That article has been completely debunked because of outlandish starting assumptions. The amount of metal they modeled for a car electric motor was orders of magnitude off. Even a brief scan of that study reveals just complete and utter incompetence that I wonder if it was a high school project.
As for electricity generation, the beauty is that the mix of energy sources for electricity generation can change over time. It can be oil, coal, natural gas, nuclear, geothermal, hydro, solar, biogas, tide, wind, or more. For a gasoline car, it's basically oil + a bit of ethanol for the entire lifespan of the car. Even as it stands, where coal and natural gas are the primary sources of fuel, electric vehicles have lower pollution levels and lower greenhouse gas emissions (mainly due to a shift to natural gas as the predominate producer). It also matter where the pollution occurs, and power plants are typically not downtown. Further, in the U.S., a larger percentage of the energy mix for electricity is domestically produced and not subject to the pricing whims of other governments that are not friendly to the U.S.
Slander/defame anyone, and they might sue! Including you! I bet there are scenarios where you would be willing to go to court for redress. No need to pretend you wouldn't do the same thing in a similar situation.
Your embarrassment and association is completely in your head. Whether or not someone is a jackass is their problem - there are plenty of jackasses out there with any number of issues. It might even be you at times.
As a journalist, we have high expectations that Mr. Broder would reports impartial facts. Since he wrote it in the New York Times, we have expectations about the journalistic integrity of the writer and the facts within the article. The article at best, is misleading and plays loose with the facts. At worst, it is a hatchet job just on the side of possibly escaping legal culpability.
First of all, he has to decide what he was trying to accomplish. He if is trying to test Tesla's supercharger network and that is the primary motivation, then Mr. Broder exceeded the test parameters. It is not that hard to successfully travel where he went using only the superchargers. However, if he wants to exceed the test parameters, then by all means he could have chosen to plug in at any number of other EV charging locations, had chosen to charge fully, or chosen to plug in overnight. The closest analogy I can think of is if a journalist is trying to verify mileage claims of say, a Prius. The mileage claim is provided given certain test parameters. If you drive too fast, you won't get that mileage. If it is too cold or too hot, it won't get the same mileage. So if you want to see if you can get that mileage, restrict yourself to only fueling near the limits of that resulting range, and then drive fast *and* choose to not fuel all the way up, then yeah, you didn't get the mileage. Whose fault is that?
Mr. Broder on several occasions noted temperatures and speeds that were not indicative of what he actually experienced throughout the drive. His writing clearly exaggerates the situation, most of which is his own doing. Further, it's nearly impossible to not see the ability to charge further. As a long time energy reporter for the New York Times, can we reasonably expect that he is this incompetent? Mr. Broder didn't need to be so loose with the facts, since the current generation of BEVs are not really ready for most people. They do need to be plugged in. They are fantastic for those that can afford it as a daily driver, mostly commuting and 2 hour round trips. Cost of ownership has dropped to roughly equivalent of gasoline power cars (battery replacement costs gas costs, probably less repair needed for BEV vs. gasoline car over time). But for road-tripping where multiple back to back full energy transfers are necessary, it isn't as convenient as a gasoline car at the moment. Mr. Broder, as a journalist writing a piece that is expected to accurately portray the facts, could have pointed this out while sticking to the facts and competently operating/handling the vehicle and he failed to do so.
I've already purchased several 3rd party Lightning cables that work well for less than $10. I know there are some awful ones out there, but the ones I bought are pretty good. This project could have just bundled those cables, which probably cost less than $6 in bulk.
As usual, there are a bunch of people spazzing about Apple... and perfectly willing to cozy up to a number of different megacorps that behave just as badly or oftentimes even worse. A number of these megacorps have more questionable patent practices, proprietary connectors, labor practices, environmental policies, and so forth. That's before you get to the damning emails in the Samsung trial after the illegal purging of a slew of relevant evidence. Sigh. I guess as long as you believe your favorite megacorp is better, than so be it, but you are likely just kidding yourself in an exercise in self validation for the product choice you made.
BTW, both mini-usb and micro-usb are very failure prone and do not provide enough charging current. Apple has never been afraid to stake new ground when they see a technical or user advantage.
Wow... you are completely insane. Do you not know Microsoft's very sketchy history? Where were you picketing Microsoft for their developer policies over the decades? Do you have any idea how much it cost to develop for and deploy on various consumer computing platforms like Xbox, Nintendo Wii, Playstation, Windows, etc? Have ever sold any software through a publisher? Obviously not.
You do realize that iOS already runs on x86, right? The simulator that is provided with Xcode is running iOS for x86... and the binary of the app tested in the simulator is an x86 binary, built for iOS.
Cut the size of government and its power and you cut the opportunity for mischief and mayhem.
If only it were that simple. If the functions that a particular government organization is performing are cut and then are merely transferred to private enterprise, then the opportunity for mischief and mayhem remain, at best, the same. In addition, private enterprise is by many metrics less transparent, less accountable, and more profit driven than government. If that function was for the public good, then going private enterprise means less accountability and more mischief and mayhem... but at least with less transparency, you and I might know less about it.
I am not advocating bigger government or smaller government. In the end, there are no easy solutions which makes public policy and the business of government very boring and unsuited to 30 second soundbites. Our system is still very flawed and the way our politicians play the game these days just makes it worse. But of course, it is the people that lets this happen and the people in the end have to decide as collective to fix it.
Ah... you should check your history on the American Revolutionary War. The Americans lost engagement after engagement and was on the brink of destruction for most of the war. The French helped a lot - Ambassador Benjamin Franklin was instrumental in getting England's primary rival involved. The Americans also pioneered assymetrical warfare and fought "unfairly" (in the eyes of the British) - helping to make it unpalatable to continue to occupy (note to the American occupiers today, ironically).
Now, the Americans did hang tough... against all odds, against all conventional reason, against a vastly superior military force for a stand on liberty, freedom, and justice. For that, we should be very proud of our forefathers. But make no mistake, Americans did not kick Britain's ass almost the entire time, and certainly not by ourselves.
It surprises you that money is the overriding concern? It is most likely that Apple is putting those jobs in Cupertino, CA, and the remark about efficiency is exactly that. Apple may find it more efficient to have the software engineers in Cupertino and the cost savings of going to India isn't worth it. If it is worth it, then by all means Apple should be there. Apple has outsourced much of its hardware manufacturing - it made sense to do so. Apple's management has a duty to its shareholders first and foremost... everything else is and should be a secondary concern.
Further, this isn't about outsourcing the software side - this is about Apple setting up another in-house development site. The arguments pro/con outsourcing is mostly irrelevant here. The discussion here also wasn't about the call support centers of which Apple runs a bunch from a variety of countries.
Rude and unpleasant describes almost everyone in customer support. I've dealt with both good and bad Indian customer support. It comes down to the companies, not the nationality or race. If the company wants to cut costs at the expense of customer satisfaction, then whether or not it is India, Buglaria, Ireland, China, Mexico, or anywhere else, it doesn't matter. In the end, as a customer, it is better to deal with people that can actually get stuff done. Getting employees to actually care about the customer is hard enough as it is, even in the U.S. or other affluent countries. Getting it done with rather stilted and limited call center scripts and techniques is even harder.
This will, of course, all shake out over time. The question really is, will the U.S. consumer be willing to spend more money, on a consistent basis, for better support and customer service or not?
First, the BW article speculates... the author doesn't really know. So was it high wages? Was it something else? "We" don't know yet.
However, there are several issues with setting up in India that probably make it less attractive for Apple.
1) Worker loyalty: while all tech workers probably seem like mercenaries these days, it is even more so in India's white hot tech areas. The workers will leave for what we, in the U.S., would consider miniscule salary differences. 2) Worker training: Indian workers are often broad brush trained in "popular" technologies - finding software engineers trained in non-Windows, non-Oracle, non-SAP, or non-J2EE tech is probably much harder to find at a cost effective salary. Again, this is an issue in the U.S. too, but more pronounced in India and many other non-U.S. technology boom areas. 3) Best of the best: Apple is small (workforce numbers) and tends to follow the hire the best of the best (even if they don't give them the best of the best resources to work with). Those that are really good are probably already working in the U.S., or would not find it all that hard to make it into the U.S. The number left of the best of the best in India probably aren't much cheaper these days (one would often have to be 4:1 to 8:1 cheaper to outweigh the below). 4) Big costs (not just money): Apple doesn't have huge projects that require a thousand or thousands of engineers on a single project that might be able to amortize the costs/issues of temporal and geographical displacement. Apple has most of its software engineering done in Cupertino, and it would take a big shift to deal with significant outsourcing or remote development. 5) Core strength: software engineering is Apple's bread and butter, it is what differentiates the hardware, it is its own profit center. Messing with this too much is not a good idea. Apple can't treat this as a commodity item on a balance sheet. 6) Expansion deals went through in CA: Apple bought a large data center and has plans to build another campus in CA - and the review of those deals going through probably meant that this Indian effort doesn't make sense for Apple right now.
None of this particularly means anything with respect to India, India's tech boom, IBM in India, outsourcing to India, etc. This is merely Apple's evaluation on whether or not it makes sense for Apple. These issues have been there, will continue to be there. It is strange that Apple started and effort but then pulled out, but that is better that they are contantly critically re-evaluating rather than what we've seen from some other U.S. companies that have staked huge efforts on "hot trends" that some CIO/CFO/CEO reads in a trade mag, rather than doing true critical analysis. Going to India may make sense for lots of companies, but certainly not to the level we've seen it lately.
Spinning mechanical devices with motor actuated heads is going to fail. It maybe tomorrow, it may be 3 years from now, but it will fail. SCSI, SATA, or UATA. So "not to break in the first place" is just not true. Reliability has nothing to do with the drive interface anyways - it's just that "enterprise" class drives are typically only available with FC or SCSI interfaces. That is not necessarily the case today.
Further, looking at the reliability of a single drive w/o context is meaningless and old school thinking too. You have to consider heat, vibration, the # of spindles necessary to store the required data set size while maintaining protection from single drive failure. Modern ATA RAID setups with 3 x 500GB server class drives to store 1TB of data is going to be more reliable in many use cases than 5 x 300GB SCSI drives. Remember, each additional spindle decreases your overall reliability. So comparing a single drive to another single drive is meaningless by itself. Plus, the cost of the drives alone is 4x more for the SCSI setup.
Besides, old school IT professionals use parallel SCSI. Anything actually critical has been Fibre Channel for some time now - going to fully clustered or at least zone mapping failover to secondary server. In either case, parallel SCSI is, well, ancient. If you truly care about performance and/or resiliency, it's all FC. Otherwise, you don't and SATA or UATA is probably good enough with enough care in design.
Wow... having to carry around two separate laptops is the "best of both worlds." Really?
Does your $1200 laptop have Firewire target disk mode? Does it have a 6 pin Firewire port so that it can power bus powered Firewire devices like iPods and webcams? No? Are there differences? Yes.
The argument you gave with the Dell Lattitude could be given against an IBM Thinkpad. Exactly which Dell Latitude for a total cost of $1200 is comparable in the ways that are important to your boss? I know you have low expectations, but your boss clearly doesn't. Ever compare the difference between setup and daily use of Wifi and Bluetooth between Windows XP, Linux, and Mac OS X? You'll be hard pressed to find the _balance_ of size, weight, features, and operating experience for the same price as a PowerBook, even as Freescale has not given Apple faster G4's. You obviously don't consider things like size, weight, wake from sleep, connectivity, and so forth... much less the issues of spyware, viruses, and so forth that requires security software that can negate the CPU performance difference.
You teach students? That's frightening given the logic issues you have with your post.
1) Cost of computer != cost of parts. 2) Apple is public stock company and their profit margins are filed with the SEC and are under 30% gross including high profit areas like software sales. Hardware sales have a lower profit margin. Your examples are meaningless. 3) 75% of folks include anyone that self selects as "doing serious media stuff" and is hardly a useful metric except to see who reads Digital Media Net's drivel and responds. 4) You do not know what Apple is going to ship and when, either PowerPC or Intel based. You do not know if/when Apple ships AMD CPUs. 5) CPU performance != productivity. 6) encoding does not have to done on the workstation machine, nor does it have to be done on the same platform or CPU. (server farms come to mind) 7) Quicktime's H.264 won't be the only encoder available for Mac OS X on x86 since the performance optimizations for the Windows versions will mostly be able to translate over to the Mac OS X x86 version. 8) Apple has participated in H.264 interoperability testing... the snear over monopolizing was not just unjustified, it didn't make any sense at all. 9) Apple's dual dual cores G5 workstations are starting to ship. Please note that the retail price of Apple's dual dual core G5 tower is lower priced than many if not most Tier 1/Tier 2 dual core Opteron setups from the likes of IBM, HP, Sun, Boxx, and so forth. Note that buying a quality dual core AMD Opteron setup with parts from NewEgg (Tyan motherboard, approved power supplies, etc), you will still have a hard time getting the price under Apple's retail price (assuming you buy memory elsewhere and you buy the same display, both of which are easy and customary).
It is frightening that people allow you to talk to students.
I worked one over at WWDC for 2 hours... our stuff doesn't need six or 9 months to port, as we mostly have Java or Cocoa Obj-C code. However, we do need it for a short period of time for testing. It would be nice to be able to ARD into a Macintel for testing, but $999 for a 1.5 year lease is a bit steep when we won't be able to effectively use the box for very long.
The Opteron systems I've seen that support > 8GB of RAM do so with registered ECC 2GB DIMMs. Until recently, it wasn't easy to find 2GB DIMMs. The cost is somewhere between $450 to $1200 per DIMM (for DDR333), and you'll need 8 of them. You can find some by Transcend on NewEgg. Crucial carries them at > $800/DIMM.
So even though there have been quite a few Opteron motherboards that have 16GB support on the datasheet, vendors haven't had 2GB DIMMs to fill them out readily.
Has anyone tried a 2GB DIMM in an Apple G5 system?
A large portion of the difference is tax. In the U.S., prices are almost always listed w/o sales tax which is added later. The tax varies by state and whether or not the seller does business in that state - if the seller does not, then the seller does not need to collect the tax and pay it - it is up to the purchaser to report the purchase and pay the tax (you can imagine how often that happens).
Then you have local import duties... and possible delay in price updates (in other words, when Apple set the price, the exchange rate was different than it is now). Right now, the iMac 17" 1.8 costs $1,605.74 ex VAT converting pounds to $, and $1,688.92 ex VAT converting euros to $ using a recent exchange rate. Still a big difference from $1,499 in the U.S., but not as big of a difference as your original comparison. Complain to your local Apple sales office and get them to re-price. The machine comes from Asia or Ireland (not sure, but Apple has manufacturing in both), so Apple isn't importing from the U.S.
Have you ever developed software for Mac OS X? Have you ever written an IOKit device driver?
You write with such confidence... and yet you are so wrong.
First, Apple's G5 workstations are price competitive ($0-$1000 cheaper) as compared to dual Xeon or Opteron workstations - so the PowerPC does have cheap and powerful solutions.
Second, you don't have infinite combination of hardware on the x86 side... and any driver written for Windows can be written for Mac OS X. Where do you get off saying that Mac OS X can't "even handle adapting to all that hardware" ??? First of all, Microsoft doesn't write all those device drivers. 3rd parties do and give it to Microsoft for inclusion (in some cases) or provide CD/downloads of the drivers.
Then you go off on the "void your warranty if you open your box" crap. Absolute BS on the Mac. "only the ones that Apple approves" is more crap. Where do you get this crap?
Here is the starting point for Apple's documentation for hardware developers: http://developer.apple.com/hardware/
Here is the starting point for Apple's Mac OS X I/O Kit documentation: http://developer.apple.com/documen tation/DeviceDri vers/devicedrivers.html
Here is a partial list of the hardware available for the Mac (over 4,500 items) most of which did not require "approval" in any way from Apple: http://guide.apple.com/ushardware.lasso
Ah... dual Apple G5 workstations are cheaper than the equivalent Xeon or Opteron workstations, not to mention that dual 2.5GHz G5's can outperform any available x86 workstation at some tasks.
Throw in the cost of Fibre Channel interconnects, and the price difference is even greater.
The only place where Macs are not price competitive is the high performance single processor gaming rig w/o a monitor. That's because Apple doesn't have one, instead Apple chooses to target consumers with an all-in-one design. Even that AIO is competitive against equivalent x86 offerings, but you can't remove the high quality and expensive LCD.
Dantz owns a patent, 5,150,473 Data storage format for addressable or sequential memory media which essentially covers the use of a on-disk catalog to record what is written to tape for faster retrieval and creating incrementals.
This patent can be very cumbersome for companies trying to enter the Mac OS X backup market. With that said, there are quite a few backup solutions available or coming to Mac OS X - BakBone, Avail, SGL, Tolis Group, and more. I know that the Tolis Group doesn't use a catalog the same way and doesn't do point-in-time incremental snapshots like Retrospect does. I don't know if anyone else coming to Mac OS X does.
It is rumored that OmniGroup's OmniBackup was killed over this patent issue. Too bad, since that was the only tape backup application for Mac OS X Server at the time.
I doubt Linux has more software available than Mac OS X. After all, lots of "Linux" apps easily port to Mac OS X. Some are an easy re-compile or packages may already be available (Fink, DarwinPorts, Gentoo Portage trees too). There are relatively few that are Linux only or even fewer that are only available for Linux on PowerPC in binary form. Mac OS X, on the other hand, has many more software titles available written against Cocoa or Carbon.
"But it makes no sense to have 100 dissimilar systems on the network that 10-20 full time techs have to support, instead of 100 identical systems that a single tech can handle part time."
As a blanket statement, this is wrong. If 100 identical systems end up costing more in support, lack of features, or lack of performance, then it is still inadequate.
Also, efficiency for IT is not necessary efficiency for the business or organization. IT doesn't rule... it serves. Too many IT folks forget that.
This is exactly the problem with the monoculture and the monopoly.
You have no clue what the Macs are capable of doing... and yet you are confident of what they can't do. You have bought into the Microsoft monopoly so completely that it is sickening.
1) Administration - Macs do play well on TCP/IP standard networks with several kinds of directory services. Macs do interoperate with Active Directory at various levels and standard things like LDAP, Kerberos, NIS and so forth. Not sure what you are talking about with hard-drive-based backup of network storage... near line storage options are abound if that is what you mean. Web caching and filtering can and should be done in a transparent manner - squid among other do this well w/o imposing operating system dependencies. In other words, you don't have to jump through hoops to get Macs to work in most systems unless you specifically exclude them.
2) Mac OS X supports NetBoot and NetInstall as well as Apple System Restore _out_of_the_box_. For a single shell script you can re-image and change the boot volume and restart for multiple machines across the network if you choose to do it that way. Matter of fact, network or removable storage based re-imaging is extremely flexible and low cost (free) - a decided advantage over PC's. Plus, Macs can do target disk mode which gives you yet another option for imaging and recovery. If you didn't know this, you didn't know diddly about administrating Macs and you should reserve judgement. This goes all back to that monoculture problem - you are blind.
3) Cost - there are many ways to count cost and CPU performance is only one factor. eMacs start at $599 in single unit quantity, and Apple reps typically work hard at matching competitive offers. Macs are also available in factory refurbs.
4) Your Dell PC's can't run Final Cut Pro HD and since you are not running Linux on those Dells, you can't run Shake either. You don't get DVD Studio Pro, you don't get Motion, you don't get Logic Pro or Logic Express. And if you want to talk about color - accurate color being and issue for graphics/multimedia, then have fun with those Dells getting them right. Of course, you don't actually want your multimedia and graphics students to work with equipment that they would encounter in the real pro world, right? After all, most pro media production involves Macs to some degree if not competely done on Macs - no reason to expose them to some of it, right?
If resellers think that they can survive on a diet of iMac/eMac sales margins, overpriced memory upgrades and cables, then they'll die. Evolution. There's not enough margin in that kind of business unless you're freaking huge - and even then, CompUSA, Best Buy, and Circuit City all live on a razor's edge. The inventory management as well as investement necessary to have constant attempts to attract shoppers makes it difficult for a small independent to make money on low end consumers. PC shops usually only survive on a diet of repairs, upgrades, and fixing Windows (especially virus/malware infestations). The Mac market needs less virus/malware cleanups and historically, do less upgrades and need less repairs. Local PC vendors make no real money on the hardware.
Another way to look at this is that Apple sells everything at full retail with a pretty restrictive return policy and further, maintains a very limited selection of products. Plus Apple is paying the most premium rent you can imagine for retail space as well as maintaining a very high level of staffing. If you can't compete against that, then you don't deserve to be in business.
With that said, Apple does need to play fair - if they aren't, then Apple deserves to get slapped.
Basically, resellers need to know how to do the Value Add portion of the VAR acronym. That's having video and DTP specialists, having people on staff that can lay out a SAN and configure a complete server rack. People who can implement VPN, custom configure Mac OS X Server's SMB services, integrate Open Directory with Red Hat, etc, etc. In other words, the money is in the consulting and value add services in the professional and enterprise space for a small independent. Those kinds of resellers are plentiful selling Linux, Sun, Dell, HP/Compaq, white box servers, etc. However, many of the "legacy" Apple resellers don't have such expertise and have no prayer of getting to that point - so they will die. Evolution.
Sounds like you need to look at Tyan Opteron motherboard based systems. Or, just get a Newisys based system like a Sun V20z or others like the IBM eServer 325, or various 2nd tier vendors like Pengiun Computing that all offer 1U and 2U rackmount dual Opteron servers that expressly support Linux (and may come with it pre-installed and tested).
Depending on your algorithm and how much tweaking one can do with it, the Apple Xserve G5 or G5 desktop may also be very compelling from a price/performance standpoint (dual Opterons and Xeons can cost quite a bit more than Apple's top end G5s).
As for Intel systems, there's always Itanium2 based systems too... if you have the dough and need possibly the fastest floating point capability inside a single CPU. Possibly fastest because it may or may not be able to achieve such performance depending on your algorithm and how well the compiler can churn out code for it - it is particularly picky this way, even if the axiom applies to all platforms.
Slashdot Mirror
That article has been completely debunked because of outlandish starting assumptions. The amount of metal they modeled for a car electric motor was orders of magnitude off. Even a brief scan of that study reveals just complete and utter incompetence that I wonder if it was a high school project.
As for electricity generation, the beauty is that the mix of energy sources for electricity generation can change over time. It can be oil, coal, natural gas, nuclear, geothermal, hydro, solar, biogas, tide, wind, or more. For a gasoline car, it's basically oil + a bit of ethanol for the entire lifespan of the car. Even as it stands, where coal and natural gas are the primary sources of fuel, electric vehicles have lower pollution levels and lower greenhouse gas emissions (mainly due to a shift to natural gas as the predominate producer). It also matter where the pollution occurs, and power plants are typically not downtown. Further, in the U.S., a larger percentage of the energy mix for electricity is domestically produced and not subject to the pricing whims of other governments that are not friendly to the U.S.
Slander/defame anyone, and they might sue! Including you! I bet there are scenarios where you would be willing to go to court for redress. No need to pretend you wouldn't do the same thing in a similar situation.
Your embarrassment and association is completely in your head. Whether or not someone is a jackass is their problem - there are plenty of jackasses out there with any number of issues. It might even be you at times.
As a journalist, we have high expectations that Mr. Broder would reports impartial facts. Since he wrote it in the New York Times, we have expectations about the journalistic integrity of the writer and the facts within the article. The article at best, is misleading and plays loose with the facts. At worst, it is a hatchet job just on the side of possibly escaping legal culpability.
First of all, he has to decide what he was trying to accomplish. He if is trying to test Tesla's supercharger network and that is the primary motivation, then Mr. Broder exceeded the test parameters. It is not that hard to successfully travel where he went using only the superchargers. However, if he wants to exceed the test parameters, then by all means he could have chosen to plug in at any number of other EV charging locations, had chosen to charge fully, or chosen to plug in overnight. The closest analogy I can think of is if a journalist is trying to verify mileage claims of say, a Prius. The mileage claim is provided given certain test parameters. If you drive too fast, you won't get that mileage. If it is too cold or too hot, it won't get the same mileage. So if you want to see if you can get that mileage, restrict yourself to only fueling near the limits of that resulting range, and then drive fast *and* choose to not fuel all the way up, then yeah, you didn't get the mileage. Whose fault is that?
Mr. Broder on several occasions noted temperatures and speeds that were not indicative of what he actually experienced throughout the drive. His writing clearly exaggerates the situation, most of which is his own doing. Further, it's nearly impossible to not see the ability to charge further. As a long time energy reporter for the New York Times, can we reasonably expect that he is this incompetent? Mr. Broder didn't need to be so loose with the facts, since the current generation of BEVs are not really ready for most people. They do need to be plugged in. They are fantastic for those that can afford it as a daily driver, mostly commuting and 2 hour round trips. Cost of ownership has dropped to roughly equivalent of gasoline power cars (battery replacement costs gas costs, probably less repair needed for BEV vs. gasoline car over time). But for road-tripping where multiple back to back full energy transfers are necessary, it isn't as convenient as a gasoline car at the moment. Mr. Broder, as a journalist writing a piece that is expected to accurately portray the facts, could have pointed this out while sticking to the facts and competently operating/handling the vehicle and he failed to do so.
I've already purchased several 3rd party Lightning cables that work well for less than $10. I know there are some awful ones out there, but the ones I bought are pretty good. This project could have just bundled those cables, which probably cost less than $6 in bulk.
As usual, there are a bunch of people spazzing about Apple... and perfectly willing to cozy up to a number of different megacorps that behave just as badly or oftentimes even worse. A number of these megacorps have more questionable patent practices, proprietary connectors, labor practices, environmental policies, and so forth. That's before you get to the damning emails in the Samsung trial after the illegal purging of a slew of relevant evidence. Sigh. I guess as long as you believe your favorite megacorp is better, than so be it, but you are likely just kidding yourself in an exercise in self validation for the product choice you made.
BTW, both mini-usb and micro-usb are very failure prone and do not provide enough charging current. Apple has never been afraid to stake new ground when they see a technical or user advantage.
Wow... you are completely insane. Do you not know Microsoft's very sketchy history? Where were you picketing Microsoft for their developer policies over the decades? Do you have any idea how much it cost to develop for and deploy on various consumer computing platforms like Xbox, Nintendo Wii, Playstation, Windows, etc? Have ever sold any software through a publisher? Obviously not.
You do realize that iOS already runs on x86, right? The simulator that is provided with Xcode is running iOS for x86... and the binary of the app tested in the simulator is an x86 binary, built for iOS.
Cut the size of government and its power and you cut the opportunity for mischief and mayhem.
If only it were that simple. If the functions that a particular government organization is performing are cut and then are merely transferred to private enterprise, then the opportunity for mischief and mayhem remain, at best, the same. In addition, private enterprise is by many metrics less transparent, less accountable, and more profit driven than government. If that function was for the public good, then going private enterprise means less accountability and more mischief and mayhem... but at least with less transparency, you and I might know less about it.
I am not advocating bigger government or smaller government. In the end, there are no easy solutions which makes public policy and the business of government very boring and unsuited to 30 second soundbites. Our system is still very flawed and the way our politicians play the game these days just makes it worse. But of course, it is the people that lets this happen and the people in the end have to decide as collective to fix it.
Ah... you should check your history on the American Revolutionary War. The Americans lost engagement after engagement and was on the brink of destruction for most of the war. The French helped a lot - Ambassador Benjamin Franklin was instrumental in getting England's primary rival involved. The Americans also pioneered assymetrical warfare and fought "unfairly" (in the eyes of the British) - helping to make it unpalatable to continue to occupy (note to the American occupiers today, ironically).
Now, the Americans did hang tough... against all odds, against all conventional reason, against a vastly superior military force for a stand on liberty, freedom, and justice. For that, we should be very proud of our forefathers. But make no mistake, Americans did not kick Britain's ass almost the entire time, and certainly not by ourselves.
It surprises you that money is the overriding concern? It is most likely that Apple is putting those jobs in Cupertino, CA, and the remark about efficiency is exactly that. Apple may find it more efficient to have the software engineers in Cupertino and the cost savings of going to India isn't worth it. If it is worth it, then by all means Apple should be there. Apple has outsourced much of its hardware manufacturing - it made sense to do so. Apple's management has a duty to its shareholders first and foremost... everything else is and should be a secondary concern.
Further, this isn't about outsourcing the software side - this is about Apple setting up another in-house development site. The arguments pro/con outsourcing is mostly irrelevant here. The discussion here also wasn't about the call support centers of which Apple runs a bunch from a variety of countries.
Rude and unpleasant describes almost everyone in customer support. I've dealt with both good and bad Indian customer support. It comes down to the companies, not the nationality or race. If the company wants to cut costs at the expense of customer satisfaction, then whether or not it is India, Buglaria, Ireland, China, Mexico, or anywhere else, it doesn't matter. In the end, as a customer, it is better to deal with people that can actually get stuff done. Getting employees to actually care about the customer is hard enough as it is, even in the U.S. or other affluent countries. Getting it done with rather stilted and limited call center scripts and techniques is even harder.
This will, of course, all shake out over time. The question really is, will the U.S. consumer be willing to spend more money, on a consistent basis, for better support and customer service or not?
First, the BW article speculates... the author doesn't really know. So was it high wages? Was it something else? "We" don't know yet.
However, there are several issues with setting up in India that probably make it less attractive for Apple.
1) Worker loyalty: while all tech workers probably seem like mercenaries these days, it is even more so in India's white hot tech areas. The workers will leave for what we, in the U.S., would consider miniscule salary differences.
2) Worker training: Indian workers are often broad brush trained in "popular" technologies - finding software engineers trained in non-Windows, non-Oracle, non-SAP, or non-J2EE tech is probably much harder to find at a cost effective salary. Again, this is an issue in the U.S. too, but more pronounced in India and many other non-U.S. technology boom areas.
3) Best of the best: Apple is small (workforce numbers) and tends to follow the hire the best of the best (even if they don't give them the best of the best resources to work with). Those that are really good are probably already working in the U.S., or would not find it all that hard to make it into the U.S. The number left of the best of the best in India probably aren't much cheaper these days (one would often have to be 4:1 to 8:1 cheaper to outweigh the below).
4) Big costs (not just money): Apple doesn't have huge projects that require a thousand or thousands of engineers on a single project that might be able to amortize the costs/issues of temporal and geographical displacement. Apple has most of its software engineering done in Cupertino, and it would take a big shift to deal with significant outsourcing or remote development.
5) Core strength: software engineering is Apple's bread and butter, it is what differentiates the hardware, it is its own profit center. Messing with this too much is not a good idea. Apple can't treat this as a commodity item on a balance sheet.
6) Expansion deals went through in CA: Apple bought a large data center and has plans to build another campus in CA - and the review of those deals going through probably meant that this Indian effort doesn't make sense for Apple right now.
None of this particularly means anything with respect to India, India's tech boom, IBM in India, outsourcing to India, etc. This is merely Apple's evaluation on whether or not it makes sense for Apple. These issues have been there, will continue to be there. It is strange that Apple started and effort but then pulled out, but that is better that they are contantly critically re-evaluating rather than what we've seen from some other U.S. companies that have staked huge efforts on "hot trends" that some CIO/CFO/CEO reads in a trade mag, rather than doing true critical analysis. Going to India may make sense for lots of companies, but certainly not to the level we've seen it lately.
Spinning mechanical devices with motor actuated heads is going to fail. It maybe tomorrow, it may be 3 years from now, but it will fail. SCSI, SATA, or UATA. So "not to break in the first place" is just not true. Reliability has nothing to do with the drive interface anyways - it's just that "enterprise" class drives are typically only available with FC or SCSI interfaces. That is not necessarily the case today.
Further, looking at the reliability of a single drive w/o context is meaningless and old school thinking too. You have to consider heat, vibration, the # of spindles necessary to store the required data set size while maintaining protection from single drive failure. Modern ATA RAID setups with 3 x 500GB server class drives to store 1TB of data is going to be more reliable in many use cases than 5 x 300GB SCSI drives. Remember, each additional spindle decreases your overall reliability. So comparing a single drive to another single drive is meaningless by itself. Plus, the cost of the drives alone is 4x more for the SCSI setup.
Besides, old school IT professionals use parallel SCSI. Anything actually critical has been Fibre Channel for some time now - going to fully clustered or at least zone mapping failover to secondary server. In either case, parallel SCSI is, well, ancient. If you truly care about performance and/or resiliency, it's all FC. Otherwise, you don't and SATA or UATA is probably good enough with enough care in design.
Wow... having to carry around two separate laptops is the "best of both worlds." Really?
Does your $1200 laptop have Firewire target disk mode? Does it have a 6 pin Firewire port so that it can power bus powered Firewire devices like iPods and webcams? No? Are there differences? Yes.
The argument you gave with the Dell Lattitude could be given against an IBM Thinkpad. Exactly which Dell Latitude for a total cost of $1200 is comparable in the ways that are important to your boss? I know you have low expectations, but your boss clearly doesn't. Ever compare the difference between setup and daily use of Wifi and Bluetooth between Windows XP, Linux, and Mac OS X? You'll be hard pressed to find the _balance_ of size, weight, features, and operating experience for the same price as a PowerBook, even as Freescale has not given Apple faster G4's. You obviously don't consider things like size, weight, wake from sleep, connectivity, and so forth... much less the issues of spyware, viruses, and so forth that requires security software that can negate the CPU performance difference.
You teach students? That's frightening given the logic issues you have with your post.
1) Cost of computer != cost of parts.
2) Apple is public stock company and their profit margins are filed with the SEC and are under 30% gross including high profit areas like software sales. Hardware sales have a lower profit margin. Your examples are meaningless.
3) 75% of folks include anyone that self selects as "doing serious media stuff" and is hardly a useful metric except to see who reads Digital Media Net's drivel and responds.
4) You do not know what Apple is going to ship and when, either PowerPC or Intel based. You do not know if/when Apple ships AMD CPUs.
5) CPU performance != productivity.
6) encoding does not have to done on the workstation machine, nor does it have to be done on the same platform or CPU. (server farms come to mind)
7) Quicktime's H.264 won't be the only encoder available for Mac OS X on x86 since the performance optimizations for the Windows versions will mostly be able to translate over to the Mac OS X x86 version.
8) Apple has participated in H.264 interoperability testing... the snear over monopolizing was not just unjustified, it didn't make any sense at all.
9) Apple's dual dual cores G5 workstations are starting to ship. Please note that the retail price of Apple's dual dual core G5 tower is lower priced than many if not most Tier 1/Tier 2 dual core Opteron setups from the likes of IBM, HP, Sun, Boxx, and so forth. Note that buying a quality dual core AMD Opteron setup with parts from NewEgg (Tyan motherboard, approved power supplies, etc), you will still have a hard time getting the price under Apple's retail price (assuming you buy memory elsewhere and you buy the same display, both of which are easy and customary).
It is frightening that people allow you to talk to students.
I worked one over at WWDC for 2 hours... our stuff doesn't need six or 9 months to port, as we mostly have Java or Cocoa Obj-C code. However, we do need it for a short period of time for testing. It would be nice to be able to ARD into a Macintel for testing, but $999 for a 1.5 year lease is a bit steep when we won't be able to effectively use the box for very long.
The Opteron systems I've seen that support > 8GB of RAM do so with registered ECC 2GB DIMMs. Until recently, it wasn't easy to find 2GB DIMMs. The cost is somewhere between $450 to $1200 per DIMM (for DDR333), and you'll need 8 of them. You can find some by Transcend on NewEgg. Crucial carries them at > $800/DIMM.
So even though there have been quite a few Opteron motherboards that have 16GB support on the datasheet, vendors haven't had 2GB DIMMs to fill them out readily.
Has anyone tried a 2GB DIMM in an Apple G5 system?
A large portion of the difference is tax. In the U.S., prices are almost always listed w/o sales tax which is added later. The tax varies by state and whether or not the seller does business in that state - if the seller does not, then the seller does not need to collect the tax and pay it - it is up to the purchaser to report the purchase and pay the tax (you can imagine how often that happens).
Then you have local import duties... and possible delay in price updates (in other words, when Apple set the price, the exchange rate was different than it is now). Right now, the iMac 17" 1.8 costs $1,605.74 ex VAT converting pounds to $, and $1,688.92 ex VAT converting euros to $ using a recent exchange rate. Still a big difference from $1,499 in the U.S., but not as big of a difference as your original comparison. Complain to your local Apple sales office and get them to re-price. The machine comes from Asia or Ireland (not sure, but Apple has manufacturing in both), so Apple isn't importing from the U.S.
Have you ever developed software for Mac OS X? Have you ever written an IOKit device driver?
n tation/DeviceDri vers/devicedrivers.html
You write with such confidence... and yet you are so wrong.
First, Apple's G5 workstations are price competitive ($0-$1000 cheaper) as compared to dual Xeon or Opteron workstations - so the PowerPC does have cheap and powerful solutions.
Second, you don't have infinite combination of hardware on the x86 side... and any driver written for Windows can be written for Mac OS X. Where do you get off saying that Mac OS X can't "even handle adapting to all that hardware" ??? First of all, Microsoft doesn't write all those device drivers. 3rd parties do and give it to Microsoft for inclusion (in some cases) or provide CD/downloads of the drivers.
Then you go off on the "void your warranty if you open your box" crap. Absolute BS on the Mac. "only the ones that Apple approves" is more crap. Where do you get this crap?
Here is the starting point for Apple's documentation for hardware developers: http://developer.apple.com/hardware/
Here is the starting point for Apple's Mac OS X I/O Kit documentation:
http://developer.apple.com/docume
Here is a partial list of the hardware available for the Mac (over 4,500 items) most of which did not require "approval" in any way from Apple:
http://guide.apple.com/ushardware.lasso
Ah... dual Apple G5 workstations are cheaper than the equivalent Xeon or Opteron workstations, not to mention that dual 2.5GHz G5's can outperform any available x86 workstation at some tasks.
Throw in the cost of Fibre Channel interconnects, and the price difference is even greater.
The only place where Macs are not price competitive is the high performance single processor gaming rig w/o a monitor. That's because Apple doesn't have one, instead Apple chooses to target consumers with an all-in-one design. Even that AIO is competitive against equivalent x86 offerings, but you can't remove the high quality and expensive LCD.
Dantz owns a patent, 5,150,473 Data storage format for addressable or sequential memory media which essentially covers the use of a on-disk catalog to record what is written to tape for faster retrieval and creating incrementals. This patent can be very cumbersome for companies trying to enter the Mac OS X backup market. With that said, there are quite a few backup solutions available or coming to Mac OS X - BakBone, Avail, SGL, Tolis Group, and more. I know that the Tolis Group doesn't use a catalog the same way and doesn't do point-in-time incremental snapshots like Retrospect does. I don't know if anyone else coming to Mac OS X does. It is rumored that OmniGroup's OmniBackup was killed over this patent issue. Too bad, since that was the only tape backup application for Mac OS X Server at the time.
Sybase Announces Availability of Enterprise RDBMS on MAC OS X Panther
Oracle Announces Plans to Make Oracle 10g Technology Available to Apple Developers. Oracle Database 10g Early Adopters Release 2 (10.1.0.3) for Apple Mac OS X
I doubt Linux has more software available than Mac OS X. After all, lots of "Linux" apps easily port to Mac OS X. Some are an easy re-compile or packages may already be available (Fink, DarwinPorts, Gentoo Portage trees too). There are relatively few that are Linux only or even fewer that are only available for Linux on PowerPC in binary form. Mac OS X, on the other hand, has many more software titles available written against Cocoa or Carbon."But it makes no sense to have 100 dissimilar systems on the network that 10-20 full time techs have to support, instead of 100 identical systems that a single tech can handle part time."
As a blanket statement, this is wrong. If 100 identical systems end up costing more in support, lack of features, or lack of performance, then it is still inadequate.
Also, efficiency for IT is not necessary efficiency for the business or organization. IT doesn't rule... it serves. Too many IT folks forget that.
This is exactly the problem with the monoculture and the monopoly.
You have no clue what the Macs are capable of doing... and yet you are confident of what they can't do. You have bought into the Microsoft monopoly so completely that it is sickening.
1) Administration - Macs do play well on TCP/IP standard networks with several kinds of directory services. Macs do interoperate with Active Directory at various levels and standard things like LDAP, Kerberos, NIS and so forth. Not sure what you are talking about with hard-drive-based backup of network storage... near line storage options are abound if that is what you mean. Web caching and filtering can and should be done in a transparent manner - squid among other do this well w/o imposing operating system dependencies. In other words, you don't have to jump through hoops to get Macs to work in most systems unless you specifically exclude them.
2) Mac OS X supports NetBoot and NetInstall as well as Apple System Restore _out_of_the_box_. For a single shell script you can re-image and change the boot volume and restart for multiple machines across the network if you choose to do it that way. Matter of fact, network or removable storage based re-imaging is extremely flexible and low cost (free) - a decided advantage over PC's. Plus, Macs can do target disk mode which gives you yet another option for imaging and recovery. If you didn't know this, you didn't know diddly about administrating Macs and you should reserve judgement. This goes all back to that monoculture problem - you are blind.
3) Cost - there are many ways to count cost and CPU performance is only one factor. eMacs start at $599 in single unit quantity, and Apple reps typically work hard at matching competitive offers. Macs are also available in factory refurbs.
4) Your Dell PC's can't run Final Cut Pro HD and since you are not running Linux on those Dells, you can't run Shake either. You don't get DVD Studio Pro, you don't get Motion, you don't get Logic Pro or Logic Express. And if you want to talk about color - accurate color being and issue for graphics/multimedia, then have fun with those Dells getting them right. Of course, you don't actually want your multimedia and graphics students to work with equipment that they would encounter in the real pro world, right? After all, most pro media production involves Macs to some degree if not competely done on Macs - no reason to expose them to some of it, right?
If resellers think that they can survive on a diet of iMac/eMac sales margins, overpriced memory upgrades and cables, then they'll die. Evolution. There's not enough margin in that kind of business unless you're freaking huge - and even then, CompUSA, Best Buy, and Circuit City all live on a razor's edge. The inventory management as well as investement necessary to have constant attempts to attract shoppers makes it difficult for a small independent to make money on low end consumers. PC shops usually only survive on a diet of repairs, upgrades, and fixing Windows (especially virus/malware infestations). The Mac market needs less virus/malware cleanups and historically, do less upgrades and need less repairs. Local PC vendors make no real money on the hardware.
Another way to look at this is that Apple sells everything at full retail with a pretty restrictive return policy and further, maintains a very limited selection of products. Plus Apple is paying the most premium rent you can imagine for retail space as well as maintaining a very high level of staffing. If you can't compete against that, then you don't deserve to be in business.
With that said, Apple does need to play fair - if they aren't, then Apple deserves to get slapped.
Basically, resellers need to know how to do the Value Add portion of the VAR acronym. That's having video and DTP specialists, having people on staff that can lay out a SAN and configure a complete server rack. People who can implement VPN, custom configure Mac OS X Server's SMB services, integrate Open Directory with Red Hat, etc, etc. In other words, the money is in the consulting and value add services in the professional and enterprise space for a small independent. Those kinds of resellers are plentiful selling Linux, Sun, Dell, HP/Compaq, white box servers, etc. However, many of the "legacy" Apple resellers don't have such expertise and have no prayer of getting to that point - so they will die. Evolution.
Sounds like you need to look at Tyan Opteron motherboard based systems. Or, just get a Newisys based system like a Sun V20z or others like the IBM eServer 325, or various 2nd tier vendors like Pengiun Computing that all offer 1U and 2U rackmount dual Opteron servers that expressly support Linux (and may come with it pre-installed and tested).
Depending on your algorithm and how much tweaking one can do with it, the Apple Xserve G5 or G5 desktop may also be very compelling from a price/performance standpoint (dual Opterons and Xeons can cost quite a bit more than Apple's top end G5s).
As for Intel systems, there's always Itanium2 based systems too... if you have the dough and need possibly the fastest floating point capability inside a single CPU. Possibly fastest because it may or may not be able to achieve such performance depending on your algorithm and how well the compiler can churn out code for it - it is particularly picky this way, even if the axiom applies to all platforms.