Almost any expansion onto other planets that we can easily forsee is just that, an expansion, not a migration.
I suspect that we've read many SF books in common, but without teleportation or Warp Drive, there is no way that more than 1,000 people a year will ever leave Earth orbit.
Indeed, it's not clear to me how you can get a million people into space per year without really quite impressive environmental impact.
There's nowhere to go where we can live without exotic engineering, so the vast bulk of our descendants for the next few millennia are going to be on Earth.
A self sufficent Mars colony, based upon what we know is centuries away, and unless it is self sufficient buys the race nothing in terms of survival.
Even then, many of the big scary things we know about (novas, various large scale radiation events, perturbations of orbits) would make them more vulnerable than us.
Burning crap has it's own environmental problems...
Indeed for me the reson to move away from Carbon has always been based upon the other polluting effects, and of course the depletion of cheap crap to burn.
But the "burning crap" position is essentially where we are going to need to be. The Eath is not "naturally" habitable by humans, or even comfortable for us.
Over the long term we are going to need to manage it like some sort of garden if we are to survive. If we look at those subsets of humanity that live "in tune with nature", not only are their individual lives short, but these groups have a tendency to suddenly be wiped out when nature changes her mind.
I'm not disagreeing with your basic point, but in the case of Solar cycles, by the time we've reduced our nice warm blanket of greenhouse gases, the cycle may be on a down turn, when we've just spent trillions making the Earth colder. Ooops.
It's easy to suspect that fertiliser and pesticide run offs are altering which microbes prosper and thus affect climate. Bio-fuels require yet more intensive agriculture, and so may make the problem worse, or not. Again a focus on CO2 may be doing more harm than good.
Also, there is the random "background noise" of vulcanism. This can be a big term. In the last couple of centuries we've seen at least one event that lost our planeet an entire summer, and for a while entirely overwhelmed any possible human effect. A few big volcanoes randomly going off could make the most hysterical predicitons of arts-graduate Greens look like a rainy afternoon.
The nature of climate variation affects deeply how we should respond. We can't do much about volcanoes, except maybe set one off as a frantic reaction to short term global warming:) If the forces are outside our control, like Solar, or Chinese industralisation then the optimum may be to make it less bad and not spend money on trying to slow it down. I'm not saying this is the case, I'm saying that a religious belief in a single monolithic truth is extravagantly unlikely to deliver the best choices.
That's a valid point.
Holocaust denyers, climate change rubbishers , creationists et al do us a service by making themselves easy to spot.
We risk falling into Al Gore's stupidity of "facts are a kind of pollution" if we censor those who don't emit politically correct statements.
To me, the whole point of tax funded science is to find new and interesting ways of showing that what most people believe is wrong. If it's nice warm fuzzy confirmations of our world view, then it's more suitable for the Discovery Channel or a corporate R&D lab.
I guess the odds are 99% that it's humans that are causing all this stuff we see.
But what of the other 1% ?
What if we spend some vast amount of money changing the global economy to find that it's a subtle change in the Sun's output, a random downturn in Vulcanism, or somehow to do with deep sea bacteria ?
I don't believe any of those ideas much, but I don't 100% believe they are wrong either.
America is now paying the price of having the highest relgious observance in the developed world. The damage is not the general superstious attack on evolution, or even the rape of thousands of children by priests, but a more insidious view that there is a SINGLE TRUTH, and EVERYONE WHO SAYS ANYTHING ELSE IS EVIL.
Such is the invasive nature of religion that even poeple with scientific training are prey to the condition.
Part of our filter is checking people understand English.
You failed.
I explictly said we don't care about this stuff.
As it happens, although the field is only about 3% female, 17% of the people we put forward in 2006 were female.
I'm not even going to claim this as virtue, just applied greed. Discrimination would cost us money, since we'd place fewer people, so we don't do it.
It might not even be anything to do with us since our research implies more women are trying to get in.
I'm a headhunter, and it's not always easy to give feedback.
We do high end programming jobs for banks, and screen candidates before they go in.
Some are under the deluded impression that if they argue forcefully that they can undo a poor assessment of their suitablity.
A common position is "I can learn that", which may or may not be true. I tend to suggest the book they can read, but I'm still not sending them in.
It can get very tiresome to politely say "but you didn't know about iterators", and be loudly told that this doesn't matter, especially if the candidate perceives that if he keeps this up long enough we'll give in.
We don't insist on Jeeves like smoothness from candidates, but some are simply inappropriate for any job in a bank to a degree that it's a mystery how they got hired for their current role, but not why they are leaving. People who say "that's a stupid question", or give the impression that their current boss is a flaky criminal mastermind don't score points here.
Also there is the threat of being sued. Our equal opportunity policy is basically that we do this for the money, and don't give a toss about their skin colour, sex age, or which side they fought on in the Gulf War. But some people are just not good enough, or more ambiguously not "right". Sometimes that is stuff that's hard to specifically form "he has no experience of X", especially since you usually don't meet them unless they have such terms on their CV
I've been counselled not to use terms that imply a preference for a given age, and work in an field that is 97% male, so it's natural, but dangerous for me to use terms that imply I only want men. Most managers can't do this with the reliability of someone who does this every day, and so risk getting sued, so often HR departments mandate that they don't say anything, and especially not in an email.
Also feedback from clients is a sensitive issue. We occasionally get quite robust feedback about the failings of a given candidate, and we're certainly not going to get a client into legal shit. Occasionally the hiring manager will evaluate a candidate entirely unfairly and incorrectly. As it happens I'm a bit of a techie myself, and so it's a bit of a challenge to put a good face on bizarre and useless interview questions, and yes that includes the cases where the interviewer's "correct answer" was wrong.
We do better feedback than other pimps, but honestly, none of us make it all the way to good.
So ultimately, you're not going to get quality feedback. What you need to do is find some friend who's done some hiring, buy him a beer or dinner and get him to check you out.
I agree it is a good question, but I'd spin it slightly.
I'd ask him *how* to doctor a screen, and how trivial it would be to fiddle records that showed
the defendant had a given IP address.
It would take very little time for a competent person to do this, indeed to ridicule the RIAA position,I could take a couple of days with an average 10 year old would leave them able to do this, a smart 8 year old could do it in a morning.
Ask him if he's conducted a review of ISP logging s/w, as in read the source code, not as in sent an email asking if it was "OK". Would bet good money he hasn't.
Actually the ISP's aren't likely to sayt their s/w is 100%.
a) Because it's a lie which no one will believe
b) they don't know if it works, and don't care enough to check.
Ask him why the records sent by ISPs say in big letters words to the effect "we've no bloody idea if this record is accurate, hell we can't even get change of billing address right, or get the accounts to add up, you think we trust these records ? Dream on. We sent them because we don't want to go to jail, not because they are correct."
One question I'd ask him as an educator is
If you had a student that could not change this data to support the RIAA case, would he award them a good grade ?
Maybe follow up by asking him how many people have such training (my guess is that there are more people capable of this in the USA than firing a gun competently. Would you convict on the grounds that the prosecutor said "almost no one can shoot a target as small as a person at 25 metres"
I would follow this pattern for any of the evidence produced by the RIAA
Get him to explain as their expert how it could be faked. When he claims something cannot, come to Slashdot, and I am very confident that not only can we find an "expert" who can fix it, but possibly more usefully a 13 year old with no formal CompSci education to demo how trivial it is.
There is no computer record used by ISPs or almost anyone else that cannot be faked if you have the password.
My background includes records stored by banks and a major government, and they use tapes and disks of the same brand and configuration as everyone else. Tedious, but not hard.
Even the access logs that record such changes are themselves very fragile, and are simply entries in a different easily malleable list, typically on the same system, and it's far from unknown for the access level required for the audit list to be reachable with the standard system admin password. This is the default for nearly all database systems.
If his track record is accurate, then he will have the options of either admitting the evidence could be fake, or lying.
Next question is to ask him the typical failure rate of IT systems. Ask him the difference between mission critical computing like you see on aircraft and medical systems and the famously buggy and bizarre scareware the utilities blunder with.
Ask him if he'd convict a friend of a serious crime based upon ISP records.
No one with any integrity would do this.
Then ask him what level of crime/penalty he'd accept. Good odds he'll pick music piracy.
In particular it is important that you get him to acknowledge that the records say that this IP address matched an account, not a computer.
This is very much not the same as saying "this computer did this".
If you're lucky and this twerp does'nt read slashdot, he will say the MAC address unqiuely identifies a computer.
One typically assumes this in many applications, but it is a standard documented function of many devices such as routers to take whatever MAC address you tell them.
I think someone needs to do the maths on this. Assume their numbers are correct. Take a mid sized city like Paris, (according to a bit of googling) about 1,000 square miles corresponding to a circle of about 18 miles radius. 1,000 square miles is about 27,878,400,000,000 square feet, or as we techies say "a lot". But of course it's vaguely fractal, and the surface area of a building can be thought of as a cube stuck on the ground, with 5 surfaces open to the air to give us a nice round 100,000,000,000,000 ft2 To get the 50% reduction, we need 15% coverage or 15,000,000,000,000 ft2 Not all of this area is covered by buildings or roads of course, but to compensate many buildings are taller than simple cubes.
Thus if this stuff costs (say) $1 a square more than concrete, you are talking about consuming something like the entire US GDP to reduce Paris air pollution by 50%. I know things haven't been right between the USA and France for a while, but even so I don't think this makes any sense whatsoever. There's a chance that I've missed/gained a zero here or there, so it may only be stupidly expensive, rather than amazingly stupidly expensive.
For this to be applied to the 10 largest cities globally, you are talking about consuming the entire world's GDP. For a reduction in pollution that helps far less than 1% of it's population. Also, let's do some chemistry as well. This stuff is a catlyst, and I guess most of us here remember enough of school chemistry to know that a catalyst is not consumed by the reaction it helps. But, in real world chemistry we note that catalysts get poisoned. They react with other stuff, break down, or in the case of building surfces crud will land on them, meaning they get no air flow. Many quite stable compounds last very well in a box, but leave them out in sunlight and rain for 5 years, and they are screwed, UV is serious stuff over the life of a building. Thus this compound will degrade. I know this is Businesswekk, a mag for the hard of thinking, but is it not the case that roads and pavements have peope driving and walking on thiem ? Are we to believe this stuff won't wear off ?
The inventor quotes some price he made up. Note he does not include labour. Anyone willing to paint 5 storey buildings for nothing ? My guess is $3-5K per building, which means for our example of Paris, we are talking about employing every man in Russia for a year or so. Would certainly help their economy, but... The structure of Paris actually favours this idea more than (say) NY. How do you paint a very tall building ? Won't be cheap. And of course the places that need the pollution controlled most are those with the biggest buildings. Maybe they could paint the Eiffel tower ? It's surface area is huge, given that it's not a simple surface. Tat sounds silly doesn't it ? So does the whole idea.
The place for catalysts is in the exhaust system, or in the process by which fuels are made or burned.
A big part of our business is taking maths/science people into banks to become quants, and we're a well known name in this game, and I'm a director of the firm.
So I do have a good perspective on this.
Banks are (surprisingly enough) careful about the financial background of the people they employ, and yes they do credit checks.
Reason #1 is that they are very very cheap, $10 will generally do it.
They confirm that you are probably who you say you are, indicate that you don't have much of a criminal record, and yes that you don't have a flagrant contempt for money.
Certain types of credit malfunction indicate chemical abuse or gambling issues.
Banks aren't keen on any of the above, and I think we all can assume they are about the worst case scenario for this sort of problem.
But...
A good % of students had "interesting" credit histories. Banks won't knowing hire bankrupts, and there is an array of credit "issues" that means they
are forbidden to hire you in any useful capacity.
People do get hired by banks who have had problems, and by other types of firms.
Forgetting the occasional bill, is not a big deal.
Usually, I can't guarantee anything 100%.
The trick is to tell your pimp.
It's our job to deal with this sort or crap, and we've had more than one person with this problem.
Enough that I have taken the trouble of getting specific confirmation from senior HR at household name banks that provided the courts weren't involved they simply don't care.
Of course pimps need to be managed like everyone else. I can honestly say that my gang won't care, but will gently drop it into the ear of HR once they say they want you.
We don't get paid if you don't get hired. That might be obvious, but always remember this in dealing with recruitment people.
But I wouldn't tell you to put this on your CV, or mention it until they are positively interested.
You *must* do it before they offer, and you must tell them the truth.
Nothing makes an employer lose interest faster than nasty surprises. Tell him up front, and get your pimp to help spin it.
They will be a bit sad you didn't tell them beforehand, but we get a lot of worse grief, and we do it for the cash.
I read the article to mean if everyone changed all the bulbs it would be equal to 1.5 million.
I don't see how it can be based on changing 1 out of the 50 bulbs you cite.
For that to be true, something like 75% of home energy usage would be lighting, and given the suck from aircon, heat, computers, aircon for the computers, fridges etc, I can't see that.
But does it really matter ?
Apparently if the USA totally goes for these, that's the same as 1.5 million people's consumption.
That's 0.5% of USA consumer usage of electricity, or abut 0.1 to 0.2% of total american energy usage.
Global energy usage is growing waaay faster than this, so roughly it buys the human race one month before whatever you expect to happen.
Maybe 2 weeks, maybe 5.
Imagine Americans gave up using any electricity in their homes.
All of it.
Not one joule.
And they do it at no cost, and in one day.
What effect would that have ?
Depending upon whose numbers you believe, the rest of the world would get us to the level we are at now in a few years, I guess around 18 months. The the bog standard exponential curve will keep on accelerating.
You can do other numbers, maybe you can get all the way to 2 years, maybe even 3 if you can think of something even more bizarrely
optimisitc than my base case.
But to me the implication is clearly that e cannot get out of the mess we're in by any plausible conservation measure by any country, even the USA.
Unles you assume that within that time we do something outrageously smart.
Trouble is there is nothing on any drawing board that I know of, or combination of new technologies that will do this.
We can't buy time to get to fusion, even if you believe it will ever work.
Orbital solar would need NASA to be able to launch a shuttle more than once a year. You'd also need a different shuttle that a) worked, and b) could reach anywhere useful. Any strategy involving NASA would be better replaced by wishful thinking and celebrity endorsements.
Solar is crap. Maybe in 50-100 years, maybe not. Many cells never even make back the energy used to make them.
Wind is mildly useful. Round about the level of Irish peat burning. Also assumes you are happy to exterminate vast number of bird species.
Wave ditto.
What's left ?
The lawyers are only "obeying orders", and am disappointed by the fact that so many people here blame them for simply carrying out a difficult job in difficult circumstances, for what I am sure is only modest financial reward from those fine defenders of probity, the RIAA.
In particular I gain the impression that some here are thinking of taking direct action of some kind against these kindly, well intentioned folks. I would urge you not to do this.
I note that the links lead to the application by the lawyers, who helpfully include their email address.
We should treat this information with the respect it deserves and should not cause them to receive any spam, and we should not put matthew@srkllp.com in the email field of every popup window that offers up free porn, Viagra, or free Ipods. Signing them up for services is malicious and may be for all I know, illegal.
Also, the huge numbers of Slashdot readers should not send him or his firm emails, because that would be a bad thing, and might upset their email service.
The fact that their Managing Partner (dick@srkllp.com) likes to be referred to as a "Dick" should be treated as a personal choice, and in the spirit of a diverse and respectful society, I urge all you all to refrain from sending him jokes about his name. He's probably heard them all before. Except of course if he's deaf. Deaf Dick lawyers have feelings too, be kind.
I call upon everyone here to respect the good intentions of the undoubtedly excellent and obviously totally ethical firm of Soble Rowe Krichbaum LLP
I note that their firm specialises in "Complex cases" and "mediation". Perhaps if there is a legitimate criticism of these fine men (and possibly a female secretary), is that their obvious talents are being wasted. Whatever view you have of current affairs in the Middle East, might it not be resolved with less pain if these fine men (and the lady who makes them coffee) were instead to use their obvious talents on a global scale ?
Having sold secure stuff I wrote to the UK government, I see several reasons here.
Big one is that without the source code interfacing to new non-US made gear will be very hard. Britain will want to buy and build other systems. Also looks at firm like Computer Associates who've done bad things to customers through object code, and made good money from it. Ever had Oracel threaten to withdraw support unless you pay a grossly inflated bill ?
Also the US will presumably sell these planes to Israel, and I think we can assume that even if the code is not
given the Israelis are easily smart enough to get it anyway. Israel has supplied ex-British/US technology to countries actively engaged in fighting UK forces. They are a US ally, not a British one.
Recall that Iran was armed with the best American technology befor it went tits up, and modern stealth technology is based upon the fact that the CIA was successful in acquiring working models of Soviet Radars. Hint: von Braun wasn't always a loyal US citizen:)
Short version technology leaks to your opposition faster than you might think.
But in a body of code that size, one could of course hide a backdoor in plain sight. I've done code reviews of operating systems, and it's bloody hard to be 100% that someone hasn't screwed up, let alone planted a logic bomb. One really nasty bugoid was that a Microsoft C was generating undocumented opcodes, which was fine until Intel released a new revision where the "useful" instruction was gone.
Source code != object code != what it actually does.
So you need the full plans, microcode et al for the processors, compilers, and of course debugger.
Also on a political level it must be remembered that America is not an ally of Britain. Yes, Britain is an ally of America, but it's not symmetrical. Each time since WWII when Brtain has been attacked by either conventional forces like in Argentina, or terrorists America has either "tried to be friends to both sides", or in the case of Irish terrorists actively helped the bad guys. Read up on when Britain, France and Israel allied against Soviet backed Egypt to see how "useful" it is to have America as an ally. Ditto Grenada.
These fighters will be in service 20-25 years from now. Presumably Britain will want to go and kill someone in that time, what if one of the Bush family is making good money out of that country ?
MS will no doubt be up for fighting a SCO-like kamizaze attack.
But I can't be the only person who spotted the way that MS will only do it if it doesn't cost much ?
According to Techweb (and many other sites), 25% of Office users will suffer the pain of having to alter apps
because Microsoft didn't own some of the technology it shipped. http://www.techweb.com/wire/software/179100683
Users are currently infringing patents now. But come the next update apparently MS will disable these features.
Thus your applications will go bang. I see that as bad.
Recall how MS issued the WMF vulnerability patch by remotely rebooting people's machines, including servers ?
If you buy an MP3 or DVD player which breaks someone's patent, it's hard to care if the manufacturer gets sued. But if you look at the Blackberry debacle, we see that a patent holder can reach into a user's equipment and disable it.
RIM is not a trivial firm, albeit smaller than MS, it has fought hard, but is losing.
MS lost it's case, and recall all the fights over tech in IE ?
Perhaps this is the real reason the MS is pandering to the copyright holders in the MPAA and RIAA for Vista ?
Thus we have both a legal and technical framework by which you can be shafted. MS can't protect you.
Like with viruses, it's very size makes it more vulnerable, and a more attractive target.
If you were a SCOlike entity would you rather extort $1 from every copy of Windows or Gentoo Linux ?
There are (I guess) >50,000 patentable things in Microsoft's product line. Most of which are of course prior art, or a few licenced or owned. But MS cannot afford even to buy off all the patent holders who might come after it.
Thus MS protection is very limited.
Part of my education was during the Cold War about the role of the British nuclear deterrent, which was a distant 3rd after Russia & the US. It couldn't destroy Russia, but could make an awful mess.
To me it points to what the open source movement should be moving towards with future versions of GPL etc.
By patenting a slew of s/w features, MS, Oracle et al now have something to fear.
The gangs of lawyers on contingency unleashed if MS tries to screw the O/S user base have the same deterrence as as the UK nukes. Not the end of the world, but the end of your world, which is good enough.
Since the lawyers are being used weapons, not as revenue generators, you can cut a deal where they get to keep all the money.
This means they will be more effective than the SCO gang, because their goal is to make a profit, not mutually assured destruction.
At Google there will be people who look at the packet traffic, and occasionally look inside them.
Also most development shops work by using "real" data somewhere in the test/code cycle. They shouldn't of course, but generating
realistic fake data is actually quite hard, so that corner is cut.
Many people are a big vague about the real data, even when there is no restriction...
Making a system safe from developers and sysops roughly adds 50% to it's cost.
That sounds high, but it's lots of little things. If you're doing the job properly, you need entirely separate infrastructure for development and deployment. You need to encrypt everything, not just data but "interesting" file names and of course the executables are locked down big time.
What % of your problems in live systems have you solved by spotting that certain inputs blow it up ? Or that this file grows very much too fast or gets locked ? That is taken away from you, and to make the cost higher of course you have to pay smart people to do the taking away.
Of course you can use dumb people to manage this. In Britain EDS has procedures vaguely like this for big government projects, and the cost overruns alone make the British army presence in Iraq look cheap.
Often you get more problems from the fact that people can't quite see what's going on than from acts of malice.
At one sensitive site, I was allowed higher access privileges, but at the price of not ever knowing my own password:)
I'd guess >90% of the money spent online is through MS IE. What if they demanded a rake off ?
Wouldn't be hard to implement. A "security patch" that makes Amazon et al unable to take credit card #'s, or which spits
out warnings when you access the site.
No doubt someone at MS has though of this, but anti-trust is too scary for them currently.
However what about Mac users ?
They trust Apple a lot. No I can't work out why either given it's record, but Apple looks like it will own the market for Mac browsers now that MS has lost interest.
What's the betting that real soon now Apple starts demanding protection money ? Called something like iSecurity ?
Agreed, 3G providers have a very tight grip on the content, meaning that it's so poor & expensive that I never buy any.
Telcos have long had the view that any money made that they don't get a cut from is somehow "stolen".
I agree that IP would be useful for my phone, and if I was allowed to use it properly I'd do all the stuff the 2.5G guy does as well.
I have a 3G phone. It can do IP, and run Java.
I never use these facilities.
Because it's in a "walled garden". You want to send data to my phone you have to pay big bucks.
Thus the content is junk, high priced sports clips, stupidly expensive music videos, and pretty much nothing else, not even proper emaill.
The big money in mobile phones is ring tones.
Think that one though, long ago you used to be able to buy sets of sound effects for your PC, and you could set the noises to be FX from Star Trek, and have windows opening sound like the doors, etc.
That's the future of the tiered internet, where DRM will charge you $10 to download the opening music of windows. There won't be anything else.
The dream scenario for telcos is a sort of balkanisation. For tiering to work you've got to have cross ISP tariffs, like we did with copper line telephone calls.
Then they charge each other, plus a service fee for colleting the money on "our" behalf. Any ISP that doesn't charge finds that it has to pay other telcos to carry traffic, so all have to follow.
The RIAA and MPAA will love it, since of course high prices = high quality downloads. So will Apple, that's why they've remained silent. Also why the BBC article cited will be the last. There only difference between the BBC and the PR department of Apple is that Apple's PR statements are bounded by advertising laws.
In ancy case, avoid 3G phones, trust me on this.
It is important politically to show that you've made a decision, not that you are an advocate.
You do not want to be seen as working for the open source community, any more than a decision maker
wants to be seen as in the pockets of the hardware vendor.
Feel free to fight media with media, most of it is very O/S friendly, and even the MS centric press reports on Windows vulnerabilities pretty much daily. Also do not neglect PR here. Make sure you have some idea of what your key decision makers have on their home PCs. A quiet one to one warning that Winamp has a scary hole gains you points, and feel free to emphasise the Win bit of Winamp.
Also many managers are bovine herd animals, many of them even play golf.
Saying "we're copying the strategy of Amazon", or a competitor scorese big points. Saying "we're going to pioneer Apache load balancing" would set off alarm bells.Also show your commercial skills, the game here is to push your agenda, which includes you personally.
If you show interest in O/S then Oracle, MS et al may well suddenly become more flexible on prices and licencing. When dealing with managment it's quite legitimate (and wise) to say "we pay $XXXXXXXX to MS, and although using O/S will only save $XXXXXXXX / 50 it means that when we negotiate with them they will know we can shift business away from them". Most decision makers will think better of both you and your views because of this commercial acumen. You can research tales of how the big vendors have marketing pools just for this purpose.
Fact is that many non-IT managers perceive us to be in the pockets of vendors, not necessarily in a corrupt way, but worse on ideological views. A good % of us have paid money Linux/Apple/Intel t-shirts, toys et al, some of which we wear to work. I think it's a safe bet to assume that the finance director is unlikely to appear in a Price Waterhouse t-shirt.
However, I must say I'm surprised at the basic issue here. Before I became a pimp I was an IT director, and non-IT management had if anything a higher impression of Linux et al than it merited for what we were doing. More than once I was specifically asked why we weren't using Linux. And yes at one point I over rode the people who wanted to deploy a horde of Linux desktops.
My view of Linux is that in most practical ways the business perspective is much the same as for MS stuff.
Although with my techie hat on the access to source is nice, fact is that it's usually not all that wise to screw around with it unless you know what you're doing, and most people don't.
Thus with both open source and proprietary, you pay people to make it do what you want, and to nurse it when it gets sick.
Thus the variable is the cost and quality of developers and support people. At the risk of being shot at, MS stuff is typically easier to drive at least 80% of the time. Ironically of course this means that anyone who can get past the occasionally quite grotesque open source interfaces is usually smarter, and often cheaper.
Re-reading that, perhaps this is a candidate for the shortest sentence that will offend absoutely anyonne, but it's the way I've made decisions:)
In an ideal world you have someone wise in the ways of each product you depend upon. Saldy real != ideal
More than once I've told someone to go and look at a misbehaving system they're not familiar with, simply because they were the guy available, hell it's happened to all of us.
Must say that as a manager I'd be less fearful of doing that with a MS system than most open source.
Thus whatever solution you propose to your management, it needs to address risks, support and costs. Risks are things like not having a dependance upon one person, costs have structure and uncertainty, and support is often a function of the relationships you can build.
DCFC The Pimp
To make s/w reliable you have to assume that it isn't. Although NASA lost the plot decades ago, it's worth looking at how IBM engineers built systems that could survive the failures of these chimps. IBMers assumed that everything could go wrong. There was great pride in the team when they found a stupidly obscure and massively unlikely bug. There was no notion of "having to decide whether you were an engineer or a manager".
IBM stuff managed to survive things like motherboards being melted in flight with microgee solder balls shorting out the rest of the system.
Even in Challenger the IT was working as it hit the water on the way down.
That culture is far more important than any single technique.
Implicit is the notion that no one can really check their own work. You need someone who takes a positive joy in pointing out your errors.
That needs to be backed up by managment, who must not only avoid trating bug finders as trouble makers, but positively reward them, and yes that means cash.
I think we can already see why so much commercial s/w has stupid levels of bugs...
If you're serios then I'd suggesting voting as an architecture
Multiple modules to take decisions.
To do this properly, each needs to have different code. The gold standard is separate code, each in a different language written by a different programmer. The mulitple language precept is because two programmers in (say) C++ may actually make the same error from the same specification. A classic example is "between one and ten items".
Different people may interprest that as including 1 and 10, or excluding them, or in the case of some C++ developers 0..9
VB programmers "know" that Dim q(10) actually has 11 elements in it.
Voting puts a bound on the effect of a failure.
Although your employer is behaving badly, you have to make a business decision. If you invest $X and Y hours, what will the return be ? To you, personally ?
The next calculation is whether you can get a better return on the investment by learning some other skill with about the same level of up front spend.
In neither of these cases should you assume you are going to stay in the same firm. There are plenty of jobs boards out there where you can work out your market value, or you can talk to a pimp. I'd offer to help myself, but we don't do network people.
I don't see that you are under any moral obligation to spend your time & money on helping your employer undo the effect of bad decisions. Taking what you say at face value, you're going to screw up. You are now in charge of a mildly complex system that you don't understand. Although I assume you're a competent individual in general, you are now not equipped to do your job. Bad things will happen.
You may well get blamed for these screwups, and even if you do the training, there will be a lag between study and competence. I recall the first time I had to build a LAN. Wasn't pretty or reliable and took >10 times as long to build compared to the work of someone who knew what they were doing. (think how long it would take you to discover about crossover cables if no one had ever told you:(
Thus you need to start some pre-emptive blame managment. You need to have emails saying that without proper training, you cannot guarantee the integrity of a resource that is critical to your firm. A good next step is to say that you need to hire in a consultant because there's things you can't do. The costs of this will be non-trivial, and inevitably a lot more than paying or training you. Done right, they may even require that you go on training.
But however it works out, do not invest your time or money for free.
Oh dear. Let's do some really simple economics here...
Migration moves people from where they can work cheaply, to a place where they can charge more for their labour.
These people won't stop existing if they are not allowed to migrate. They will do IT work in their home country.
Since India, China et al have lower wage rates, he will work for less.
Thus the world supply of IT people will be sold at a lower average rate.
Migration actually holds up the price of labour for those skills which are portable across countries.
It is also the case that an IT person costs quite a bit of money to build. This comes from a combination of parents and the home state.
However if they migrate, the return in the form of taxes and the general economic use of their work is for the benefit of the country that he now lives in. In effect countries that do not retain their skilled people are transferring wealth to those that do.
Also, at the risk of political incorrectness. Look at economic migrants in terms of their quality. Those who have been born with some grave issue affecting their physical health or general intelligence, find it much harder to move.
Thus they are on average better people to parent the next generation of citizens, and it is that generation who will be working to support us when we retire.
If we look at history over the last 200 years, we can observe the fate of those countries who kept out foreigners.
China, which was abjectly garbage. It's current growth is based upon low wages. Crap pay is a sign of past economic failure, not success.
Other socialist countries like Russia and E.Europe kept people out for a long time. Anyone want to emulate them in any way whatsoever ?
Japan kept people out, stayed backward until the Americans forced it open.
Conversely look at America, Oz, NZ et al.
Easy.
Fact is that it's now a global market for labour. When I left university having done computers, IT people were so rare that if 2 strangers in this area bumped into each other they often struck up conversation because they felt part of a small group. I guess there's more than 80 million full time IT people in the world. In the early 80s, most people on Earth worked in mud hut or gulag economies. Effective world work force of about 250 million, in W.Europe, N.America, and various fragments of the former British empire, like Oz, HK, Nz, and Singapore.
Now there's nearer two billion people in the world economy, and growing.
Lots more jobs, but you have to make sure you are equipped to compete.
If I understand your process, you have some poor guy whose job is to handle a horde of little changes.
Very hard for someone to keep track of these things to work out their overall consequences; it should be automated.
But the implication is that it's not and I'm guessing he uses his wit and wisdom to spot obvious bad things.
Worse, you may have a team following someone else's rules which they neither understand nor care about. Also a helpdesk team finds it harder than an individual to keep a view of the overall system.
It's easy to see that going wrong by simple human error, which is more common when you have to have a relatively smart person doing
a job that 99% of the time is dull.
I bet there's more staff turnover in this role, eventually of course the point of stabiity is when you end up with someone who can't get a better job, hard to see that as good.
How do you keep them sharp for the
It seems to me that a rules based system could capture the essence of this job, and something that will make your security audit peple happy is that the software can be tested.
I wonder if there exists such a tool already ?
(feel free to reduce my ignorance here)
It strikes me as a rather good candidate for an open source collaboration.
Another reason is that being a programmer is like placing bets on your future. Soonmer or later you make a crap one, and it all goes wrong.
I was lucky to do C/Unix at college, did OS/2 for a while which was good whilst it lasted, but suddenly went down the drain.
C++ and VB were good choices, but I missed the boat on the easy money for SAP and Powerbuilder. Should have spent more effort on SQL, and wasted less time on Lisp.
This game is very buzzword driven. Doubt there's anything that bleeps that I can't program given a manual, and I'm far from unique in this. But it's hard to change track to a more fashionable technology, since many pimps and HRs just see the words, not the underlying skills.
After 20 years of programming and a bit of bossing programmers around I found I could make more money and work sensible hours by becoming a headhunter. Some people prefer dealing with a pimp who doesn't think Iterators are the evil mechanoids from Stargate SG1, we also can toake job specs as informal as "a bit like Steve, but maybe a little more maths".
I still do a bit of coding, and our most successful marketing tool has been my "C++ for interviews" document where I go through several dozen commonly asked interview questions, and answer them. A couple of banks have acquired copies, merely so they can upgrade their interview process:)
Management like "team players", and Indians often have this attribute, if you define it as not ever arguing with the boss. I've managed very good programmers and they can be a real pain. I would not characterise Indians as "yes men", but some don't warn me when I'm going to do something stupid. A good employee can warn his boss of impending doom, without it being confrontational, so the optimum is somewhere in between.
Also, many pimps and HR types do not like to hire people older than the boss of that area.
This is bollocks of course. As a former IT manager I've hired people who were older because my success was dependant upon the quality of my team, and frankly I don't give a toss about people's age provided they don't die at their desks.
Without exception, the managers I've spoken with, have pretty much the same view. They will indicate that a given job is for a newbie, but if they can do the work they don't care.
What you want is Markov chains.
For any given statement you can easily construct the set of statmenets that follow it, together with their number.
We will observe that given
for
There are a lot of branches to the tree, with
(i =0;
being very common
and// Woo woo
being quite rare.
We can easily generate more code by picking statements at random, but in proportion to their observed frequency.
But we can do better.
Given pairs of statements, we can generate a table where given the pair of statements
for( i=0;
we know the probability of
i !=
This because we're dealing with pairs from a large set, the branching will be quite low, ie it will be much more constrained, and thus a lot more like the original code.
This is analagous to being able to say in English, given "Space, the"
we mostly get "final" as the most probable next symbol, and now we have "the final", the next word wmay be|"frontier" but given that "Space" is not being thought about any more, we might have "cut".
Text generated like this is quite like English, and if you include punctuation, can generate things that are as grammatically correct as many people's posting. It can be quite funny. I have used it to merge Terry Pratchett and Microsoft marketing blurb, and certainly the MS stuff improves...
Same applies to C++.
It will take a little experimentation to find out the best length for the chain, but assuming that the input code is syntactically correct, so the output will often compile.
Quite what it will do is another question, but this sort of chain is rather like how people learn.
If you listen to babies, they start off making random noises, then they make sounds that have very roughly the same frequency of ocurrence as their parents language, then they burble in things that sound like English, but clearly are not. Their "singing" at 18 month is usually quite free of real words, but they've heard "twinkle twinlke little star" so often that the chain of "twinnle twinnle lipple sarr" is carved into their neural net.
Given that the spec here is for "useful" I propose this as an AI test.
Slashdot Mirror
Almost any expansion onto other planets that we can easily forsee is just that, an expansion, not a migration.
I suspect that we've read many SF books in common, but without teleportation or Warp Drive, there is no way that more than 1,000 people a year will ever leave Earth orbit. Indeed, it's not clear to me how you can get a million people into space per year without really quite impressive environmental impact.
There's nowhere to go where we can live without exotic engineering, so the vast bulk of our descendants for the next few millennia are going to be on Earth.
A self sufficent Mars colony, based upon what we know is centuries away, and unless it is self sufficient buys the race nothing in terms of survival.
Even then, many of the big scary things we know about (novas, various large scale radiation events, perturbations of orbits) would make them more vulnerable than us.
Burning crap has it's own environmental problems...
Indeed for me the reson to move away from Carbon has always been based upon the other polluting effects, and of course the depletion of cheap crap to burn.
But the "burning crap" position is essentially where we are going to need to be. The Eath is not "naturally" habitable by humans, or even comfortable for us.
Over the long term we are going to need to manage it like some sort of garden if we are to survive. If we look at those subsets of humanity that live "in tune with nature", not only are their individual lives short, but these groups have a tendency to suddenly be wiped out when nature changes her mind.
I'm not disagreeing with your basic point, but in the case of Solar cycles, by the time we've reduced our nice warm blanket of greenhouse gases, the cycle may be on a down turn, when we've just spent trillions making the Earth colder. Ooops. :) If the forces are outside our control, like Solar, or Chinese industralisation then the optimum may be to make it less bad and not spend money on trying to slow it down. I'm not saying this is the case, I'm saying that a religious belief in a single monolithic truth is extravagantly unlikely to deliver the best choices.
It's easy to suspect that fertiliser and pesticide run offs are altering which microbes prosper and thus affect climate. Bio-fuels require yet more intensive agriculture, and so may make the problem worse, or not. Again a focus on CO2 may be doing more harm than good.
Also, there is the random "background noise" of vulcanism. This can be a big term. In the last couple of centuries we've seen at least one event that lost our planeet an entire summer, and for a while entirely overwhelmed any possible human effect. A few big volcanoes randomly going off could make the most hysterical predicitons of arts-graduate Greens look like a rainy afternoon.
The nature of climate variation affects deeply how we should respond. We can't do much about volcanoes, except maybe set one off as a frantic reaction to short term global warming
That's a valid point.
Holocaust denyers, climate change rubbishers , creationists et al do us a service by making themselves easy to spot.
We risk falling into Al Gore's stupidity of "facts are a kind of pollution" if we censor those who don't emit politically correct statements. To me, the whole point of tax funded science is to find new and interesting ways of showing that what most people believe is wrong. If it's nice warm fuzzy confirmations of our world view, then it's more suitable for the Discovery Channel or a corporate R&D lab.
I guess the odds are 99% that it's humans that are causing all this stuff we see. But what of the other 1% ? What if we spend some vast amount of money changing the global economy to find that it's a subtle change in the Sun's output, a random downturn in Vulcanism, or somehow to do with deep sea bacteria ? I don't believe any of those ideas much, but I don't 100% believe they are wrong either.
America is now paying the price of having the highest relgious observance in the developed world. The damage is not the general superstious attack on evolution, or even the rape of thousands of children by priests, but a more insidious view that there is a SINGLE TRUTH, and EVERYONE WHO SAYS ANYTHING ELSE IS EVIL.
Such is the invasive nature of religion that even poeple with scientific training are prey to the condition.
Part of our filter is checking people understand English. You failed. I explictly said we don't care about this stuff. As it happens, although the field is only about 3% female, 17% of the people we put forward in 2006 were female. I'm not even going to claim this as virtue, just applied greed. Discrimination would cost us money, since we'd place fewer people, so we don't do it. It might not even be anything to do with us since our research implies more women are trying to get in.
I'm a headhunter, and it's not always easy to give feedback.
We do high end programming jobs for banks, and screen candidates before they go in.
Some are under the deluded impression that if they argue forcefully that they can undo a poor assessment of their suitablity. A common position is "I can learn that", which may or may not be true. I tend to suggest the book they can read, but I'm still not sending them in. It can get very tiresome to politely say "but you didn't know about iterators", and be loudly told that this doesn't matter, especially if the candidate perceives that if he keeps this up long enough we'll give in.
We don't insist on Jeeves like smoothness from candidates, but some are simply inappropriate for any job in a bank to a degree that it's a mystery how they got hired for their current role, but not why they are leaving. People who say "that's a stupid question", or give the impression that their current boss is a flaky criminal mastermind don't score points here.
Also there is the threat of being sued. Our equal opportunity policy is basically that we do this for the money, and don't give a toss about their skin colour, sex age, or which side they fought on in the Gulf War. But some people are just not good enough, or more ambiguously not "right". Sometimes that is stuff that's hard to specifically form "he has no experience of X", especially since you usually don't meet them unless they have such terms on their CV
I've been counselled not to use terms that imply a preference for a given age, and work in an field that is 97% male, so it's natural, but dangerous for me to use terms that imply I only want men. Most managers can't do this with the reliability of someone who does this every day, and so risk getting sued, so often HR departments mandate that they don't say anything, and especially not in an email.
Also feedback from clients is a sensitive issue. We occasionally get quite robust feedback about the failings of a given candidate, and we're certainly not going to get a client into legal shit. Occasionally the hiring manager will evaluate a candidate entirely unfairly and incorrectly. As it happens I'm a bit of a techie myself, and so it's a bit of a challenge to put a good face on bizarre and useless interview questions, and yes that includes the cases where the interviewer's "correct answer" was wrong.
We do better feedback than other pimps, but honestly, none of us make it all the way to good. So ultimately, you're not going to get quality feedback. What you need to do is find some friend who's done some hiring, buy him a beer or dinner and get him to check you out.
I agree it is a good question, but I'd spin it slightly. I'd ask him *how* to doctor a screen, and how trivial it would be to fiddle records that showed the defendant had a given IP address.
It would take very little time for a competent person to do this, indeed to ridicule the RIAA position,I could take a couple of days with an average 10 year old would leave them able to do this, a smart 8 year old could do it in a morning.
Ask him if he's conducted a review of ISP logging s/w, as in read the source code, not as in sent an email asking if it was "OK". Would bet good money he hasn't. Actually the ISP's aren't likely to sayt their s/w is 100%. a) Because it's a lie which no one will believe
b) they don't know if it works, and don't care enough to check.
Ask him why the records sent by ISPs say in big letters words to the effect "we've no bloody idea if this record is accurate, hell we can't even get change of billing address right, or get the accounts to add up, you think we trust these records ? Dream on. We sent them because we don't want to go to jail, not because they are correct."
One question I'd ask him as an educator is
If you had a student that could not change this data to support the RIAA case, would he award them a good grade ?
Maybe follow up by asking him how many people have such training (my guess is that there are more people capable of this in the USA than firing a gun competently. Would you convict on the grounds that the prosecutor said "almost no one can shoot a target as small as a person at 25 metres" I would follow this pattern for any of the evidence produced by the RIAA
Get him to explain as their expert how it could be faked. When he claims something cannot, come to Slashdot, and I am very confident that not only can we find an "expert" who can fix it, but possibly more usefully a 13 year old with no formal CompSci education to demo how trivial it is.
There is no computer record used by ISPs or almost anyone else that cannot be faked if you have the password.
My background includes records stored by banks and a major government, and they use tapes and disks of the same brand and configuration as everyone else. Tedious, but not hard.
Even the access logs that record such changes are themselves very fragile, and are simply entries in a different easily malleable list, typically on the same system, and it's far from unknown for the access level required for the audit list to be reachable with the standard system admin password. This is the default for nearly all database systems. If his track record is accurate, then he will have the options of either admitting the evidence could be fake, or lying. Next question is to ask him the typical failure rate of IT systems. Ask him the difference between mission critical computing like you see on aircraft and medical systems and the famously buggy and bizarre scareware the utilities blunder with. Ask him if he'd convict a friend of a serious crime based upon ISP records.
No one with any integrity would do this. Then ask him what level of crime/penalty he'd accept. Good odds he'll pick music piracy. In particular it is important that you get him to acknowledge that the records say that this IP address matched an account, not a computer. This is very much not the same as saying "this computer did this". If you're lucky and this twerp does'nt read slashdot, he will say the MAC address unqiuely identifies a computer. One typically assumes this in many applications, but it is a standard documented function of many devices such as routers to take whatever MAC address you tell them.
I think someone needs to do the maths on this.
Assume their numbers are correct.
Take a mid sized city like Paris, (according to a bit of googling) about 1,000 square miles corresponding to a circle of about 18 miles radius.
1,000 square miles is about 27,878,400,000,000 square feet, or as we techies say "a lot".
But of course it's vaguely fractal, and the surface area of a building can be thought of as a cube stuck on the ground, with 5 surfaces open to the air
to give us a nice round 100,000,000,000,000 ft2
To get the 50% reduction, we need 15% coverage or 15,000,000,000,000 ft2
Not all of this area is covered by buildings or roads of course, but to compensate many buildings are taller than simple cubes.
Thus if this stuff costs (say) $1 a square more than concrete, you are talking about consuming something like the entire US GDP to reduce Paris
air pollution by 50%. I know things haven't been right between the USA and France for a while, but even so I don't think this makes any sense whatsoever.
There's a chance that I've missed/gained a zero here or there, so it may only be stupidly expensive, rather than amazingly stupidly expensive.
For this to be applied to the 10 largest cities globally, you are talking about consuming the entire world's GDP. For a reduction in pollution that helps
far less than 1% of it's population.
Also, let's do some chemistry as well. This stuff is a catlyst, and I guess most of us here remember enough of school chemistry to know that a catalyst is not consumed by the reaction it helps.
But, in real world chemistry we note that catalysts get poisoned. They react with other stuff, break down, or in the case of building surfces crud will land on them, meaning they get no air flow. Many quite stable compounds last very well in a box, but leave them out in sunlight and rain for 5 years, and they are screwed, UV is serious stuff over the life of a building.
Thus this compound will degrade.
I know this is Businesswekk, a mag for the hard of thinking, but is it not the case that roads and pavements have peope driving and walking on thiem ?
Are we to believe this stuff won't wear off ?
The inventor quotes some price he made up. Note he does not include labour. Anyone willing to paint 5 storey buildings for nothing ?
My guess is $3-5K per building, which means for our example of Paris, we are talking about employing every man in Russia for a year or so.
Would certainly help their economy, but...
The structure of Paris actually favours this idea more than (say) NY. How do you paint a very tall building ? Won't be cheap. And of course the places that need
the pollution controlled most are those with the biggest buildings.
Maybe they could paint the Eiffel tower ?
It's surface area is huge, given that it's not a simple surface.
Tat sounds silly doesn't it ?
So does the whole idea.
The place for catalysts is in the exhaust system, or in the process by which fuels are made or burned.
A big part of our business is taking maths/science people into banks to become quants, and we're a well known name in this game, and I'm a director of the firm. So I do have a good perspective on this. Banks are (surprisingly enough) careful about the financial background of the people they employ, and yes they do credit checks. Reason #1 is that they are very very cheap, $10 will generally do it. They confirm that you are probably who you say you are, indicate that you don't have much of a criminal record, and yes that you don't have a flagrant contempt for money. Certain types of credit malfunction indicate chemical abuse or gambling issues. Banks aren't keen on any of the above, and I think we all can assume they are about the worst case scenario for this sort of problem. But... A good % of students had "interesting" credit histories. Banks won't knowing hire bankrupts, and there is an array of credit "issues" that means they are forbidden to hire you in any useful capacity. People do get hired by banks who have had problems, and by other types of firms. Forgetting the occasional bill, is not a big deal. Usually, I can't guarantee anything 100%. The trick is to tell your pimp. It's our job to deal with this sort or crap, and we've had more than one person with this problem. Enough that I have taken the trouble of getting specific confirmation from senior HR at household name banks that provided the courts weren't involved they simply don't care. Of course pimps need to be managed like everyone else. I can honestly say that my gang won't care, but will gently drop it into the ear of HR once they say they want you. We don't get paid if you don't get hired. That might be obvious, but always remember this in dealing with recruitment people. But I wouldn't tell you to put this on your CV, or mention it until they are positively interested. You *must* do it before they offer, and you must tell them the truth. Nothing makes an employer lose interest faster than nasty surprises. Tell him up front, and get your pimp to help spin it. They will be a bit sad you didn't tell them beforehand, but we get a lot of worse grief, and we do it for the cash.
I read the article to mean if everyone changed all the bulbs it would be equal to 1.5 million. I don't see how it can be based on changing 1 out of the 50 bulbs you cite. For that to be true, something like 75% of home energy usage would be lighting, and given the suck from aircon, heat, computers, aircon for the computers, fridges etc, I can't see that.
But does it really matter ? Apparently if the USA totally goes for these, that's the same as 1.5 million people's consumption. That's 0.5% of USA consumer usage of electricity, or abut 0.1 to 0.2% of total american energy usage. Global energy usage is growing waaay faster than this, so roughly it buys the human race one month before whatever you expect to happen. Maybe 2 weeks, maybe 5. Imagine Americans gave up using any electricity in their homes. All of it. Not one joule. And they do it at no cost, and in one day. What effect would that have ? Depending upon whose numbers you believe, the rest of the world would get us to the level we are at now in a few years, I guess around 18 months. The the bog standard exponential curve will keep on accelerating. You can do other numbers, maybe you can get all the way to 2 years, maybe even 3 if you can think of something even more bizarrely optimisitc than my base case. But to me the implication is clearly that e cannot get out of the mess we're in by any plausible conservation measure by any country, even the USA. Unles you assume that within that time we do something outrageously smart. Trouble is there is nothing on any drawing board that I know of, or combination of new technologies that will do this. We can't buy time to get to fusion, even if you believe it will ever work. Orbital solar would need NASA to be able to launch a shuttle more than once a year. You'd also need a different shuttle that a) worked, and b) could reach anywhere useful. Any strategy involving NASA would be better replaced by wishful thinking and celebrity endorsements. Solar is crap. Maybe in 50-100 years, maybe not. Many cells never even make back the energy used to make them. Wind is mildly useful. Round about the level of Irish peat burning. Also assumes you are happy to exterminate vast number of bird species. Wave ditto. What's left ?
The lawyers are only "obeying orders", and am disappointed by the fact that so many people here blame them for simply carrying out a difficult job in difficult circumstances, for what I am sure is only modest financial reward from those fine defenders of probity, the RIAA.
In particular I gain the impression that some here are thinking of taking direct action of some kind against these kindly, well intentioned folks.
I would urge you not to do this.
I note that the links lead to the application by the lawyers, who helpfully include their email address.
We should treat this information with the respect it deserves and should not cause them to receive any spam, and we should not put matthew@srkllp.com in the email field of every popup window that offers up free porn, Viagra, or free Ipods. Signing them up for services is malicious and may be for all I know, illegal.
Also, the huge numbers of Slashdot readers should not send him or his firm emails, because that would be a bad thing, and might upset their email service.
The fact that their Managing Partner (dick@srkllp.com) likes to be referred to as a "Dick" should be treated as a personal choice, and in the spirit of
a diverse and respectful society, I urge all you all to refrain from sending him jokes about his name.
He's probably heard them all before.
Except of course if he's deaf.
Deaf Dick lawyers have feelings too, be kind.
I call upon everyone here to respect the good intentions of the undoubtedly excellent and obviously totally ethical firm of Soble Rowe Krichbaum LLP
I note that their firm specialises in "Complex cases" and "mediation". Perhaps if there is a legitimate criticism of these fine men (and possibly a female secretary), is that
their obvious talents are being wasted. Whatever view you have of current affairs in the Middle East, might it not be resolved with less pain if these fine men (and the lady who makes them coffee) were instead to use their obvious talents on a global scale ?
Having sold secure stuff I wrote to the UK government, I see several reasons here.
:)
Short version technology leaks to your opposition faster than you might think.
Big one is that without the source code interfacing to new non-US made gear will be very hard. Britain will want to buy and build other systems. Also looks at firm like Computer Associates who've done bad things to customers through object code, and made good money from it. Ever had Oracel threaten to withdraw support unless you pay a grossly inflated bill ?
Also the US will presumably sell these planes to Israel, and I think we can assume that even if the code is not given the Israelis are easily smart enough to get it anyway. Israel has supplied ex-British/US technology to countries actively engaged in fighting UK forces. They are a US ally, not a British one.
Recall that Iran was armed with the best American technology befor it went tits up, and modern stealth technology is based upon the fact that the CIA was successful in acquiring working models of Soviet Radars. Hint: von Braun wasn't always a loyal US citizen
But in a body of code that size, one could of course hide a backdoor in plain sight. I've done code reviews of operating systems, and it's bloody hard to be 100% that someone hasn't screwed up, let alone planted a logic bomb. One really nasty bugoid was that a Microsoft C was generating undocumented opcodes, which was fine until Intel released a new revision where the "useful" instruction was gone. Source code != object code != what it actually does.
So you need the full plans, microcode et al for the processors, compilers, and of course debugger.
Also on a political level it must be remembered that America is not an ally of Britain. Yes, Britain is an ally of America, but it's not symmetrical. Each time since WWII when Brtain has been attacked by either conventional forces like in Argentina, or terrorists America has either "tried to be friends to both sides", or in the case of Irish terrorists actively helped the bad guys. Read up on when Britain, France and Israel allied against Soviet backed Egypt to see how "useful" it is to have America as an ally. Ditto Grenada.
These fighters will be in service 20-25 years from now. Presumably Britain will want to go and kill someone in that time, what if one of the Bush family is making good money out of that country ?
MS will no doubt be up for fighting a SCO-like kamizaze attack. But I can't be the only person who spotted the way that MS will only do it if it doesn't cost much ?
According to Techweb (and many other sites), 25% of Office users will suffer the pain of having to alter apps because Microsoft didn't own some of the technology it shipped.
http://www.techweb.com/wire/software/179100683 Users are currently infringing patents now. But come the next update apparently MS will disable these features.
Thus your applications will go bang. I see that as bad.
Recall how MS issued the WMF vulnerability patch by remotely rebooting people's machines, including servers ?
If you buy an MP3 or DVD player which breaks someone's patent, it's hard to care if the manufacturer gets sued. But if you look at the Blackberry debacle, we see that a patent holder can reach into a user's equipment and disable it.
RIM is not a trivial firm, albeit smaller than MS, it has fought hard, but is losing.
MS lost it's case, and recall all the fights over tech in IE ?
Perhaps this is the real reason the MS is pandering to the copyright holders in the MPAA and RIAA for Vista ?
Thus we have both a legal and technical framework by which you can be shafted. MS can't protect you.
Like with viruses, it's very size makes it more vulnerable, and a more attractive target.
If you were a SCOlike entity would you rather extort $1 from every copy of Windows or Gentoo Linux ?
There are (I guess) >50,000 patentable things in Microsoft's product line. Most of which are of course prior art, or a few licenced or owned. But MS cannot afford even to buy off all the patent holders who might come after it.
Thus MS protection is very limited.
Part of my education was during the Cold War about the role of the British nuclear deterrent, which was a distant 3rd after Russia & the US. It couldn't destroy Russia, but could make an awful mess.
To me it points to what the open source movement should be moving towards with future versions of GPL etc.
By patenting a slew of s/w features, MS, Oracle et al now have something to fear.
The gangs of lawyers on contingency unleashed if MS tries to screw the O/S user base have the same deterrence as as the UK nukes. Not the end of the world, but the end of your world, which is good enough.
Since the lawyers are being used weapons, not as revenue generators, you can cut a deal where they get to keep all the money.
This means they will be more effective than the SCO gang, because their goal is to make a profit, not mutually assured destruction.
DCFC the Pimp. http://pauldominic.com/
At Google there will be people who look at the packet traffic, and occasionally look inside them. Also most development shops work by using "real" data somewhere in the test/code cycle. They shouldn't of course, but generating realistic fake data is actually quite hard, so that corner is cut.
:)
Many people are a big vague about the real data, even when there is no restriction...
Making a system safe from developers and sysops roughly adds 50% to it's cost.
That sounds high, but it's lots of little things. If you're doing the job properly, you need entirely separate infrastructure for development and deployment. You need to encrypt everything, not just data but "interesting" file names and of course the executables are locked down big time.
What % of your problems in live systems have you solved by spotting that certain inputs blow it up ? Or that this file grows very much too fast or gets locked ? That is taken away from you, and to make the cost higher of course you have to pay smart people to do the taking away.
Of course you can use dumb people to manage this. In Britain EDS has procedures vaguely like this for big government projects, and the cost overruns alone make the British army presence in Iraq look cheap.
Often you get more problems from the fact that people can't quite see what's going on than from acts of malice.
At one sensitive site, I was allowed higher access privileges, but at the price of not ever knowing my own password
I'd guess >90% of the money spent online is through MS IE. What if they demanded a rake off ? Wouldn't be hard to implement. A "security patch" that makes Amazon et al unable to take credit card #'s, or which spits out warnings when you access the site.
No doubt someone at MS has though of this, but anti-trust is too scary for them currently.
However what about Mac users ?
They trust Apple a lot. No I can't work out why either given it's record, but Apple looks like it will own the market for Mac browsers now that MS has lost interest.
What's the betting that real soon now Apple starts demanding protection money ? Called something like iSecurity ?
DCFC the Pimp
Agreed, 3G providers have a very tight grip on the content, meaning that it's so poor & expensive that I never buy any. Telcos have long had the view that any money made that they don't get a cut from is somehow "stolen". I agree that IP would be useful for my phone, and if I was allowed to use it properly I'd do all the stuff the 2.5G guy does as well.
I have a 3G phone. It can do IP, and run Java.
I never use these facilities.
Because it's in a "walled garden". You want to send data to my phone you have to pay big bucks.
Thus the content is junk, high priced sports clips, stupidly expensive music videos, and pretty much nothing else, not even proper emaill.
The big money in mobile phones is ring tones. Think that one though, long ago you used to be able to buy sets of sound effects for your PC, and you could set the noises to be FX from Star Trek, and have windows opening sound like the doors, etc.
That's the future of the tiered internet, where DRM will charge you $10 to download the opening music of windows. There won't be anything else.
The dream scenario for telcos is a sort of balkanisation. For tiering to work you've got to have cross ISP tariffs, like we did with copper line telephone calls.
Then they charge each other, plus a service fee for colleting the money on "our" behalf. Any ISP that doesn't charge finds that it has to pay other telcos to carry traffic, so all have to follow.
The RIAA and MPAA will love it, since of course high prices = high quality downloads. So will Apple, that's why they've remained silent. Also why the BBC article cited will be the last. There only difference between the BBC and the PR department of Apple is that Apple's PR statements are bounded by advertising laws.
In ancy case, avoid 3G phones, trust me on this.
It is important politically to show that you've made a decision, not that you are an advocate. :)
In an ideal world you have someone wise in the ways of each product you depend upon. Saldy real != ideal
More than once I've told someone to go and look at a misbehaving system they're not familiar with, simply because they were the guy available, hell it's happened to all of us.
Must say that as a manager I'd be less fearful of doing that with a MS system than most open source.
Thus whatever solution you propose to your management, it needs to address risks, support and costs. Risks are things like not having a dependance upon one person, costs have structure and uncertainty, and support is often a function of the relationships you can build.
DCFC The Pimp
You do not want to be seen as working for the open source community, any more than a decision maker wants to be seen as in the pockets of the hardware vendor.
Feel free to fight media with media, most of it is very O/S friendly, and even the MS centric press reports on Windows vulnerabilities pretty much daily. Also do not neglect PR here. Make sure you have some idea of what your key decision makers have on their home PCs. A quiet one to one warning that Winamp has a scary hole gains you points, and feel free to emphasise the Win bit of Winamp.
Also many managers are bovine herd animals, many of them even play golf. Saying "we're copying the strategy of Amazon", or a competitor scorese big points. Saying "we're going to pioneer Apache load balancing" would set off alarm bells.Also show your commercial skills, the game here is to push your agenda, which includes you personally.
If you show interest in O/S then Oracle, MS et al may well suddenly become more flexible on prices and licencing. When dealing with managment it's quite legitimate (and wise) to say "we pay $XXXXXXXX to MS, and although using O/S will only save $XXXXXXXX / 50 it means that when we negotiate with them they will know we can shift business away from them". Most decision makers will think better of both you and your views because of this commercial acumen. You can research tales of how the big vendors have marketing pools just for this purpose. Fact is that many non-IT managers perceive us to be in the pockets of vendors, not necessarily in a corrupt way, but worse on ideological views. A good % of us have paid money Linux/Apple/Intel t-shirts, toys et al, some of which we wear to work. I think it's a safe bet to assume that the finance director is unlikely to appear in a Price Waterhouse t-shirt. However, I must say I'm surprised at the basic issue here. Before I became a pimp I was an IT director, and non-IT management had if anything a higher impression of Linux et al than it merited for what we were doing. More than once I was specifically asked why we weren't using Linux. And yes at one point I over rode the people who wanted to deploy a horde of Linux desktops. My view of Linux is that in most practical ways the business perspective is much the same as for MS stuff. Although with my techie hat on the access to source is nice, fact is that it's usually not all that wise to screw around with it unless you know what you're doing, and most people don't. Thus with both open source and proprietary, you pay people to make it do what you want, and to nurse it when it gets sick. Thus the variable is the cost and quality of developers and support people. At the risk of being shot at, MS stuff is typically easier to drive at least 80% of the time. Ironically of course this means that anyone who can get past the occasionally quite grotesque open source interfaces is usually smarter, and often cheaper. Re-reading that, perhaps this is a candidate for the shortest sentence that will offend absoutely anyonne, but it's the way I've made decisions
To make s/w reliable you have to assume that it isn't. Although NASA lost the plot decades ago, it's worth looking at how IBM engineers built systems that could survive the failures of these chimps. IBMers assumed that everything could go wrong. There was great pride in the team when they found a stupidly obscure and massively unlikely bug. There was no notion of "having to decide whether you were an engineer or a manager". IBM stuff managed to survive things like motherboards being melted in flight with microgee solder balls shorting out the rest of the system. Even in Challenger the IT was working as it hit the water on the way down. That culture is far more important than any single technique. Implicit is the notion that no one can really check their own work. You need someone who takes a positive joy in pointing out your errors. That needs to be backed up by managment, who must not only avoid trating bug finders as trouble makers, but positively reward them, and yes that means cash. I think we can already see why so much commercial s/w has stupid levels of bugs... If you're serios then I'd suggesting voting as an architecture Multiple modules to take decisions. To do this properly, each needs to have different code. The gold standard is separate code, each in a different language written by a different programmer. The mulitple language precept is because two programmers in (say) C++ may actually make the same error from the same specification. A classic example is "between one and ten items". Different people may interprest that as including 1 and 10, or excluding them, or in the case of some C++ developers 0..9 VB programmers "know" that Dim q(10) actually has 11 elements in it. Voting puts a bound on the effect of a failure.
Although your employer is behaving badly, you have to make a business decision.
:(
If you invest $X and Y hours, what will the return be ?
To you, personally ?
The next calculation is whether you can get a better return on the investment by learning some
other skill with about the same level of up front spend.
In neither of these cases should you assume you are going to stay in the same firm. There are plenty of jobs boards out there where you can work out your market value, or you can talk to a pimp. I'd offer to help myself, but we don't do network people.
I don't see that you are under any moral obligation to spend your time & money on helping your employer undo the effect of bad decisions.
Taking what you say at face value, you're going to screw up. You are now in charge of a mildly complex system that you don't understand. Although I assume you're a competent individual in general, you are now not equipped to do your job. Bad things will happen.
You may well get blamed for these screwups, and even if you do the training, there will be a lag between study and competence. I recall the first time I had to build a LAN. Wasn't pretty or reliable and took >10 times as long to build compared to the work of someone who knew what they were doing.
(think how long it would take you to discover about crossover cables if no one had ever told you
Thus you need to start some pre-emptive blame managment.
You need to have emails saying that without proper training, you cannot guarantee the integrity of a resource that is critical to your firm.
A good next step is to say that you need to hire in a consultant because there's things you can't do.
The costs of this will be non-trivial, and inevitably a lot more than paying or training you.
Done right, they may even require that you go on training.
But however it works out, do not invest your time or money for free.
DCFC the pimp
Oh dear. Let's do some really simple economics here...
Migration moves people from where they can work cheaply, to a place where they can charge more for their labour.
These people won't stop existing if they are not allowed to migrate. They will do IT work in their home country.
Since India, China et al have lower wage rates, he will work for less.
Thus the world supply of IT people will be sold at a lower average rate.
Migration actually holds up the price of labour for those skills which are portable across countries.
It is also the case that an IT person costs quite a bit of money to build. This comes from a combination of parents and the home state.
However if they migrate, the return in the form of taxes and the general economic use of their work is for the benefit of the country that he now lives in. In effect countries that do not retain their skilled people are transferring wealth to those that do.
Also, at the risk of political incorrectness. Look at economic migrants in terms of their quality. Those who have been born with some grave issue affecting their physical health or general intelligence, find it much harder to move.
Thus they are on average better people to parent the next generation of citizens, and it is that generation who will be working to support us when we retire.
If we look at history over the last 200 years, we can observe the fate of those countries who kept out foreigners.
China, which was abjectly garbage. It's current growth is based upon low wages. Crap pay is a sign of past economic failure, not success.
Other socialist countries like Russia and E.Europe kept people out for a long time. Anyone want to emulate them in any way whatsoever ?
Japan kept people out, stayed backward until the Americans forced it open.
Conversely look at America, Oz, NZ et al.
Easy.
Fact is that it's now a global market for labour. When I left university having done computers, IT people were so rare that if 2 strangers in this area bumped into each other they often struck up conversation because they felt part of a small group. I guess there's more than 80 million full time IT people in the world. In the early 80s, most people on Earth worked in mud hut or gulag economies. Effective world work force of about 250 million, in W.Europe, N.America, and various fragments of the former British empire, like Oz, HK, Nz, and Singapore. Now there's nearer two billion people in the world economy, and growing.
Lots more jobs, but you have to make sure you are equipped to compete.
If I understand your process, you have some poor guy whose job is to handle a horde of little changes. Very hard for someone to keep track of these things to work out their overall consequences; it should be automated. But the implication is that it's not and I'm guessing he uses his wit and wisdom to spot obvious bad things. Worse, you may have a team following someone else's rules which they neither understand nor care about. Also a helpdesk team finds it harder than an individual to keep a view of the overall system. It's easy to see that going wrong by simple human error, which is more common when you have to have a relatively smart person doing a job that 99% of the time is dull. I bet there's more staff turnover in this role, eventually of course the point of stabiity is when you end up with someone who can't get a better job, hard to see that as good. How do you keep them sharp for the It seems to me that a rules based system could capture the essence of this job, and something that will make your security audit peple happy is that the software can be tested. I wonder if there exists such a tool already ? (feel free to reduce my ignorance here) It strikes me as a rather good candidate for an open source collaboration.
Another reason is that being a programmer is like placing bets on your future. Soonmer or later you make a crap one, and it all goes wrong. I was lucky to do C/Unix at college, did OS/2 for a while which was good whilst it lasted, but suddenly went down the drain. C++ and VB were good choices, but I missed the boat on the easy money for SAP and Powerbuilder. Should have spent more effort on SQL, and wasted less time on Lisp. This game is very buzzword driven. Doubt there's anything that bleeps that I can't program given a manual, and I'm far from unique in this. But it's hard to change track to a more fashionable technology, since many pimps and HRs just see the words, not the underlying skills. After 20 years of programming and a bit of bossing programmers around I found I could make more money and work sensible hours by becoming a headhunter. Some people prefer dealing with a pimp who doesn't think Iterators are the evil mechanoids from Stargate SG1, we also can toake job specs as informal as "a bit like Steve, but maybe a little more maths". I still do a bit of coding, and our most successful marketing tool has been my "C++ for interviews" document where I go through several dozen commonly asked interview questions, and answer them. A couple of banks have acquired copies, merely so they can upgrade their interview process :)
Management like "team players", and Indians often have this attribute, if you define it as not ever arguing with the boss. I've managed very good programmers and they can be a real pain. I would not characterise Indians as "yes men", but some don't warn me when I'm going to do something stupid. A good employee can warn his boss of impending doom, without it being confrontational, so the optimum is somewhere in between.
Also, many pimps and HR types do not like to hire people older than the boss of that area.
This is bollocks of course. As a former IT manager I've hired people who were older because my success was dependant upon the quality of my team, and frankly I don't give a toss about people's age provided they don't die at their desks.
Without exception, the managers I've spoken with, have pretty much the same view. They will indicate that a given job is for a newbie, but if they can do the work they don't care.
What you want is Markov chains. For any given statement you can easily construct the set of statmenets that follow it, together with their number. We will observe that given for There are a lot of branches to the tree, with (i =0; being very common and // Woo woo
being quite rare.
We can easily generate more code by picking statements at random, but in proportion to their observed frequency.
But we can do better.
Given pairs of statements, we can generate a table where given the pair of statements
for( i=0;
we know the probability of
i !=
This because we're dealing with pairs from a large set, the branching will be quite low, ie it will be much more constrained, and thus a lot more like the original code.
This is analagous to being able to say in English, given "Space, the"
we mostly get "final" as the most probable next symbol, and now we have "the final", the next word wmay be|"frontier" but given that "Space" is not being thought about any more, we might have "cut".
Text generated like this is quite like English, and if you include punctuation, can generate things that are as grammatically correct as many people's posting. It can be quite funny. I have used it to merge Terry Pratchett and Microsoft marketing blurb, and certainly the MS stuff improves...
Same applies to C++.
It will take a little experimentation to find out the best length for the chain, but assuming that the input code is syntactically correct, so the output will often compile.
Quite what it will do is another question, but this sort of chain is rather like how people learn.
If you listen to babies, they start off making random noises, then they make sounds that have very roughly the same frequency of ocurrence as their parents language, then they burble in things that sound like English, but clearly are not. Their "singing" at 18 month is usually quite free of real words, but they've heard "twinkle twinlke little star" so often that the chain of "twinnle twinnle lipple sarr" is carved into their neural net.
Given that the spec here is for "useful" I propose this as an AI test.