My friend was working on what he thought was a video game audio/video sharing app, I'm not a gamer but I know there are several apps now that side by side video of you playing and what's going on in the game. Turns out it was really cam-girl software.
I agree with what your saying and I'm not saying it was Apple. I will say that when I've noticed it the topics were so unique there has to be something more to it then randomness. With the college example, it wasn't an ad, it ended up being a week of ads plastered everywhere.
Nope, no FB or FB owned apps installed, which still doesn't mean another app isn't part of the larger FB ad network. I also have location services turned off for most everything, I don't run very many apps. I'm not saying it was Apple. I really have no theories on how it happened.
I've had several experiences where random run ins with people I don't know have started yielding lots of ads related to obscure topics we've talked about. For example in an elevator I ran into someone wearing the shirt of a small college I was looking at but passed on, we talked about 1 minute about the school then went our separate ways. Within 15 minutes I started seeing ads for the school even though I lived 1000 miles away and ran into the guy 2500 miles away. It might very well be an option I clicked through on an app, but it's happened on multiple occasions at this point.
If you look at Manhattan one of the main reasons the medallion system was setup was to reduce congestion from an excessive number of taxis on the street. During the great depression people who had cars but no job just became independent taxi companies. At most times of the day there were more cabs parked or driving then potential passengers grinding traffic to a halt and eliminating street parking for most who were taxi drivers. There were some safety, pricing, and quality issues baked into the medallion design as well but the overall point was to provide as safe and consistent travel experience as possible while not over-congesting streets and parking. The medallion system was far from perfect mainly because modifications to the system and responses to changes in consumption move at a glacial pace but the historical precedent is there.
I'm watching some pre-election TV shows I recorded, when the campaign ads come on I can totally understand why Hillary lost, her ads fall into that really bad spot from a production quality standpoint where it's better then the sort of cheesy and comfortable videos and worse then a top notch production. Her messages are all over the place and change almost week to week, except they all tend to focus on the negative. Trumps ads all seem to be very cheesy or masterful productions, and he stays pretty much on his MAGA point. When you watch the Trump videos you actually feel kind of good, even if you do like him. When you watch the Hillary videos you feel bad even if you like her.
There is a fundamental difference between this cycle and previous ones, the outcome might be the same, vast new fields of jobs are created, but there is one fundamental difference and that is technological advance has significantly slowed. If we look at this through a Rip van winkle lens pre and post steam engine it was so different the pre steam engine person would be lost, again for automobiles. If you look through most periods for the past 300+ years taking someone 25-35 years in the future would almost always blow their minds. If we go back 25 years now and look at the differences it's more look at how slick the internet has become, wow look someone merged a Palm, with a cellphone, with a GPS, that's slick. The only thing they'll probably lose their mind over is the lack of privacy from social media. But in prior periods you would have to be retrained in how to live. So if we fast forward 8 years, and we have self driving cars and trucks, support that's really the next generation siri and alexa answering 99% of calls what's rally different. It's not doubling the amount of food I can produce, it's not really giving me any more free time, and it's probably not increasing my ability to produce. All it's really doing is taking out a lower section of the food chain. Heck most of this are things there were thought of in the original Star Trek series in 66, so you could go backover 50 years and people would just go, wow, it's just like Star Trek.
The "don't have enough workers" argument is a poor one. The unemployment rate is between 4-5%, which is near full employment, but it's not at all time lows and the labor participation rate is near a historic low. The number of people engaging in retail sales and food service jobs is near an all time high. If there was pent up demand for more skilled jobs there is ample workforce to take it. Automating trucks, cabs, cars, ordering food, shopping online is going to cut out a bunch of lower wage, lower skilled jobs.
But, is it all bad. If robotic automated farms lower the cost of grain. If automated cattle ranches with lower priced grain lower the price beef. If automated trucks lower the cost of transport. If automated trucks, farms, and ranches lower the price of a burger at McDonalds, and kiosks lower the cost of sale it's not bad. I would be more then happy to share my job with an unemployed trucker, especially if I could cut him off a piece enough to maintain his previous standard of living and mine while we both get more spare time though the lower prices of automation. I have a multitude of side projects I would like to do, books to read, trails to hike, and maybe a game or two to play. Maybe we'll get back to Keynes vision of a shorter work week.
I've seen a couple studies now that really suggest the longer out of med shoot, not necessarily age, has a huge impact on doctor effectiveness and peak effectiveness is like 4/5 years out of Med school. It has to do with the exceptionally slow adoption of new SOPs in the medical field. The old SOPs aren't wrong the new ones are just better and if it's not officially broke people are slow to change. One study looked at Cardiac incident survival rates based off doctors age at the same hospitals so same pool of patients and they saw a pretty signifacent correlation to of age to outcomes and looking though treatment notes it all came down minor tweaks in treatment.
So for comparisons sake my unemployment benefits last for a maximum of 20 weeks and are capped at a maximum of $350/week, out of which taxes are taken. After 10 weeks you must accept any job that pays 120% or $10.50/hour. So you get unemployment for 10 weeks, then you're at McDonalds. It takes 6+ weeks to get started and you have to submit proof of applying for jobs, something like 25/week, or you don't get paid. Any week without proper documentation will result in you case being closed.
There is no easy way to do this compare for two big reasons and a bunch of little ones. The two big reasons are taxes between two states are almost as varied ( less the federal component) as two countries. Some states have no income tax, some have no sales tax, some have no taxes and generate their revenue off of oil extraction rights. So when you start talking US averages which vary greatly you start short changing states. Comparing to other countries also starts bringing out all other weird compares. Our VAT is 15 your sales tax is only 7 on average, but that country might have very low property tax. As someone who recently thought about taking an opportunity in Europe I have to say US taxes were lower for my salary ( trying to take everything into account ) but it wasn't like I kept 80% in the US and only 45% in France. It wasn't that big of a difference.
You also have to look at bang for the buck. Like you have a huge VAT but you aren't paying $700/month for healthcare. When you factor in a lot of what we don't get here or the better version you get there, I think it's pretty easy to see the US taxes are too high argument even though the rate might be lower.
To use an analogy if I have to pay $5 for a meal of 2 slices of bread and warm water and you pay $10 for fresh meat, vegetables, and a beer, you could make the argument that the $5 meal is cheaper and you shouldn't complain about paying more, but you could also say the $5 is crap and not worth $5.
Starting with the Pentium Pro the internal processor core of Intel processors has been essentially 100% RISC. They take the CISC commands and break them into a series of RISC commands and execute. So externally you are correct the commands are 100% CISC but that goes away after decode.
The Federal government pays a lot of money for research and development in a lot of areas, architecture, bridges, roads, jet engines, custom ASICs, etc etc. I used to design custom racks, brackets, conduit routing, power/heating/cooling systems for electronics. Think Humvees with quarter racks to a full mobile data center. It eventually got to the point where we were only doing something new/innovative every 3rd or 4th deal. Every other deal was use the bracket designed for A, the rack from B, the generator from C, etc. If we had to release those cad drawings we would have had no competitive edge. If they're saying code should be reusable across agencies and parts should be made open, when stop just at software?
In regards to the women on women ism. I heard an interview with Claudia Goldin, professor at Harvard who studies gender equality. She holds that most of the pay gap can be controlled for. The interesting part is when they came to that small part that remained the interviewer asked, if woman are the victim who is the oppressor? Goldin said the natural response is to look at it in the binary, if women are the victims then not women, or men, must be the oppressors, but this isn't the correct way to see it, because this implies that women can't or won't impact earnings from other women and that's just isn't true from the statistics. Women both from their internal and external behaviors contribute, whether is asking for more flexible hours in exchange for lower pay or maintain that hourly flexibility should result in reduced wages.
Not all people are sexist or racist, so at some point the non-sexist person would see a female applicant, see them for the quality they were, and see the deal, 72 cents on the dollar, they could even split the difference, save 14% and have a competitive advantage, rinse repeat. This would ultimately create a mostly female, reasonably yet underpaid company with a massive economic advantage. I remember reading once that someone even tried to do it. Staff a chunk of his business with 72 cent on the dollar woman. He couldn't find them. While I believe sexism is still there and there is still some gender based wage discrimination in some businesses and some industries the gap number just isn't a valid apples to apples comparison.
A couple of reasons. First as others have mentioned IBM still needs the lines. IBM's processor design is fairly integrated. It needs custom circuits which really on their fabrication technology. Their chip design process is the antithesis of fabless development. So they can't just shut off lights to the fab without crippling the Power, Mainframe, and high end storage business for years to come. The other issue is there are customers getting chips manufactured. If you shut that down there are typically very steep contract termination fees. This is really 6 year wind down with higher costs every year. The fab business is very cutthroat, you have to hit fairly high yields and have the line near full capacity or else you lose a boatload because the fixed costs are so high. They've been getting out of this business for years, especially on the low end. This is just the last step.
IBM is deploying system I on power now but they are still making custom z processors for the mainframe, although the mainframe can also have some power and x86 processors they really aren't for mainframe processing just hi RAS local access for offloading certain workloads.
It's going to be interesting what arm does. There are multiple vendors looking at arm based servers with hp already having one out.
I'm regurgitating this from an article I read recently, so hopefully I'll do it justice. By and far the bundles are really the doing of cable companies as a way to try and control per channel costs. The stronger stances the channels have taken on their cut more recently are because they don't want to subdivide their channels anymore. Now the following scenario didn't play out for everyone but it played out for many.
The way the bundles started was because channel wanted $x per month because they had y viewer hours. The cable company didn't want to set a precedent that Y viewable hours get $x so they said if you come out with a channel-2 we'll give you $.7x for channel-1 and $.3x for channel-2. This initially looked like a deal. Advertising money was skewed toward TV and the cable landscape was sparse. The channel didn't want the sister network to be a total dog because there was a cost and they needed to hit minimum bars to get some good advertising revenue so they shifted some content from channel-1 to channel-2 diluting the brand a little but cashed in for a ton of combo cash. The cable companies won as well. They got more channels, weaker individual channels, some of those channel-2's were exclusive for a period of time and now when a lesser channel came the could say the more popular channel only gets $.7x why do you deserve more?
Now TV advertising has lost a bunch of share to web based advertising and the cable landscape has so many channels the -2,3,and 4s are just getting lost, so they no longer want to subdivide their brands, they want what they see as their fair share based of the number of hours viewers watch the channel. When you look at how some of the fights break down the cable companies portray it as greedy network wants a 100% increase. What they leave out it's a 100% increase on a rate established 6 years ago and this is the last rate increase they'll get for the next 5. They also leave out the network is just trying to get money more inline with what other networks are getting for the same amount of views. Neither is really at fault but neither is innocent. The predication is for cable prices to stay somewhere between flat to mildly up, network prices to increase faster, but the number of channels to start to decrease.
The article should read the low end of the x86 business. IBM has already picked over the best parts of System X and moved them into PureSystems and has also started co-designing x86 server hardware with Hitachi for PureSystems. So they are going to be focusing on integrated server, networking, and storage plays instead of just plain standalone servers. Really trying to mimic the success EMC and NetApp have had partnering with Cisco and their UCS platform.
The problem is Mainframe performance is growing faster then demand for MIPS, so most customers or consolidating more and more Mainframes with every release. Once they get to one and have room to spare the upgrade slow to a halt.
I asked some PMs from Intel who they thought the next big competition was and everyone thought Samsung had all the tech and talent to turn into a major adversary over the next couple years.
IBM sold their hard drive division to Hitachi, storage is still in STG
Lexmark was essentially a spin-off of the consumer printer group
Ricoh purchased IBM's enterprise printer business
Lenovo purchased IBM's PC group
Network Hardware Group was sold to Cisco, NHD was drunk on token ring, although IBM is back into networking by purchasing BNT.
Point of Sales systems was sold to Fujitsu
IBM sold their low end PowerPC business to Applied Micro.
Lenovo's also been OEMing low end IBM servers for years and if I understand it correctly selling them for a lower price at higher margin.
One man's trash is another man's treasure. Selling off PC group was a huge win for both companies. IBM shed a low margin business, margins that were so low investing the money the put into PC group into t-bills would have yielded more profit. Lenovo had a leaner operating structure and different business options being a Chinese company that would let them run higher margin and they've made more then enough profit to pay off the acquisition several times over. IBM also got a nice revenue stream from licensing IP to Lenovo as well as the services for running Lenovo's first line support as well as coordinating their break fix.
Slashdot Mirror
My friend was working on what he thought was a video game audio/video sharing app, I'm not a gamer but I know there are several apps now that side by side video of you playing and what's going on in the game. Turns out it was really cam-girl software.
I agree with what your saying and I'm not saying it was Apple. I will say that when I've noticed it the topics were so unique there has to be something more to it then randomness. With the college example, it wasn't an ad, it ended up being a week of ads plastered everywhere.
Nope, no FB or FB owned apps installed, which still doesn't mean another app isn't part of the larger FB ad network. I also have location services turned off for most everything, I don't run very many apps. I'm not saying it was Apple. I really have no theories on how it happened.
I've had several experiences where random run ins with people I don't know have started yielding lots of ads related to obscure topics we've talked about. For example in an elevator I ran into someone wearing the shirt of a small college I was looking at but passed on, we talked about 1 minute about the school then went our separate ways. Within 15 minutes I started seeing ads for the school even though I lived 1000 miles away and ran into the guy 2500 miles away. It might very well be an option I clicked through on an app, but it's happened on multiple occasions at this point.
If you look at Manhattan one of the main reasons the medallion system was setup was to reduce congestion from an excessive number of taxis on the street. During the great depression people who had cars but no job just became independent taxi companies. At most times of the day there were more cabs parked or driving then potential passengers grinding traffic to a halt and eliminating street parking for most who were taxi drivers. There were some safety, pricing, and quality issues baked into the medallion design as well but the overall point was to provide as safe and consistent travel experience as possible while not over-congesting streets and parking. The medallion system was far from perfect mainly because modifications to the system and responses to changes in consumption move at a glacial pace but the historical precedent is there.
I'm watching some pre-election TV shows I recorded, when the campaign ads come on I can totally understand why Hillary lost, her ads fall into that really bad spot from a production quality standpoint where it's better then the sort of cheesy and comfortable videos and worse then a top notch production. Her messages are all over the place and change almost week to week, except they all tend to focus on the negative. Trumps ads all seem to be very cheesy or masterful productions, and he stays pretty much on his MAGA point. When you watch the Trump videos you actually feel kind of good, even if you do like him. When you watch the Hillary videos you feel bad even if you like her.
There is a fundamental difference between this cycle and previous ones, the outcome might be the same, vast new fields of jobs are created, but there is one fundamental difference and that is technological advance has significantly slowed. If we look at this through a Rip van winkle lens pre and post steam engine it was so different the pre steam engine person would be lost, again for automobiles. If you look through most periods for the past 300+ years taking someone 25-35 years in the future would almost always blow their minds. If we go back 25 years now and look at the differences it's more look at how slick the internet has become, wow look someone merged a Palm, with a cellphone, with a GPS, that's slick. The only thing they'll probably lose their mind over is the lack of privacy from social media. But in prior periods you would have to be retrained in how to live. So if we fast forward 8 years, and we have self driving cars and trucks, support that's really the next generation siri and alexa answering 99% of calls what's rally different. It's not doubling the amount of food I can produce, it's not really giving me any more free time, and it's probably not increasing my ability to produce. All it's really doing is taking out a lower section of the food chain. Heck most of this are things there were thought of in the original Star Trek series in 66, so you could go backover 50 years and people would just go, wow, it's just like Star Trek.
The "don't have enough workers" argument is a poor one. The unemployment rate is between 4-5%, which is near full employment, but it's not at all time lows and the labor participation rate is near a historic low. The number of people engaging in retail sales and food service jobs is near an all time high. If there was pent up demand for more skilled jobs there is ample workforce to take it. Automating trucks, cabs, cars, ordering food, shopping online is going to cut out a bunch of lower wage, lower skilled jobs.
But, is it all bad. If robotic automated farms lower the cost of grain. If automated cattle ranches with lower priced grain lower the price beef. If automated trucks lower the cost of transport. If automated trucks, farms, and ranches lower the price of a burger at McDonalds, and kiosks lower the cost of sale it's not bad. I would be more then happy to share my job with an unemployed trucker, especially if I could cut him off a piece enough to maintain his previous standard of living and mine while we both get more spare time though the lower prices of automation. I have a multitude of side projects I would like to do, books to read, trails to hike, and maybe a game or two to play. Maybe we'll get back to Keynes vision of a shorter work week.
I've seen a couple studies now that really suggest the longer out of med shoot, not necessarily age, has a huge impact on doctor effectiveness and peak effectiveness is like 4/5 years out of Med school. It has to do with the exceptionally slow adoption of new SOPs in the medical field. The old SOPs aren't wrong the new ones are just better and if it's not officially broke people are slow to change. One study looked at Cardiac incident survival rates based off doctors age at the same hospitals so same pool of patients and they saw a pretty signifacent correlation to of age to outcomes and looking though treatment notes it all came down minor tweaks in treatment.
So for comparisons sake my unemployment benefits last for a maximum of 20 weeks and are capped at a maximum of $350/week, out of which taxes are taken. After 10 weeks you must accept any job that pays 120% or $10.50/hour. So you get unemployment for 10 weeks, then you're at McDonalds. It takes 6+ weeks to get started and you have to submit proof of applying for jobs, something like 25/week, or you don't get paid. Any week without proper documentation will result in you case being closed.
There is no easy way to do this compare for two big reasons and a bunch of little ones. The two big reasons are taxes between two states are almost as varied ( less the federal component) as two countries. Some states have no income tax, some have no sales tax, some have no taxes and generate their revenue off of oil extraction rights. So when you start talking US averages which vary greatly you start short changing states. Comparing to other countries also starts bringing out all other weird compares. Our VAT is 15 your sales tax is only 7 on average, but that country might have very low property tax. As someone who recently thought about taking an opportunity in Europe I have to say US taxes were lower for my salary ( trying to take everything into account ) but it wasn't like I kept 80% in the US and only 45% in France. It wasn't that big of a difference. You also have to look at bang for the buck. Like you have a huge VAT but you aren't paying $700/month for healthcare. When you factor in a lot of what we don't get here or the better version you get there, I think it's pretty easy to see the US taxes are too high argument even though the rate might be lower. To use an analogy if I have to pay $5 for a meal of 2 slices of bread and warm water and you pay $10 for fresh meat, vegetables, and a beer, you could make the argument that the $5 meal is cheaper and you shouldn't complain about paying more, but you could also say the $5 is crap and not worth $5.
Starting with the Pentium Pro the internal processor core of Intel processors has been essentially 100% RISC. They take the CISC commands and break them into a series of RISC commands and execute. So externally you are correct the commands are 100% CISC but that goes away after decode.
The Federal government pays a lot of money for research and development in a lot of areas, architecture, bridges, roads, jet engines, custom ASICs, etc etc. I used to design custom racks, brackets, conduit routing, power/heating/cooling systems for electronics. Think Humvees with quarter racks to a full mobile data center. It eventually got to the point where we were only doing something new/innovative every 3rd or 4th deal. Every other deal was use the bracket designed for A, the rack from B, the generator from C, etc. If we had to release those cad drawings we would have had no competitive edge. If they're saying code should be reusable across agencies and parts should be made open, when stop just at software?
If you fed it the scripts from some Rom-coms you would have some weird stalker bot. I don't think movies and fiction could get much more anti reality.
In regards to the women on women ism. I heard an interview with Claudia Goldin, professor at Harvard who studies gender equality. She holds that most of the pay gap can be controlled for. The interesting part is when they came to that small part that remained the interviewer asked, if woman are the victim who is the oppressor? Goldin said the natural response is to look at it in the binary, if women are the victims then not women, or men, must be the oppressors, but this isn't the correct way to see it, because this implies that women can't or won't impact earnings from other women and that's just isn't true from the statistics. Women both from their internal and external behaviors contribute, whether is asking for more flexible hours in exchange for lower pay or maintain that hourly flexibility should result in reduced wages.
Not all people are sexist or racist, so at some point the non-sexist person would see a female applicant, see them for the quality they were, and see the deal, 72 cents on the dollar, they could even split the difference, save 14% and have a competitive advantage, rinse repeat. This would ultimately create a mostly female, reasonably yet underpaid company with a massive economic advantage. I remember reading once that someone even tried to do it. Staff a chunk of his business with 72 cent on the dollar woman. He couldn't find them. While I believe sexism is still there and there is still some gender based wage discrimination in some businesses and some industries the gap number just isn't a valid apples to apples comparison.
A couple of reasons. First as others have mentioned IBM still needs the lines. IBM's processor design is fairly integrated. It needs custom circuits which really on their fabrication technology. Their chip design process is the antithesis of fabless development. So they can't just shut off lights to the fab without crippling the Power, Mainframe, and high end storage business for years to come. The other issue is there are customers getting chips manufactured. If you shut that down there are typically very steep contract termination fees. This is really 6 year wind down with higher costs every year. The fab business is very cutthroat, you have to hit fairly high yields and have the line near full capacity or else you lose a boatload because the fixed costs are so high. They've been getting out of this business for years, especially on the low end. This is just the last step.
IBM is deploying system I on power now but they are still making custom z processors for the mainframe, although the mainframe can also have some power and x86 processors they really aren't for mainframe processing just hi RAS local access for offloading certain workloads. It's going to be interesting what arm does. There are multiple vendors looking at arm based servers with hp already having one out.
Scissors cut money.
I'm regurgitating this from an article I read recently, so hopefully I'll do it justice. By and far the bundles are really the doing of cable companies as a way to try and control per channel costs. The stronger stances the channels have taken on their cut more recently are because they don't want to subdivide their channels anymore. Now the following scenario didn't play out for everyone but it played out for many. The way the bundles started was because channel wanted $x per month because they had y viewer hours. The cable company didn't want to set a precedent that Y viewable hours get $x so they said if you come out with a channel-2 we'll give you $.7x for channel-1 and $.3x for channel-2. This initially looked like a deal. Advertising money was skewed toward TV and the cable landscape was sparse. The channel didn't want the sister network to be a total dog because there was a cost and they needed to hit minimum bars to get some good advertising revenue so they shifted some content from channel-1 to channel-2 diluting the brand a little but cashed in for a ton of combo cash. The cable companies won as well. They got more channels, weaker individual channels, some of those channel-2's were exclusive for a period of time and now when a lesser channel came the could say the more popular channel only gets $.7x why do you deserve more? Now TV advertising has lost a bunch of share to web based advertising and the cable landscape has so many channels the -2,3,and 4s are just getting lost, so they no longer want to subdivide their brands, they want what they see as their fair share based of the number of hours viewers watch the channel. When you look at how some of the fights break down the cable companies portray it as greedy network wants a 100% increase. What they leave out it's a 100% increase on a rate established 6 years ago and this is the last rate increase they'll get for the next 5. They also leave out the network is just trying to get money more inline with what other networks are getting for the same amount of views. Neither is really at fault but neither is innocent. The predication is for cable prices to stay somewhere between flat to mildly up, network prices to increase faster, but the number of channels to start to decrease.
COS Play?
The article should read the low end of the x86 business. IBM has already picked over the best parts of System X and moved them into PureSystems and has also started co-designing x86 server hardware with Hitachi for PureSystems. So they are going to be focusing on integrated server, networking, and storage plays instead of just plain standalone servers. Really trying to mimic the success EMC and NetApp have had partnering with Cisco and their UCS platform.
The problem is Mainframe performance is growing faster then demand for MIPS, so most customers or consolidating more and more Mainframes with every release. Once they get to one and have room to spare the upgrade slow to a halt.
I asked some PMs from Intel who they thought the next big competition was and everyone thought Samsung had all the tech and talent to turn into a major adversary over the next couple years.
IBM sold their hard drive division to Hitachi, storage is still in STG Lexmark was essentially a spin-off of the consumer printer group Ricoh purchased IBM's enterprise printer business Lenovo purchased IBM's PC group Network Hardware Group was sold to Cisco, NHD was drunk on token ring, although IBM is back into networking by purchasing BNT. Point of Sales systems was sold to Fujitsu IBM sold their low end PowerPC business to Applied Micro. Lenovo's also been OEMing low end IBM servers for years and if I understand it correctly selling them for a lower price at higher margin.
One man's trash is another man's treasure. Selling off PC group was a huge win for both companies. IBM shed a low margin business, margins that were so low investing the money the put into PC group into t-bills would have yielded more profit. Lenovo had a leaner operating structure and different business options being a Chinese company that would let them run higher margin and they've made more then enough profit to pay off the acquisition several times over. IBM also got a nice revenue stream from licensing IP to Lenovo as well as the services for running Lenovo's first line support as well as coordinating their break fix.