here"/"tried that two years ago"/"Joe didn't like that the last time someone suggested it"/etc.
Work is only as boring and lifeless as you wish to make it. "Just whistle while you work"..those darn seven dwarves.
Dangerous to store with HDDs. They are more likely to be incompatible with future interfaces (even USB) than you'd expect. I had a very reputable manufacturer-built computer, an IBM 386 from 1991, that I couldn't reuse the HDD in my 1998 Dell machine because the interface had changed already (maybe there was a converter...or not). That's without even going into the mechanical and electrical issues with HDD. Optical will be "better" but they have their own issues. The only thing known to last for thousands of years are pyramids and sculptures in the desert.
Difference in interpretation, Consumerreports is on what you should realistically consider for yourself. The previous post is the historic mortgage lender/banker model. It's how people get upsized and in trouble but tend to muddle through in non-bubble markets.
The hours stem from the underdog mission. You've got to work hard at all hours to beat the evil empire(s). Anything less and you're showing your not loyal enough to the cause.
Money is just the baseline for a job - if it's ridiculously low then people leave, if it's reasonably close then they stay based on other factors (a lot of research, Maslow and others).
Probably the main reason people stay is the "sense of passion" for the product - they are proud to be working on them which leads into... They still have the sense of a mission or struggling underdog to fight their way beyond 5% market share against the evil empire(s).
Get enough people that want to join such a mission and supply/demand kicks in and you can hire a good engineer for next to nothing. There are a dozen other equally talented engineers willing to work for nothing to be part of something.
Then remember, it's not a computer company anymore... When Wall Street marks up the value of Apple shares it's because of the number of iPods projected to sell in the next quarter, not computers.
Cost of cameras and video/sound editing (especially Linux & Open Source) have really dropped. The internet is a definite distribution medium. Get some people together and make short and long movies. Start a web site. Look for fan sites and fan films (star wars to star trek) for passion and methods. Go to youtube. With costs falling - it's very possible to launch your own brand. I saw where George Lukas was said to have modeled his rebellion against the movie empire. He was in the business and wanted a hit to get out to make his own stuff. (why o why jarr-jarr?). Have some fun like the guys doing "Chad Vader". So make your own and bootstrap up.
Only real protection is to have lots of great ideas. You might have one or two lifted, but eventually you'll win.
One of the problems with many of the light distros is you load up a program and there isn't a quick nor obvious way to add the program to the "start menu"... Some do automatically, but most of the lighter ones you have to go searching... /bin..nope.. /sbin...nope...nope...not there... etc.
I did a Xubuntu install recently and then loaded up many of the lighter weight window managers via apt-get to try them out ( I was going to do another install so I wasn't worried about borking the system). Enlightenment, fluxbox, flwm, icewm, jwm, openbox, and saphire. Plan was to do a server install and an alternate WM.
A couple of them I liked, but I ended up back at Xfce on standard Xubuntu because it still had the best (of the lighter WMs) program finding menu without a bunch of internet searches to find the "tricks". Though Xfce is not as good for menus as KDE (so when recommending someone upgrade their old PC to try out Linux I have them start with Kubuntu).
Go with Ubuntu "server-install" - better updates to future-protect yourself. Hardcore-types might fiddle with Debian, FreeBSD, Gentoo, etc. But there is a growing userbase of the X/K/Ubuntu's that provide many hints and fixes and forums to get and keep you going.
Look for 'perfect server' setups on google. here's an older version of Ubuntu LAMP setup http://ubuntuguide.org/wiki/Ubuntu:Edgy/Servers
Most P1's don't have the "energy star" ratings... and can be problematic to idle HDDs or go into standby. The PIIs and newer will generally be better about this (I swapped out a P1 motherboard for a PII because the NAS software at the time couldn't spin down the HDD's with the P1).
However, if you build a NAS (like FreeNAS.org) or router (pfsense/monowall) on a stripped out computer it will use something like 32watts or less running (I just finished a pfsense AP on a PII-300Mhz and this was a real reading using a Kill-a-watt unit), and with idle I've had thin clients (LTSP.org) down to 15watts in sleep mode. A dedicated newly purchased router might only be 12-15watts (but that assumes you "recycled" the old unit and spent how much energy manufacturing and shipping the new router?).
The new pc builders also demand 400-500Watt power supplies as "you have to have that much for peak performance" - then they laugh at anyone trying to use a 200Watt ps with modern power hungry CPU's, but 200watt is what was in the P1's and PII's.
Most of the discussion on pre-installed systems here seems to be relative to Dell,... last I saw both Lenovo and HP were also providing Linux desktop (Ubuntu) systems.
,br> For even more savings _and_ send the Linux message at the same time.. look for the Eee laptop pc... a second slashdot article today shows Asus is releasing a desktop version too.
And if you're really into saving... find someone who is upgrading from XP to Vista and can't use their old computer any more "too slow!" - or go to Ebay. Reformat the hard drive with Ubuntu.com 8.04 and you're on your way. This way you'll save the environment from making a new machine (one statistic is 2 tons of raw materials are needed to make a new pc...not including the energy to transform it from materials into something).
I use a P4-2.4Ghz for email, surfing, office documents for my consulting business (running Kubuntu). I run a refurbished P4-3Ghz machine for Finite Element Analysis (on PCLinuxOS) - heavy number crunching. There's a 1.7Ghz machine running Xubuntu with a better graphics card to play FPS's. All machines were rescued from neighbors replacing/upgrading their windows machines. The only purchases were that I ebay'd a faster processor and additional RAM for the 3Ghz machine.