He switches back and forth between the Mac Mini and the MacBook throughout the article, and explains it was because he needed to return the Mini to the Apple Store in the standard 14 days.
He worked out some deal with a retail outlet to take a later return on the MacBook.
He never "configured" anything, he didn't even try, because he was dealing with retail outlets, had zero budget, and had to return everything to get refunds at the end of the article. He was more interested in what he could get off-the-shelf than trying to contact Apple PR for a loaner or something better... I guess.
Doesn't help that no one needs a review of OSX 6 years after it was first released.
He claims he was thinking about keeping the MacBook, but didn't appear to actually do that... who knows?
ADD poster child... The article's structure about what he's reviewing when is very poor.
They're a profit-making company that makes their own applications to do things their customer base generally needs. Why would they "help" the community making competitive products to their own? They're more than helpful with developer tools, etc... (XCode... yadda yadda), documentation, etc... for the cost of a free signup to the Apple Developer website...
They have no business reason/need to care about open-source, but they DO help it along, considering that they don't have to provide free dev tools. There once was a time when all dev tools for any OS weren't free, ya know? It's the defacto standard to make them free today, one of the reasons Apple does it, I'm sure... but they didn't HAVE to...
The Apple open source community seems to do just fine "without" them, really.
He went way out of his way to make sure he had a free machine for 30 days, thus cutting himself off from any "normal" customizations or upgrades he might have done online via the Apple Store to the Mini, missing the mid-range iMac completely, and various other problems.
He's obviously trying to write a review that looks legitimate with zero budget. Nothing wrong with that, and it's easy to see in the story.
Had he started the article something sooner than six years after OSX came out, and worked with Apple PR and had some real journalistic credentials, he might have gotten some better hardware to review, eh?
You missed one of the oldest and best text editors on the Mac platform, originally designed and released under OS 6. BBEdit. It and it's counterpart TextWrangler are very good, and TextWrangler is free.
During the OS load there's an introductory video to all the "non-standard" type things a Mac has that people overlook. He must have skipped it completely.
Don't read the manual, or at least watch the nifty little show... you can't expect the OS to remind you later that you chose to ignore it. Or the OS would get a bad review if it did.
"We'd start testing on Apple's lowest-end computer, a $600 Mac Mini. We wanted to see if a low-end computer could handle the Mac OS X operating system. We would then move to a higher-end $1500 MacBook."
Both are the bottom of the barrel for performance in their respective classes of machine. One in the desktop category, the other in the laptop category. He didn't even hit "mid-grade" in the performance curve of the overall Apple hardware lineup (presumably because he couldn't afford them), but complaints about performance in the article must fall on deaf reader's ears who understand that Apple has ALWAYS under-powered their lower-priced machines.
Second:
"Additionally, each program does not terminate when you close all its windows. To do so, one has to either choose to quit the application from the menu bar, or right click on the icon in the dock to quit it. This is a boon when you want to keep an application resident in memory because you know you'll use it frequently, and a bane when you close out the windows and forget to close the application."
A skill learned by most regular Mac users is keyboard shortcuts. (Truthfully, by any user of any OS... if they're smart.) A simple Apple-Tab (analogous to ALT-TAB in Windows and in virtually the exact same keyboard location, not much "learning" to do there) and Apple-Q to Quit the application selected, and it's gone in two keystroke sequences.
Just like Windows... ALT-TAB, ALT-F4.
Plus, the OS will swap out anything that's not truly running/doing anything... any modern OS will -- it's not sitting in active RAM making your machine sluggish, unless you left the application DOING something...
Third:
"Yes, there is right clicking in Mac OS X - there has been for some time - and Apple even sells two button mice now. If you're on a notebook without a mouse, holding Ctrl while clicking the trackpad works as well. Right clicking in the dock brings up a list of commands, which include quitting an application. Holding down the alt button while doing so brings up an alternate list of commands - including a "force quit" option for misbehaving applications."
He also never read the manual or looked at the online help -- all Mac laptops today ship with touchpads that understand multiple finger-presses. Drop two fingers on the pad, and hit the single mouse button, instant "right-click" functionality. (Working on my wife's older iBook which doesn't have this functionality or my work IBM/Lenovo laptop that also doesn't have it drives me crazy now...)
Fourth:
"Though the Mac Mini does not have a DVD burner, there was an option to save it as a disc image (an.img file)."
You can buy Mac Mini's with DVD burners, or add external ones. No brainer. Apple's consistent use of "Combo Drive" for CD burning drives, and "SuperDrive" for DVD burners, is a bit obtuse, I'll admit. Clicking and reading the specs on the two choices for Mini's makes the option abundantly clear for anyone REALLY shopping for a Mac.
Fifth:
"Had Apple sold a computer configuration that was easily upgradeable at a lower price point than the quad-core Mac Pro line, I probably would have made the decision to go with it for my evaluation."
He completely disregarded the iMac, probably on aesthetic, not technical merits. Then complains that there's no "mid-range" machine he could have purchased.
Summary:
He didn't do a very good job for a professional reviewer... but we're all used to that from tech writers these days.
Not if you buy the right hardware. Too many people think "pee-cees" are real computers, while they're hideously bottlenecked for I/O.
There ARE solutions out there for absolutely flawless rebuilds of RAID arrays, reconfiguration of them on-the-fly, etc... but they're not in a price range most home users will EVER be able to afford.
The best you can do is LOOK for good in-between solutions. Typically these are still in the "expensive" range for home users. They're out there. They change too rapidly to recommend any of them.
There are some inherent problems with Ethernet as it relates to RF also... many Ethernet devices that are cheaply designed far exceed the "incidental radiation" regulations for RF in some bands.
Holding the antenna of a good quality spectrum analyzer up to an operating piece of Cat 5e carrying "standard" 100 Mb/s Ethernet is a lesson in spectrum analysis and management.
When you find any RF technology that can completely ignore every other unwanted signal "on the wire" that the receiver is receiving other than the one you wanted to receive, let me know.
None of this is "FCC myth", it's cold hard engineering fact. Yes, your flashlight can pass through the laser, but you're talking about frequencies that don't PROPAGATE well.
The original discussion is about 700 MHz, not light. Stay on topic.
The only technology that comes close to being able to really "share" frequencies is CDMA with it's "code hopping" and even CDMA engineers KNOW that only a certain number of transmitters can share a channel that can be "seen" by a specific receiver.
Noise is noise, no matter how you try to filter it. If the S/N ratio, be it real (analog) or digital (mathematical filters) isn't high enough, the signal doesn't get through. Real world receivers have limitations (think automatic gain control here...) if a high power transmitter comes on the air nearby, no matter what you do... you can't force your receiver without losing gain by adding selectivity... to "hear" your weak signal you want further away.
Selectivity and Sensitivity are mutually exclusive in the RF engineering world. Read up on it. The statement stands, "In unlicensed free-for-all's, he who dies with the most and biggest transmitters... wins." It's a statement of physics fact.
Software-defined radio shows promise (currently at a very high cost) in doing better filtering than many other receiver designs, but a strong signal nearby will still push a SDR into limiting and overload without AGC, and once the AGC kicks in, you'll lose the weakest signals below the AGC floor. No matter what.
The people touting no control at all, also have no metrics or basis for their claims. Your analysis is as close as it comes when we talk about unlicensed free-for-alls, and if 900 MHz and 2.4 GHz are examples... trying to do real services that people pay for in unlicensed uncontrolled spectrum would be a joke. Whoever had the most money for the most transmitters and amplifiers, would win.
And considering that there are still LICENSED users of those bands who've all but had to abandon them to the noise floor created by the Part 15 unlicenced gadgets also adds more fuel to your comments.
900 MHz, and 2.4 GHz are already overcrowded wastelands, and spread spectrum technology somewhat covers up the mess that's been made there for the end-users. There are now 15 (most open, unsecured) 802.11 access points accessible from my suburban driveway. We're all interfering with each other, most of the end-users just don't know it. They think the performance numbers they get today are normal. Early adopters have seen it go drastically downhill.
Think in terms of data entry for mobile users or something like that... you pull out your machine, add some data to a system, and later the app can "sync" with the webserver/database. The web app doesn't croak just because it can't talk to the webserver.
Always run more than one controller, is the lesson you missed somewhere. In truly mission-critical systems, a single controller (or anything else single) is NEVER used.
Multiple servers (clustering of whatever form you like), multiple controllers, multiple cables, multiple arrays, multiple power supplies, multiple disks.
Home and low-end users tend to only focus on the latter... multiple disks, and ooh and ahh over boring things like RAID.
Super high-end systems might even have multiple internal backplanes in all of the above.
How much redundancy do you want/need is balanced by available budget. Systems with enough revenue flowing always pay for themselves.
They also offer SLA's and business contracts that would allow a company to sue their assess off if their storage solutions fail. This is the oft-overlooked reason that large proprietary systems are used by businesses. They want "one throat to choke" if something bad happens.
Saying that your $2K "developed by some guy down the hall" storage solution lost your most important data to your customers, is far more painful PR than saying "the EMC went down and we've requested that they come fix it and honor their SLA by refunding our money."
In many cases it's not about the technology -- It's about the realities of a company's fiscal liability if something goes wrong.
I love the cheap route, and I think ZFS is niftier than sliced bread, from a purely technological standpoint.
But I don't think Sun has started writing SLA's and offering free data recovery support on-site for the $2K SATA Terrabyte "cheap-ass-engineering" solution the guy down the hallway built with OpenSolaris... yet.
Would I build one or two and replicate my data across them for redundancy to save $50K? Probably. Would I put something mission-critical that pulls down $50,000 an hour on it? Probably not... but maybe.
That's the kind of decision big corporate management is making when they buy a brand-name high-end product that lives in their data center... "Can my people make a phone call and expect the problem will be fixed within the allotted down-time limitations without hassle or having to keep hideously expensive people on staff to babysit the thing? And if the thing blows up, burns down the data center, whatever... does the vendor have deep enough pockets for me to pay lawyers to go recoup some/most/all of my losses?"
There's a reason there's always a "top-three" in any industry, even if it seems the technology is simple enough "anyone" could build one. Sure they can... but can they SUPPORT it and back up that talk and put their money where their mouths are when the excrement hits the rotating air oscillator.
Re:Why is DNS so complicated anyway ?
on
DNS Complexity
·
· Score: 1
You're mixing up DNS with the implementation of DNS in software, specfically BIND.
There *are* DNS systems that use relational databases which you can query and/or update with standard SQL statements.
BIND and its syntax (while I'm fluent in it and very good at it, so I don't bother with above-mentioned RDBMS-based systems) is NOT DNS.
I'm tired of people claiming that someone else, doing what they want with their property is "wrong" and then forcing that person to do something else with their property.
You want to do something with his property, he will happily SELL it to you and then you can do whatever the hell you feel like.
Instead of invoking the big bad government (EPA), just buy the books and dispose of them however you like.
Step up, or step off. Private property is private property.
Buy books and send them to prisons if you think that's what's right, don't just bitch that someone else didn't do it for you, or worse, try to make it look like your opinion should be heeded and try to force someone else to your will by peer pressure.
Big government is not the answer to some little guy burning his books. Christ. Why does everyone want the government to play mommie and daddy all the time now?
His books, his choice. Period. None of your business. (Which could probably be a reply to this whole stupid article.) He didn't hurt a damn thing.
Why aren't you acting like every other company out there and firing all your (now "worthless") staff and outsourcing to some big Linux geek company? LOL!
(It's a joke, kinda.)
Seriously though -- if the skill-set of the Field Engineering staff doesn't match the products anymore, you *might* have some culling to do. They can learn Linux on their own.
If they're not learning it on their own already, it might be a sign that they're not interested. Find people that are.
Just avoid the checkpoints altogether and get your pilot's license. I've seen people blow more money on a ski trip than a Private rating, and anyway -- it's fun.
Get some Instrument flying skills and that rating too, and you're even better off.
It's not cheap, and no corporation on the planet wants to know that you're flying yourself somewhere, but who cares?
You get to leave when you want, (weather permitting), take along what you want (yes, you're allowed to leave your Leatherman on your belt), and you even get to taxi out ahead of airplanes full of sheep who had to go through the "inquisition" inside the terminal building, and... oh yeah, it's fun!
Side-note... what quality level of hardware carries each type of circuit at your end-points?
If you're bitchin' that you have to regularly reboot a $25 "router" vs. $1000 worth of T1 interconnect equipment, you left out a major variable in your analysis.
Generally I agree with your assessment, but many people leave out that most "DSL routers" are crap. You *can* buy high-quality DSL cards from the likes of Cisco, etc... and you really should compare apples to apples...
You'd like to believe your T1 is circuit-switched end-to-end, wouldn't you? Maybe ten years ago, but not today. Your telco works hard to make sure you don't notice, but if it's voice, it's likely a path that looks something like this...
Your telco does NOT guarantee you remain circuit-switched on ANY circuit any more... It's not cost-effective. Ask 'em, if you must.
Most carriers have started refusing to purchase circuit-switched gear for their CO's, and must approve it at VP levels or higher in cases where circuit-switched gear is required for the job at hand, and they'll hunt high and low for a packet-switched device to take its place before committing to purchase anything circuit-switched to put in their CO's.
Only the INTERFACE to you looks like your traditional circuit-switched network, so you don't see a difference or have to change your equipment at your end for them to reap the rewards of packet-switching the backbone links.
Slashdot Mirror
He switches back and forth between the Mac Mini and the MacBook throughout the article, and explains it was because he needed to return the Mini to the Apple Store in the standard 14 days.
He worked out some deal with a retail outlet to take a later return on the MacBook.
He never "configured" anything, he didn't even try, because he was dealing with retail outlets, had zero budget, and had to return everything to get refunds at the end of the article. He was more interested in what he could get off-the-shelf than trying to contact Apple PR for a loaner or something better... I guess.
Doesn't help that no one needs a review of OSX 6 years after it was first released.
He claims he was thinking about keeping the MacBook, but didn't appear to actually do that... who knows?
ADD poster child... The article's structure about what he's reviewing when is very poor.
They're a profit-making company that makes their own applications to do things their customer base generally needs. Why would they "help" the community making competitive products to their own? They're more than helpful with developer tools, etc... (XCode... yadda yadda), documentation, etc... for the cost of a free signup to the Apple Developer website...
They have no business reason/need to care about open-source, but they DO help it along, considering that they don't have to provide free dev tools. There once was a time when all dev tools for any OS weren't free, ya know? It's the defacto standard to make them free today, one of the reasons Apple does it, I'm sure... but they didn't HAVE to...
The Apple open source community seems to do just fine "without" them, really.
He went way out of his way to make sure he had a free machine for 30 days, thus cutting himself off from any "normal" customizations or upgrades he might have done online via the Apple Store to the Mini, missing the mid-range iMac completely, and various other problems.
He's obviously trying to write a review that looks legitimate with zero budget. Nothing wrong with that, and it's easy to see in the story.
Had he started the article something sooner than six years after OSX came out, and worked with Apple PR and had some real journalistic credentials, he might have gotten some better hardware to review, eh?
You missed one of the oldest and best text editors on the Mac platform, originally designed and released under OS 6. BBEdit. It and it's counterpart TextWrangler are very good, and TextWrangler is free.
http://www.barebones.com/products/textwrangler/
There is no PowerMac line anymore. Mac Pro is what you have to get, but otherwise, you're correct.
During the OS load there's an introductory video to all the "non-standard" type things a Mac has that people overlook. He must have skipped it completely.
Don't read the manual, or at least watch the nifty little show... you can't expect the OS to remind you later that you chose to ignore it. Or the OS would get a bad review if it did.
First:
.img file)."
"We'd start testing on Apple's lowest-end computer, a $600 Mac Mini. We wanted to see if a low-end computer could handle the Mac OS X operating system. We would then move to a higher-end $1500 MacBook."
Both are the bottom of the barrel for performance in their respective classes of machine. One in the desktop category, the other in the laptop category. He didn't even hit "mid-grade" in the performance curve of the overall Apple hardware lineup (presumably because he couldn't afford them), but complaints about performance in the article must fall on deaf reader's ears who understand that Apple has ALWAYS under-powered their lower-priced machines.
Second:
"Additionally, each program does not terminate when you close all its windows. To do so, one has to either choose to quit the application from the menu bar, or right click on the icon in the dock to quit it. This is a boon when you want to keep an application resident in memory because you know you'll use it frequently, and a bane when you close out the windows and forget to close the application."
A skill learned by most regular Mac users is keyboard shortcuts. (Truthfully, by any user of any OS... if they're smart.) A simple Apple-Tab (analogous to ALT-TAB in Windows and in virtually the exact same keyboard location, not much "learning" to do there) and Apple-Q to Quit the application selected, and it's gone in two keystroke sequences.
Just like Windows... ALT-TAB, ALT-F4.
Plus, the OS will swap out anything that's not truly running/doing anything... any modern OS will -- it's not sitting in active RAM making your machine sluggish, unless you left the application DOING something...
Third:
"Yes, there is right clicking in Mac OS X - there has been for some time - and Apple even sells two button mice now. If you're on a notebook without a mouse, holding Ctrl while clicking the trackpad works as well. Right clicking in the dock brings up a list of commands, which include quitting an application. Holding down the alt button while doing so brings up an alternate list of commands - including a "force quit" option for misbehaving applications."
He also never read the manual or looked at the online help -- all Mac laptops today ship with touchpads that understand multiple finger-presses. Drop two fingers on the pad, and hit the single mouse button, instant "right-click" functionality. (Working on my wife's older iBook which doesn't have this functionality or my work IBM/Lenovo laptop that also doesn't have it drives me crazy now...)
Fourth:
"Though the Mac Mini does not have a DVD burner, there was an option to save it as a disc image (an
You can buy Mac Mini's with DVD burners, or add external ones. No brainer. Apple's consistent use of "Combo Drive" for CD burning drives, and "SuperDrive" for DVD burners, is a bit obtuse, I'll admit. Clicking and reading the specs on the two choices for Mini's makes the option abundantly clear for anyone REALLY shopping for a Mac.
Fifth:
"Had Apple sold a computer configuration that was easily upgradeable at a lower price point than the quad-core Mac Pro line, I probably would have made the decision to go with it for my evaluation."
He completely disregarded the iMac, probably on aesthetic, not technical merits. Then complains that there's no "mid-range" machine he could have purchased.
Summary:
He didn't do a very good job for a professional reviewer... but we're all used to that from tech writers these days.
Mod parent up.
Nicely stated. And 100% true.
Not if you buy the right hardware. Too many people think "pee-cees" are real computers, while they're hideously bottlenecked for I/O.
There ARE solutions out there for absolutely flawless rebuilds of RAID arrays, reconfiguration of them on-the-fly, etc... but they're not in a price range most home users will EVER be able to afford.
The best you can do is LOOK for good in-between solutions. Typically these are still in the "expensive" range for home users. They're out there. They change too rapidly to recommend any of them.
Getting there. Yep.
There are some inherent problems with Ethernet as it relates to RF also... many Ethernet devices that are cheaply designed far exceed the "incidental radiation" regulations for RF in some bands.
Holding the antenna of a good quality spectrum analyzer up to an operating piece of Cat 5e carrying "standard" 100 Mb/s Ethernet is a lesson in spectrum analysis and management.
When you find any RF technology that can completely ignore every other unwanted signal "on the wire" that the receiver is receiving other than the one you wanted to receive, let me know.
None of this is "FCC myth", it's cold hard engineering fact. Yes, your flashlight can pass through the laser, but you're talking about frequencies that don't PROPAGATE well.
The original discussion is about 700 MHz, not light. Stay on topic.
The only technology that comes close to being able to really "share" frequencies is CDMA with it's "code hopping" and even CDMA engineers KNOW that only a certain number of transmitters can share a channel that can be "seen" by a specific receiver.
Noise is noise, no matter how you try to filter it. If the S/N ratio, be it real (analog) or digital (mathematical filters) isn't high enough, the signal doesn't get through. Real world receivers have limitations (think automatic gain control here...) if a high power transmitter comes on the air nearby, no matter what you do... you can't force your receiver without losing gain by adding selectivity... to "hear" your weak signal you want further away.
Selectivity and Sensitivity are mutually exclusive in the RF engineering world. Read up on it. The statement stands, "In unlicensed free-for-all's, he who dies with the most and biggest transmitters... wins." It's a statement of physics fact.
Software-defined radio shows promise (currently at a very high cost) in doing better filtering than many other receiver designs, but a strong signal nearby will still push a SDR into limiting and overload without AGC, and once the AGC kicks in, you'll lose the weakest signals below the AGC floor. No matter what.
The people touting no control at all, also have no metrics or basis for their claims. Your analysis is as close as it comes when we talk about unlicensed free-for-alls, and if 900 MHz and 2.4 GHz are examples... trying to do real services that people pay for in unlicensed uncontrolled spectrum would be a joke. Whoever had the most money for the most transmitters and amplifiers, would win.
And considering that there are still LICENSED users of those bands who've all but had to abandon them to the noise floor created by the Part 15 unlicenced gadgets also adds more fuel to your comments.
900 MHz, and 2.4 GHz are already overcrowded wastelands, and spread spectrum technology somewhat covers up the mess that's been made there for the end-users. There are now 15 (most open, unsecured) 802.11 access points accessible from my suburban driveway. We're all interfering with each other, most of the end-users just don't know it. They think the performance numbers they get today are normal. Early adopters have seen it go drastically downhill.
Think in terms of data entry for mobile users or something like that... you pull out your machine, add some data to a system, and later the app can "sync" with the webserver/database. The web app doesn't croak just because it can't talk to the webserver.
Always run more than one controller, is the lesson you missed somewhere. In truly mission-critical systems, a single controller (or anything else single) is NEVER used.
Multiple servers (clustering of whatever form you like), multiple controllers, multiple cables, multiple arrays, multiple power supplies, multiple disks.
Home and low-end users tend to only focus on the latter... multiple disks, and ooh and ahh over boring things like RAID.
Super high-end systems might even have multiple internal backplanes in all of the above.
How much redundancy do you want/need is balanced by available budget. Systems with enough revenue flowing always pay for themselves.
They also offer SLA's and business contracts that would allow a company to sue their assess off if their storage solutions fail. This is the oft-overlooked reason that large proprietary systems are used by businesses. They want "one throat to choke" if something bad happens.
Saying that your $2K "developed by some guy down the hall" storage solution lost your most important data to your customers, is far more painful PR than saying "the EMC went down and we've requested that they come fix it and honor their SLA by refunding our money."
In many cases it's not about the technology -- It's about the realities of a company's fiscal liability if something goes wrong.
I love the cheap route, and I think ZFS is niftier than sliced bread, from a purely technological standpoint.
But I don't think Sun has started writing SLA's and offering free data recovery support on-site for the $2K SATA Terrabyte "cheap-ass-engineering" solution the guy down the hallway built with OpenSolaris... yet.
Would I build one or two and replicate my data across them for redundancy to save $50K? Probably. Would I put something mission-critical that pulls down $50,000 an hour on it? Probably not... but maybe.
That's the kind of decision big corporate management is making when they buy a brand-name high-end product that lives in their data center... "Can my people make a phone call and expect the problem will be fixed within the allotted down-time limitations without hassle or having to keep hideously expensive people on staff to babysit the thing? And if the thing blows up, burns down the data center, whatever... does the vendor have deep enough pockets for me to pay lawyers to go recoup some/most/all of my losses?"
There's a reason there's always a "top-three" in any industry, even if it seems the technology is simple enough "anyone" could build one. Sure they can... but can they SUPPORT it and back up that talk and put their money where their mouths are when the excrement hits the rotating air oscillator.
You're mixing up DNS with the implementation of DNS in software, specfically BIND.
There *are* DNS systems that use relational databases which you can query and/or update with standard SQL statements.
BIND and its syntax (while I'm fluent in it and very good at it, so I don't bother with above-mentioned RDBMS-based systems) is NOT DNS.
The author of Freakonomics came to much the same economic conclusion.
I'm tired of people claiming that someone else, doing what they want with their property is "wrong" and then forcing that person to do something else with their property.
You want to do something with his property, he will happily SELL it to you and then you can do whatever the hell you feel like.
Instead of invoking the big bad government (EPA), just buy the books and dispose of them however you like.
Step up, or step off. Private property is private property.
Buy books and send them to prisons if you think that's what's right, don't just bitch that someone else didn't do it for you, or worse, try to make it look like your opinion should be heeded and try to force someone else to your will by peer pressure.
Big government is not the answer to some little guy burning his books. Christ. Why does everyone want the government to play mommie and daddy all the time now?
His books, his choice. Period. None of your business. (Which could probably be a reply to this whole stupid article.) He didn't hurt a damn thing.
And under your contracts, most companies now own everything you write, at or away from work. Luckily, they rarely exercise that option.
Ahh, don't worry. They just needed a way to say, "We're not giving you a raise again this year."
It has little to do with what they ACTUALLY want you doing with your time at the office.
Why aren't you acting like every other company out there and firing all your (now "worthless") staff and outsourcing to some big Linux geek company? LOL!
(It's a joke, kinda.)
Seriously though -- if the skill-set of the Field Engineering staff doesn't match the products anymore, you *might* have some culling to do. They can learn Linux on their own.
If they're not learning it on their own already, it might be a sign that they're not interested. Find people that are.
Just avoid the checkpoints altogether and get your pilot's license. I've seen people blow more money on a ski trip than a Private rating, and anyway -- it's fun.
Get some Instrument flying skills and that rating too, and you're even better off.
It's not cheap, and no corporation on the planet wants to know that you're flying yourself somewhere, but who cares?
You get to leave when you want, (weather permitting), take along what you want (yes, you're allowed to leave your Leatherman on your belt), and you even get to taxi out ahead of airplanes full of sheep who had to go through the "inquisition" inside the terminal building, and... oh yeah, it's fun!
Finding a "new way" to do something on the same hardware, simply means that you didn't do it right the first time, by definition.
The hardware didn't change, your level of understanding did.
You didn't know the way to tell the computer to do what you wanted accurately enough, the first time, if you "found a new way to do it".
Side-note... what quality level of hardware carries each type of circuit at your end-points?
If you're bitchin' that you have to regularly reboot a $25 "router" vs. $1000 worth of T1 interconnect equipment, you left out a major variable in your analysis.
Generally I agree with your assessment, but many people leave out that most "DSL routers" are crap. You *can* buy high-quality DSL cards from the likes of Cisco, etc... and you really should compare apples to apples...
You'd like to believe your T1 is circuit-switched end-to-end, wouldn't you? Maybe ten years ago, but not today. Your telco works hard to make sure you don't notice, but if it's voice, it's likely a path that looks something like this...
Customer Premesis -> Outside Plant -> DAX -> Switch -> DAX -> Sonus (or similar "soft switch" gear) -> Telco backbone IP cloud
If it's data, it probably looks like this...
Customer Premesis -> Outside Plant -> DAX -> Router -> Telco backbone IP cloud
Your telco does NOT guarantee you remain circuit-switched on ANY circuit any more... It's not cost-effective. Ask 'em, if you must.
Most carriers have started refusing to purchase circuit-switched gear for their CO's, and must approve it at VP levels or higher in cases where circuit-switched gear is required for the job at hand, and they'll hunt high and low for a packet-switched device to take its place before committing to purchase anything circuit-switched to put in their CO's.
Only the INTERFACE to you looks like your traditional circuit-switched network, so you don't see a difference or have to change your equipment at your end for them to reap the rewards of packet-switching the backbone links.