Slashdot is powered by your submissions, so send in your scoop


Forgot your password?
DEAL: For $25 - Add A Second Phone Number To Your Smartphone for life! Use promo code SLASHDOT25. Also, Slashdot's Facebook page has a chat bot now. Message it for stories and more. Check out the new SourceForge HTML5 Internet speed test! ×

Comment Re:Can't a magic 8 ball replace most CEOs? (Score 1) 236

In my experience with ceo's and senior executives- the "numbers" have a funny habit of changing to match projections. I saw them waste at least 6.5 billion dollars over 5 years in failed project after failed project. All based on unrealistic assumptions. And in every case, the failures were redefined as successes except for the failed SAP rollout.

A large corporation can cover some terrible errors and CEO's (and executives) are paid for changing things- not for running them well as they are. So you can have a good business practice and it will be removed and replaced with something else which is much, much worse.

If the change works- yea! Big bonus for the CEO and/or executives. If it fails- yea! Big golden parachute. If it's in the muddy middle- there will be a lot of pressure to say the change worked. Because deadlines, definitions- sometimes reality- is subject to intense pressure and manipulation to say the change worked.

Actual reality doesn't set in until they leave or the company goes tits up.

Comment Re:The pain starts in 2030... (Score 1) 236

OTH, as of 2024, the boomers will already be dying in large numbers. Roughly 4 million a year by that point.
60% of men and 43% of women boomers born in 1946 will be dead by 2026. Actually, those who are not well off are more likely to be dead at that rate by 2023 and that's 80% of the population.

But going by the more conservative figures...

By 2030, almost 5 million boomers will have already died.
By 2040, about 30 million boomers will have already died.

The biggest crisis we really need to address is medical care expenses for the last 90 days of life. Speaking as a boomer myself, we cannot afford to spend over a thousand dollars a day to keep someone alive for an extra three months. That's were we need to cap our spending and limit our losses. If someone wants to spend their own money or has insurance- fine. Otherwise, we have to make some hard rational choices.

However, if we destroy 30% of jobs for young people with automation and AI, then there won't be tax revenue to pay.

Comment Re:Like what? (Score 1) 236

Actually, there has only been a job if you speak of humans generically.

Sure, the generation after the luddites had jobs. But most of the luddites died homeless or exposure and starvation years earlier than they should have.

We *may* have jobs again 30 years after this hits but the most likely scenario is 33% unemployment for decades. And almost all manual labor jobs suitable for people with low drive or low iq being replaced by robots. And many high intelligence jobs being replaced by automation. And many skilled jobs replaced by a combination of automation and robots.

Over the next 10 years, it looks likely that 10% of jobs will simply go away. This includes m ost retail clerks (4 million jobs) (as you simply pick up the goods and walk out of the store with an rfid charging you automatically) or you order it online (for less- and no gasoline or miles on your car and no going to the store to find it is sold out), drivers (3 million jobs), most manufacturing jobs, and most manual labor jobs.

In the end- taxes will have to be collected from those who still have incomes. And overall our productivity will be higher than it is today- so providing for everyone should take a smaller share of gdp. But if people are left without jobs and without safety net benefits- it could become a hell hole and possibly fall to civil disorder and even civil war.

Comment Re:Like what? (Score 1) 236

Climate change is serious but not as serious as the limits to growth coming up. They are going to hit harder and sooner.

No more cheap stainless steel will be a big one.

We could see a billion people die ahead of time over a single decade sometime between 2050 and 2100. Likely to threaten civilization, provoke wars, and be a period when the carrying capacity of the earth drops by a couple billion people over 50 years.

Our usage of many resources continues to grow exponentially as the population continues to grow and the standard of living continues to rise. We consumed more chromium in 2014 than we did from 1901 to 2000 combined. And similar for many, many other resources.

At the same time climate change might destroy our ability to raise grain as it destroys viability of many of the worlds major growing zones and pushes the climate into areas not capable of growing. And changes the pattern of rainfall.

Climate change is dangerous and big- but limits to growth are much bleaker. And the most likely scenario is they will hit way to fast for us to mitigate them so we'll still be accelerating as we hit them head on.

Comment Re: Electric, or Jet? (Score 1) 169

Actually... I knew exactly what he meant.

I think the person being pedantic is you. He might being imprecise. I think you are the one who's being ostentatious about your learning or overly concerned with minute details or formalisms.



ostentatious in one's learning.
overly concerned with minute details or formalisms, especially in teaching.

Comment Re:Which is why bars serve peanuts pretzels and ch (Score 1) 78

For three thousand years, everyone "knew" that the only places to play go pieces in the early game was on columns 3 and 4.

Then A.I. played on column 5 in the shoulder position on move 37 and it payed off enormously with control of the middle of the board later in the game... which it won.

Salty snacks has seemed to make sense for a long time. But .. then a study comes along and what you always "knew to be true" may be shown to not be true.

Comment The Tautological President (Score 3) 619

President Trump has made a habit of signing executive orders which say we will do what we are doing.

And that's all the latest executive order does. It literally doesn't change anything. It simply says we will do what we've been doing.

And, really, that's all the president can do except veto things. Mr. Trump shouldn't even be able to order acts of war without prior approval of congress. It's one thing to attack terrorists (for which president's have some approval). It's illegal and unconstitutional to attack the legitimate military forces of sovereign nations without explicit approval by congress.

When are people going to catch on to the fact that Mr. Trump isn't really achieving *anything* with most of his executive orders except publicity? He's dependent on congress to budget money and to change laws.

And since his government is still largely unstaffed after nearly 90 days in office, he lacks the people to implement his policies. There is a serious disconnect between the presidency and the departments right now created by thousands of unfilled upper level positions.

Comment Re:I was most frustrated by ... (Score 1) 149

Developers deliver kludged, hacked up, cowboy code.

Maintenance programmers fix the bugs in that code and should also be able to make it easier to maintain.

I managed both development and maintenance teams and I was a maintenance programmer at heart from the first code I took over in the early 1980s.

The original rule was no one can touch the program INVTUP. After they grew confident in my abilities, they let me make a series of changes to the program that didn't change functionality but which did improve legibility and maintainability. It took about 15 months but at the end I'd reduced the program from over 30,000 lines of ill documented spaghetti code to about 12,000 lines of well documented, well structured code which was trivial for any programmer to maintain..

In many cases, I've helped other programmers debug their programs- often merely by having them change the variable names to be more descriptive. Likewise, using eclipse refactoring tools to extract subroutines can turn a block of hundreds of lines of code into a dozen subroutines and a block of code that each fits on one screen.

Once SOX came in, you could no longer do this kind of code improvement during downtime. Instead, you did nothing until something was approved. And in some cases, it was literally over a week of sitting doing nothing because you couldn't even check out the code without an approved ticket or project. And once things were approved, then you could start. And sometimes, they'd spent so long approving it, that you were very close to the point release date.

Comment Re:I was most frustrated by ... (Score 1) 149

Sure I could have done that- but as a low level manager (promoted after SOX), that would have made no difference unless i was assigned to a project to work on our controls and procedures. Complaining about the procedures beyond simple grousing would simply get you on the shit list as having a bad attitude.

Comment Re:I was most frustrated by ... (Score 1) 149

I live and love in Texas. I saw how age discrimination went in computer science back in the early 80's before age discrimination protections were gutted in 2009.

So I saved hard and I've been retired for several years now. The only programming I do is for fun (star fleet battles, minecraft, noodling).

The last employer was only cheap in some ways. As I said, they dropped about 1.5 billion on a failed SAP implementation. As implemented at the company the CEO and CIO were legally responsible for changes so they wanted assurances about every change from someone. We also had auditors who reviewed a about 2% of the changes every quarter and confirmed the process was being followed.

As I indicated, there simply wasn't enough time to approve everything for quite a while after Sox came out. We went 6 months where it was a real night mare and getting approvals was tight for another year or so. And then sap came along and consumed even more bandwidth (despite statements it wouldn't occur, there was a lot of bidirectional double development in the old and new system... and in the new and old system.).

If a coder made a change without approval- it was a firing offense. You could gamble and risk it I suppose but it wasn't worth the risk.

Comment Re:I was most frustrated by ... (Score 1) 149

Almost completely true. But if there was any change noted (even if not a problem) then it was a firing offense. After a couple people got burned, everyone else just gave up.

It's like when they tried to go to SAP. Many of us could see the way they were approaching it was doomed and actually - probably literally impossible. After two people firmly fought against it and said it was very likely to fail and were fired, everyone else just shut up and did their jobs and after 18 glorious months of design they started rolling it out and it utterly failed when it was under 7% rolled out and had to be rolled back. Last I heard the cost was at least 1.5 billion dollars.

Comment Re:I was most frustrated by ... (Score 1) 149


Pre-Sox, you could refactor code and if it had the same outputs for the same inputs, you were fine unless there was a major failure in production (which happened about 20x more often for new development than it did for maintenance and 20x more often for existing code getting a new unexpected condition or data that killed it.

And for larger changes, you could get approval from your team lead- who might mention it to the manager. If they were good then that was it.

And then there was another scale for senior directors.

AFTER Sox- everything had to be approved by everyone up to the CIO- and tho I had no visibility to it, I'm pretty sure there was a CIO/CEO meeting too because ultimately the CEO's job was at stake.

So the cost overhead for changes was so high that small changes could no longer be approved.

And at first it was horrific. We had change meetings where 290 projects would come in and then only 60-70 would be approved in the 90 minute meeting and *that was it* for the week. You had to come back next week. But #71 wasn't #1 next time. Nope- it might be #41 or even #85 if higher priority projects got ahead of it. And after missing 6 change meetings, you missed the release. And that was it for three months. And at the same time, the people blocking you were screaming at your managers to get the medium priority projects approved and about low priority projects when that particular 13 million dollar customer called in upset that their problem wasn't being addressed.

it was very unpleasant.. and it made me unhappy.

Slashdot Top Deals

Work continues in this area. -- DEC's SPR-Answering-Automaton