Microsoft's betting on Azure being the next IBM mainframe-style lock in device for IT. It seems to me like their goal is to get IT people thinking in Azure terms whenever they design anything, such that it becomes one of only a couple of ways to get anything deployed. Look at Windows Server 2016 and the upcoming Azure Stack -- Microsoft is basically telegraphing that the days of an on-site server not controlled by the Azure resource manager are on the way out. I'm betting Server 2016 is one of the last "monolithic" server releases, and the rest is going to be an Azure-y collection of services that you turn on and off either in the cloud or in your own datacenter.
Given that, and given Germany's privacy laws, it makes perfect sense that they would essentially build a "Public Azure Stack" to work around that detail. Whether every single company decides they're not afraid of the public cloud or not is in question, but Microsoft's looking to control that conversation and slowly bring everyone into the ongoing monthly charges model. Makes sense too -- either collect one fee for Windows Server one time, or sell it over and over again in monthly installments forever -- the choice seems obvious!
I think it's going to take a much longer time for one simple reason -- cloud computing.
Think about it, all these startups have to do now is pay Amazon or Microsoft a monthly fee to run their IT infrastructure. They don't have to build out $100 million data centers or make huge equipment investments. Companies in the 90s had to do all this, and it was great when the crash came around -- cheap gear on eBay as far as the eye could see. The cloud lets these companies live longer on less money, and the VC money can now go to paying insane salaries and going on hiring binges.
That said, it's definitely on the wane. Look at the tech press on a random day -- startups are starting to lay off people again. I just think it's going to be a much slower deflation this time, but it's deflating.
It's really amusing to watch this whole dotcom bubble from the late 90s being replayed almost exactly the same way. VC valuations lead to IPOs that lead to temporary market insanity, and it all comes crashing down when people realize it can't last forever. And just like the first dotcom boom, the products are websites, phone apps and other software.
I guess the thing SF and California in general have going for them is the climate, so it's not like San Francisco is going to become some Rust Belt city when the bottom falls out. But, the reality distortion field around SF, SV and Los Angeles is really powerful. Coming from a place where a Lincoln Town Car was an aspirational vehicle, and seeing 25 year old kid CEOs driving Maseratis and Mercedes is a big shocker.
I do feel for people who have normal jobs or are artsy types in SF. Can you imagine being, say, a cop or a civil servant in the county clerk's office making the statewide civil service wage, and having to compete for housing with someone who's making $250K working for Google or Apple, and just wants to live in hipster land? (That's another interesting phenomenon -- these techies could easily afford a house in SV closer to work, but they choose a multi-hour commute so they can live in a hipster loft.
One of the things that has always kept me away from development and more on the systems side has been the overwhelming evidence that the job category is shrinking. Some aspects of development, such as developing in the Web Framework of the Moment, are very abstract from the actual operations performed, and are mostly gluing together libraries and API calls. It's amazing how little many developers have to do to get something to work. Phone apps are another example -- huge SDKs do almost everything for the developer; they just have to signal intent.
The thing that's complex, and that requires talent, is writing all of those frameworks, libraries, APIs and abstractions. Knowing how the full stack of a system works and what is actually happening is a very useful skill. This is why embedded developers are generally not low-level guys -- those libraries and other niceties don't fit into the tiny CPU and RAM constraints on many devices.
Then again, who knows -- cloud is killing a lot of the expert-level systems jobs as well. I've been very careful to stay a generalist, but I know lots of my colleagues who spent enormous amounts of effort learning things like Cisco networking, various VM hypervisors and SAN storage inside and out, front and back, and the cloud is slowly eating away at all of that. The days of being an EMC genius, or Exchange guru, and making massive amounts of money are numbered unfortunately -- we're experiencing similar salary reductions due to commoditization that developers are facing because of H-1Bs and other factors.
"An Uncle of mine owns a paid for house in Nashua, NH and has worked for IBM near there for decades."
I remember reading somewhere that IBM's former strategy during the golden years was to put their engineering and manufacturing operations somewhere removed from the big cities, but not too far removed. The idea was that they would be insulated from some of the worst of the HQ politics while still being within a short flight or drive of one of the major installations. Nashua's a good example, Burlington, VT and Rochester, MN are other good ones. It was kind of a "nearshoring" program because they would be able to hire educated engineers who would be able to afford a house and make a good living on less than they would have to pay in NYC/Westchester County.
That said, the term "company town" comes to mind. Binghamton's not doing so well these days, not sure how many System i customers are being supported by Rochester residents, and the Burlington chip fab was sold off to GlobalFoundaries along with their other fabs. The problem is that when the good jobs from one or a couple of employers dries up, there has to be something there to take their place.
"when I think of "IBM", I think of 15-30 people around me - co-workers and immediate layers of management - who I know, enjoy, and trust. "
^^^ This! I know no one will see this but thought you deserved a reply. Everyone chooses their work environment for different reasons. If people are happy _and_ growing in a workplace, so much the better. Not everyone is a 20-something hipster working for a social media startup and hopping jobs every 6 months for 20% salary increases. I just turned 40 and have 2 little kids -- stability is way more important to me now than when I was 25 and on my own. I feel like I'm in a good spot, even while knowing I'm not being paid top dollar for our local job market.
I work for a medium-sized faceless corporation, but the group of engineers I work with is a refreshing change from the rest of the company, which I freely admit is a mess. We get lots of work done, products built and problems solved even while working within the massive bureaucratic disaster that is the larger company. I imagine it's like this in some of the more interesting arms of IBM (Research, etc.) People who succeed at our location are capable of tuning out the mess and focusing on what we can change. We have gone through more than a few new hires who just can't accept the fact that they won't change the overall culture of the company. What they don't realize is that the company leaves us alone for the most part because we do good work. When this changes, I'll be one of the first to go, but for now life is OK.
The thing that sucks is that a lot of people don't take the time to develop themselves after falling into a safe comfortable spot. I really hate it when I see someone in their mid 50s get tossed out years before retirement age, and know that they're not going to find new work easily because they spent the last 20+ years doing the exact same thing. My goal for the second half of my career is to stay useful. Lots of people advocate climbing the corporate ladder, but that's only good if you're suited for that sort of thing, Plus, lots of your effort gets wasted on navigating that particular company's politics and bureaucracy. That's only useful within that company's structure, and leads to a lot of "lifers" spending more time reading the tea leaves and politicking than doing useful work.
I'm a 20 year IT veteran, and have been through the offshoring thing a few times. Companies are starting to get that there are some tasks that can be scripted and handed off and some that require intelligence, but it's a slow process. I hope the trend stabilizes before all of the upside is sucked out of an IT career.
I've mentioned my support for a professional organization for IT and software dev, complete with apprenticeships, real barriers to entry, and consequences for screwing up. Rather than throwing support behind a politician, people should be supporting ideas like this. All the money and time spent propping up a candidate is better spent buying a way out of the current H-1B cycle. It's way more effective to send your team of lobbyists to Congress with brown paper bags full of money, in exchange for the chance to write legislation favorable to your group. Businesses do it all the time, and so do other professions' organizations.
I think the blatant abuse of the H-1B program has to stop. The program itself is a good stopgap for _real_ shortages of skilled workers. What I don't like is Cognizant, Infosys, IBM, HP, Accenture, Xerox, and all the other outsourcers swapping in H-1Bs in an arms-length transaction to the target company. This allows the Disneys and Southern California Edisons of the world to pull a "Pontius Pilate" and wash their hands of the IT department -- "It wasn't us! It was our offshore business partners!"
Don't forget that you're most likely not going to be able to retrofit your '72 Olds Cutlass or similar with self-driving technology. Future cars are likely to be electric and very energy efficient, so the greenhouse gases would be generated at the point of power generation, which is also getting more efficient year on year.
I'm a little more concerned about other things surrounding autonomous driving: - Atrophy of driving skills, so if the computer does end up causing a problem, the human won't be able to recover - Knocking yet another entire class of labor (taxi drivers, truck drivers, bus drivers, delivery drivers, and so on) out of the ranks of the employed. No one wants to address this fact, but the reality is that driving is often a "job of last resort" for some people, and often the only job paying a decent wage that the holder is qualified to do. - Privacy -- Remember Google seems to be the ones driving this the hardest. The second they get a key patent that locks out competitors, it's game over for privacy. Your driving history will be for sale to advertisers who will bombard you with ads while you drive past their establishments.
" Punishing people for speech, as you are advocating, is dangerous and tyrannical. And I don't understand why people just don't get it. Freedom is the best thing to ever happen, and the soft squishy types like yourself only seem to want to put it back in the bottle. It makes me sick."
I'm not advocating punishing free speech. I am advocating not espousing the belief that angry, harassing, hate-inciting speech is the preferred method of discourse. Two examples: - How would you like it if everyone, in every interaction, acted like the "racist uncle" or "sexist grandpa" or "angry conservative father" or "hippie pinko liberal commie sister" that lots of people have in their own families? And did so because society encourages it on the basis of..."Murica! Free speech! Woohoo!!!"? - Sexual harassment. Really, how hard is it to act professional in a workplace and refrain from speech or actions that someone might find harassing? It's always funny listening to large corporates' CYA harassment presentations, but there's a reason they exist. Because at some point, some ex-fratboy salesman or executive felt they were empowered to say whatever they wanted, no matter how offensive.
All I'm saying is that shouting each other down isn't a way to solve problems long-term, and encouraging everyone to act like that, as some of us do now, isn't a good idea.
I'm 40, so way out of college, but mainstream media targeted at us old folks loves to hype up the "soft squishy hypersensitive Milennial" meme. It's presented in the form of "look at these damn kids, they can't handle the real world, and everyone should be able to say whatever they want to each other without any consequences."
I have the opposite view -- I feel there needs to be some sort of consequences for inflammatory speech. Look at how awful political discourse is now, on both the left and the right. Everyone is hyper-focused on their opinions, partially because social media and targeted advertising continuously reinforces it. I really don't want a country of 300 million angry loudmouth Donald Trump clones walking around. People should be sensitive to others' feelings and opinions. Even tenured professors need to operate within an authority structure, as do most of us. Anyone who has worked for a large corporation with crappy office politics knows that you don't get far by shooting your mouth off at every turn.
They're going to leave some of the farmers in place, but the goal is to have 60% or more of the population living in cities by 2020. Just like the US has less than 2% of the population doing agriculture, China seems like it wants to start down that road and build a consumer class. When you're able to do top-down stuff like this with zero debate, massive projects like this are possible.
I think it's a little different from the Cultural Revolution or similar movements, because you're not sending city dwellers and intellectuals "down to the countryside" with no training in how to farm. You're sending underserved, poor peasants to cities in the hope that they're going to get non-farming jobs and stoke domestic demand, or at least be easier to provide services to in an urban setting. So, at least this experiment will probably not end up with a massive famine.
China's going through a very interesting transition period, and they're doing a lot of things that the average citizen might not agree with. It kind of makes sense in their society to crack down further on dissent at this point. For example, it's coming to light now that those "ghost cities" that the West laughed off as pyramid-building are actually part of a mass-urbanization movement. China's going to take hundreds of millions of rural farmers and move them to cities to jump-start their consumer-driven phase of economic development. Pulling something like that off requires total control over the population and the messaging around it. It will be very interesting to see if this can be done successfully -- the Cultural Revolution or Great Leap Forward didn't produce the expected results, and the Soviet crash program of industrialization had major side effects.
Now, how in the world do you enforce a ban like this? I guess the Chinese versions of internationally-owned news services are off limits now?
Whenever I go to a new city on business, I always make it a point to quickly study a zoomed-out map of the area just so I know a little bit about where I'm going. The thing about GPS is that it's always zoomed into the immediate area you're driving in, and the only info you (should be) looking at while driving is the distance to the next turn. Trying to navigate a dark road at night after a multi-hour flight while having the GPS barking orders at you is stressful enough, but not knowing anything about where you are right then is even more disorienting.
Any navigational aid (or other tool for that matter) should be sanity checked before relying on it. How many IT people do you know who can't troubleshoot without Internet access, or programmers who need StackOverflow to get anything done?
I've been doing electronic tax filing since the days of yore, even back when the tax software was generating a special machine-readable "1040PC" form with all your data on one page. If I remember correctly, the PIN was supposed to be a replacement for your physical signature on the return, since the rules say you need to certify that you are submitting a true return and acknowledge the penalties for not doing so. So, I'm not sure it was a secret PIN in that sense.
BUT -- these e-filing services shouldn't be so insecure that someone can just sniff traffic and collect the PINs. I always assumed that it worked something like this -- IRS hands out TLS certificate to "authorized e-file providers" who operate the tax payment gateways and communicate the return data from the program, to the gateway, to the IRS. Hopefully they're not just FTPing the data around:)
The site does appear to list more details than traditionally were provided, but I'm hoping they're planning on giving more details. It's great to know "something" was fixed with "some component" but previous granular Windows Update packages often had references to the KB articles prompting the release of the hotfix.
I know the goal is to get customers on a completely stateless iPhone-style device, but there are a lot of use cases that need the power and control of a traditional PC for whatever reason. Legacy code isn't going away, and sometimes you just need to run things locally because of network constraints or security. Knowing that "applying this KB fixes X, but will break your application because of a dependency" is a very useful thing when you're supporting thousands of PCs.
Most of the "cybersecurity holes" can be tracked down to some contractor slapping in an insecure installation of -whatever- to do the bare minimum needed to keep the contract. This is what needs to be fixed -- contracts need to be monitored closely and terminated in cases of poor performance. Security is a human error thing mostly: - Not removing default passwords and accounts - Leaving ports open and services running that aren't necessary - Not keeping up with product versions and patch cycles - Leaving unencrypted disks full of data on trains or in cars that get broken into
The problem is that even big companies can't manage to get this right, let alone government agencies. Big companies fall prey to the same mentality of just hiring contractors. Even the NSA did this -- if there was ever an organization that needed to do their own in-house IT, that's definitely #1 on the list. Employees will care about security when employers start demanding it.
The solution, which is nearly impossible to implement, is to make everyone involved step their game up. Hire real, full time employees who are committed to the agencies' or companies' missions at a level slightly above "I can keep my job." Make sure everyone is trained and double-check work.
I remember elementary and high school math from the 80s and early 90s. It was an endless cycle of memorization of procedures and formulas, with very little emphasis on the real utility of it all. In particular, I remember plane geometry proofs that barely made sense to me -- I can't imagine what someone who was bad at math or disinterested thought of those. That, and the algebra manipulation phase (factoring, quadratic equations, etc.) I will always remember that x = (-b +/- sqrt(b^2 - 4ac)) / 2a -- for some reason.:-)
Here's a question for math lovers -- what needs to be taught differently in early math so that students will enjoy it? I know the only time I ever got interested in math was later on, using it in science courses to solve actual problems. Everything before that was just operations. The problem was that being behind in math kept me from doing well in engineering coursework. Contrast this with my eventual degree in chemistry -- I had a great high school chemistry teacher and really caught onto it immediately, probably because it wasn't as math heavy until physical chemistry and analysis courses. Most people barely understand chemistry and consider it something they pass once and never see again. Is it really just as simple as good initial teaching? What makes math interesting?
Yes, yes, flame away, but I feel that the world isn't PC enough given the changes in the way discourse is handled.
Before the Internet and easy-to-use social media platforms, people who had social issues could only offend a limited number of people within their local sphere of influence. By this I'm talking about the people who don't have a filter and just let their mouths run without thinking about how they sound or who they're talking to. I know many, and giving people like this access to Facebook, Twitter and the other social platforms just makes them worse. They also tend to pull in more people around them who are attracted to their abrasive style.
PC is required because the loud-mouth crowd is using the concept of free speech as a license to say whatever comes to their minds with no repercussions. If people would simply follow the golden rule of "don't be a total asshat to one another" we wouldn't need it.
The problem with a situation like this is that the loudest mouth wins, whether or not what they have to say is worth listening to. Look at people like Trump, angry conservative talk show hosts, or radical leftists for that matter, and you'll see how extreme positions affect the public narrative. Putting reasonable limits on what people can say to one another is a good idea in my opinion.
Layoffs should always be a last resort in positions that require technical talent. This goes double if your staff are more experienced and able to see the writing on the wall. I've chosen companies carefully over the years and have had long tenures at places I've worked. But, when a talented person who can get a job somewhere else sees the layoff balloons going up, the immediate thought is whether or not they'll be next. This causes everyone good to head for the exits, and you're left with the low-talent people who are just hanging on hoping they don't get it in the next round. This has happened to me twice, and I've carefully considered my options both times, opting to leave before things got worse.
This is partially driven by the (irrational) preference to hire only employed people. I know a lot of talented people who've just been blindsided by a sudden layoff, capricious firing or even the business going bankrupt. The road back for these people is very hard and they often have to take lower-salary work or work for companies with less-than-ideal working conditions.
There have been a lot of stories like this over the brief history of technology. IBM is a really good example. Their senior management is doing everything they can to sell off the company bit by bit while collecting money, and they still can't kill it. Microsoft is another excellent example, riding Windows and Office through to their current states. They're currently poised to pull the ultimate vendor lock-in trick with Azure and subscription software because they have loads of money to spend. Some companies, especially those with huge cash balances, can manage through transitions. Others will just keep beating money out of their cash cows for as long as possible (again, IBM is the perfect example.) Others, like Sun, end up getting bought at fire sale prices. All of the companies mentioned were absolutely dominant at one time or another. IBM is a total joke these days, but in the 70s/80s they represented the state of the art in all things computing.
Apple's problem is that they are now too consumer-focused and don't have a pipeline of expensive gadgets to sell them. Whether they'll use that huge pile of cash they have to buy into the next trend remains to be seen.
There are so many vulnerable SCADA systems, device-specific Ethernet adapters and other stuff out there, and it just chugs along for years and years. Especially with public sector stuff, multiple layers of contractors put gear in, barely document it and hand it over to the operating authority. The problem is that since no one permanent knows the ins and outs of the system, it can stay vulnerable for ages. Even if a vendor does release patches, the "don't touch it or 500K customers lose power" mentality around critical infrastructure means they barely ever get applied.
Anything IoT is going to have to be secure by default, as in, hard to get working instead of open and easy. I doubt the "just contract it out" mentality is ever going to go away in the public sector -- I've inherited systems where the only documentation is a statement of work from 5 years back that the contractor cut and pasted from the vendor's manuals.
It's amazing how much cabling gets forgotten about when you have a chaotic lab environment and new stuff coming in all the time (we do hardware evaluations and other systems integration work.) There's never any money left over for structured cabling once it's been spent on all the fancy new hardware. Even if we invested in structured cabling it would turn into an unstructured mess quickly. I have racks that look like those Magic Eye pictures; the only thing that will solve it is unplugging everything. I'm sure world class scientists can't be bothered to label anything if we can't!
I saw this back in the late 90s. People I knew with very shaky skills were getting paid 6 figures to design website back-ends, simply because the demand was so astronomical. Come 2001, a lot of those people were unemployed or were being paid a lot less. The point is that the bubble is coming to an end: - CS enrollments are at an all time high (just in time for grads to get out into the nonexistent job market...) - Companies are paying insane salaries due to the bubble and hype around apps, social media, etc. - More and more semi-skilled people are jumping on the bandwagon, getting into the "exciting world of development"
As a counterpoint, look at the story about Disney's H-1B replacement workers still on the front page. That seems to be what's coming for the low end of the market. The high end is cyclical -- BS artist consultants on the latest fad come and go, really good consultants and employees can still command a good salary if they know how to market themselves correctly.
It's interesting to see a new angle on this, and to see a group actually fighting back against such a large employer. But...lawsuits won't fix this long term. What is going to fix this is a professional organization with a little more teeth than something like the IEEE or ACM. IT Professionals (developers, systems guys, DevOps people, whatever) need to start standing up against stuff like this before any hope of combating it goes away.
I walk the line between worker and manager in a lead position, so I see both sides of an employers' argument. Here's the uncomfortable truth -- there really is a shortage of qualified people, always has been. You need to find and hang on to qualified people for dear life, because you're not going to get a department full of superstars. The problem is that a lot of unqualified people can BS their way to a $150K+ job, and employers often don't know the difference between good and bad. Because of this, they're always looking to cut costs. So when Tata or Infosys comes in, and tells the CIO to write them a monster check to make their lazy good-for-nothing IT department go away, the argument holds water. Anyone working in an offshored IT environment knows that it never works out, but we do a very poor job of communicating our value to the business in some cases.
Other professionals are much smarter than we are about this. They saw companies moving to limit their power and formed professional organizations. The AMA pays for legislation, makes political campaign donations, and ensures its members still continue to command high salaries. If they ever let up, United Healthcare or similar would buy a law saying that nurses or medical assistants could perform advanced procedures for 1/10 the cost. Same thing with engineers, accountants, etc. There is an accepted barrier to entry (medical school, accreditation, licensure, etc.) to weed out the first-level BS artists. Imagine if an IT professional with X years' experience came with a full well-rounded education in computing fundamentals and their speciality, as opposed to graduating from a certification bootcamp. Or if a developer could be guaranteed to know something other than the JQuery and Python scripting he was taught in Coder Academy. As an employer, I'd pay for that instead of having to cycle endlessly through crappy onshore and offshore employees.
The point is that both sides have to give a little. Employers need to stop offshoring to the lowest bidder long enough to allow a talent pool to grow domestically, and IT professionals need to embrace the idea of a profession with salary progression commensurate with experience. If I were king and were able to form the IT Professionals Association tomorrow, here's what would happen: - A huge collection would have to be taken up from members to purchase legislation banning the most obvious abuses of the current visa system. (Not an outright ban, because the original idea is good.) - Some fundamental standards and practices would need to be established. This is the really hard part, because everyone is used to things going a million miles an hour and vendors promoting lock-in at every turn. But we're big boys and girls now, and computers are a part of our daily lives; their use should be more like a branch of engineering than a mad scientists' lab or skunkworks. - Experience levels would need to be set, and training requirements to reach the next level would need to be established. Yes, this includes the idea of licensure, and at the lower levels, the dirty word "apprenticeship." This would allow employers to pay less for lower-skilled domestic labor. Does that sound like a skilled trade? It should -- the fundamentals of computing are becoming skilled labor now, and the creative engineering work should be done higher up the stack by people who have done the grunt work before. - Members of the profession would need to start taking responsibility for their work, PE or medical malpractice style. It infuriates me when I've walked into projects where someone messed things up
I've worked in a few corporate environments where they were extremely paranoid about e-discovery (back when this was a new thing.) Almost always, the answer was to set the retention policy to 30 days, as in, no email backups older than 30 days, no (sanctioned) way to archive email, and everything older than 30 days was purged from mailboxes. This allowed the company to say with a straight face, "I'd love to give you the messages relevant to such-and-such business deal gone bad 5 years ago, but I simply cannot."
It sounds a lot like what Apple's doing -- they purposely built the encryption system with no way to bypass it so they can push it right back on the police and courts -- "Sorry, can't help you!" That gets them tons of great customer PR, as opposed to Google/Android, so it makes sense.
Slashdot Mirror
Microsoft's betting on Azure being the next IBM mainframe-style lock in device for IT. It seems to me like their goal is to get IT people thinking in Azure terms whenever they design anything, such that it becomes one of only a couple of ways to get anything deployed. Look at Windows Server 2016 and the upcoming Azure Stack -- Microsoft is basically telegraphing that the days of an on-site server not controlled by the Azure resource manager are on the way out. I'm betting Server 2016 is one of the last "monolithic" server releases, and the rest is going to be an Azure-y collection of services that you turn on and off either in the cloud or in your own datacenter.
Given that, and given Germany's privacy laws, it makes perfect sense that they would essentially build a "Public Azure Stack" to work around that detail. Whether every single company decides they're not afraid of the public cloud or not is in question, but Microsoft's looking to control that conversation and slowly bring everyone into the ongoing monthly charges model. Makes sense too -- either collect one fee for Windows Server one time, or sell it over and over again in monthly installments forever -- the choice seems obvious!
I think it's going to take a much longer time for one simple reason -- cloud computing.
Think about it, all these startups have to do now is pay Amazon or Microsoft a monthly fee to run their IT infrastructure. They don't have to build out $100 million data centers or make huge equipment investments. Companies in the 90s had to do all this, and it was great when the crash came around -- cheap gear on eBay as far as the eye could see. The cloud lets these companies live longer on less money, and the VC money can now go to paying insane salaries and going on hiring binges.
That said, it's definitely on the wane. Look at the tech press on a random day -- startups are starting to lay off people again. I just think it's going to be a much slower deflation this time, but it's deflating.
It's really amusing to watch this whole dotcom bubble from the late 90s being replayed almost exactly the same way. VC valuations lead to IPOs that lead to temporary market insanity, and it all comes crashing down when people realize it can't last forever. And just like the first dotcom boom, the products are websites, phone apps and other software.
I guess the thing SF and California in general have going for them is the climate, so it's not like San Francisco is going to become some Rust Belt city when the bottom falls out. But, the reality distortion field around SF, SV and Los Angeles is really powerful. Coming from a place where a Lincoln Town Car was an aspirational vehicle, and seeing 25 year old kid CEOs driving Maseratis and Mercedes is a big shocker.
I do feel for people who have normal jobs or are artsy types in SF. Can you imagine being, say, a cop or a civil servant in the county clerk's office making the statewide civil service wage, and having to compete for housing with someone who's making $250K working for Google or Apple, and just wants to live in hipster land? (That's another interesting phenomenon -- these techies could easily afford a house in SV closer to work, but they choose a multi-hour commute so they can live in a hipster loft.
One of the things that has always kept me away from development and more on the systems side has been the overwhelming evidence that the job category is shrinking. Some aspects of development, such as developing in the Web Framework of the Moment, are very abstract from the actual operations performed, and are mostly gluing together libraries and API calls. It's amazing how little many developers have to do to get something to work. Phone apps are another example -- huge SDKs do almost everything for the developer; they just have to signal intent.
The thing that's complex, and that requires talent, is writing all of those frameworks, libraries, APIs and abstractions. Knowing how the full stack of a system works and what is actually happening is a very useful skill. This is why embedded developers are generally not low-level guys -- those libraries and other niceties don't fit into the tiny CPU and RAM constraints on many devices.
Then again, who knows -- cloud is killing a lot of the expert-level systems jobs as well. I've been very careful to stay a generalist, but I know lots of my colleagues who spent enormous amounts of effort learning things like Cisco networking, various VM hypervisors and SAN storage inside and out, front and back, and the cloud is slowly eating away at all of that. The days of being an EMC genius, or Exchange guru, and making massive amounts of money are numbered unfortunately -- we're experiencing similar salary reductions due to commoditization that developers are facing because of H-1Bs and other factors.
"An Uncle of mine owns a paid for house in Nashua, NH and has worked for IBM near there for decades."
I remember reading somewhere that IBM's former strategy during the golden years was to put their engineering and manufacturing operations somewhere removed from the big cities, but not too far removed. The idea was that they would be insulated from some of the worst of the HQ politics while still being within a short flight or drive of one of the major installations. Nashua's a good example, Burlington, VT and Rochester, MN are other good ones. It was kind of a "nearshoring" program because they would be able to hire educated engineers who would be able to afford a house and make a good living on less than they would have to pay in NYC/Westchester County.
That said, the term "company town" comes to mind. Binghamton's not doing so well these days, not sure how many System i customers are being supported by Rochester residents, and the Burlington chip fab was sold off to GlobalFoundaries along with their other fabs. The problem is that when the good jobs from one or a couple of employers dries up, there has to be something there to take their place.
"when I think of "IBM", I think of 15-30 people around me - co-workers and immediate layers of management - who I know, enjoy, and trust. "
^^^ This! I know no one will see this but thought you deserved a reply. Everyone chooses their work environment for different reasons. If people are happy _and_ growing in a workplace, so much the better. Not everyone is a 20-something hipster working for a social media startup and hopping jobs every 6 months for 20% salary increases. I just turned 40 and have 2 little kids -- stability is way more important to me now than when I was 25 and on my own. I feel like I'm in a good spot, even while knowing I'm not being paid top dollar for our local job market.
I work for a medium-sized faceless corporation, but the group of engineers I work with is a refreshing change from the rest of the company, which I freely admit is a mess. We get lots of work done, products built and problems solved even while working within the massive bureaucratic disaster that is the larger company. I imagine it's like this in some of the more interesting arms of IBM (Research, etc.) People who succeed at our location are capable of tuning out the mess and focusing on what we can change. We have gone through more than a few new hires who just can't accept the fact that they won't change the overall culture of the company. What they don't realize is that the company leaves us alone for the most part because we do good work. When this changes, I'll be one of the first to go, but for now life is OK.
The thing that sucks is that a lot of people don't take the time to develop themselves after falling into a safe comfortable spot. I really hate it when I see someone in their mid 50s get tossed out years before retirement age, and know that they're not going to find new work easily because they spent the last 20+ years doing the exact same thing. My goal for the second half of my career is to stay useful. Lots of people advocate climbing the corporate ladder, but that's only good if you're suited for that sort of thing, Plus, lots of your effort gets wasted on navigating that particular company's politics and bureaucracy. That's only useful within that company's structure, and leads to a lot of "lifers" spending more time reading the tea leaves and politicking than doing useful work.
I'm a 20 year IT veteran, and have been through the offshoring thing a few times. Companies are starting to get that there are some tasks that can be scripted and handed off and some that require intelligence, but it's a slow process. I hope the trend stabilizes before all of the upside is sucked out of an IT career.
I've mentioned my support for a professional organization for IT and software dev, complete with apprenticeships, real barriers to entry, and consequences for screwing up. Rather than throwing support behind a politician, people should be supporting ideas like this. All the money and time spent propping up a candidate is better spent buying a way out of the current H-1B cycle. It's way more effective to send your team of lobbyists to Congress with brown paper bags full of money, in exchange for the chance to write legislation favorable to your group. Businesses do it all the time, and so do other professions' organizations.
I think the blatant abuse of the H-1B program has to stop. The program itself is a good stopgap for _real_ shortages of skilled workers. What I don't like is Cognizant, Infosys, IBM, HP, Accenture, Xerox, and all the other outsourcers swapping in H-1Bs in an arms-length transaction to the target company. This allows the Disneys and Southern California Edisons of the world to pull a "Pontius Pilate" and wash their hands of the IT department -- "It wasn't us! It was our offshore business partners!"
Don't forget that you're most likely not going to be able to retrofit your '72 Olds Cutlass or similar with self-driving technology. Future cars are likely to be electric and very energy efficient, so the greenhouse gases would be generated at the point of power generation, which is also getting more efficient year on year.
I'm a little more concerned about other things surrounding autonomous driving:
- Atrophy of driving skills, so if the computer does end up causing a problem, the human won't be able to recover
- Knocking yet another entire class of labor (taxi drivers, truck drivers, bus drivers, delivery drivers, and so on) out of the ranks of the employed. No one wants to address this fact, but the reality is that driving is often a "job of last resort" for some people, and often the only job paying a decent wage that the holder is qualified to do.
- Privacy -- Remember Google seems to be the ones driving this the hardest. The second they get a key patent that locks out competitors, it's game over for privacy. Your driving history will be for sale to advertisers who will bombard you with ads while you drive past their establishments.
" Punishing people for speech, as you are advocating, is dangerous and tyrannical. And I don't understand why people just don't get it. Freedom is the best thing to ever happen, and the soft squishy types like yourself only seem to want to put it back in the bottle. It makes me sick."
I'm not advocating punishing free speech. I am advocating not espousing the belief that angry, harassing, hate-inciting speech is the preferred method of discourse. Two examples:
- How would you like it if everyone, in every interaction, acted like the "racist uncle" or "sexist grandpa" or "angry conservative father" or "hippie pinko liberal commie sister" that lots of people have in their own families? And did so because society encourages it on the basis of..."Murica! Free speech! Woohoo!!!"?
- Sexual harassment. Really, how hard is it to act professional in a workplace and refrain from speech or actions that someone might find harassing? It's always funny listening to large corporates' CYA harassment presentations, but there's a reason they exist. Because at some point, some ex-fratboy salesman or executive felt they were empowered to say whatever they wanted, no matter how offensive.
All I'm saying is that shouting each other down isn't a way to solve problems long-term, and encouraging everyone to act like that, as some of us do now, isn't a good idea.
I'm 40, so way out of college, but mainstream media targeted at us old folks loves to hype up the "soft squishy hypersensitive Milennial" meme. It's presented in the form of "look at these damn kids, they can't handle the real world, and everyone should be able to say whatever they want to each other without any consequences."
I have the opposite view -- I feel there needs to be some sort of consequences for inflammatory speech. Look at how awful political discourse is now, on both the left and the right. Everyone is hyper-focused on their opinions, partially because social media and targeted advertising continuously reinforces it. I really don't want a country of 300 million angry loudmouth Donald Trump clones walking around. People should be sensitive to others' feelings and opinions. Even tenured professors need to operate within an authority structure, as do most of us. Anyone who has worked for a large corporation with crappy office politics knows that you don't get far by shooting your mouth off at every turn.
A couple of sources describe this:
http://www.wsj.com/articles/SB...
http://www.thenatureofcities.c...
http://www.bloomberg.com/bw/ar...
They're going to leave some of the farmers in place, but the goal is to have 60% or more of the population living in cities by 2020. Just like the US has less than 2% of the population doing agriculture, China seems like it wants to start down that road and build a consumer class. When you're able to do top-down stuff like this with zero debate, massive projects like this are possible.
I think it's a little different from the Cultural Revolution or similar movements, because you're not sending city dwellers and intellectuals "down to the countryside" with no training in how to farm. You're sending underserved, poor peasants to cities in the hope that they're going to get non-farming jobs and stoke domestic demand, or at least be easier to provide services to in an urban setting. So, at least this experiment will probably not end up with a massive famine.
China's going through a very interesting transition period, and they're doing a lot of things that the average citizen might not agree with. It kind of makes sense in their society to crack down further on dissent at this point. For example, it's coming to light now that those "ghost cities" that the West laughed off as pyramid-building are actually part of a mass-urbanization movement. China's going to take hundreds of millions of rural farmers and move them to cities to jump-start their consumer-driven phase of economic development. Pulling something like that off requires total control over the population and the messaging around it. It will be very interesting to see if this can be done successfully -- the Cultural Revolution or Great Leap Forward didn't produce the expected results, and the Soviet crash program of industrialization had major side effects.
Now, how in the world do you enforce a ban like this? I guess the Chinese versions of internationally-owned news services are off limits now?
Whenever I go to a new city on business, I always make it a point to quickly study a zoomed-out map of the area just so I know a little bit about where I'm going. The thing about GPS is that it's always zoomed into the immediate area you're driving in, and the only info you (should be) looking at while driving is the distance to the next turn. Trying to navigate a dark road at night after a multi-hour flight while having the GPS barking orders at you is stressful enough, but not knowing anything about where you are right then is even more disorienting.
Any navigational aid (or other tool for that matter) should be sanity checked before relying on it. How many IT people do you know who can't troubleshoot without Internet access, or programmers who need StackOverflow to get anything done?
I've been doing electronic tax filing since the days of yore, even back when the tax software was generating a special machine-readable "1040PC" form with all your data on one page. If I remember correctly, the PIN was supposed to be a replacement for your physical signature on the return, since the rules say you need to certify that you are submitting a true return and acknowledge the penalties for not doing so. So, I'm not sure it was a secret PIN in that sense.
BUT -- these e-filing services shouldn't be so insecure that someone can just sniff traffic and collect the PINs. I always assumed that it worked something like this -- IRS hands out TLS certificate to "authorized e-file providers" who operate the tax payment gateways and communicate the return data from the program, to the gateway, to the IRS. Hopefully they're not just FTPing the data around :)
The site does appear to list more details than traditionally were provided, but I'm hoping they're planning on giving more details. It's great to know "something" was fixed with "some component" but previous granular Windows Update packages often had references to the KB articles prompting the release of the hotfix.
I know the goal is to get customers on a completely stateless iPhone-style device, but there are a lot of use cases that need the power and control of a traditional PC for whatever reason. Legacy code isn't going away, and sometimes you just need to run things locally because of network constraints or security. Knowing that "applying this KB fixes X, but will break your application because of a dependency" is a very useful thing when you're supporting thousands of PCs.
Most of the "cybersecurity holes" can be tracked down to some contractor slapping in an insecure installation of -whatever- to do the bare minimum needed to keep the contract. This is what needs to be fixed -- contracts need to be monitored closely and terminated in cases of poor performance. Security is a human error thing mostly:
- Not removing default passwords and accounts
- Leaving ports open and services running that aren't necessary
- Not keeping up with product versions and patch cycles
- Leaving unencrypted disks full of data on trains or in cars that get broken into
The problem is that even big companies can't manage to get this right, let alone government agencies. Big companies fall prey to the same mentality of just hiring contractors. Even the NSA did this -- if there was ever an organization that needed to do their own in-house IT, that's definitely #1 on the list. Employees will care about security when employers start demanding it.
The solution, which is nearly impossible to implement, is to make everyone involved step their game up. Hire real, full time employees who are committed to the agencies' or companies' missions at a level slightly above "I can keep my job." Make sure everyone is trained and double-check work.
I remember elementary and high school math from the 80s and early 90s. It was an endless cycle of memorization of procedures and formulas, with very little emphasis on the real utility of it all. In particular, I remember plane geometry proofs that barely made sense to me -- I can't imagine what someone who was bad at math or disinterested thought of those. That, and the algebra manipulation phase (factoring, quadratic equations, etc.) I will always remember that x = (-b +/- sqrt(b^2 - 4ac)) / 2a -- for some reason. :-)
Here's a question for math lovers -- what needs to be taught differently in early math so that students will enjoy it? I know the only time I ever got interested in math was later on, using it in science courses to solve actual problems. Everything before that was just operations. The problem was that being behind in math kept me from doing well in engineering coursework. Contrast this with my eventual degree in chemistry -- I had a great high school chemistry teacher and really caught onto it immediately, probably because it wasn't as math heavy until physical chemistry and analysis courses. Most people barely understand chemistry and consider it something they pass once and never see again. Is it really just as simple as good initial teaching? What makes math interesting?
Yes, yes, flame away, but I feel that the world isn't PC enough given the changes in the way discourse is handled.
Before the Internet and easy-to-use social media platforms, people who had social issues could only offend a limited number of people within their local sphere of influence. By this I'm talking about the people who don't have a filter and just let their mouths run without thinking about how they sound or who they're talking to. I know many, and giving people like this access to Facebook, Twitter and the other social platforms just makes them worse. They also tend to pull in more people around them who are attracted to their abrasive style.
PC is required because the loud-mouth crowd is using the concept of free speech as a license to say whatever comes to their minds with no repercussions. If people would simply follow the golden rule of "don't be a total asshat to one another" we wouldn't need it.
The problem with a situation like this is that the loudest mouth wins, whether or not what they have to say is worth listening to. Look at people like Trump, angry conservative talk show hosts, or radical leftists for that matter, and you'll see how extreme positions affect the public narrative. Putting reasonable limits on what people can say to one another is a good idea in my opinion.
Layoffs should always be a last resort in positions that require technical talent. This goes double if your staff are more experienced and able to see the writing on the wall. I've chosen companies carefully over the years and have had long tenures at places I've worked. But, when a talented person who can get a job somewhere else sees the layoff balloons going up, the immediate thought is whether or not they'll be next. This causes everyone good to head for the exits, and you're left with the low-talent people who are just hanging on hoping they don't get it in the next round. This has happened to me twice, and I've carefully considered my options both times, opting to leave before things got worse.
This is partially driven by the (irrational) preference to hire only employed people. I know a lot of talented people who've just been blindsided by a sudden layoff, capricious firing or even the business going bankrupt. The road back for these people is very hard and they often have to take lower-salary work or work for companies with less-than-ideal working conditions.
There have been a lot of stories like this over the brief history of technology. IBM is a really good example. Their senior management is doing everything they can to sell off the company bit by bit while collecting money, and they still can't kill it. Microsoft is another excellent example, riding Windows and Office through to their current states. They're currently poised to pull the ultimate vendor lock-in trick with Azure and subscription software because they have loads of money to spend. Some companies, especially those with huge cash balances, can manage through transitions. Others will just keep beating money out of their cash cows for as long as possible (again, IBM is the perfect example.) Others, like Sun, end up getting bought at fire sale prices. All of the companies mentioned were absolutely dominant at one time or another. IBM is a total joke these days, but in the 70s/80s they represented the state of the art in all things computing.
Apple's problem is that they are now too consumer-focused and don't have a pipeline of expensive gadgets to sell them. Whether they'll use that huge pile of cash they have to buy into the next trend remains to be seen.
There are so many vulnerable SCADA systems, device-specific Ethernet adapters and other stuff out there, and it just chugs along for years and years. Especially with public sector stuff, multiple layers of contractors put gear in, barely document it and hand it over to the operating authority. The problem is that since no one permanent knows the ins and outs of the system, it can stay vulnerable for ages. Even if a vendor does release patches, the "don't touch it or 500K customers lose power" mentality around critical infrastructure means they barely ever get applied.
Anything IoT is going to have to be secure by default, as in, hard to get working instead of open and easy. I doubt the "just contract it out" mentality is ever going to go away in the public sector -- I've inherited systems where the only documentation is a statement of work from 5 years back that the contractor cut and pasted from the vendor's manuals.
It's amazing how much cabling gets forgotten about when you have a chaotic lab environment and new stuff coming in all the time (we do hardware evaluations and other systems integration work.) There's never any money left over for structured cabling once it's been spent on all the fancy new hardware. Even if we invested in structured cabling it would turn into an unstructured mess quickly. I have racks that look like those Magic Eye pictures; the only thing that will solve it is unplugging everything. I'm sure world class scientists can't be bothered to label anything if we can't!
I saw this back in the late 90s. People I knew with very shaky skills were getting paid 6 figures to design website back-ends, simply because the demand was so astronomical. Come 2001, a lot of those people were unemployed or were being paid a lot less. The point is that the bubble is coming to an end:
- CS enrollments are at an all time high (just in time for grads to get out into the nonexistent job market...)
- Companies are paying insane salaries due to the bubble and hype around apps, social media, etc.
- More and more semi-skilled people are jumping on the bandwagon, getting into the "exciting world of development"
As a counterpoint, look at the story about Disney's H-1B replacement workers still on the front page. That seems to be what's coming for the low end of the market. The high end is cyclical -- BS artist consultants on the latest fad come and go, really good consultants and employees can still command a good salary if they know how to market themselves correctly.
It's interesting to see a new angle on this, and to see a group actually fighting back against such a large employer. But...lawsuits won't fix this long term. What is going to fix this is a professional organization with a little more teeth than something like the IEEE or ACM. IT Professionals (developers, systems guys, DevOps people, whatever) need to start standing up against stuff like this before any hope of combating it goes away.
I walk the line between worker and manager in a lead position, so I see both sides of an employers' argument. Here's the uncomfortable truth -- there really is a shortage of qualified people, always has been. You need to find and hang on to qualified people for dear life, because you're not going to get a department full of superstars. The problem is that a lot of unqualified people can BS their way to a $150K+ job, and employers often don't know the difference between good and bad. Because of this, they're always looking to cut costs. So when Tata or Infosys comes in, and tells the CIO to write them a monster check to make their lazy good-for-nothing IT department go away, the argument holds water. Anyone working in an offshored IT environment knows that it never works out, but we do a very poor job of communicating our value to the business in some cases.
Other professionals are much smarter than we are about this. They saw companies moving to limit their power and formed professional organizations. The AMA pays for legislation, makes political campaign donations, and ensures its members still continue to command high salaries. If they ever let up, United Healthcare or similar would buy a law saying that nurses or medical assistants could perform advanced procedures for 1/10 the cost. Same thing with engineers, accountants, etc. There is an accepted barrier to entry (medical school, accreditation, licensure, etc.) to weed out the first-level BS artists. Imagine if an IT professional with X years' experience came with a full well-rounded education in computing fundamentals and their speciality, as opposed to graduating from a certification bootcamp. Or if a developer could be guaranteed to know something other than the JQuery and Python scripting he was taught in Coder Academy. As an employer, I'd pay for that instead of having to cycle endlessly through crappy onshore and offshore employees.
The point is that both sides have to give a little. Employers need to stop offshoring to the lowest bidder long enough to allow a talent pool to grow domestically, and IT professionals need to embrace the idea of a profession with salary progression commensurate with experience. If I were king and were able to form the IT Professionals Association tomorrow, here's what would happen:
- A huge collection would have to be taken up from members to purchase legislation banning the most obvious abuses of the current visa system. (Not an outright ban, because the original idea is good.)
- Some fundamental standards and practices would need to be established. This is the really hard part, because everyone is used to things going a million miles an hour and vendors promoting lock-in at every turn. But we're big boys and girls now, and computers are a part of our daily lives; their use should be more like a branch of engineering than a mad scientists' lab or skunkworks.
- Experience levels would need to be set, and training requirements to reach the next level would need to be established. Yes, this includes the idea of licensure, and at the lower levels, the dirty word "apprenticeship." This would allow employers to pay less for lower-skilled domestic labor. Does that sound like a skilled trade? It should -- the fundamentals of computing are becoming skilled labor now, and the creative engineering work should be done higher up the stack by people who have done the grunt work before.
- Members of the profession would need to start taking responsibility for their work, PE or medical malpractice style. It infuriates me when I've walked into projects where someone messed things up
I've worked in a few corporate environments where they were extremely paranoid about e-discovery (back when this was a new thing.) Almost always, the answer was to set the retention policy to 30 days, as in, no email backups older than 30 days, no (sanctioned) way to archive email, and everything older than 30 days was purged from mailboxes. This allowed the company to say with a straight face, "I'd love to give you the messages relevant to such-and-such business deal gone bad 5 years ago, but I simply cannot."
It sounds a lot like what Apple's doing -- they purposely built the encryption system with no way to bypass it so they can push it right back on the police and courts -- "Sorry, can't help you!" That gets them tons of great customer PR, as opposed to Google/Android, so it makes sense.