Become a fan of Slashdot on Facebook


Forgot your password?

Comment Re:Your simple solution doesn't work (Score 3, Informative) 218

Eliminate the H1-B program. For the few real geniuses we want have a lengthy peer review process to prevent the diploma mill graduate loophole. Raise taxes on import goods made with slave labor. Redistribute the wealth in the form of socialized medicine, free education, infrastructure programs and basic income. Lather, rinse repeat. These things work. That's why the 1% is frightened of them; and it's why they want you to be frightened too.

All potentially good ideas, but first we have to find a way to get the 99% to get out of the "temporarily embarrassed millionaire" mindset and get them to see how they're often being used to the benefit of the 1%.

Comment Re:It's pretty obvious what happened to them (Score 4, Interesting) 218

If the salary discrepancy between a US citizen and (let's say) an outsourced Indian employee was less than $4k/year, companies probably wouldn't bother outsourcing in the first place.

Sure they would. In addition to the (illegal) discrepancies in pay that some companies implement, H-1Bs are perceived as more compliant - the employer can always hang the threat of cancelling their visa sponsorship over their heads, so they won't talk back and can't just leave for another U.S. company.

Comment Re:A better use of educational dollars (Score 1) 157

Lots of students struggle with math above arithmetic. They could see why arithmetic is useful, but anything above that? Not so much. So by having the students apply math concepts to accomplish something (ie programming), they'll improve in High School level maths.

Algebra is pretty much necessary, as is a basic understanding of probability/statistics. Trigonometry can be quite helpful as well. Calculus is handy to know at times, but most people get along just fine without it. Learning how to do stuff like decomposing polygons to calculate area, and deriving proofs in a basic algebra/trig course is going to give them most of the benefit of learning to think algorithmically. Any student that can't see where algebra/trig would be useful in their everyday lives has a poor teacher, because there are *tons* of examples.

What math concepts are they going to learn in a dumbed-down high school CS class that will help them enough to justify that kind of expense?

Comment Re:Surprised? (Score 1) 570

For me I would still choose a Christmas party over a SAN upgrade, as one is guaranteed to cause trouble, whereas the other is only a maybe.

I might reinforce the fact that the SAN *was failing* (often multiple times per day, causing outages of 45 minutes to an hour each time). It wasn't a matter of if it was going to die permanently, but when. Replacing the bad controller (which I ended up having to go on eBay to acquire since it was a part that had been EOL for years) was a band-aid at best, and I still had to fight to get the money for even that. Putting valuable data that has been entrusted to you at risk is not acceptable, IMO.

This sound like every business I've ever worked for.
Given the choice of doing things right and taking ages, and missing a limited window of opportunity, or going to market half-baked and winging it, and possibly making it work, I will choose the latter every time.
Every successful business does this (Apple, MS, Google etc).

Every business you've worked for engages in blatant copyright infringement and considers that okay as a normal course of business? As a side note, I left the company as part of this debacle, and they weren't able to get the feature implemented until a year and a half later.

This doesn't sound like a management problem to me. This is pure Ops.

No, it was pure management. The president wouldn't release the funds to get the necessary subassemblies built, and knew for a fact that the machine wasn't going to be even close to ready when the acceptance was scheduled. They were successfully sued for a lot of money not long afterwards as a result.

eg can you imagine if the company announced they are cancelling the annual Christmas party in order to upgrade some IT thing that no-one else knows what it does? Yeah you'd get a working SAN, but every other person in the company would probably resign. No manager will ever make that choice.

They'd understand that "no working SAN" == "no more customers", and that after the lawsuits were done there wouldn't be a company to work for.

Comment A better use of educational dollars (Score 1) 157

I see this big push for CS literacy in schools, and it puzzles me why we're talking about spending so much money to teach a skill that most kids aren't going to use in their everyday lives while we're dealing with stuff like the Common Core silliness and teachers/parents are having to pay for basic classroom supplies out of their own pocket. They have money to buy tablets/laptops for kids (and the associated IT support costs), but they don't seem to have the money to make sure that teachers have enough whiteboard markers, copy/printer paper, and other fundamental stuff that every school provided for the classroom 30 years ago.

IMO. time spent teaching to code would be better spent on the 3 Rs, and far more useful to the kids. Intramural programs for teaching how computers interact with and affect society might be a good thing too, but not at the expense of skills that they *need* in order to function in their adult lives.

Comment Re:Surprised? (Score 1) 570

It's easy to pick good and bad ideas after the fact. It's picking them in advance that is difficult.

Sometimes, but a little common sense goes a long way. Here are a few of the "good ideas" management has had at places I've worked that have failed spectacularly, after having had them pointed out ahead of time and been burned at the stake for it:

1.) Allowing their in-house cloud infrastructure that ran critical lockbox services for two dozen banks to run off a single homebuilt SAN with a failing RAID controller for months because they didn't want to pay for a new SAN (or even a replacement controller until the customers started talking about legal action for all the downtime), but we still had cash for a lavish two-day Christmas party.

2.) Basing a new high-profile product release around the use of a complex proprietary third-party library that they had not licensed and had no documentation for, and expecting the integration to go smoothly on target for a release six weeks later.

3.) Bringing a customer in for an on-site acceptance test, when the machine was missing a power supply and RF driver, and thus was not functional *at all*. Bonus points: the customer was from Japan and flew all the way to Florida for this acceptance test.

I get that it's hard to predict the future, and some decisions are hard to make ahead of time because there's not enough information available. That isn't what I'm talking about.

Comment Re:Surprised? (Score 3, Insightful) 570

Anyone who does immediately gets the bum's rush: incompetence, insubordination, bad judgement, blamed for someone else's incompetence or malfeasance, face doesn't fit, socially inept, politically incorrect... the list goes one for ever.

It's not just big companies where this happens, and it's not limited to the C-levels and their minions. In my experience, there are far too many in management at all levels that can't deal with the blow to the ego of being told that choices that they've made aren't good ones. Rather than actually think about what they've been told, they perceive it as unwarranted personal criticism even in the face of overwhelming objective evidence.

Comment Promised throughput (Score 3, Insightful) 154

Wouldn't it be nice if ISPs wrote a rebate check each month to reflect the percentage of their promised throughput that was actually available?

They do if you want to negotiate a SLA that guarantees it, but that tends to be kinda expensive for the average residential customer. Otherwise you get best-effort.

Comment Re:Pure crap (Score 1) 268

During this stepping process, you had to stop the machine and manually rotate the spindle to check the sizes and ensure the sensors were reading properly which is not a problem at all (done all the time on manual lathes). If an errant signal is sent, it could move the tool, start spinning the lathe or a host of other uncontrolled things unexpectedly.

Which is why we have interlocks on things like that. If the machine doesn't have some physical means of being rendered safe, then it's not acceptable for the shop floor IMO. Back when I worked for a laser integrator, we had one machine series that had the shutter controlled solely by software, and the computer would close the shutter when it saw the door I/O or shutter switch signals go low. I raised hell about that until the electrical team put in a real, no-shit interlock that was physically and directly connected to the shutter solenoid. Of course, all of the encoders, optical switches, etc. were still active when the interlock was open since as you mention, you sometimes need to check to see if things are working properly, but the machine couldn't produce a beam or move any of the fixturing, conveyors, etc. if the cabinet was opened or a light curtain was tripped no matter what the software told it to do. Of course, if you willfully defeat the interlocks that *are* there, it's on you if you get hurt.

Slashdot Top Deals

The primary function of the design engineer is to make things difficult for the fabricator and impossible for the serviceman.