Follow Slashdot stories on Twitter

 



Forgot your password?
typodupeerror

Comment Low cost chip, high cost support (Score 4, Interesting) 91 91

I'm sure the hardware itself will be cheap. Oracle's hardware is like IBM's mainframes -- they'll practically give away the hardware if you'll burn up MIPS on a regular basis. Even if "give away" is thousands per socket, it's a drop in the bucket compared to the fees for support and any OS licensing. Our relatively large company is a decent sized Oracle DB customer (lots and lots of hosted J2EE enterprisey applications) and the maintenance fees alone, just to be able to run the software, are eye watering.

The problem is that licensing like that keeps all but the most well heeled customers off SPARC, and hence the popularity will never get much higher than it is. Ever since Linux on x86 became a viable alternative, companies without a real need to run SPARC and by extension Solaris on SPARC are migrating away. Even Debian dropped support for its SPARC port.

Whether it's the high cost keeping people off SPARC, or the niche nature of Itanium keeping people off Itanium, a system architecture needs a critical mass of customers with a continued need to run on it to be successful.

Comment Nothing exposed to the Internet is private. (Score 3, Interesting) 18 18

I've been doing desktop computing stuff for ages, and one of the things you need to take into account is this -- Nothing will ever stop one idiot end user from double-clicking on an attachment, following a link to a cat video, or giving their password to someone over the phone. This could be anyone from the CEO (actually, more likely to be them...) to the lowliest call center person working on what you think is a locked down desktop/Citrix session. Microsoft has gotten better over the years by making the OS and applications usable by a non-administrator, but that's only one piece of the problem. Most large organizations have a hard time patching regular vulnerabilities in their OSes, let alone emergency patching a zero day exploit.

I've always wondered when companies are going to just say "screw it" and give workers back the 2015 version of a green screen terminal to do their work on. VDI is vulnerable, Citrix is -very- vulnerable, and standalone desktops are extremely hard to secure. These "security researchers" have way more resources than an overburdened, understaffed, underfunded and often outsourced IT department. Most companies can't afford to re-architect their network in a "trust-nothing" fashion, or don't want to pay for it because IT is seen as a cost center. What makes this worse is that companies get away with it all the time -- as long as they have their PCI and/or HIPAA audit box checked, they can shrug their shoulders and say "we're powerless to stop them, see, we did everything you asked!" Then, their insurance just pays off the credit card companies and it's business as usual again until the next big hack.

When you can "fix" a security problem by giving away a useless credit monitoring service, there's no incentive to fix the problem.

Comment Right idea, but a big shift (Score 4, Insightful) 315 315

I've been using Windows 10 for quite a while. The thing that's going to really change on the "enterprisey" side of things is the need to buy the Enterprise version so you can get the Long Term Stable servicing branch, and thus you'll be forced into volume licensing rather than OEM licensing. If you don't, you run the risk of Microsoft introducing a new change in the Current Branch for Business that breaks your applications, with a ticking clock counting down to the time you're forced to accept it. Unlike phones, PCs in businesses typically run applications that, for whatever reason, can't easily be upgraded. I've worked in end user computing for years, and it happens everywhere, in large and small businesses. Entire departments live and die by Excel macros and Access databases. Web applications that are too expensive to upgrade have to keep working. And on and on...

I think the biggest thing that Microsoft needs to get right is stability. Rolling out new features all the time sounds like a really great idea, more Agile, etc. etc. The problem is that to do this with an operating system, those feature changes need to be solid and not break existing functionality. If they got rid of all their QA staff, I hope they're not relying on Windows Insiders to test key functionality. Insiders are generally not running the legacy junk applications that businesses need to keep supported and alive. Insiders are running their general Office workstations, maybe some web browsing, but usually not legacy applications.

One of the things from the past that was nice about a definitive "RTM" line in the sand was that the code was declared feature complete, and most showstopper bugs were squashed before the OS was allowed to be released. Back in the day, it was because you were pressing a million DVDs and your customers couldn't easily download patches, so it had to work. Now, the "ship it, we'll just rush out a patch later" mentality is dominant everywhere. The other nice thing was that when Version X came out, features didn't change until X.1 was ready. With this continuous upgrade cycle, I can see some problems. Maybe this is part of Microsoft's long term strategy -- just kill desktop applications and make everyone run VDI in Azure.

Comment Re:Autism and future employment trends (Score 1) 36 36

"Where are all the buggy makers going to work when people stop driving buggies?"
Service jobs.
"Where are all of the film developers going to work when people stop using film?"
Service jobs.
"Where are all of the steel workers going to work when we ship our foundries to China?"
Service jobs.
"Where are all of the assembly line workers going to work when we replace them with robots?"
Service jobs.
"Where are all of the secretaries from the pool going to work when we replace them with computers and software?"
Service jobs.

Where are all of the workers in service jobs going to work when the work is automated or offshored? .....

I don't think we've thought through that last one yet, and it's going to be a very messy transition. I highly doubt we'll jump straight to Star Trek universe where everyone has work that's meaningful and rewarding to them. The problem is that this time around, there are no higher-level jobs to retrain for that the average worker can handle. You can't take a factory worker and turn him into a doctor.

Comment Re:Autistic-friendly business environment (Score 4, Insightful) 36 36

I sure would and I'm not autistic. I can't stand the new "cafeteria table" style open plan workplaces. You can't concentrate on anything, hold a phone conversation with a customer/vendor, or do anything that doesn't involve talking to the 10 other colleagues crammed into your little workspace.

Comment Autism and future employment trends (Score 2) 36 36

Interviews like this bring up interesting points, especially the "Businesses and autism" part. I'm nowhere near autistic/Aspergers/whatever, but I feel for those who do because I certainly tend towards being an introvert. One thing to think about is this -- with the increasing numbers of people being diagnosed on the autism spectrum, where are we going to employ them?

Software dev and IT used to be perfect places for introverts to work -- good pay, interesting work that doesn't involve a lot of personal interaction, etc. One thing I worry about is that with the current offshoring/outsourcing trend, businesses will continue letting these task-oriented IT jobs move somewhere else rather than have to deal with the "weird IT/dev guys." Increasingly, you need to be an extroverted person to be in IT, because often you're the last man standing in the "onshore team" who has to answer for the offshore teams' latest screwups.

I'm guessing the last places for employment for those who don't want to act like salesmen are going to be in scientific research...unfortunately that's a field that many IT or dev people wouldn't be able to deal with.

Comment Lesson - never chase fads with your education (Score 1, Troll) 67 67

I graduated in 1997, just in time to watch the tech bubble inflate to full capacity and pop. I wasn't a CS grad, but wound up in IT. I did notice that a lot of people were starting CS majors while I was in school. Are there really only 24,000 undergrad CS students? Maybe they're just talking about people who finish.

There was an NPR piece a while back about undergrad Petroleum Engineering programs being super-hot and producing grads that got 6-figure salaries at the top of the fracking boom. Cue the stories of undergrads taking huge loans out, spending years studying a field that has reduced employment prospects when they get out. (Almost every law school grad is experiencing this now due to some of the same factors we in IT have, such as offshoring and wage deflation.)

The new grad market is ruthless and demand spikes get flattened out way faster than a typical education cycle. Remember the huge boom in the healthcare field for nurses and allied health professions? It's still there but nowhere near what it was a while back. Now with all the insurance companies merging, I'd hate to be a medical office assistant as doctors figure out they can lay off some of the billers and coders.

While it's true that it's foolhardy to "do what you love and the money will follow," students paying big bucks for education need to focus on fundamentals. Take a challenging subject, figure out what you like to do, and work that into your entry level job search plan. Shortcuts to the huge salaries/signing bonuses are only temporary. If you get caught out, and hate what you studied, you're really stuck.

I'm not bragging or trying to hold myself out as a huge success story, but slow and steady has worked well for me. I watched all the dotcom millionaires from a relatively boring, old-line job where I learned a ton of fundamental knowledge that continues to serve me well. Now we're seeing the bubble ready to pop again, complete with the Silicon Valley companies funded with imaginary VC cash catering to new grads with adult preschool work environments. Now is not the time to go into CS -- 5 years ago was the time.

Comment This is a smart move for them (Score 5, Interesting) 170 170

One problem with Google+ in my opinion was that G+ would be like Facebook if Facebook had perfect information on every one of your habits/actions/locations. When you have to use it to sign in to everything, it approaches the creepy line that a lot of people have.

It's similar to how a lot of discussion forums, etc. encourage or force Facebook logins to post. You'd think that would keep people at bay, but all that vitriolic hatred you see on news forums, etc. is right there next to people's pictures and occupations -- "Joe User - 7th Grade Social Studies teacher at Somewhereville Middle School." I once thought, "No one could be that dumb, posting trash like that as a public figure with their real ID." Sure enough, go look for the person on LinkedIn, there he is with a matching picture, etc. People really have zero knowledge about how social media works, what the companies use the data for, or anything about online privacy.

Comment Re:IT and SW development need better training (Score 1) 132 132

Show me another Western industrialized country where one can more easily dismiss an employee who isn't working out.
You're partially right. It is really easy for a manager to just say "You're fired, pack up your stuff and leave." My experience working with large companies is that (a) this only happens in very rare circumstances, and (b) the cost to advertise, recruit for, rehire and retrain is high, so unless a worker is a serious drain on productivity or is poisoning morale for everyone else, a lot is done to work with that individual. It's usually only the tyrannical small business owners who pull a Donald Trump in front of everyone, consequences be damned. Even though they know they'll win, companies don't want to spend money or in-house counsel time on employment-related lawsuits. Personal example - I know a manager at the large company I work for who spent a year gathering enough evidence for the dismissal of someone who just wasn't learning their job or making an effort to do so.

This is why large companies love contract labor. They can get rid of them tomorrow if they wish with no repercussions. The downside is a disposable workforce who has nothing invested in the future of the company except for their invoices being paid. Personally, I think a lot of the security problems happening lately are because of the split that happens between "the business" and the offshored/outsourced IT. The outsourcer doesn't care about what happens to the company's data as long as they don't go out of business. Not that FTEs are guaranteed to care either, but the lack of involvement can generate this situation.

Comment IT and SW development need better training (Score 3, Insightful) 132 132

I really don't want to sound like an H-1B apologist, but I do understand at least partially where companies are coming from. This comes from being on both the worker side of the fence and the "influencing hiring decisions" side, as well as about 20 years' experience in IT. Some people end up doing incredibly well at a job despite the first impression they give, and others really disappoint after a great first impression.

I do think they're going about this "fix" the wrong way, but I can understand why a company would be reluctant to pull someone off the street that they didn't know already in today's hiring environment and just sort of hope they work out. In my experience, the problem is that there are lots of domestic talented people out there who just can't sell themselves to hiring managers. Either they can't write a resume to save their lives, or they interview very poorly. Conversely, the extroverted schmoozers and posers interview incredibly well, especially in front of the management making the hiring decisions. These guys end up getting the jobs, not performing as expected, and we get the "we can't find any domestic talent" meme. The other two strikes against domestic hires are the perceived wage premium, although it usually takes way more in consulting dollars to clean up offshored or H-1B messes, and the fact that there is the offshoring/outsourcing safety valve that allows companies to ignore the first problem (inability to identify and keep talented people.) Bring the wild west of "expert IT recruiters" in and it's a huge mess.

Techies would never even consider unionizing, but I think a professional guild is a way to combat this. Standardize training, and find a way to equitably weed out the empty suits from the really talented who just don't interview well. The problem is that the H-1B or outsourcing route has to be closed off enough to give domestic hiring a shot at working.

Comment Corporate equivalent = Shadow IT (Score 4, Interesting) 154 154

I have almost exclusively worked for large corporations. In almost every one of them, there has been a central purchasing department that does nothing more than forward orders to a pre-approved supplier. I think you become a pre-approved supplier by kicking back a certain percentage of sales to the purchasing manager.

When faced with this, every place I have worked at has had a shadow IT department. Back in the pre-cloud days, this was the department buying equipment that IT didn't know about simply because the quoted price was too much or it took too long. These days, it's a manager whipping out the credit card and putting company data out on AWS or Azure. The usual "better to ask for forgiveness than beg per permission" applies here, and IT ends up supporting it anyway. Centralized purchasing doesn't work for IT stuff -- it *may* save you money on toilet paper and light bulbs, but IT is too complex to reduce to a line item in a PO.

This is just the government equivalent. The only reason we know about it is because the records are public.

Comment Newsflash, the desperate have computers too (Score 5, Insightful) 176 176

I just turned 40 and am a happily married guy, so I haven't been "on the market" lately. But, I do know a lot of people, men and women, who are increasingly desperate and affected with the "urge to merge." $300K is excessive, and I think most reasonable people would have seen the light sooner. But I can definitely see this demographic being a good target for con artists. This guy even pushed the Italy button -- what lovesick middle aged woman doesn't dream of some crazy Tuscan romance fantasy?

That said, things are different now and it is harder for older people to find suitable partners -- they're fishing in a dwindling pool full of:
- Unpleasant, bitter divorcees who have had their personalities permanently ruined
- The unmarryable -- men and women -- who haven't been able to attract anyone due to serious flaws of one kind or another
- The permanently single -- aka the creepy 55 year old guy still hitting on women in the bar with no intention of settling down or even being honest

Every woman around my age mentions this as their problem. Some might say they're being too picky, but I definitely see their point. If your choices are limited, and someone suddenly comes along who isn't a player, doesn't live in Mom's basement, and isn't an RMS clone, I could see being very vulnerable.

Comment Re:Cloud or no cloud, you still need smart people (Score 1) 138 138

Sure, but this would be after getting rid of 90% of their IT department, selling the data center, and realizing they have to pay Amazon or Microsoft a not-insignificant sum to move petabytes of data/VMs back out. Data transfer to the cloud is free, transfers out cost a lot of money.

I have been through a couple of offshoring exercises. What usually happens is the company is sold a dream, the reality fails to materialize, and the company needs to just wait out the contract because they no longer have the expertise to run everything in house. They'll never admit that things failed - they'll just quietly rebuild things while the contract is running and try to mitigate damage. This is why offshore providers are still able to sell crappy service, no one who's been burned wants to talk about it. CIOs and other execs just see the lower number on the spreadsheet, say to themselves "Gsrtner thinks they're wonderful, so they must be!" and sign the contract.

Comment Cloud or no cloud, you still need smart people (Score 1) 138 138

I've been doing systems work for quite a while, and The Cloud isn't making things easier for IT workers -- it's making them more complex as there are now more moving parts you don't control to consider. Our company is still mainly an on-premises shop because we deploy stuff in areas where The Cloud can't be accessed at a reasonable speed for a reasonable price. But, I would say that virtualization in general has made things a lot more...fluid...than before. What's needed now is more people who know integration and the end-to-end nature of a system. I'm not talking about master black belt CCIE MCSE RHCE whatever savant experts...just people who have the ability to break a problem up into parts, troubleshoot what they can, and know who can help with what they can't.

In previous days, you had the Storage Guys (or Girls) that would do the magical incantations to convince a SAN to provision a LUN to your machine, the Server Guys who would manage the operating system, the Application Guys who would manage the program running on the server, the Network Guys who did all the connectivity magic, and the Data Center Guys who would install and fix physical equipment. Each one of those was a specialty, and still is to some extent. But, as more and more small VMs can be squeezed onto fewer and fewer boxes, there's less of a need for an infrastructure guy. As storage gets more virtual and easier to self-provision, the storage guys become more of a commodity. And if your company goes AWS, Azure or similar, all those Guys get replaced with a web interface and it becomes someone else's problem. I'm still totally amazed how many machines fit on a single HP DL380p physical server compared to what was possible even 5 years ago. And the public cloud services are even more interesting -- multi-football field size isolated rural data centers with thousands of machines and 4 employees to swap parts/install more nodes.

I think the future in IT is going to be less on the front lines and more cross-specialty, regardless of whether your data is onsite or offsite. In house coders are probably going to have problems because every single company is being sold the cloudy Salesforce or other ERP system as the cure for all its ills, so dev jobs are going to shift more towards software companies. Infrastructure guys will still be needed, but they'll be working at a higher level doing design/architecture rather than physical server management. There will still be analyst and project manager jobs, but I think those will be even less technical than they are now. Analysts will solely be an interface between "the business" and "the cloud guys". PMs will be secretaries who beg people to get things done. Add in the constant threat of offshoring, and salaries are definitely going to drop. I think they're probably going to go bimodal -- even lower pay for basic tasks, but similar or maybe even more pay for engineers/designers/architects who can successfully make the transition.

No one is capable of stopping The Cloud. The vendors will continue to sell companies on how wonderful it is, and the companies will find out after a while that it costs too much to get their data back and rebuild their own capacity on-site. I'm just hoping that good people will be allowed to work remotely so there won't be some massive migration that IT guys need to do to survive.

Comment The system isn't the same as it was (Score 1) 265 265

I think a lot of people are thinking about the 1950s style of institutionalizing the mentally ill as being the norm now. It's not. Mentally ill people are pretty much left to fend for themselves, and there is an extremely high bar to putting someone in a mental hospital. States closed almost all of their hospitals in the 70s through the 90s, mostly because of budget problems. (Long term patients used to work as part of their rehab, and when it was ruled they couldn't work for free anymore the model collapsed.) Personally I think some of the capacity should have been retained for medium term stays -- let people work their problems out without having to be put in prison where there is no chance of recovery.

Most truly disturbed people wind up in jail for acting out, or they are homeless on the street. Less disturbed people can be treated without locking them up these days. It's not like it was, when people would be sent away for years or lobotomized because of a panic attack or depression. This is happening because Scientologists don't believe in psychiatry or that mental illness is a physical disease. Most mentally ill people are fine living on their own - it's the occasional shooting, kidnapping or pushing people off subway platforms that gets the most attention.

Economists state their GNP growth projections to the nearest tenth of a percentage point to prove they have a sense of humor. -- Edgar R. Fiedler

Working...