Re: Ho hum
>Resilience costs money.
And it is grossly inefficient. Right up to the point when you need it.
IT is often way too focused on efficiency.
4748 posts • joined 4 Dec 2007
>Resilience costs money.
And it is grossly inefficient. Right up to the point when you need it.
IT is often way too focused on efficiency.
>No, but my laptop, if you're including that in devices, is often displaying a document I'm working on and one of the documents I'm working from side by side.
Better yet, a system with a couple of 24" screens. Even with a laptop, having a desk to work on is optimal. Fewer devices means a clearer desk, not no desk.
Having a printer/multi-page scanner is also not dependent on the pc/tablet debate.
My desktop sits under the drawers of my desk. Assuming the desk is required, even a mid-tower pc can have zero additional footprint.
>The best way to fight extremists and their supporters is not to ban them and censor them ... but to drag them out into the daylight and force them to discuss their views with people outside their own little circle of haters.
Oh for a thousand upvotes to give.
Ok, maybe dragging them out forcibly is the wrong approach, but why not encourage discussion of various values? Our increasingly PC legal and social environment is so preoccupied with making sure someone's feelings are not upset that it forces the disengagement of those with less fashionable views from the rest of society. Forced to keep their beliefs locked away, they fester in the dark, reinforced by people who may be less extreme, but also have to insulate their beliefs from the tempering influence of the outside world.
This appears to be driven mostly be secularists trying to eradicate religion, sometimes overtly, sometimes under the guise of "equality for all". It is a misguided strategy. What we need to be able to do is to defend our values and our positions. We need not just dialogue, but debate. We need to evaluate Islamic, Christian, Jewish, Buddhist, Hindu, New Age and Secular positions and apply some critical thinking. Rather than name-calling, we should encourage people to elucidate why their position is the best and should be adopted. "Should we punch a Nazi?" "Should we eat pork?" Sometimes the answer will be, "because X said so," an we can give that the attention it deserves, but we should encourage, not stifle the debate. Name-calling and trying to exclude people by "no-platforming" is a sign of a cause without reason.
Sadly I don't think it will ever happen. Increasingly the debate platforms are advertising-funded and they thrive on eternal conflict and true-believers. As the protagonists become professional and/or commercial, their economic interests become aligned with with "keeping the conflict going." Even when they've won, they can't let the issues go.
The aim should be to win your opponent to your position, not to try to demonstrate a third party that you are more intelligent or pious than the other man. What we need is increased tolerance - the ability to disagree with people without trying to destroy them just because they hold "sub-optimal" beliefs.
An assault is carried out by a threat of bodily harm coupled with an apparent, present ability to cause the harm. It is both a crime and a tort and, therefore, may result in either criminal or civil liability. Generally, the common law definition is the same in criminal and Tort Law. There is, however, an additional Criminal Law category of assault consisting of an attempted but unsuccessful Battery.
Saying mean stuff is not assault, whatever you may feel.
I would be the last one to defend hateful speech, but the scope for abuse of such legislation is massive. We can't always safely legislate away everything bad.
How do we decide what is "fake news"? Who is going to do the research to determine the truth? How long will it take to sort out the arguments? Who foots the bill?
Machines are really good at working on a narrowly defined problem. The classic example is chess. A massive possible data set, but the rules for what can be (how pieces move) are very strict and very few. There aren't that many areas where that kind problem domain and definition holds true.
The problem with AI is that the promise of AI is that you can replace human analysis with machines. Unfortunately machines can easily miss what humans may not, because humans have real intelligence and intuition. How will AI end up being used - to augment humans and do what they can't, or to try to replace humans and so end up losing the more intelligent part of the team?
>who needs "the store" when you have your own web site already?
People who want to sell to people running Windows S?
>Importantly, it will cover both indirect communication and the creation of fake accounts.
I think the accounts are real, the data provided by the creators just isn't what the provider can effectively turn to cash.
In other news, freely available stuff on the internet gets gamed!
Next you'll suggest putting an AI chat bot online.
As for revenge porn. Stupid laws to protect stupid people. For all the talk of the age of reason, teaching science in schools and rational humanism, the concept of cause and effect seems to have eluded a lot of people. The scope for abuse far exceeds the good these laws can do. I'm sure there are worthy edge cases, but don't go there. Life is harsh, take the precautionary principle.
Except on NCIS, where you can even find the Less Spotted Windows Phone in its last remaining natural habitat - the sponsored pitch.
"Let me Bing that for you."
>The NHS IT is run on a very constrained budget, and what you're suggesting needs someone to look at what's on the network, and work out a plan for sorting it out.
The incompetence doesn't necessarily refer to the techies, but would more properly be assigned to those who made the decision that funds would be better allocated on something other than security.
Now where is their cost saving?
>So what happens when a patch for a Cat I vulnerability broke something critical in the process, creating a dilemma because the critical machine was inoperable either way?
You reduce the attack surface by making sure critical systems are segregated and get extra protection. For example, you don't run web browsers or email clients on critical servers. Maybe you don't map drives to large swathes of critical files, make sure write access is only granted to those who really need it; maybe provide terminal server access for things which are important, so you can control the environment more easily.
Standard security precautions really. How much have you saved by not hiring a security team? Are you sure you've saved money?
While I like the reasoning, does it hold up?
Are uber drivers beholden to uber or can they also work for lyft? (I don't know) If they can, then they would be economically independent from Uber.
"the provider supplies the whole service... so that the two services form an inseparable whole..." is that different from any employment agency going out to market via its website and saying, "I know of a network admin you can have at $x/day. Let me know if you want me to send him for an interview and I'll take a cut of the final payments"?
Its late and I haven't thought it through, but what seem obvious doesn't always turn out to be so.
>With the disclaimer that I may be tripping up on semantics, how is that not also a 'client and server' model?
It is the terminal model. Client server puts logic on the client and data on the server. Endpoints these days just display the data and take user input. You wouldn't have a phone executing SQL queries across the wire(less).
In other news, "Pushing people's buttons generates page impressions."
And later, "Why the 'targeted advertising' model might be driving partisanship."
Ah, Facebook - that well-known seething pit of pro-Trump, pro-Brexit alt-rightism!
I'd be curious to know how many people switched their opinions in the run-up to the election, but couldn't say why. My guess would be almost no-one. I doubt even the explicit arguments made by the campaigns influenced many people. People already had their opinions and kept them.
I find it slightly amusing that the guardian, so much in favour of having our lives governed by unelected foreigners should write such a lengthy expose on a conspiracy regarding how our lives are being run by, er, unelected foreigners. Whether you think that is good, bad or indifferent, the lack of awareness speaks of a complete irony detection failure or extreme partisanship, turning the news into the image of American news.
>Can we just, finally, let it die and put fiber in the ground.
Is it just the Oz press getting the wrong end of the stick based on DSL being a long(ish) range tech and this is by the same person?
Surely this is more for "inside-the-data-centre" where the exponential drop-off isn't an issue?
>I'd dispute "modern" and "philosophy"
My observation is that most of the problems lie in GUI software.
The first is "blob" input. When input is text, parsing and checking is relatively simple and the dodgy inputs are not expected to get past the person entering the data. Passing a pointer to spreadsheet object leaves the receiver wide open to a great deal more abuse.
Where I see a great failure on modern OS design is the inability to set resource capabilities at run-time from the OS. MSOffice's "protected view" etc is all well and good, but how about telling the OS to open execute Excel without network capabilities and disk write capabilities except to a quick-provisioned RAM disk for it and its sub-processes?
That would kill attack vectors dead, even if the application is vulnerable to corruption.
>trying to predict the future using the past.
Worse. Humans can come to recognise bias when its bad because we have morality.
The whole purpose of AI is to implement bias and it has no morality.
AI isn't intelligence, it is merely statistical analysis. Complicated stats, but still just stats. It can make correlations, but not assess causation. It is incredibly dumb.
Stupid and morality-free. It is only useful when you really don't care too much about the outcome.
If AI is used to determine if you get parole, it is because the judicial system doesn't really care about the outcome, only that the process is cheap. I think its fairly easy to assess the morality of those who commissioned that.
What happens when everyone moves to AI and we no longer have humans doing the job? Where do you get your training data? No-one can tell how decisions are made and there's no way of measuring quality of the output. How do you know if someone has found an algorithmic flaw and is exploiting it? How do you catch the outliers?
AI has a place, but there are dangers. One that immediately comes to mind is that we try to do too much and it pushes policy to places we shouldn't go. Why check only the fingerprints of criminals when you can check everyone's? What happens when systems trained on a little data from California are exported to Kenya? Does anyone know? What happens if you go back the other way - take your data from Kenya and use it in California?
Part of the problem is that the vendors hawking AI systems have no vested interest in their correct use. Problems are compounded when those buying and using such systems don't have too much interest in the outcome either, just as long as an outcome is reached and more cheaply than a human could do it.
>Quite a few of us realised this up front.
Techies perhaps. Everyone else just thinks it is web-based email. The seductiveness of receiving email was documented way back by "You've Got Mail." The more you share, the more "mail" you get. The larger the social network, the more constant the stream of affirmation.
It is a sad reflection of where people get their self-worth, even before you get to the privacy invasion.
Hence skype is "Cloud" not p2p anymore. MS has too much to lose if leaned upon.
So no large telcoms provider can provide e2e encryption. That means you have to do your own. I'm not sure that changes much, if you are at all interested in privacy.
In short, you probably shouldn't trust anyone with a significant amount of money to lose from non-compliance. It doesn't matter how much encryption your application does if the OS taps the microphone.
Commercial products have been hacked together bits of code for years.
Ah pity da fool who thinks otherwahse
Icon: Close, but no cigar
>Where's the coup's? Where's the bombs?
The Crimea? Ukraine?
I suspect they laugh at the US' ham-fisted and massively ill-judged attempts at controlling the middle east.
But in this case, yes, I think its just for the lulz.
>It's an orchestrated campaign by Russia
So much concern about the actors and so little concern with the content. Are we really complaining about greater (if still imperfect) transparency for political parties? It is a good thing to find out their inner motivations before you let them into power.
What we should be asking is, "When we know what these people are like, do we still want them in office?"
Unless there is something really damaging in the dump (in which case they deserve to be found out) then its really too late for Le Pen to do anything with the information apart from laughing at their security. I don't care who did it or what their motivations were. If it does tilt the election and undo the 20 point advantage, that indicates that the French people happen to agree with the Russians about the candidates. There's nothing wrong with that.
This assumes that the dump is real and has not been tampered with. Then we're in to a whole different ball game. However, we still have to respect democracy. If people want to believe what they see on the internet more than what they hear the politicians say, then perhaps that says something about our politicians. Maybe politicians need to build an honourable reputation for themselves such that people dismiss lies about them because they have a proven track record of honesty and integrity. Perhaps if you live by lies and half-truths, you die by lies and half-truths.
What we want is a duress PIN. Wipes local data.
Er, sorry officer I got them mixed up.
>Windows users uses it because they can find the tools they need
In the MS App store? Really?
Ok, I agree with much of your sentiment but when you look at how MS, Oracle, Apple, the security vendors etc behave its no wonder no-one wants to play with them any more. They are likely to stab you in the back and take your toys.
Google gets a semi-pass because although they are horribly creepy, they make their money by giving you stuff for free which makes life easier. The others could make your life easy - OS development is a sunk cost - but choose not to. See people turn on Google too when they start restricting Android patches and so on to force upgrades.
It used to be the case that MS made money by releasing a new OS with new and better features. Now they've released a new OS Which Does Less! YAY! Or not.
Linux, on the other hand, adds features to each new release. You may not like systemd but there is no denying that it is a technical/feature improvement in some way, even if you don't need or want that particular trade-off.
>Can it run all of my WIN32 software?
I think the article said, "Only if its in the Windows Application store."
My 130w 3930k cores sit at 35-45 C but spikes to around 50 C under load when transcoding or something like that. Go-go water cooling!
It would sit at 60C before the water cooling and I was concerned, but that could have been a poorly installed fan.
It also returns a 20% higher cpubenchmark average than the (4 generations later) i7-7700 <pointlessly-smug/>
I have yet to understand why Windows Update uses up so much CPU and takes so very, very long to think about the problem of patches.
But then, I don't run it for personal use, so its all billable time :)
>Chipzilla also wants that branding to reach the names that clouds apply to their servers-for-hire so that users can more easily understand the performance offered by different cloudy VMs
Isn't VM performance dependent on contention?
>Until then, they are toys for the curious.
Voice input is just generally bad. Even if it worked perfectly, it is *slow* ... and it doesn't work perfectly. Would you rather navigate your bank's voice prompts or sit in front of computer and click/type through the login sequence?
Voice recognition is best when you've already decided on voice interaction (calling the bank) but the bank is desperate not to pay for someone to talk to you. It could also work for car-based SMS systems. As a general interface, however, its pretty poor.
>It's also to administer and not to administrate.
Perhaps the author was thinking ahead.
"What should I do with this MS system? I'll defenestrate it!"
>They replaced it with the even lighter MacBook.
I'm not sure that's a replacement. Replacing an i7 with SD card, thunderbolt and USB ports and maglink power with a system with a rubbish cpu with a single USB/charging combo?
Everything is worse than it used to be. Apple have exited sanity, never mind education.
Maybe it isn't about the porn, but about getting another datapoint on people to help with mass surveillance.
Or maybe its just morality theatre.
Except that it does actually work... which surprised me.
I had a friend who advertised a new product through fb with particular demographics specified. They got increased sales from the campaign and given that it was pretty much their only marketing, its reasonably safe to say it worked.
Sure, the response rate was tiny compared to the reach, but that really doesn't matter. It only takes a vary small percentage of people to be interested to make it worthwhile and it was far, far cheaper than a magazine advert which would also be mostly "wasted." That's why we have a couple of very large companies - as an advertiser you want maximum coverage, knowing full well that almost all of the advertising will be wasted. But "wasted" is just a cost of doing business. The advert buyers know its mostly wasted, just like it is with billboards and TV. That doesn't really matter.
I'm less concerned with adverts per se, than the really annoying ones which expand to cover large parts of the the screen, but that isn't just adverts, I hate menus on websites which do the same thing.
>So guided missiles are low tech now?
How else would you describe a book called "How to Throw a Rock"?
>YouTube has adverts? Since when?
I leave the adverts running on channels I want to support. Generally on mute in the background though and what I'm watching generally runs from 40 minutes upwards so it is a relatively short amount of time.
Watch out El Reg! It seems the home office is after your tagline!
>whilst on the other side we have tiny boutique operations like, err, Google...
Two wrongs do not make a right. We should be opposing all abuses of power, not excusing them on the basis that google is large and politically powerful too.
I'm not giving anyone free access to my personal life. My diary and phone is full of cryptic and re-used appointment and alarm text. Internet browsing uses FF because of noscript and some trust. Corporate intranet often gets chrome because of rubbish apps, process isolation and zero personal care factor on slurping corporate data. IE and Edge I find just... ugly whitespace hogs and are a last resort on the corporate webapp side of things. I don't care if the browser rendering is a little unexpected, I don't care much about the speed. The main care factors are privacy, compatibility and efficient screen usage.
And did the article author really resurrect the 90's idea that its a good to give your web browser r/w URL access to the file system? Does he remember what data MS' browsers send to the cloud on Windows 10? Does he want to include all local file access in that list? Not that I care. I've got nothing personal under Windows, All my stuff is under Suse.
Fun fact: HMS Victory was never decommissioned and is still on the "active" list.
The active crew is 600 - what's a couple of zero's between friends? A third of the crew will be required to weigh anchor before you get started. You may be better of cutting and running!
>LibreOffice is your friend.
Unless you need Visio.
Or a mail system which allows delegation.
LibreOffice is great, but we still have more work to do.
>I don't think they mean actual fingerprints from fingers, more a "fingerprint" as in a profile of the user
True, but if you wipe the door handle between customers, you can probably pick up their fingerprints too.
Only an authenticated user can run this attack?
>it is important for chatbots and intelligent assistants to be able to understand when a person is being sarcastic
Who said Americans don't understand sarcasm?
There will be cake when you get to the other side.
>What on earth do these chips have that is necessary for (ARM) Windows 10, that their previous chips don't?
x86 emulation? So you can run Windows' massive legacy software base.
MS was stuffed by Intel's exit from the mobile market. Without their legacy software MS has little to offer.
Intel's problem is that they don't want to do cheap. Why create a cheap Atom system when an i5 is so much more lucrative? MS' problem is that most users don't want underpowered Windows and those that do, don't want to pay full-fat licensing. It is a lethal combination for any mobile Windows solution. Essentially MS will have to give mobile licenses away as a bundle with something else. Hence... Cloud. With cloud the low-power cpu issue goes away and you can tie licenses to named users, preventing mobile licenses from being used simultaneously with desktop ones by different users.
>Forget the fact that these might be old versions of the OS. Forget the fact that they're not patched. Why are these idiots exposing SMB services on the open net?
Realistically, if SMB is open on the net, they've already been pwned. Its just now that the script kiddies have got hold of the machines and are making it obvious.
>hat low-end machine is vastly more powerful than my desktop 10 years ago, so there's no excuse for not running a fullfat OS.
And did I read the spec's correctly, the 64 bit version gets double the RAM?
That implies it's 32bit by default!
As in, it isn't a guard dog, it just sits there an watches things?
Biting the hand that feeds IT © 1998–2017