×

Announcing: Slashdot Deals - Explore geek apps, games, gadgets and more. (what is this?)

Thank you!

We are sorry to see you leave - Beta is different and we value the time you took to try it out. Before you decide to go, please take a look at some value-adds for Beta and learn more about it. Thank you for reading Slashdot, and for making the site better!

Comments

top

EFF Unveils Plan For Ending Mass Surveillance

Pseudonym Re:Now using TOR after WH threats to invade homes (205 comments)

Will the EFF be the ones who apologize to the families of those killed by attacks that could have been stopped?

Really? Do tell. What are these attacks that have been stopped by mass surveillance and could not have been stopped by good old-fashioned detective work?

Terror attacks are rare in the United States. They are remarkable precisely because they are rare. This is why anti-terrorism powers are overwhelmingly used to investigate non-terrorism offences, and the vast majority of terrorist attacks foiled are ones that they made up.

2 hours ago
top

Ask Slashdot: Is Pascal Underrated?

Pseudonym Re:Discussion is outdated (468 comments)

It's a little too C++-ish for my tastes [...]

I have the opposite problem: Qt isn't C++ enough for my tastes. As a general rule, you can't program both in Qt and in modern C++ in the same program. Not without an insulation layer, anyway.

yesterday
top

Ask Slashdot: Is Pascal Underrated?

Pseudonym Re:Modula-3 FTW! (468 comments)

There is a Modula-3?

Indeed there is, and it is an extremely influential language. Java 1.0 is pretty much Modula-3 semantics with a C syntax.

yesterday
top

Interviews: Alexander Stepanov and Daniel E. Rose Answer Your Questions

Pseudonym Re:Second System Effect (42 comments)

It is perfectly consistent to say that C++ is a "elephantine, feature-laden monstrosit[y]" compared to C, but still prefer C++ to C. Nobody is disputing that it is feature-laden, after all.

Stepanov said it himself: Simula-style inheritance is unsound.

3 days ago
top

Is D an Underrated Programming Language?

Pseudonym Re:COBOL (382 comments)

Sorry, left out the cxu. So much for mian putron Esperanton.

about a week ago
top

Is D an Underrated Programming Language?

Pseudonym Re:COBOL (382 comments)

Kial Esperanto uzas malfacilan Polan prononcon?

about a week ago
top

Gunmen Kill 12, Wound 7 At French Magazine HQ

Pseudonym Re:islam (1350 comments)

First of all the attack on the Pakistani schools would not count since the victims were muslim.

I'm floored. Something like 90% of victims of Islamist terrorism are Muslim. Attacking non-Muslims is the relatively rare case. What you're saying is that criticising the vast majority of heinous acts by Muslim extremists doesn't count as criticising Muslim extremists. That makes no sense at all.

But since you asked, there's plenty of criticism going around about this crime. Do you want links, or can you use Google?

I'm talking about the local papers explaining that some attack was wrong and counter to islam, that extremist justifications and teachings are counter to islam; I'm talking about the local imams preaching these things in the local mosques; etc.

What research did you do to verify that this isn't happening? Did you read some local papers, attend Friday prayers in some of these local mosques?

Having said that, do remember that groups like ISIS and the Taliban are something akin to organised crime or an occupying force in places where they control. Much of the resistance is underground because it has to be.

about three weeks ago
top

Gunmen Kill 12, Wound 7 At French Magazine HQ

Pseudonym Re:islam (1350 comments)

You can't protect or refuse to criticize muslim extremists merely because they share your faith. When non-muslims have legitimate grievances against muslim extremists then moderate muslims need to side with the non-muslims.

But they do! Everywhere except, apparently, in stories reported by the mainstream media. Gotta keep fear alive.

about three weeks ago
top

Interviews: Ask Alexander Stepanov and Daniel E. Rose a Question

Pseudonym Re:STL (80 comments)

Merely using information hiding doesn't make something object-oriented. There is virtually no OOAD in the STL.

(Pun intended; grep for virtual in the C++ standard library some time.)

about three weeks ago
top

Interviews: Ask Alexander Stepanov and Daniel E. Rose a Question

Pseudonym Re:C++ (80 comments)

The first implementation was in Ada. It wasn't the full STL as we know and love it today, of course.

about three weeks ago
top

Interviews: Ask Alexander Stepanov and Daniel E. Rose a Question

Pseudonym Re:STL (80 comments)

Related question: C++ was originally conceived as "C + Simula", but something that is interesting about the STL is how non-object-oriented it is, in particular using no inheritance.

If we were designing a new "better C" today, one that you'd be happy to implement a STL-like system in, knowing what we know now, would we bother with Simula-style objects at all?

about three weeks ago
top

If the Programmer Won't Go To Silicon Valley, Should SV Go To the Programmer?

Pseudonym Re:Mirrors (294 comments)

And you define critical thinking as different from logic?

Uhm... yes.

Logic is a framework for constructing individual theories which can then be used to model real-world situations. Critical thinking is a system and methodology for understanding and analysing arguments made by others, and to ensure that your own thoughts are clear and reasoned. These are obviously distinct areas, though they are not disjoint. And they are both "philosophy".

The critical thinking subject which I did didn't touch on linear logic (for example) at all. Plus, a lot of it was analysing the semantics and pragmatics of language.

Well now, I guess Trig must not be math either.. because you know.. it's "trig" right?

Trigonometry isn't calculus, even though there are plenty of places where they touch. But they are both "mathematics".

about three weeks ago
top

If the Programmer Won't Go To Silicon Valley, Should SV Go To the Programmer?

Pseudonym Re:Mirrors (294 comments)

Blindly claiming "Dunning Kruger" when a person has at least 7 years of University knowledge is telling.

I had some high-profile examples in mind as I was writing that. Richard Dawkins is probably the clearest example of someone who is an undoubted world expert in one field (biology, and evolutionary biology specifically), who vastly overrates his competence in pretty much any other field.

Philosophy is in essence logic [...]

I'm going to say "no". Philosophy is, in essence, critical and systematic thinking. Its boundaries are fuzzy, but it is the primordial soup from which new fields of human endeavour form. These fields eventually graduate to be new faculties and departments of their own. Logic is one such field, but it's not the only one.

So let me be more clear on this.

I would not turn my nose up at anyone with a PhD in philosophy from a non-fake institution of higher learning. That person is very likely to be highly valuable. However, the context is hiring an engineer to implement and maintain industrial control systems. I don't know specifically what the job was, but do remember that SCADA systems are often the second line of defence against an industrial accident. I've been engineering for 20 years, and I consider myself unqualified to work on safety-critical systems.

If all you knew about someone was that they had a PhD in philosophy and described themselves as a "self-taught Java guru", what are the chances that they have the appropriate knowledge and methodology (whether formally acquired or not) for this specific job?

Does this guy have what it takes to learn the requisite knowledge? That seems likely; in fact, he probably has more aptitude for working in this area than most software engineering graduates. Has he acquired the knowledge during his "self-taught Java guru" training? That seems unlikely.

about three weeks ago
top

If the Programmer Won't Go To Silicon Valley, Should SV Go To the Programmer?

Pseudonym Re:Exactly this. (294 comments)

Are you one of those idiots who thinks programming is harder than philosophy?

I am not the OP, and I don't think this, but someone with a high qualification in one field may, thanks to the Dunning-Kruger effect, overestimate their aptitude for a different field. Plus, "self-taught java guru" is a red flag by any measure.

Yeah, I'd consider them if they had a relevant referee or a portfolio, like a github repository that I could inspect. A smart person is a smart person and formal qualifications aren't everything. Besides, their PhD might have been in logic.

about three weeks ago
top

Red Hat Engineer Improves Math Performance of Glibc

Pseudonym Re:Yikes ! (226 comments)

This is MPI code, not IEE754 code.

As I understood it, the issue is that glibc's libm occasionally falls back to a multi-precision version of some transcendental functions if the "pure IEEE-754" version isn't good enough on some inputs.

I did look at the code, and the implementation of exp() on 64-bit floats does indeed fall back to a slow path on extreme inputs. I can't remember if it's the MP code or not.

about three weeks ago
top

Red Hat Engineer Improves Math Performance of Glibc

Pseudonym Re:Obligatory Carmack (226 comments)

You mean obligatory Walsh. Of course, it's obsolete now that RSQRTSS is ubiquitous.

about three weeks ago
top

Red Hat Engineer Improves Math Performance of Glibc

Pseudonym Re:Yikes ! (226 comments)

What kind of horror story is this??

Don't tell me that you've never written an algorithm which uses speculation before. It's quite a common scenario that you have a common fast path and an uncommon slow path, and the cost of deciding which path to use is a significant fraction of the cost of the fast path + checking the result.

In the case of libm, there are a lot of code paths which are there only to maintain strict compliance; no numeric analyst would ever call exp or pow (or even round) with arguments in that range. Not on purpose, anyway.

about three weeks ago
top

Red Hat Engineer Improves Math Performance of Glibc

Pseudonym Re:C versus Assembly Language (226 comments)

Not just every architecture. In general, you may need to write it for every major revision of every architecture. As CPU pipelines and instruction sets change, the hand-crafted assembler may no longer be optimal.

(Exercise: Write an optimal memcpy/memmove.)

about three weeks ago
top

Red Hat Engineer Improves Math Performance of Glibc

Pseudonym Re:C versus Assembly Language (226 comments)

Indeed it is, however it's still rare that you have to go to ASM in those cases. In simple cases the compiler already generates SIMD code on code which can benefit from it, and for almost all other cases, there are C intrinsics.

about three weeks ago

Submissions

Pseudonym hasn't submitted any stories.

Journals

Pseudonym has no journal entries.

Slashdot Login

Need an Account?

Forgot your password?