It's hard to know what the future holds. Much of the leveling off of China's growth isn't about changes in management style but about slowing exports and a lack of internal demand.
It seems axiomatic that there will be attempts made to consolidate power by Xi Jinping given some of the weakness of Hu Jintao and the perceived threats from the likes of Bo or other who may follow him.
There are so many challenges to PRC leadership it's hard to know where to start or what could really hobble them in ways they don't know how to fix. The economy is a big one, there's so many places for it to go wrong (currency manipulation, corrupt/inept banking, bubbles, etc etc) and a lot of players with money on the table The environment. Foreign relations (Japan, the rest of Asia, the US...). Resources. Internal dissent.
I think the economy is a major issue; mismanaged, a lot of people who were doing very well could be doing a lot less well. This could be a complex threat for CCP leadership, as a major economic downturn (or a poorly managed one) could result in a twisted alliance between officials with investments and entrepreneurs vs. CCP leadership. And then let's not forget the sizable new and aspirational classes benefitting or hoping to benefit from the economy. Their loyalty to the government may be tested in a downturn.
I suspect that it isn't that they haven't faced the risk of going under, it's that they are too worried about going under and losing what they have and therefore unwilling to do anything that risks their current holdings.
I don't know if anyone has written it, but I suspect there's a great PhD thesis to be written studying the relationship between employee stock ownership, stock options and company innovation and risk taking.
I would wager that as more of the leadership has stock and options in otherwise successful companies, the more risk averse they are and the more willing they are to resist innovation because it threatens what they have (or may soon get).
For unsuccessful companies or those not successful it probably has the reverse motivation -- the stock isn't worth anything until they are successful, so the risk is not innovating.
Re:Ballmer is evidence of the role of luck in life
on
Ballmer To Retire
·
· Score: 1
He may not have been selling used cars, but would he have been in a senior leadership position at a giant software company based solely on his math SATs and his Harvard degrees?
Relatively speaking, a lot of 800s have been awarded and Harvard has likely matriculated a number of math/econ grads.
Maybe he would have become a mid-level guy at a big 6 accounting/consulting company bullying junior associates into working 80 hour weeks while taking mid-level clients out for overpriced steak dinners, cocktails and a trips to the strip club?
Or would he have ended up the Math professor at a midwest lib arts college?
It seems unlikely his individual qualifications would have automatically made him top leadership material.
I've heard that geese are pretty smart and will learn to avoid hunters or areas where they have been hunted. I think the lesson was "hunt early, and as far north as you can" because the birds you encounter will have seen fewer hunters and will be more inclined to land for feed and decoys.
After they have been shot at a few times if the setup looks like what they've experienced before they will not fly low enough to be shot at nor land.
There's also no telling the geese "coming back" are actually coming back, or whether they are different geese who see an empty beach and figure its up for grabs.
I don't know how much AI they throw at code analysis to ensure compliance with their rules. Are they able to identify or at least flag as suspicious possible interpreters? Having iOS apps with interpreters has been an issue in the past, so it wouldn't surprise me if they could ID them.
Being able to include code you wanted to execute without having it show as code at all (ie, hidden in included images, etc) seems like it would sneak past the censors more easily.
Let's say you submit an app to the app store, and like many it's designed to do something fairly idiotic that today's kids find funny, say, take a picture and then superimpose the picture onto a set of background images included with the app.
Now, let's say the app writer has steganographically embedded "naughty" code in the background images, maybe even going so far as to spread the code across all the images, encrypt, etc. to make it difficult to find.
Can the app modify itself by taking its hidden code from the images and actually execute it? Can you download "new" code from the internet, even if its steganographically hidden? It seems like you shouldn't be able to do this, like the apps should be sandboxed from modifying their own code just to prevent importing unapproved code.
Obviously with FC you're talking a physically distinct network, although I'm sure someone, somewhere uses FC for data networking.
With iSCSI, most places I do work at have converged network infrastructure with iSCSI traffic and "normal" network traffic on the same hardware; the separation of storage traffic is purely logical (VLANs) and not based on a specific physical isolation into discrete hardware. This means less equipment and more flexibility in getting access to storage.
Only in really large setups do you find a lot of dedicated Ethernet for iSCSI, and that's more about maintaining sanity of the physical layout, usually with traffic split over multiple switches, LACP links between and LACP uplinks to the core.
The core reason why is to avoid the shitty reliability nightmare that contemporary mechanical HDDs represent and to get a bump in performance.
My current environment is 6x2TB Seagate 7200s in RAID-10 and I find with a virtualization workload good throughput dies off pretty quickly. Sure, a single contiguous large write or read can saturate the link, but 2 ESXi hosts and 6-8 busy VMs really brings up the latency and brings down the performance.
After setting this up last fall, I find I made it bigger than I really need and that it doesn't deliver the performance I'd like. With at least one dead disk in a year, 512GB SSDs (5xRAID-5, with one hotspare) seems like a reasonable performance boost for the kind of active disk space that makes sense.
Now I just need to swallow the $1600 and headache to go with SSDs.
A RAID array is not a SAN but I have yet to see an actual SAN configured JBOD only, if it's even an option. I'm not sure how you would aggregate the storage of single SSDs without RAID.
And I don't know what's moronic about home SANs, mine has ~7 TB storage and volumes exported via iSCSI and NFS to 3-4 systems.
What's wrong with RAID redundancy techniques for SSDs? Between the aggregation required for larger LUNs, I would think you would want to hedge the risk of a device failure.
While I sort of share in the nostalgia for drive-ins (I first saw Star Wars at a drive in!), they seem very anachronistic, relying on cars, large amounts of real estate used really inefficiently, etc.
It makes me wonder, why is there no other innovation in the world of movie theaters? Incorporating good restaurant food, bars, better seating, etc? I can think of one theater with a bar and better food.
I previously would have maybe wanted this but not been willing to spend the money or expose my storage to disk failure with consumer SSD.
I'm thinking now it's getting to the point where it might be reasonable. I usually do RAID-10 for the performance (rebuild speed on RAID-5 with 2TB disks scares me) with the penalty of storage efficiency.
With 512GB SSD sort of affordable, I can switch to RAID-5 for the improved storage efficiency and still get an improvement in performance.
I know I've read analyses from economists that have focused on tuition and student loans as basically a feedback loop, with student loan amounts and tuition ratcheting up at equal paces.
While I understand the link between university administrators and student loans, I hadn't thought of the link between building contractors and student loans. Obviously administrators stand to gain from bigger budgets, since it means more responsibility and higher salaries (or moves to new jobs to demonstrate how they can grow programs...)
It's a great political story because contractors and building trades unions are a major political force and might actually be capable of working both sides, getting legislatures to encourage universities to expand and encouraging legislatures to increase student loans.
Personally, I'm shocked at the scale and amount of new buildings at the University of Minnesota,relative to when I went there in 1985. When I was student there I think I could count on one hand the number of buildings that had been built on campus in the previous 20 years and it seems like there are more than double the number of 20 year or less old buildings now.
People don't want "the real story" of Apple or even of Jobs.
They don't care about Woz's tech wizardry making Apple computers what they were or the other people (or institutions, *cough*PARC*cough*) that made Apple what it is.
They want a story about a hero, a guy who through sheer force of personality made an iconic computer company and then came back and "saved it" and made it even better than it was, creating the iPhone, etc.
A story about a narcissist who serially manipulated people, refused to support his pregnant girlfriend, and relied on the achievements of others to promote himself and his company and created an image as the guy who was responsible for all of it isn't what people want in a story.
(You could also take a Chomesky-esque view, and say that this is what the corporate power structure wants -- people to believe that CEOs are singular geniuses, alone responsible for a company's success...)
Well, there's a postcard version of the Amish in their button-free clothes, hand-making all their stuff and living bucolic lives.
And then there's the real world version of the Amish, where young people smoke, drink, and drive cars before they become full members of the church, the internecine religious conflicts involving sects, beard cutting, etc.
I'm pretty sure that despite the awesome appearance of tech-free Amish life, there's a lot of psychological stress maintaining such an existence in the face of the modern world, along with all the stuff that goes on behind closed doors. I have to believe the Amish have their own problems with violence, sex abuse, etc -- we just don't hear about it because their culture keeps a lid on it.
There's always someone who says "we're not that different" and then sets up a completely bogus comparison. Soldiers and civilians aren't in the same league in terms of rights.
The reality is that the UK and its byproducts don't really have freedoms like we have, and various examples of government misconduct related to whatever paranoia is getting votes these days don't make us "like them."
Will Microsoft shareholders ever tolerate pruning for growth?
Split Microsoft into three companies and spin them off as seperate entities with perpetual cross-licensing agreements for all MS technologies released and planned for release in the next 3 years.
One, the business entity selling Office, SQL, Exchange, Windows.
Two, the consumer entity selling Xbox, Phone, Surface, Zune, and whatever else they are pushing at consumers.
Three, the nascent services entity selling web mail, hosted Exchange, Azure, all cloud-based products.
Company one remains Microsoft. It will probably shrink some as PC sales slide but it may become stronger (the way a tree does when pruned right) because it has a strong focus on products with a lot of history and near-universal buy-in in the business world. They can adapt without worrying about trying to be all things to all platforms and focus on what their specific market wants.
Company two can begin to truly innovate without old guard apparatchiks hobbling it. They may build on Winphone or Surface in ways that aren't tied to the old company and actually challenge Android or Apple.
Company three has the biggest uphill battle but also possibly the widest possible future/horizon as it represents services and products with a lot of growth potential.
I know MS wants to try to integrate all of this, but I wonder if maybe it's just not possible to coordinate that much technology across so many platforms. Even the military has trouble doing that and they have even MORE money than MS and, well, military discipline and force behind them.
I misspoke, my SSD replacement was a 500GB Samsung, not a 750. One reason (besides cost) I held out so long is I thought I *needed* a 750 -- I cluttered a bunch of shit (a dozen SBS 2003 ISO images, for one) and found the 500 would work for me.
I agree that it is a meaningful improvement over a pure platter drive, but for me I seemed to deviate just a little too often from the same-tasks, different-day kinds of usage pattern and when I did performance fell off, especially for boot up and some application launches.
My sense is that Seagate is right that about 9 GB flash is the point of diminishing returns for ordinary users and block-based caching, but I also think that more flash would equal better performance, especially if there was some way to dedicate flash to specific programs or operating system files.
I think it would be interesting to have a combo hybrid drive:
64 GB Flash, 1 TB platter. Flash would be adjustable 0-64 GB SSD with the unused portion used as a cache for the platter. The SSD would be write mirrored to a non-OS visible partition and in case of SSD failure could be made visible.
So in one package you could have high data storage densities, flash usable as you saw fit with HDD backing for pure flash operations.
I've seen RAID controllers (the SATA/SAS kind) that will support "read-biased" mirroring. The idea being that you have a RAID-1 set that writes more or less as per normal, but biases reads to the SSD.
I'm not sure how it compares in performance to other solutions.
Another idea would be one of the dedicated parity RAID configurations and have the physical disk be the parity disk. With a decent amount of cache and a snappy controller CPU you ought to get the best of all worlds, fast reads and writes with the safety net of physical HDD parity should you lose an SSD.
I had a Seagate Momentus XT (750 GB hybrid) and I replaced it with a Samsung 750 GB SSD. The pure SSD solution is noticeably faster in all respects, especially in boot up, and this is with a machine now using Truecrypt whole disk encyption (wasn't using it on the Momentus).
The Momentus was a good upgrade until SSDs in the size I wanted were reasonably priced, but performance wise it isn't in the same league as a SSD.
The hybrid SSD solution really shows its weakness when you deviate from "normal" behavior, and this can be anything from an application upgrade, running Windows updates, or accessing stuff you don't use that much. Performance just seems back to dismal levels and I suspect that it takes a while for the cache to re-optimize if the deviating disk activity is at all intensive.
I think the hybrid concept is interesting, but I think you need more cache and a way to optimize the cache not just not most recently accessed blocks but for the operating system and applications in use, too.
I would think a large-scale project like this would contribute to economic growth, not instability or stagnation, much like wartime economics helped pull the US out of the depression.
It's not like a bunch of rich people are going to magically transform $900 billion into a space station. It would take millions of workers directly working on the project and millions more working indirectly for suppliers, raw materials, etc.
A project like this won't happen, but I've often thought it would make sense to have a very large scale engineering project to both solve a specific problem and to create a lot of jobs. The first thing that always pops into my mind is a terawatt scale nuclear power plant in the Arizona desert used to pump and desalinate sea water from the gulf of California.
I can only guess, but I would imagine that a system like this would partition its storage into a "system" area and have a power-on sequence that would run an "initialization" sequence that would blank the system area and then re-load it if desired.
Well, pre-installed ROMs aren't exactly the same thing. Apple ][s would run a flavor of BASIC from ROM, but software was executed from RAM. It was also immutable, so any variables or data had to be stored in RAM and were subject to loss with power. DOS required booting from media, although I'm sure somebody figured out a way to pack DOS into maybe INTBASIC and burn to an EEPROM that would be the active boot ROM in slot 0.
I don't recall any of my early Macs (starting with the Mac Plus) bootable to Mac OS in ROM, although large parts of what we'd now call system libraries (Mac OS Toolbox) were in ROM but were commonly relocated to RAM with patches and upgrades. I still think they required booting from an OS boot media and I don't think Finder was in ROM.
There was a ROM monitor, but I don't think it did much besides let you view memory locations and possibly issue a "goto $hex" type command to continue execution of whatever was addressable.
I've long suspected it's at least as much about control and compliance for management as it is the wages. Given the overhead involved in getting H1Bs versus native workers and the reasonably well accepted performance penalty (weaker talent, communications, bad code) involved in using non-native labor, I think the cost saving is probably a wash if you looked at the labor costs associated with a finished project.
H1Bs have two things going for them as management sees them:
1) Most are from a country with deep social stratification and an in-built deference to those in a higher social strata. You're not just getting less expensive labor, you're getting someone who has been raised in a culture where they have been taught since youth that they are subservient to their betters.
2) The strong desire to stay in America, even if it means living 12 to a shitty 2 bedroom apartment.
Both of those things result in a compliant workforce that eats overtime for free and is happy to switch from a good, interesting job to shit work during shit hours without complaining.
Native-born IT workers usually see themselves at least as intelligent, if not more so, than management and quite often easily challenge management decisions regarding IT, frustrating management control and prestige. With high IT salary demands, it becomes harder for middle tier managers to control their employees and achieve status since their income isn't enough more than their employees to make wealth a strong differentiator.
Slashdot Mirror
It's hard to know what the future holds. Much of the leveling off of China's growth isn't about changes in management style but about slowing exports and a lack of internal demand.
It seems axiomatic that there will be attempts made to consolidate power by Xi Jinping given some of the weakness of Hu Jintao and the perceived threats from the likes of Bo or other who may follow him.
There are so many challenges to PRC leadership it's hard to know where to start or what could really hobble them in ways they don't know how to fix. The economy is a big one, there's so many places for it to go wrong (currency manipulation, corrupt/inept banking, bubbles, etc etc) and a lot of players with money on the table The environment. Foreign relations (Japan, the rest of Asia, the US...). Resources. Internal dissent.
I think the economy is a major issue; mismanaged, a lot of people who were doing very well could be doing a lot less well. This could be a complex threat for CCP leadership, as a major economic downturn (or a poorly managed one) could result in a twisted alliance between officials with investments and entrepreneurs vs. CCP leadership. And then let's not forget the sizable new and aspirational classes benefitting or hoping to benefit from the economy. Their loyalty to the government may be tested in a downturn.
Enron was a trading company in many of the same ways that investment banks, hedge funds, are trading companies. Risk is how you make money.
Microsoft is a products company -- they make products people buy, so for them risk is "Will people buy this?"
I suspect that it isn't that they haven't faced the risk of going under, it's that they are too worried about going under and losing what they have and therefore unwilling to do anything that risks their current holdings.
I don't know if anyone has written it, but I suspect there's a great PhD thesis to be written studying the relationship between employee stock ownership, stock options and company innovation and risk taking.
I would wager that as more of the leadership has stock and options in otherwise successful companies, the more risk averse they are and the more willing they are to resist innovation because it threatens what they have (or may soon get).
For unsuccessful companies or those not successful it probably has the reverse motivation -- the stock isn't worth anything until they are successful, so the risk is not innovating.
He may not have been selling used cars, but would he have been in a senior leadership position at a giant software company based solely on his math SATs and his Harvard degrees?
Relatively speaking, a lot of 800s have been awarded and Harvard has likely matriculated a number of math/econ grads.
Maybe he would have become a mid-level guy at a big 6 accounting/consulting company bullying junior associates into working 80 hour weeks while taking mid-level clients out for overpriced steak dinners, cocktails and a trips to the strip club?
Or would he have ended up the Math professor at a midwest lib arts college?
It seems unlikely his individual qualifications would have automatically made him top leadership material.
I've heard that geese are pretty smart and will learn to avoid hunters or areas where they have been hunted. I think the lesson was "hunt early, and as far north as you can" because the birds you encounter will have seen fewer hunters and will be more inclined to land for feed and decoys.
After they have been shot at a few times if the setup looks like what they've experienced before they will not fly low enough to be shot at nor land.
There's also no telling the geese "coming back" are actually coming back, or whether they are different geese who see an empty beach and figure its up for grabs.
I don't know how much AI they throw at code analysis to ensure compliance with their rules. Are they able to identify or at least flag as suspicious possible interpreters? Having iOS apps with interpreters has been an issue in the past, so it wouldn't surprise me if they could ID them.
Being able to include code you wanted to execute without having it show as code at all (ie, hidden in included images, etc) seems like it would sneak past the censors more easily.
Datagate?
Let's say you submit an app to the app store, and like many it's designed to do something fairly idiotic that today's kids find funny, say, take a picture and then superimpose the picture onto a set of background images included with the app.
Now, let's say the app writer has steganographically embedded "naughty" code in the background images, maybe even going so far as to spread the code across all the images, encrypt, etc. to make it difficult to find.
Can the app modify itself by taking its hidden code from the images and actually execute it? Can you download "new" code from the internet, even if its steganographically hidden? It seems like you shouldn't be able to do this, like the apps should be sandboxed from modifying their own code just to prevent importing unapproved code.
Obviously with FC you're talking a physically distinct network, although I'm sure someone, somewhere uses FC for data networking.
With iSCSI, most places I do work at have converged network infrastructure with iSCSI traffic and "normal" network traffic on the same hardware; the separation of storage traffic is purely logical (VLANs) and not based on a specific physical isolation into discrete hardware. This means less equipment and more flexibility in getting access to storage.
Only in really large setups do you find a lot of dedicated Ethernet for iSCSI, and that's more about maintaining sanity of the physical layout, usually with traffic split over multiple switches, LACP links between and LACP uplinks to the core.
The core reason why is to avoid the shitty reliability nightmare that contemporary mechanical HDDs represent and to get a bump in performance.
My current environment is 6x2TB Seagate 7200s in RAID-10 and I find with a virtualization workload good throughput dies off pretty quickly. Sure, a single contiguous large write or read can saturate the link, but 2 ESXi hosts and 6-8 busy VMs really brings up the latency and brings down the performance.
After setting this up last fall, I find I made it bigger than I really need and that it doesn't deliver the performance I'd like. With at least one dead disk in a year, 512GB SSDs (5xRAID-5, with one hotspare) seems like a reasonable performance boost for the kind of active disk space that makes sense.
Now I just need to swallow the $1600 and headache to go with SSDs.
A RAID array is not a SAN but I have yet to see an actual SAN configured JBOD only, if it's even an option. I'm not sure how you would aggregate the storage of single SSDs without RAID.
And I don't know what's moronic about home SANs, mine has ~7 TB storage and volumes exported via iSCSI and NFS to 3-4 systems.
What's wrong with RAID redundancy techniques for SSDs? Between the aggregation required for larger LUNs, I would think you would want to hedge the risk of a device failure.
While I sort of share in the nostalgia for drive-ins (I first saw Star Wars at a drive in!), they seem very anachronistic, relying on cars, large amounts of real estate used really inefficiently, etc.
It makes me wonder, why is there no other innovation in the world of movie theaters? Incorporating good restaurant food, bars, better seating, etc? I can think of one theater with a bar and better food.
In the 2-5 TB range?
I previously would have maybe wanted this but not been willing to spend the money or expose my storage to disk failure with consumer SSD.
I'm thinking now it's getting to the point where it might be reasonable. I usually do RAID-10 for the performance (rebuild speed on RAID-5 with 2TB disks scares me) with the penalty of storage efficiency.
With 512GB SSD sort of affordable, I can switch to RAID-5 for the improved storage efficiency and still get an improvement in performance.
I know I've read analyses from economists that have focused on tuition and student loans as basically a feedback loop, with student loan amounts and tuition ratcheting up at equal paces.
While I understand the link between university administrators and student loans, I hadn't thought of the link between building contractors and student loans. Obviously administrators stand to gain from bigger budgets, since it means more responsibility and higher salaries (or moves to new jobs to demonstrate how they can grow programs...)
It's a great political story because contractors and building trades unions are a major political force and might actually be capable of working both sides, getting legislatures to encourage universities to expand and encouraging legislatures to increase student loans.
Personally, I'm shocked at the scale and amount of new buildings at the University of Minnesota,relative to when I went there in 1985. When I was student there I think I could count on one hand the number of buildings that had been built on campus in the previous 20 years and it seems like there are more than double the number of 20 year or less old buildings now.
People don't want "the real story" of Apple or even of Jobs.
They don't care about Woz's tech wizardry making Apple computers what they were or the other people (or institutions, *cough*PARC*cough*) that made Apple what it is.
They want a story about a hero, a guy who through sheer force of personality made an iconic computer company and then came back and "saved it" and made it even better than it was, creating the iPhone, etc.
A story about a narcissist who serially manipulated people, refused to support his pregnant girlfriend, and relied on the achievements of others to promote himself and his company and created an image as the guy who was responsible for all of it isn't what people want in a story.
(You could also take a Chomesky-esque view, and say that this is what the corporate power structure wants -- people to believe that CEOs are singular geniuses, alone responsible for a company's success...)
Well, there's a postcard version of the Amish in their button-free clothes, hand-making all their stuff and living bucolic lives.
And then there's the real world version of the Amish, where young people smoke, drink, and drive cars before they become full members of the church, the internecine religious conflicts involving sects, beard cutting, etc.
I'm pretty sure that despite the awesome appearance of tech-free Amish life, there's a lot of psychological stress maintaining such an existence in the face of the modern world, along with all the stuff that goes on behind closed doors. I have to believe the Amish have their own problems with violence, sex abuse, etc -- we just don't hear about it because their culture keeps a lid on it.
There's always someone who says "we're not that different" and then sets up a completely bogus comparison. Soldiers and civilians aren't in the same league in terms of rights.
The reality is that the UK and its byproducts don't really have freedoms like we have, and various examples of government misconduct related to whatever paranoia is getting votes these days don't make us "like them."
Will Microsoft shareholders ever tolerate pruning for growth?
Split Microsoft into three companies and spin them off as seperate entities with perpetual cross-licensing agreements for all MS technologies released and planned for release in the next 3 years.
One, the business entity selling Office, SQL, Exchange, Windows.
Two, the consumer entity selling Xbox, Phone, Surface, Zune, and whatever else they are pushing at consumers.
Three, the nascent services entity selling web mail, hosted Exchange, Azure, all cloud-based products.
Company one remains Microsoft. It will probably shrink some as PC sales slide but it may become stronger (the way a tree does when pruned right) because it has a strong focus on products with a lot of history and near-universal buy-in in the business world. They can adapt without worrying about trying to be all things to all platforms and focus on what their specific market wants.
Company two can begin to truly innovate without old guard apparatchiks hobbling it. They may build on Winphone or Surface in ways that aren't tied to the old company and actually challenge Android or Apple.
Company three has the biggest uphill battle but also possibly the widest possible future/horizon as it represents services and products with a lot of growth potential.
I know MS wants to try to integrate all of this, but I wonder if maybe it's just not possible to coordinate that much technology across so many platforms. Even the military has trouble doing that and they have even MORE money than MS and, well, military discipline and force behind them.
I misspoke, my SSD replacement was a 500GB Samsung, not a 750. One reason (besides cost) I held out so long is I thought I *needed* a 750 -- I cluttered a bunch of shit (a dozen SBS 2003 ISO images, for one) and found the 500 would work for me.
I agree that it is a meaningful improvement over a pure platter drive, but for me I seemed to deviate just a little too often from the same-tasks, different-day kinds of usage pattern and when I did performance fell off, especially for boot up and some application launches.
My sense is that Seagate is right that about 9 GB flash is the point of diminishing returns for ordinary users and block-based caching, but I also think that more flash would equal better performance, especially if there was some way to dedicate flash to specific programs or operating system files.
I think it would be interesting to have a combo hybrid drive:
64 GB Flash, 1 TB platter. Flash would be adjustable 0-64 GB SSD with the unused portion used as a cache for the platter. The SSD would be write mirrored to a non-OS visible partition and in case of SSD failure could be made visible.
So in one package you could have high data storage densities, flash usable as you saw fit with HDD backing for pure flash operations.
I've seen RAID controllers (the SATA/SAS kind) that will support "read-biased" mirroring. The idea being that you have a RAID-1 set that writes more or less as per normal, but biases reads to the SSD.
I'm not sure how it compares in performance to other solutions.
Another idea would be one of the dedicated parity RAID configurations and have the physical disk be the parity disk. With a decent amount of cache and a snappy controller CPU you ought to get the best of all worlds, fast reads and writes with the safety net of physical HDD parity should you lose an SSD.
I had a Seagate Momentus XT (750 GB hybrid) and I replaced it with a Samsung 750 GB SSD. The pure SSD solution is noticeably faster in all respects, especially in boot up, and this is with a machine now using Truecrypt whole disk encyption (wasn't using it on the Momentus).
The Momentus was a good upgrade until SSDs in the size I wanted were reasonably priced, but performance wise it isn't in the same league as a SSD.
The hybrid SSD solution really shows its weakness when you deviate from "normal" behavior, and this can be anything from an application upgrade, running Windows updates, or accessing stuff you don't use that much. Performance just seems back to dismal levels and I suspect that it takes a while for the cache to re-optimize if the deviating disk activity is at all intensive.
I think the hybrid concept is interesting, but I think you need more cache and a way to optimize the cache not just not most recently accessed blocks but for the operating system and applications in use, too.
I would think a large-scale project like this would contribute to economic growth, not instability or stagnation, much like wartime economics helped pull the US out of the depression.
It's not like a bunch of rich people are going to magically transform $900 billion into a space station. It would take millions of workers directly working on the project and millions more working indirectly for suppliers, raw materials, etc.
A project like this won't happen, but I've often thought it would make sense to have a very large scale engineering project to both solve a specific problem and to create a lot of jobs. The first thing that always pops into my mind is a terawatt scale nuclear power plant in the Arizona desert used to pump and desalinate sea water from the gulf of California.
I can only guess, but I would imagine that a system like this would partition its storage into a "system" area and have a power-on sequence that would run an "initialization" sequence that would blank the system area and then re-load it if desired.
Well, pre-installed ROMs aren't exactly the same thing. Apple ][s would run a flavor of BASIC from ROM, but software was executed from RAM. It was also immutable, so any variables or data had to be stored in RAM and were subject to loss with power. DOS required booting from media, although I'm sure somebody figured out a way to pack DOS into maybe INTBASIC and burn to an EEPROM that would be the active boot ROM in slot 0.
I don't recall any of my early Macs (starting with the Mac Plus) bootable to Mac OS in ROM, although large parts of what we'd now call system libraries (Mac OS Toolbox) were in ROM but were commonly relocated to RAM with patches and upgrades. I still think they required booting from an OS boot media and I don't think Finder was in ROM.
There was a ROM monitor, but I don't think it did much besides let you view memory locations and possibly issue a "goto $hex" type command to continue execution of whatever was addressable.
I've long suspected it's at least as much about control and compliance for management as it is the wages. Given the overhead involved in getting H1Bs versus native workers and the reasonably well accepted performance penalty (weaker talent, communications, bad code) involved in using non-native labor, I think the cost saving is probably a wash if you looked at the labor costs associated with a finished project.
H1Bs have two things going for them as management sees them:
1) Most are from a country with deep social stratification and an in-built deference to those in a higher social strata. You're not just getting less expensive labor, you're getting someone who has been raised in a culture where they have been taught since youth that they are subservient to their betters.
2) The strong desire to stay in America, even if it means living 12 to a shitty 2 bedroom apartment.
Both of those things result in a compliant workforce that eats overtime for free and is happy to switch from a good, interesting job to shit work during shit hours without complaining.
Native-born IT workers usually see themselves at least as intelligent, if not more so, than management and quite often easily challenge management decisions regarding IT, frustrating management control and prestige. With high IT salary demands, it becomes harder for middle tier managers to control their employees and achieve status since their income isn't enough more than their employees to make wealth a strong differentiator.