>So from what discipline do you plan on recruiting people who have all the training that you so desperately need? Chemistry majors perhaps? Physics, maths, and yes the occasional chemist, or at least chemical engineer.
>Perhaps I'm misreading you, but your tone seems dismissive, as if teaching design patterns and classical algorithms was somehow unimportant. My apologies, I see them as critical, but I am dismissive of the idea that the language you choose matters all that much. If Java, fine, but you can do it in VB (not that I advocate that)
>But just because someone knows English doesn't mean they'll understand Shakespeare. Agreed. But what would you think of someone with a degree in English who had not read any Shakespeare ? My epxectation is that a graduate of a subject has a superior, deeper grasp of the subject.
To stretch the analogy perhaps too far, what would you think of an English major who had only read Shakespeare in cartoons ? "because it is easier to teach"... As it happens I have read and understood Shakespeare, and my early introduction to it was highly dumbed down. (my dimwit politically correct English teachers dealt with the anti semitism in the Merchant of Venice by ignoring it).
But I did read it in the original later, indeed one is expected to have read a large % of Shakespeare *before* starting an English degree. Chaucer is rather harder going.
Yes, by all means use any language you like to do the algorithms and structures necessary for kernel work. I first did threads in a horrible mutant Pascal, and actually found them easier to understand in C. We did O/S stuff in pseudocode, ut uitlately we did real code in a real system.
Perhaps a better analogy than English is medicine. Most doctors are not cardiac specialists, and would be as hesitant to do anything to the heart as most programmers to hack a kernel. But all doctors, even dentists know at least roughly how to spot when the cause might be the heart, or when the odds are it is something else.
It goes back to my themes of a good techie being able to think at multiple levels to be effective.
>Why would anyone need to understand the linux kernel or C++ unless they were linux kernel programmers.
Same reason wannabe doctors cut up dead bodies.
>Also your comments on Java are laughable. Java is in use in day to day objects in real life. Phones, Set top boxes, Cars. It is not just web servers. There are even >operating systems and emulators in Java which run fine. Go read up on JIT and stop living in the 90's.
OK, I'll come clean I have no problem with Java, it's Java *developers* I despise, when they can't be bothered to read my posts, and take anything which does not sing its praises as attacks on their manhood.
I really really carefully said, I had no problem with Java for many sorts of work. I explicitly agreed later that there were many jobs in it, and will go further and say that it is the right choice for many things.
Just in case my words were read by over-sensitive Java developers my post said that the success of Java in isolating you from low level detail was an issue.
I think you need to be able to think at multiple levels, even if you only code at one.
Interview questions require a finite tractable answer in a relatively short amount of time. As it happens I don't ask about bitmaps, but it strikes me as a suitable for part of evaluating an entry level developer. Nicely open ended for a superior candidate to shine.
You seem to be under the impression that a good developer knows how to do certain things. We part company on that. A good developer can work out how to do things.
The logic and coding issues behind bitmaps has applications in high end finance, but I don't expect a newbie to even be able guess what they might be. I do expect them to deal with a problem they have not seen before. I don't give a toss if they understand the genealoogy of OS/2, to Windows bitmap formats, or their relationship to VGA palettes or the transputer. I do care that they can think up a way to store data to a spec I give them.
I've had the Azure stuff demoed to me, and it looks interesting, and they coped well with the idea of a headhunter having an interest in their stuff. seemed a little pricey for what it was though, does you experience justify the cost ?
>Perhaps you should start with finding applicants who have a decent command of the English language,
Nope.
The thought makes me feel mildly ill.
I have explicit instructions from some employers that they don't care about verbal skills all that much because they are sick of getting preppies with people skills who take personal offence when the compiler criticises their code.
There are mid and low end jobs knocking up GUIs or importing data that require good people skills. We don't hire for those.
The trend is that by the end of 2008, we expect that people we work with who don't have English as a first language will be the majority.
We like people skills, but it's not in the top 5, and we will pick good maths and C++ over a nice smile every time.
>and the proven ability to interact with customers, management, and a team of programmers?
Newbies don't have that, and frankly I score down people whose CVs talk too much of "customer contact", indeed in our 140 page guide to getting a fuck off job in a bank, we tell people to scale back holiday jobs and customer skills to zero, or a vestige. We do like leading men under fire, and yes that includes people who have shot at friends of mine.
>Ensure they not only have the ability to communicate coherent thoughts in person, on the phone, in conference, or on a whiteboard.
We like articulate people. A lot. Even if the English is broken we are implacable in our demand for clear thinking. I can't use whiteboards, stopped altogether when a minion pointed out that I always drew the same horrible shape of badly drawn concentric rings with random bent radial lines. (I think it is a message from the Iterators)
>Additionally, test them for their ability to adapt to the dynamic environment that is Corporate America.
If a candidate claimed such an ability, the time taken to end the interview would be determined by my rather English politeness, not any desire to hear them bullshit any more. Again this is something we explicitly warn against in our guide, the phrase being "do not sound like you work for Accenture".
We recruit across for roles across the globe, we don't do "Corporate America", having done stints in such things I am glad I can choose never to do so again. We do American banks, and if I recommended a candidate on that basis, not only would they not be interviewed, I would lose some of my hard won credibility. (tough to do when you are pimp).
>I could walk outside right now, and find myself umpteen code monkeys to write EXACTLY what I wish them to write,
You are a lucky man to have found a job that pays you without setting the sort of demands people in my business fight.
>as opposed to the creative dialogues I seek from my team members.
I like creativity, tough to find, tougher to build. Candidates get big bonus points for this. Most don't get any.
>I want people to be enthusiastic about coding, their product, and their team.
Don't we all ?
>P.S. I'm not sure, but describing yourself as a "pimp" gives me pause as to your ability.
You are right, if I were dealing with "corporate America", it would be insane. I deal with investment banks, hedge funds, ratings agencies and the occasional government.
The mindset in my niche is quite different.
Not necessarily true, but that is my [business] perception.
Oh dear, an Apple fanboy. My issue with your rant is not whether you are right or wrong, but that the quality of the argument is so low.
I did not give any mechanism for the correlation between low programming skills and Apple ownership, and to be honest I don't know why it is so high. For all I know, it may be that the stability of the system does not attract serious programmers. For most of my life a large % of the s/w I have used has been beta, homebrew, or just terribly buggy.
>Expecting people interviewing to join an investment bank to be kernel hackers is pointless. You are confusing the housekeeping activities which are much the same at Goldman Sachs and Wal Mart and very high end work. That's not your fault, if everyone knew the shit I do, then they'd make my money.
>Expecting them to have good signal processing skills is stupid. You are wrong. The highest paid people we see taken in by banks and hedge funds know SP. By "highest paid" I mean bidding wars between the top investment banks. You aren't in that category are you ?
>I do however know how to develop exceedingly high quality business software in extraordinarily short timescales So you are a back office developer. Fine. The majority of bank developers are like you, they earn OK money. We don't handle people like you, unless they've accidentally ended up in a job far below their capabilities.
We seek out a narrow speciality combination of skills, and there's no shame or despair to be felt that you don't fit our filter. You're not a vet either, fine.
We do however recruit the people who will be giving you the "fluffy" specifications. You may or may not like that.
>So, what, you're going to hire math geeks only?
No, physicists, electronic engineers, occasional economists.
I would love to recommend more CS grads, but the quality is too low. To be sure, I don't expect CS courses to prep for my speciality, but I despair of grads who don't know O(N), or only speak one programming subject. I despise essays.
>If you're about to decree CS as a non degree, maybe you should get the input of the people who will be most brutally affected by your shortsightedness.
I'm sure I said that *I* am a CS grad, and have shared my view about them with my peers. Their reaction is typically to try and outbid me on witless newbies.
As for algorithms, I would ask you to read my post. I am entirely comfortable with Java as a language for patterns and algorithms. I said this explicitly.
If you are going to call me on this, at least do me the courtesy of reading what I say.
Same applies for explaining buffering in Python. I learned this stuff in pure pseudocode, not C/C++, bu then I got to play with a real O/S.
If my earlier post is too hard for you to read, I will summarise it as that I do not have problem with Java, I have a problem with *only* Java.
>If it was an easy subject, would changes need to be made to make it easier?
CS is not inherently easy.
But you can choose to teach the easy bits, many do. I see the results, and they are a perverse combination of horrifyingly dull.
>The point of a CS degree is to allow you to learn with phenomenal velocity--to know where to look for answers, to know how to read the papers, to know how to integrate all that into your noggin very quickly.
Agreed, where we differ is that this is an ideal that is not only in decline but is in serious danger of disappearing altogether.
You are confusing what I say about the "usefulness" of C++ in *exactly* the way I said it was not to be confused. (read my post).
I can be convinced that Java, Perl, VBA, SQL, or force of mental will staring at the screen is better than C++ for a given task.
That is not my point, and if you had read my post, you will have seen that.
As a HH, I care about C++ because my clients do, and agree it is often used when not the best tool. But in my experience I see a lot more of Java inappropriately chosen because of the poor breadth of the decision makers.
My point is that if a CS course did (say)Haskell, Perl and Java, I would not have a problem. But choosing only to do Java because the others are are hard is not the way to my heart, or a top flight job.
By "hard" you should understand I do not mean "awkward" or "tricky" but conceptually challenging. Some of C++ falls into the first two categories, some doesn't, but no single language is rich enough to educate a computer scientist well. C++ is the best in my opinion, but you'd learn more from doing Java and Haskell than by just C++
>How good ? PDEs are a minimum, and I'd see you as a potential quant developer so that would mean finite difference as well as monte carlo. More stats would be nice.
I do not think the set of language matters so much as their breadth.
I am indifferent to Ada, maybe good or bad, but alone no single language allows a good CS student to get to where he wants to be.
C++ has more concepts in it than Java, which is of course one reason it is harder to learn, so gets points from me for that.
Lisp is actually a very simple language, but being so different from Ada, C++ and Java, it helps the student learn to think about things in different ways.
Of course some never get their heads around Lisp, which is fine since it has almost no commercial value, but it should be sending a signal that this person simply isn't good at programming. Not saying they shouldbe shit hot at functional languages, but failure is worrying.
CS courses seem to be set up to make them as hard as possible to fail, yet programming is a talent. Some very smart people just can't do it. Thus allowing them to write essays, draw pretty graphics, and use Java toolkits to make applications is fraudulent.
The first page of googles on my name are all me:) A great improvement from a couple of years ago, when since my name is "catholic" I was buried underneath a list of various child rapists in the Catholic church.
I'm Dominic/at\ PaulDominic dot com, bung us your CV, and I'll get our 140 page guide to you. It's tough. There is one large firm who may be interested who have offices in London and NY, they say they want C++ but are lying, since their interviews are heavily K&R C.
Java backends are rare in our game, and we don't get involved at all, though we have a frankly demented project for back end servers running Excel VBA. (mad, bad and expensive to know).
There is not a great demand for C, though it pops up after C++, VBA, C# but above Java. We are more flexible than most on branding of university, since we work with people from everywhere ("everywhere" including a few people from places not recognised as states at all).
>You must realise that many of the great computer programmers started doing it as children I started at 12 on a computer stolen from the government:)
>high stakes is not done by magical "better" programmers or harder whipping. Thank you for pointing out something I should have said, and with which I agree 100%.
High stakes financial programming is rarely done by the best programmers, indeed I have seen stuff handling real money that would shame a Kings College CompSci graduate. We've done a 140 page guide to our line of work, and the effects of EDS on banking IT is better imagined than experienced. We want very good maths as well, and that closes the filter so tight that one has to trade off.
>is done with boring, systematic testing Actually testing does not have to be boring, and is so only because it is usally very badly managed, and these days often staffed by people who didn't make it as a mainline programmer. Much can be automated, and if you have the right culture (rare) the finder of a complex bug is a hero, not (as usually) a troublemaker.
>If your code is processing real money, and it processes the equivalent of your yearly salary in a few minutes or hours, you are "way above your head" You and I operate in different domains, and in mine the relevant unit measurement is Hertz. Our people write code that handles their annual salary ranging from 0.1 Hz to some who work at the kilohertz level, with higher peaks.
In high frequency algorithmic trading one's code may easily process megabucks/second.
Sadly I am under an obligation not share names, but there is (exactly) one system that handles G7 country sized money in real time high frequency trades written almost wholly in C#. Entertainingly, I kind of believe that Microsoft doesn't actually know about it, even though it is a very glamorous project. But everything else is done in C++, with a few C things lurking in dark alleys, preying upon the weak.
are you an illuminati of these firms ?, here to mock my incompetence ?
If not, how can you possibly know what they were interested in ? You know what they asked, but a good interviewer does not use the exam style of "using induction, prove that the sum of integers to N is...."
My equivalent question would be to ask them to add up the numbers from 1 to 661, how they do it tells me more. I would never tell them to use lambda fns or get them to draw an upside down A. To get past me, they must show they have both a good "vocabulary", but also the wisdom to choose between them when confronted with a problem.
I might ask how to find the median of a data set in linear time. Failing to know what median or linear time are being of course a non maskable interrupt. I would not presume to tell them how to solve these problems, not least because a good % are smarter than me.
The only time a candidate gets told "I was interested in X" was when they fail in such a way that I counsel them on what to go away and learn.
>This is either an absurdly basic demand to ask of programmers, or an absurdly complex one.
Thank you.
As such it is thus a perfect interview question.
A good question has an easy entry level (like working out how to store a simple bitmap), bit allows more advanced candidates to demonstrate their ability.
>so absurdly difficult that it is unlikely you'll find more than a handful of people who just got a BS in CS will know
Those are the people I want, we don't do people who write forms.
>Expecting a CS student to be an expert in signal processing is rather silly.
We don't "expect" anything, but we actively seek out those who are wise in the ways of signal processing.
As it happens whilst I was an undergrad several of colleagues independantly realised that what we now call JPG style algorithms had legs. One did it in his first year.
If a CS grad hadn't done the relevant maths for JPG, then we'd move on, but if he failed the basic bitmap, the interview would end quickly, no second chance.
Certainly I expect them to understand information theory, and have some of the algorithms involved.
I used to ask how you would crack a one time pad. The answers were quite illuminating, and the best did not come from those who happened to know the answer.
I agree about choosing the right level, and that is why I have contempt for those who only think on one.
This means doing Lisp, ML or F#, and some smattering of the Algols, Perl, and assembler.
We observe that all CS grads who own an Apple computer fail to impress us, if we continue to bother interviewing CS grads at all, this will be an early question, such is it's predictive power.
More than one deluded fool has tried to explain to me how "innovative" a little firm like Apple was, citing how they invented GUIs and portable music players. They view choosing Apple over MS as some political statement of liberty. They flatly refuse to believe me when I tell them about Apple closed architectures (those who can understand what one is), or the way Apple has oppressed developers in ways that MS would never dream of.
I don't expect them to agree with me, but I expect their opinions to be better than the BBC.
>You have zero interest in hiring kids that are in the most demand? Yes, you have exactly understood what I am saying.
The mass market is for Java, a few years back it was VB. Do you think VB developers currently enjoy good job prospects ? Think Java is any different ?
But we don't do the mass market, we do the high end.
>Not to be an elitist, but you're a fucking headhunter.
Yes, I am. I also worked on the first Intel Unix, debugged operating system code for IBM & MS, have hand carved trading systems in C++, and for two years was the most expensive person in the pay of the British government whilst director in charge of the Treasury's secure wide area network for sucking "interesting" data out of banks.
You may have a more impressive techie CV than me, many people do, but I hope you will feel I have some credibility in this domain.
>Does your boss know you're turning them away? I think he'd be upset. Actually I own a large % of the firm, so do not have that fear:) Our clients like this a lot. Our selling point is we send out an order of magnitude fewer CVs than our competitors, more than one of which have sent so many CVs without any filtering that they overflow the 100Mb limit that a lot of corporate email systems impose upon personal inboxes, and in at least one case sent so many that the bank's spam filter binned the lot, ironically including a couple of CVs that the bank actually wanted. That is why we are harsh on CS grads, most are not only outside the elite, a large % I would never have hired when I had a proper job running development teams.
>If you actually knew anything about what makes someone wise in computers, you'd be the first I've met. Indeed, the fact that everyone at my firm has a CV that would enable them to get a respectable job in the area they headhunt for makes us successful. I accept that most HHs are crap, indeed I make the claim that I was the first person to use the term "pimp" for my adopted profession, back in 1986 in IBM's labs.
>Every headhunter I've met thought SQL was a database That's not true. When hearing "see-quel" some have confused it with "see-quent" the hardware firm:)
>and didn't know there was a difference between "C-pound" and C++. Actually the Microsoft search engine on their own site only worked that out relatively recently.
>You can't know unless you've been developing 8 hours every day after work. I've done may share of 60 and 80 hour development weeks.
>Because every day I work I am 8 hours more knowledgeable in my field than the day before. I flatly do not believe that. Unless you have no goal of actually producing something, you have spend some % of your time doing shit you have done before.
>I doubt you're keeping up with me. Possibly, I do less C++ these days, though I am currently teaching a course in it which keeps me on my toes. I've kept up with the intro of Lambda fns to C++ as well as FPGA/GPU coding, but I guess some people, perhaps including you are catching up and will overtake me.
My only claim is that I understand this stuff better than any other pimp you will ever meet.
We are always hiring, but it varies what we look for. For the largest banks we have "standing mandates" for very smart people who can do maths and programming exceptionally well. We do not do "housekeeping" IT like email, or programming for data entry tasks, which I suppose Java is OK for. Hard to see knocking up forms for data as a real good job, regardless of the employer's industry.
We are increasingly interested in computer game developers, since they can do C++, and GPUs actually have a niche in very high end banking, as do FPGAs.
Even in jobs where most work is done in some mix of Perl, VBA and Java, banks and hedge funds explicitly tell us they want C++ people because it is an easy way to prove they are better. A problem with dumb Java programmers is that if something is not in Java, there is no possibility they can know it. Thus lambda functions, FPGAs and a host of coo shit is outside their understanding. They have real problems with Excel. To be sure it is a trivial system to start with, but to be shit hot, you need to actually think at least as deeply as with C++, not least because V12 is multi threaded. One bank used threats to make us find a competent Excel developer, since nearly all Excel speicalists are shit. It was horrible. A large % had to have And() explained to them, 50% could not do VBA at all, many regarded DDE as a "new feature". A CS grad should be able to apply everything from lambda functions, through database theory, Hearsay/blackboard pattern to predicate calculus, to aceing Excel but they don't because they can't. You need to think from a richer set of ideas than the one you code in.
I am a headhunter for high end roles at investment banks, and we are close to classifying CompSci as a "non degree", along with media studies, languages, etc.
Java is fine for teaching design patterns, and classical algorithms like Quicksort, or binary search. But you can't do operating systems, and the success of Java in isolating you from any notion of the hardware is actually the problem. We have already blacklisted courses like the one at Kings College, because they teach operating systems in Java. Yes, really. Their reason apparently is that it is "easier". I have zero interest in kids who have studied "easy" subjects.
The world is a bigger, more competitive place, how many jobs do you think there are for people who have an easy time at college ?
Java is part of the dumbing down of CS. A computer "expert" is not someone who knows template metaprogramming in C++, or compiler archaeology in BCPL, or the vagaries of the Windows scheduler. It is someone who understands computers at multiple levels, allowing them to choose which one illuminates the problem at hand. To be wise in computers you choose whether to think of something as a block of bytes, quadwords, a bitmap, a picture, or a buffer overflow pretending to be porn. If also have the option of understanding flash vs static RAM, virtual memory, or networked storage, all the better. I doubt if even 1% of CS grads could write code to turn this BMP into a JPG, or even explain the ideas behind this. In my experience, 50% could not work out how to develop a data structure for a bitmap that used palettes. I have interviewed CS grads with apparently good grades who could not explain any data structure beyond arrays.
Any CS grad who sends us their CV with bullshit like "computers and society" or "web design" has their CV consigned to trash with no further reading. A CS should be able to write a web server, not be an arts graduate who didn't get laid.
C++ makes you think at multiple levels, unlike Java, you simply cannot avoid thinking about your system from patterns to bytes. This may be good or bad for productivity, and I'm sure we risk a flame war here. But I am entirely convinced you need to hack your way through a "real" system.
How can someone understand the Linux kernel without C & C++ ? Is someone really fit to be called a computer scientist if like >50% of the Computer "Scientists" we interview for very highly paid jobs, show actual fear of working at that level. They have the same "way above my head" attitude that a mediocre biologist might have to applying quantum theory to skin disease.
Partly, as in the Kings College debacle it is lazy mediocre lecturers, but also CompSci grads frankly are not that smart, so they need their hands held. Although the seats get filled, they quality is in monotonic decline.
First up we have a new variant on googlebombing. The filter will be gamed by content owners to pick up on anything they possibly can. This is because of the asymmetric costs. A false positive will cost them nothing, but the poster will get zapped. Indeed blockingd free content will serve the industry quite nicely.
There are >50 content formats, and new ones keep appearing. If the "standard" filter cannot read them, then the obvious thing to do is ban them. You've now established a monopoly where only "approved" formats are allowed. Even if it is an open standard, who writes the filter for new formats ? More importantly, who pays ?
It is also an arms race, and I think we can be clear that the "standard" filter will not be open source. DRM attracts crackers in direct proportion to it's success. Many crackers may not be fans of economics, but their goals are easily modelled in economic terms. They want to take out the "big beast" current filters are small, unsucessful critters. Cracking the industry standard media filter will be more of a coup than breaking WEP, and thus inevitably be swamped.
Also, an entertaining technical/legal point is so many site use Linux so the GPL may get involved.
We don't have any OCAML roles,though bizarrely one of the upper second tier banks has gotten into F#
As for the death of C++ style languages in multi-core systems, I've heard this before, and it simply ain't true. remember Prolog ? I wrote an implementation of that the second time FLPs were going to take over the world. I missed the first one in the 1970s. Even if I'm wrong I don't care:) I'm a pimp who used to be a techie, and if the banks want hardcore C++ that's what I'll find them, even if I have to interview more brain dead CS grads.
A knowledge of Black/Scholes/Merton is good, and to an extent assumed. Obviously no undergrad course except maybe economics does BSM, but you are expected to have read the classics like Wilmott, Hull and Taleb.
There is a subset of s/w dev in banks doing trading stuff. Very highly paid, and although the hours are a bit long they are shorter than I see in some sectors like video games. You need good maths, and be very good at programming. In my experience as someone who pimps these people into banks, almost no CS grads can program, indeed I've screened a number who were so ignorant I could not devise a question they could answer. One bank came to me after they'd interviewed 37 "chimps" (their ter)m and not one could demonstrate a grasp of programming that would betray them as different from an French literature graduate who would at least make the office look prettier.
Things that confuse CS grads: Order N Square is not a Tom Clancy Novel about a Russian Mole. Shannon's limit is not a village by a pretty river in Ireland Stack Frame is not a wrestler. You can't get high at a hash table Design Patterns is not a boutique, and threads are not what they sell. Iterators aren't the little evil robots in Stargate SG1, nor indeed are Heisenbugs. Not not is is the same as no not. Yes, really I've had to explan AND to some... The brighter ones manage to look amazed at the !! term in the C++ I write, the dim ones , well I don't know what goes on in their minds. Also a CS grad knows lisp. It does a little banking but not much. However if you can't do Lisp your not a computer scientist you are merely someone who like Star Trek : Enterprise who never got a data at college.
Also I have no bloody idea how you can call yourself a CS grad if you haven't taken apart an operating system and broken/hacked it. Hint for CS grads:operating systems aren't written in HTML, I checked, and Aristotle wasn't Belgian.
They are mostly in C & C++ (not Java, a quiche langage, and if you don't recognise the term quiche you need to do some reading)
As it happens C++ is the language of high end banking, as in people straight out of college what earn three what you do after a 3 years experience. But they're not CS grads. Maybe there are some smart ones, if so, they apply this intelligence in hiding from me as a headhunter.
You've actually been done a sevice by this screwup.
This is a clear example of bad faith and high handedness. God alone knows what they'll do to you once you sign up.
I'm a pimp for banks, and if one did this to a candidate, I would already be working my way up the management chain, what's yours doing ?
You need to find a different job at a better company, you can't trust these jerks.
A bit of naming and shaming would be a service to us all.
One thing that's worth remembering is that for a hiring manager recruitment is a tedious time consuming business that you don't want to repeat if you don't have to.
You may want to give your new boss one that is exactly one chance to sort this out.
But personally, I'd run, not walk away.
Robert Heinlein and Niven/Pournelle point out that a ship that takes generations to reach a habitable planet can become far more attractive than the destination itself.
Actually that leads me to a moderately good seed for a SF theme. Perhaps the only way to make the crew leave the ship is to sabotage is so that staying onbaord is not an option...
I quite happy to take chances:)
However, my statement was over quality of argument, not whether vulcanism matters, though I must say it's growing on me:)
Given your expertise, I would have expected a better argument, not an ad-hominem attack implying that I was sending disinformaiton. I'm playing devil's advocate since I despise any form of politcal corretness/censorship even if I happen to agree with the majority.
Seems you know a lot more about this than me, but again I am forced to question the relevance of carbon concentrations.
I'm pathetically ignorant about vulcanism compared to a specialist, but my understanding is that their impact upon temperatures is through aerosol effects, not carbon.
Sulphur oxides for instance :
http://en.wikipedia.org/wiki/Volcano
I don't think anyone disputes that carbon is on a close to monotoncally increasing curve, an that includes the stooges of the oil companies and evangelical Christians.
I think it is valid to argue over the exact causes however, and of course what (if anything) we do about it. I note that few "Greens" can get their head around the fact that whatever civilised countries do, the developing world will dump as much carbon in the atmosphere as they can.
Given you have a science degree, I assume you have an intuition about exponential curves ?
Using the graphs you cite, I ask you to model the effect (if any) of a complete ban tomorrow of all SUVs from US roads for all time.
Not much is it ?
My actual position, which I have not shared i this thread yet, is that not only do I believe in bad climate change, I do not believe it is credible we can do anything that meaningfully slows it down.
Since "Greens" are more interested in sounding good to their friends and getting on TV than those horrible "numbers" thing, the best we can hope for is that we knock together technologies that enable us to survive.
In my other post I bemoan the low quality of the arguments of the people who agree with me.
I said "in the last couple of centuries".
You talk of "recent observations", maybe they are right, but I don't see the relevance ?
If you are going to use the term "thoroughly investigated" then you should be able to do better than someone who has 18 seconds and access to Wikipedia. http://en.wikipedia.org/wiki/Year_Without_a_Summer
This was one Volcano (probably) within recent well documented history. The odds of a big one in the next decade ?
Probably low, and if you had said this rather than some silly claim that I was spreading misinformation, then I'd have agreed with you.
Your quality of argument is low, implying that you are right by accident not analysis.
Just because someone doesn't agree with your position, does not mean he is spreading misinformation. I was simply enumerating other possibilities.
I'm not saying that Exxon are the good guys, and certainly not defending their honesty.
I am not attacking the validity of the consensus on climate change. I am saddened by the often low quality arguments of the people who agree with me:(
>Anyway we've paid the scientists for the work, now we don't like the answer. It's not acceptable to demand they come up with a more convenient explanation.
That's exactly the attitude I have a problem with, although I accept their answer mostly.
But I expect some them to keep on digging, not just stop in case they find inconvenient facts.
If some of them think that their time is best used on other stuff then I don't disagree.
I just want there to be some smart people with integrity and resources permanently attacking all that we hold true.
Just because the oil companies are "sowing doubt" through dishonest motives, does not mean they don't have a point.
Consider Mendel's work on genetics. Turns out that some of his results are far too good to be true. Suspicion has long pointed at his assistants who may have lied about results.
They were dishonest, but as it happens the core work was right anyway.
Short version: The oils may be right by accident, but right by accident is still right.
How do you know what the "bad" weathermen think ?
As scientists we can only observe what they say and do. With humans that is only vaguely correlated with their internal thought processes.
I personally have argued for>positions that I completely disagree with, simply because the quality of "my own side"'s arguments was so poor. I can fake superstitions like Christianity so well that some people genuinely mistake me for one. It's to make people think, and yes they resent it sometimes.
You don't known whether I believe in human drive global warming or not do you ?
It is entirely appropriate to "shun" crap science, but that is not the same as taking away someone's career because he disagrees with the establishment. Such mindless intolerance and ignorant vindictiveness is why I characterise it as "religious".
It's not my argument, it's a possibility. One that's worth looking into.
Yes of course it's hard. That why we pay people smarter than me to look at this stuff.
>What should we focus on then given CO2 levels effect the heat trapping abilities of the atmosphere? Pixies?
We should not be focusing upon CO2 to the exclusion of all else. If you read my posts you will see that I accept the high probability that CO2 is the cause, and that it is mostly a bad thing. That's not the same as saying it is the only cause we should look at, and begs awkward questions on what we should do about it.
The problem with Gore-style Greens is that they want a solution that fist their overall agenda of being "inclusive" so they can feel part of a big social movement, and preferably one that manages to blame large corporations, preferably energy ones.
Fact is that no change in behaviour by oil companies is going to make any difference whatsoever. We are going to burn all the oil as fast as we can, and if Shell/BP/Exxon or whatever try to stop us, various governments will simply take away their toys. Not saying big oil are nice people, but that they simply don't have the power to fix anything.
Here's some number crunching for you to do at home.
Google the average growth in energy consumption, pick whichever figure fits your own prejudices.
Do the same for the % of global energy consumption used by SUVs or whatever use you least approve of.
Imagine you could pray to George Bush's God, and make that consumption disappear.
How many months does that buy us until we were back where we started from ?
Imagine you could make every car in America run by the force of prayer (your congress is proably voting research funds for that as we speak).
Fact is we're on an exponential curve or acceleration
Windmills, and vote buying in the form of bio fuels are like spitting against a forest fire, and in the case of bio fuels are like dumping yet more fuel on it.
Slashdot Mirror
>So from what discipline do you plan on recruiting people who have all the training that you so desperately need? Chemistry majors perhaps?
Physics, maths, and yes the occasional chemist, or at least chemical engineer.
>Perhaps I'm misreading you, but your tone seems dismissive, as if teaching design patterns and classical algorithms was somehow unimportant.
My apologies, I see them as critical, but I am dismissive of the idea that the language you choose matters all that much.
If Java, fine, but you can do it in VB (not that I advocate that)
>But just because someone knows English doesn't mean they'll understand Shakespeare.
Agreed.
But what would you think of someone with a degree in English who had not read any Shakespeare ?
My epxectation is that a graduate of a subject has a superior, deeper grasp of the subject.
To stretch the analogy perhaps too far, what would you think of an English major who had only read Shakespeare in cartoons ?
"because it is easier to teach"...
As it happens I have read and understood Shakespeare, and my early introduction to it was highly dumbed down.
(my dimwit politically correct English teachers dealt with the anti semitism in the Merchant of Venice by ignoring it).
But I did read it in the original later, indeed one is expected to have read a large % of Shakespeare *before* starting an English degree.
Chaucer is rather harder going.
Yes, by all means use any language you like to do the algorithms and structures necessary for kernel work. I first did threads in a horrible mutant Pascal, and actually found them easier to understand in C. We did O/S stuff in pseudocode, ut uitlately we did real code in a real system.
Perhaps a better analogy than English is medicine. Most doctors are not cardiac specialists, and would be as hesitant to do anything to the heart as most programmers to hack a kernel. But all doctors, even dentists know at least roughly how to spot when the cause might be the heart, or when the odds are it is something else.
It goes back to my themes of a good techie being able to think at multiple levels to be effective.
>Why would anyone need to understand the linux kernel or C++ unless they were linux kernel programmers. Same reason wannabe doctors cut up dead bodies. >Also your comments on Java are laughable. Java is in use in day to day objects in real life. Phones, Set top boxes, Cars. It is not just web servers. There are even >operating systems and emulators in Java which run fine. Go read up on JIT and stop living in the 90's. OK, I'll come clean I have no problem with Java, it's Java *developers* I despise, when they can't be bothered to read my posts, and take anything which does not sing its praises as attacks on their manhood. I really really carefully said, I had no problem with Java for many sorts of work. I explicitly agreed later that there were many jobs in it, and will go further and say that it is the right choice for many things. Just in case my words were read by over-sensitive Java developers my post said that the success of Java in isolating you from low level detail was an issue. I think you need to be able to think at multiple levels, even if you only code at one. Interview questions require a finite tractable answer in a relatively short amount of time. As it happens I don't ask about bitmaps, but it strikes me as a suitable for part of evaluating an entry level developer. Nicely open ended for a superior candidate to shine. You seem to be under the impression that a good developer knows how to do certain things. We part company on that. A good developer can work out how to do things. The logic and coding issues behind bitmaps has applications in high end finance, but I don't expect a newbie to even be able guess what they might be. I do expect them to deal with a problem they have not seen before. I don't give a toss if they understand the genealoogy of OS/2, to Windows bitmap formats, or their relationship to VGA palettes or the transputer. I do care that they can think up a way to store data to a spec I give them.
I've had the Azure stuff demoed to me, and it looks interesting, and they coped well with the idea of a headhunter having an interest in their stuff.
seemed a little pricey for what it was though, does you experience justify the cost ?
>Perhaps you should start with finding applicants who have a decent command of the English language, Nope. The thought makes me feel mildly ill. I have explicit instructions from some employers that they don't care about verbal skills all that much because they are sick of getting preppies with people skills who take personal offence when the compiler criticises their code. There are mid and low end jobs knocking up GUIs or importing data that require good people skills. We don't hire for those. The trend is that by the end of 2008, we expect that people we work with who don't have English as a first language will be the majority. We like people skills, but it's not in the top 5, and we will pick good maths and C++ over a nice smile every time. >and the proven ability to interact with customers, management, and a team of programmers? Newbies don't have that, and frankly I score down people whose CVs talk too much of "customer contact", indeed in our 140 page guide to getting a fuck off job in a bank, we tell people to scale back holiday jobs and customer skills to zero, or a vestige. We do like leading men under fire, and yes that includes people who have shot at friends of mine. >Ensure they not only have the ability to communicate coherent thoughts in person, on the phone, in conference, or on a whiteboard. We like articulate people. A lot. Even if the English is broken we are implacable in our demand for clear thinking. I can't use whiteboards, stopped altogether when a minion pointed out that I always drew the same horrible shape of badly drawn concentric rings with random bent radial lines. (I think it is a message from the Iterators) >Additionally, test them for their ability to adapt to the dynamic environment that is Corporate America. If a candidate claimed such an ability, the time taken to end the interview would be determined by my rather English politeness, not any desire to hear them bullshit any more. Again this is something we explicitly warn against in our guide, the phrase being "do not sound like you work for Accenture". We recruit across for roles across the globe, we don't do "Corporate America", having done stints in such things I am glad I can choose never to do so again. We do American banks, and if I recommended a candidate on that basis, not only would they not be interviewed, I would lose some of my hard won credibility. (tough to do when you are pimp). >I could walk outside right now, and find myself umpteen code monkeys to write EXACTLY what I wish them to write, You are a lucky man to have found a job that pays you without setting the sort of demands people in my business fight. >as opposed to the creative dialogues I seek from my team members. I like creativity, tough to find, tougher to build. Candidates get big bonus points for this. Most don't get any. >I want people to be enthusiastic about coding, their product, and their team. Don't we all ? >P.S. I'm not sure, but describing yourself as a "pimp" gives me pause as to your ability. You are right, if I were dealing with "corporate America", it would be insane. I deal with investment banks, hedge funds, ratings agencies and the occasional government. The mindset in my niche is quite different. Not necessarily true, but that is my [business] perception.
Oh dear, an Apple fanboy.
My issue with your rant is not whether you are right or wrong, but that the quality of the argument is so low.
I did not give any mechanism for the correlation between low programming skills and Apple ownership, and to be honest I don't know why it is so high.
For all I know, it may be that the stability of the system does not attract serious programmers. For most of my life a large % of the s/w I have used has been beta, homebrew, or just terribly buggy.
>Expecting people interviewing to join an investment bank to be kernel hackers is pointless.
You are confusing the housekeeping activities which are much the same at Goldman Sachs and Wal Mart and very high end work. That's not your fault, if everyone knew the shit I do, then they'd make my money.
>Expecting them to have good signal processing skills is stupid.
You are wrong. The highest paid people we see taken in by banks and hedge funds know SP. By "highest paid" I mean bidding wars between the top investment banks.
You aren't in that category are you ?
>I do however know how to develop exceedingly high quality business software in extraordinarily short timescales
So you are a back office developer. Fine. The majority of bank developers are like you, they earn OK money.
We don't handle people like you, unless they've accidentally ended up in a job far below their capabilities.
We seek out a narrow speciality combination of skills, and there's no shame or despair to be felt that you don't fit our filter.
You're not a vet either, fine.
We do however recruit the people who will be giving you the "fluffy" specifications. You may or may not like that.
>So, what, you're going to hire math geeks only? No, physicists, electronic engineers, occasional economists. I would love to recommend more CS grads, but the quality is too low. To be sure, I don't expect CS courses to prep for my speciality, but I despair of grads who don't know O(N), or only speak one programming subject. I despise essays. >If you're about to decree CS as a non degree, maybe you should get the input of the people who will be most brutally affected by your shortsightedness. I'm sure I said that *I* am a CS grad, and have shared my view about them with my peers. Their reaction is typically to try and outbid me on witless newbies. As for algorithms, I would ask you to read my post. I am entirely comfortable with Java as a language for patterns and algorithms. I said this explicitly. If you are going to call me on this, at least do me the courtesy of reading what I say. Same applies for explaining buffering in Python. I learned this stuff in pure pseudocode, not C/C++, bu then I got to play with a real O/S. If my earlier post is too hard for you to read, I will summarise it as that I do not have problem with Java, I have a problem with *only* Java. >If it was an easy subject, would changes need to be made to make it easier? CS is not inherently easy. But you can choose to teach the easy bits, many do. I see the results, and they are a perverse combination of horrifyingly dull. >The point of a CS degree is to allow you to learn with phenomenal velocity--to know where to look for answers, to know how to read the papers, to know how to integrate all that into your noggin very quickly. Agreed, where we differ is that this is an ideal that is not only in decline but is in serious danger of disappearing altogether. You are confusing what I say about the "usefulness" of C++ in *exactly* the way I said it was not to be confused. (read my post). I can be convinced that Java, Perl, VBA, SQL, or force of mental will staring at the screen is better than C++ for a given task. That is not my point, and if you had read my post, you will have seen that. As a HH, I care about C++ because my clients do, and agree it is often used when not the best tool. But in my experience I see a lot more of Java inappropriately chosen because of the poor breadth of the decision makers. My point is that if a CS course did (say)Haskell, Perl and Java, I would not have a problem. But choosing only to do Java because the others are are hard is not the way to my heart, or a top flight job. By "hard" you should understand I do not mean "awkward" or "tricky" but conceptually challenging. Some of C++ falls into the first two categories, some doesn't, but no single language is rich enough to educate a computer scientist well. C++ is the best in my opinion, but you'd learn more from doing Java and Haskell than by just C++
>How good ?
PDEs are a minimum, and I'd see you as a potential quant developer so that would mean finite difference as well as monte carlo.
More stats would be nice.
I do not think the set of language matters so much as their breadth. I am indifferent to Ada, maybe good or bad, but alone no single language allows a good CS student to get to where he wants to be. C++ has more concepts in it than Java, which is of course one reason it is harder to learn, so gets points from me for that. Lisp is actually a very simple language, but being so different from Ada, C++ and Java, it helps the student learn to think about things in different ways. Of course some never get their heads around Lisp, which is fine since it has almost no commercial value, but it should be sending a signal that this person simply isn't good at programming. Not saying they shouldbe shit hot at functional languages, but failure is worrying. CS courses seem to be set up to make them as hard as possible to fail, yet programming is a talent. Some very smart people just can't do it. Thus allowing them to write essays, draw pretty graphics, and use Java toolkits to make applications is fraudulent.
The first page of googles on my name are all me :)
/at\ PaulDominic dot com, bung us your CV, and I'll get our 140 page guide to you. It's tough.
A great improvement from a couple of years ago, when since my name is "catholic" I was buried underneath a list of various child rapists in the Catholic church.
I'm Dominic
There is one large firm who may be interested who have offices in London and NY, they say they want C++ but are lying, since their interviews are heavily K&R C.
Java backends are rare in our game, and we don't get involved at all, though we have a frankly demented project for back end servers running Excel VBA.
(mad, bad and expensive to know).
There is not a great demand for C, though it pops up after C++, VBA, C# but above Java.
We are more flexible than most on branding of university, since we work with people from everywhere ("everywhere" including a few people from places not recognised as states at all).
>You must realise that many of the great computer programmers started doing it as children :)
I started at 12 on a computer stolen from the government
>high stakes is not done by magical "better" programmers or harder whipping.
Thank you for pointing out something I should have said, and with which I agree 100%.
High stakes financial programming is rarely done by the best programmers, indeed I have seen stuff handling real money that would shame a Kings College CompSci graduate.
We've done a 140 page guide to our line of work, and the effects of EDS on banking IT is better imagined than experienced.
We want very good maths as well, and that closes the filter so tight that one has to trade off.
>is done with boring, systematic testing
Actually testing does not have to be boring, and is so only because it is usally very badly managed, and these days often staffed by people who didn't make it as a mainline programmer. Much can be automated, and if you have the right culture (rare) the finder of a complex bug is a hero, not (as usually) a troublemaker.
>If your code is processing real money, and it processes the equivalent of your yearly salary in a few minutes or hours, you are "way above your head"
You and I operate in different domains, and in mine the relevant unit measurement is Hertz.
Our people write code that handles their annual salary ranging from 0.1 Hz to some who work at the kilohertz level, with higher peaks.
In high frequency algorithmic trading one's code may easily process megabucks/second.
Sadly I am under an obligation not share names, but there is (exactly) one system that handles G7 country sized money in real time high frequency trades written almost wholly in C#. Entertainingly, I kind of believe that Microsoft doesn't actually know about it, even though it is a very glamorous project.
But everything else is done in C++, with a few C things lurking in dark alleys, preying upon the weak.
are you an illuminati of these firms ?, here to mock my incompetence ?
If not, how can you possibly know what they were interested in ?
You know what they asked, but a good interviewer does not use the exam style of
"using induction, prove that the sum of integers to N is...."
My equivalent question would be to ask them to add up the numbers from 1 to 661, how they do it tells me more.
I would never tell them to use lambda fns or get them to draw an upside down A. To get past me, they must show they have both a good "vocabulary", but also the wisdom to choose between them when confronted with a problem.
I might ask how to find the median of a data set in linear time.
Failing to know what median or linear time are being of course a non maskable interrupt.
I would not presume to tell them how to solve these problems, not least because a good % are smarter than me.
The only time a candidate gets told "I was interested in X" was when they fail in such a way that I counsel them on what to go away and learn.
>This is either an absurdly basic demand to ask of programmers, or an absurdly complex one. Thank you. As such it is thus a perfect interview question. A good question has an easy entry level (like working out how to store a simple bitmap), bit allows more advanced candidates to demonstrate their ability. >so absurdly difficult that it is unlikely you'll find more than a handful of people who just got a BS in CS will know Those are the people I want, we don't do people who write forms. >Expecting a CS student to be an expert in signal processing is rather silly. We don't "expect" anything, but we actively seek out those who are wise in the ways of signal processing. As it happens whilst I was an undergrad several of colleagues independantly realised that what we now call JPG style algorithms had legs. One did it in his first year. If a CS grad hadn't done the relevant maths for JPG, then we'd move on, but if he failed the basic bitmap, the interview would end quickly, no second chance. Certainly I expect them to understand information theory, and have some of the algorithms involved. I used to ask how you would crack a one time pad. The answers were quite illuminating, and the best did not come from those who happened to know the answer. I agree about choosing the right level, and that is why I have contempt for those who only think on one. This means doing Lisp, ML or F#, and some smattering of the Algols, Perl, and assembler. We observe that all CS grads who own an Apple computer fail to impress us, if we continue to bother interviewing CS grads at all, this will be an early question, such is it's predictive power. More than one deluded fool has tried to explain to me how "innovative" a little firm like Apple was, citing how they invented GUIs and portable music players. They view choosing Apple over MS as some political statement of liberty. They flatly refuse to believe me when I tell them about Apple closed architectures (those who can understand what one is), or the way Apple has oppressed developers in ways that MS would never dream of. I don't expect them to agree with me, but I expect their opinions to be better than the BBC.
>You have zero interest in hiring kids that are in the most demand?
:)
:)
Yes, you have exactly understood what I am saying.
The mass market is for Java, a few years back it was VB.
Do you think VB developers currently enjoy good job prospects ? Think Java is any different ?
But we don't do the mass market, we do the high end.
>Not to be an elitist, but you're a fucking headhunter.
Yes, I am. I also worked on the first Intel Unix, debugged operating system code for IBM & MS, have hand carved trading systems in C++, and for two years was the most expensive person in the pay of the British government whilst director in charge of the Treasury's secure wide area network for sucking "interesting" data out of banks.
You may have a more impressive techie CV than me, many people do, but I hope you will feel I have some credibility in this domain.
>Does your boss know you're turning them away? I think he'd be upset.
Actually I own a large % of the firm, so do not have that fear
Our clients like this a lot.
Our selling point is we send out an order of magnitude fewer CVs than our competitors, more than one of which have sent so many CVs without any filtering that they overflow the 100Mb limit that a lot of corporate email systems impose upon personal inboxes, and in at least one case sent so many that the bank's spam filter binned the lot, ironically including a couple of CVs that the bank actually wanted.
That is why we are harsh on CS grads, most are not only outside the elite, a large % I would never have hired when I had a proper job running development teams.
>If you actually knew anything about what makes someone wise in computers, you'd be the first I've met.
Indeed, the fact that everyone at my firm has a CV that would enable them to get a respectable job in the area they headhunt for makes us successful.
I accept that most HHs are crap, indeed I make the claim that I was the first person to use the term "pimp" for my adopted profession, back in 1986 in IBM's labs.
>Every headhunter I've met thought SQL was a database
That's not true.
When hearing "see-quel" some have confused it with "see-quent" the hardware firm
>and didn't know there was a difference between "C-pound" and C++.
Actually the Microsoft search engine on their own site only worked that out relatively recently.
>You can't know unless you've been developing 8 hours every day after work.
I've done may share of 60 and 80 hour development weeks.
>Because every day I work I am 8 hours more knowledgeable in my field than the day before.
I flatly do not believe that.
Unless you have no goal of actually producing something, you have spend some % of your time doing shit you have done before.
>I doubt you're keeping up with me.
Possibly, I do less C++ these days, though I am currently teaching a course in it which keeps me on my toes.
I've kept up with the intro of Lambda fns to C++ as well as FPGA/GPU coding, but I guess some people, perhaps including you are catching up and will overtake me.
My only claim is that I understand this stuff better than any other pimp you will ever meet.
We are always hiring, but it varies what we look for.
For the largest banks we have "standing mandates" for very smart people who can do maths and programming exceptionally well.
We do not do "housekeeping" IT like email, or programming for data entry tasks, which I suppose Java is OK for.
Hard to see knocking up forms for data as a real good job, regardless of the employer's industry.
We are increasingly interested in computer game developers, since they can do C++, and GPUs actually have a niche in very high end banking, as do FPGAs.
Even in jobs where most work is done in some mix of Perl, VBA and Java, banks and hedge funds explicitly tell us they want C++ people because it is an easy way to prove they are better.
A problem with dumb Java programmers is that if something is not in Java, there is no possibility they can know it. Thus lambda functions, FPGAs and a host of coo shit is outside their understanding. They have real problems with Excel. To be sure it is a trivial system to start with, but to be shit hot, you need to actually think at least as deeply as with C++, not least because V12 is multi threaded. One bank used threats to make us find a competent Excel developer, since nearly all Excel speicalists are shit. It was horrible. A large % had to have And() explained to them, 50% could not do VBA at all, many regarded DDE as a "new feature".
A CS grad should be able to apply everything from lambda functions, through database theory, Hearsay/blackboard pattern to predicate calculus, to aceing Excel but they don't because they can't.
You need to think from a richer set of ideas than the one you code in.
I am a headhunter for high end roles at investment banks, and we are close to classifying CompSci as a "non degree", along with media studies, languages, etc.
Java is fine for teaching design patterns, and classical algorithms like Quicksort, or binary search.
But you can't do operating systems, and the success of Java in isolating you from any notion of the hardware is actually the problem.
We have already blacklisted courses like the one at Kings College, because they teach operating systems in Java.
Yes, really.
Their reason apparently is that it is "easier".
I have zero interest in kids who have studied "easy" subjects.
The world is a bigger, more competitive place, how many jobs do you think there are for people who have an easy time at college ?
Java is part of the dumbing down of CS.
A computer "expert" is not someone who knows template metaprogramming in C++, or compiler archaeology in BCPL, or the vagaries of the Windows scheduler.
It is someone who understands computers at multiple levels, allowing them to choose which one illuminates the problem at hand.
To be wise in computers you choose whether to think of something as a block of bytes, quadwords, a bitmap, a picture, or a buffer overflow pretending to be porn. If also have the option of understanding flash vs static RAM, virtual memory, or networked storage, all the better. I doubt if even 1% of CS grads could write code to turn this BMP into a JPG, or even explain the ideas behind this. In my experience, 50% could not work out how to develop a data structure for a bitmap that used palettes.
I have interviewed CS grads with apparently good grades who could not explain any data structure beyond arrays.
Any CS grad who sends us their CV with bullshit like "computers and society" or "web design" has their CV consigned to trash with no further reading.
A CS should be able to write a web server, not be an arts graduate who didn't get laid.
C++ makes you think at multiple levels, unlike Java, you simply cannot avoid thinking about your system from patterns to bytes. This may be good or bad for productivity, and I'm sure we risk a flame war here.
But I am entirely convinced you need to hack your way through a "real" system.
How can someone understand the Linux kernel without C & C++ ?
Is someone really fit to be called a computer scientist if like >50% of the Computer "Scientists" we interview for very highly paid jobs, show actual fear of working at that level.
They have the same "way above my head" attitude that a mediocre biologist might have to applying quantum theory to skin disease.
Partly, as in the Kings College debacle it is lazy mediocre lecturers, but also CompSci grads frankly are not that smart, so they need their hands held.
Although the seats get filled, they quality is in monotonic decline.
First up we have a new variant on googlebombing. The filter will be gamed by content owners to pick up on anything they possibly can.
This is because of the asymmetric costs. A false positive will cost them nothing, but the poster will get zapped. Indeed blockingd free content will serve the industry quite nicely.
There are >50 content formats, and new ones keep appearing. If the "standard" filter cannot read them, then the obvious thing to do is ban them.
You've now established a monopoly where only "approved" formats are allowed.
Even if it is an open standard, who writes the filter for new formats ? More importantly, who pays ?
It is also an arms race, and I think we can be clear that the "standard" filter will not be open source.
DRM attracts crackers in direct proportion to it's success. Many crackers may not be fans of economics, but their goals are easily modelled in economic terms.
They want to take out the "big beast" current filters are small, unsucessful critters.
Cracking the industry standard media filter will be more of a coup than breaking WEP, and thus inevitably be swamped.
Also, an entertaining technical/legal point is so many site use Linux so the GPL may get involved.
We don't have any OCAML roles ,though bizarrely one of the upper second tier banks has gotten into F#
:) I'm a pimp who used to be a techie, and if the banks want hardcore C++ that's what I'll find them, even if I have to interview more brain dead CS grads.
As for the death of C++ style languages in multi-core systems, I've heard this before, and it simply ain't true.
remember Prolog ? I wrote an implementation of that the second time FLPs were going to take over the world. I missed the first one in the 1970s.
Even if I'm wrong I don't care
A knowledge of Black/Scholes/Merton is good, and to an extent assumed. Obviously no undergrad course except maybe economics does BSM, but you are expected to have read the classics like Wilmott, Hull and Taleb.
There is a subset of s/w dev in banks doing trading stuff. Very highly paid, and although the hours are a bit long they are shorter than I see in some sectors like video games. You need good maths, and be very good at programming. In my experience as someone who pimps these people into banks, almost no CS grads can program, indeed I've screened a number who were so ignorant I could not devise a question they could answer. One bank came to me after they'd interviewed 37 "chimps" (their ter)m and not one could demonstrate a grasp of programming that would betray them as different from an French literature graduate who would at least make the office look prettier.
Things that confuse CS grads:
Order N Square is not a Tom Clancy Novel about a Russian Mole.
Shannon's limit is not a village by a pretty river in Ireland
Stack Frame is not a wrestler.
You can't get high at a hash table
Design Patterns is not a boutique, and threads are not what they sell.
Iterators aren't the little evil robots in Stargate SG1, nor indeed are Heisenbugs.
Not not is is the same as no not. Yes, really I've had to explan AND to some...
The brighter ones manage to look amazed at the !! term in the C++ I write, the dim ones , well I don't know what goes on in their minds.
Also a CS grad knows lisp. It does a little banking but not much. However if you can't do Lisp your not a computer scientist you
are merely someone who like Star Trek : Enterprise who never got a data at college.
Also I have no bloody idea how you can call yourself a CS grad if you haven't taken apart an operating system and broken/hacked it.
Hint for CS grads:operating systems aren't written in HTML, I checked, and Aristotle wasn't Belgian.
They are mostly in C & C++ (not Java, a quiche langage, and if you don't recognise the term quiche you need to do some reading)
As it happens C++ is the language of high end banking, as in people straight out of college what earn three what you do after a 3 years experience.
But they're not CS grads. Maybe there are some smart ones, if so, they apply this intelligence in hiding from me as a headhunter.
You've actually been done a sevice by this screwup.
This is a clear example of bad faith and high handedness. God alone knows what they'll do to you once you sign up.
I'm a pimp for banks, and if one did this to a candidate, I would already be working my way up the management chain, what's yours doing ?
You need to find a different job at a better company, you can't trust these jerks. A bit of naming and shaming would be a service to us all. One thing that's worth remembering is that for a hiring manager recruitment is a tedious time consuming business that you don't want to repeat if you don't have to.
You may want to give your new boss one that is exactly one chance to sort this out.
But personally, I'd run, not walk away.
Robert Heinlein and Niven/Pournelle point out that a ship that takes generations to reach a habitable planet can become far more attractive than the destination itself.
Actually that leads me to a moderately good seed for a SF theme. Perhaps the only way to make the crew leave the ship is to sabotage is so that staying onbaord is not an option...
I quite happy to take chances :)
However, my statement was over quality of argument, not whether vulcanism matters, though I must say it's growing on me :)
Given your expertise, I would have expected a better argument, not an ad-hominem attack implying that I was sending disinformaiton. I'm playing devil's advocate since I despise any form of politcal corretness/censorship even if I happen to agree with the majority.
Seems you know a lot more about this than me, but again I am forced to question the relevance of carbon concentrations. I'm pathetically ignorant about vulcanism compared to a specialist, but my understanding is that their impact upon temperatures is through aerosol effects, not carbon.
Sulphur oxides for instance : http://en.wikipedia.org/wiki/Volcano I don't think anyone disputes that carbon is on a close to monotoncally increasing curve, an that includes the stooges of the oil companies and evangelical Christians.
I think it is valid to argue over the exact causes however, and of course what (if anything) we do about it. I note that few "Greens" can get their head around the fact that whatever civilised countries do, the developing world will dump as much carbon in the atmosphere as they can.
Given you have a science degree, I assume you have an intuition about exponential curves ?
Using the graphs you cite, I ask you to model the effect (if any) of a complete ban tomorrow of all SUVs from US roads for all time.
Not much is it ?
My actual position, which I have not shared i this thread yet, is that not only do I believe in bad climate change, I do not believe it is credible we can do anything that meaningfully slows it down.
Since "Greens" are more interested in sounding good to their friends and getting on TV than those horrible "numbers" thing, the best we can hope for is that we knock together technologies that enable us to survive.
In my other post I bemoan the low quality of the arguments of the people who agree with me.r
This was one Volcano (probably) within recent well documented history. The odds of a big one in the next decade ?
Probably low, and if you had said this rather than some silly claim that I was spreading misinformation, then I'd have agreed with you.
I said "in the last couple of centuries". You talk of "recent observations", maybe they are right, but I don't see the relevance ?
If you are going to use the term "thoroughly investigated" then you should be able to do better than someone who has 18 seconds and access to Wikipedia.
http://en.wikipedia.org/wiki/Year_Without_a_Summe
Your quality of argument is low, implying that you are right by accident not analysis. Just because someone doesn't agree with your position, does not mean he is spreading misinformation. I was simply enumerating other possibilities.
I'm not saying that Exxon are the good guys, and certainly not defending their honesty. :(
I am not attacking the validity of the consensus on climate change. I am saddened by the often low quality arguments of the people who agree with me
>Anyway we've paid the scientists for the work, now we don't like the answer. It's not acceptable to demand they come up with a more convenient explanation.
That's exactly the attitude I have a problem with, although I accept their answer mostly.
But I expect some them to keep on digging, not just stop in case they find inconvenient facts.
If some of them think that their time is best used on other stuff then I don't disagree.
I just want there to be some smart people with integrity and resources permanently attacking all that we hold true.
Just because the oil companies are "sowing doubt" through dishonest motives, does not mean they don't have a point.
Consider Mendel's work on genetics. Turns out that some of his results are far too good to be true. Suspicion has long pointed at his assistants who may have lied about results.
They were dishonest, but as it happens the core work was right anyway.
Short version: The oils may be right by accident, but right by accident is still right.
How do you know what the "bad" weathermen think ?
As scientists we can only observe what they say and do. With humans that is only vaguely correlated with their internal thought processes.
I personally have argued for>positions that I completely disagree with, simply because the quality of "my own side"'s arguments was so poor. I can fake superstitions like Christianity so well that some people genuinely mistake me for one. It's to make people think, and yes they resent it sometimes. You don't known whether I believe in human drive global warming or not do you ?
It is entirely appropriate to "shun" crap science, but that is not the same as taking away someone's career because he disagrees with the establishment. Such mindless intolerance and ignorant vindictiveness is why I characterise it as "religious".
It's not my argument, it's a possibility. One that's worth looking into.
Yes of course it's hard. That why we pay people smarter than me to look at this stuff.
>What should we focus on then given CO2 levels effect the heat trapping abilities of the atmosphere? Pixies? We should not be focusing upon CO2 to the exclusion of all else. If you read my posts you will see that I accept the high probability that CO2 is the cause, and that it is mostly a bad thing. That's not the same as saying it is the only cause we should look at, and begs awkward questions on what we should do about it. The problem with Gore-style Greens is that they want a solution that fist their overall agenda of being "inclusive" so they can feel part of a big social movement, and preferably one that manages to blame large corporations, preferably energy ones.
Fact is that no change in behaviour by oil companies is going to make any difference whatsoever. We are going to burn all the oil as fast as we can, and if Shell/BP/Exxon or whatever try to stop us, various governments will simply take away their toys. Not saying big oil are nice people, but that they simply don't have the power to fix anything. Here's some number crunching for you to do at home. Google the average growth in energy consumption, pick whichever figure fits your own prejudices. Do the same for the % of global energy consumption used by SUVs or whatever use you least approve of. Imagine you could pray to George Bush's God, and make that consumption disappear.
How many months does that buy us until we were back where we started from ? Imagine you could make every car in America run by the force of prayer (your congress is proably voting research funds for that as we speak).
Fact is we're on an exponential curve or acceleration
Windmills, and vote buying in the form of bio fuels are like spitting against a forest fire, and in the case of bio fuels are like dumping yet more fuel on it.