Re: Maybe, Just maybe...
Going by the voting, half the readers didn't get your joke.
Posts by handleoclast
1287 publicly visible posts • joined 6 Jan 2012
No laptop ban on Euro flights to US... yet
The real battle of Android's future – who controls the updates
AI smut-finder is now an Android app
Three home security systems found to be vulnerable – if hackers were hiding in bushes
Thursday 18th May 2017 19:59 GMT 
Fermi Paradox Solved
Fermi posed a (so-called) paradox: "Where are they?" Meaning if life is common in the universe why haven't alien civilizations made themselves known to us one way or another?
One possibility is that alien civilizations destroy themselves before they become visible to us. I believe I have identified the mechanism.
1) Civilization invents a planet-wide networking system.
2) Civilization hooks up essential services to planet-wide networking system.
3) Civilization invents Network Of Shit technology.
4) NoS hacking effectively wipes out planet-wide network with DDoS attacks.
5) Essential services, reliant on planet-wide network, fail.
6) Civilization collapses.
Windows 10: Triumphs and tragedies from Microsoft Build
Bloke charged under UK terror law for refusing to cough up passwords
Thursday 18th May 2017 13:54 GMT
Re: The USA Equivalent Situation...
Close, but no cigar.
You're talking about the good old days, back before the war on terra.
What has happened in the US since the war on terra is the abuse of contempt of court. The judge orders you to reveal your password. You refuse, saying you forgot it (or some other reasonable excuse). The judge says he doesn't believe you and that if you don't reveal your password he'll give you 3 months for contempt of court. It's not a matter of incriminating yourself because the password is not incriminating, per se any more than the key to a safe-deposit box is incriminating. It's what they give access to that's incriminating, but you yourself are not accessing those things. Yes, that's parsing words in an effort to bypass the spirit of the constitution, but that's what they did.
Some sane people argued that this was an abuse. That if you permitted things like that the next thing you know the judge would compel you to juggle running chainsaws whilst riding a unicycle across a tightrope stretched over Niagara Falls any time he wanted an excuse to imprison you just because. The insane people won.
Over here in the UK, the sane people won the battle but lost the war. The UK used to use the contempt of court dodge. After protests over the contempt of court dodge, Part III, section 49 of RIPA was activated. Bugger protection from self-incrimination.
Thursday 18th May 2017 13:43 GMT
Re: Life sentence
So they have to use Part III, Section 49 of RIPA instead of the anti-terrorism legislation they used the first time around.
What practical difference does that make to the guy just being released? "Oh, this time I'm in gaol because of RIPA rather than anti-terrorism legislation." I doubt this pedantry will offer much in the way of comfort.
Whichever way you cut it, don't forget your password and don't use hidden-volume encyption. If possible, try to avoid being black.
Thursday 18th May 2017 10:37 GMT 
Life sentence
I've said it before, and I'll say it again. This can easily turn into a life sentence. In practise it probably won't, but it could.
Yeah, it's 3 months in prison. And when he's released, plod is waiting outside the gates to ask him his password. If he refuses, it's another 3 months. Rinse, wash, repeat.
I can understand the thinking. Without the sentencing power, Osama bin Glitter with his suicide vest of kiddy porn could just claim he's forgotten his password and there's nothing to hold him on. Of course, on the terraist side a life sentence is probably preferable to giving up details of your fellow freedom fighters and their plots. On the kiddy fiddler side, a suspected kiddy fiddler probably won't fare any better than a convicted kiddy fiddler in gaol.
The problem comes when you really do forget your password. And your phone records show that you frequently call Mr X, because like you he's a cricket fan. You met Mr X because he works at your favourite curry house. His brother goes to a mosque with a dodgy imam. 3 degrees of Kevin "I'm a muslim and don't eat" Bacon and you're under suspicion. So they ask for your phone password. And passwords for any encrypted files you may have. So don't forget your password.
Oh, and don't use TrueCrypt/VeraCrypt or any other encryption scheme with hidden volumes. The police can't prove you're using a hidden volume, but you can't prove that you're not (so you can't prove you're not refusing to reveal the password to the hidden volume you don't have). Years ago you could get around that one by actually using a hidden volume whether you needed to or not, then hand both passwords to plod. Then came a patch to permit nested hidden volumes to any level, and that means you can never prove that you've revealed all the layers.
Dell BIOS update borks PCs
Like a celeb going bonkers with botox, Google injects 'AI' into anything it can
Wednesday 17th May 2017 21:59 GMT
Another thing to delete from your phone
Or put a bit of tape over the camera lens.
Because, like google's voice-activated assistant, it will be bloody hard to completely get rid of it. And if you don't get rid of it I'm damned sure it will surreptitiously peek with your cam all the damned time (if it doesn't now, give it a few iterations). There will be times when you really, really do not want that to happen.
It was only a few hours ago, noticing the cam in a newly-acquired, cast-off laptop peering at me that I reached for the tape. After taping it, I had the thought that it would be great if there were a sliding shutter on laptop cams. Now I see it being a good idea on phones too.
Mechanically, I imagine a snap-action mechanism that causes it to be held either open or closed, as desired. Force is required to move it from either stable state to the other. Much better than relying on friction.
It needn't be integral to the laptop or phone. A glue-on device would also work. Integral is better because it would likely have a better form factor.
I hereby put this idea in the public domain. Anybody can implement it with my blessing.
Police anti-ransomware warning is hotlinked to 'ransomware.pdf'
Wednesday 17th May 2017 14:09 GMT 
How things have changed
The advice about e-mail safety these days is very different from what it used to be back when I first got on the net (dinosaurs still walked the earth in those days).
In the early days of the intertoobz, people were told it was perfectly safe to open attachments that were images, it was only executables they had to worry about.
It wasn't so long before that advice was rescinded, thanks to Microsoft "helpfully" hiding file extensions. So that "x.gif.exe" was described as "x.gif" (but had a strange icon). The advice became "make sure it's really an image before you open it."
That was superseded by "Microsoft's buggy image-handling routines mean it's no longer safe to open image attachments even if they really are images."
Then spammed malware became common so advice was amended to include not opening attachments from unknown sources.
Then malware writers clued up and started going through the user's address books. So you'd get infected spam mail from people you knew. So only opening mail from known sources was no longer a defence.
But things have changed:
Always use caution when opening (such as by double-clicking) files that come from someone you do not know
There ya go. This problem has been solved without me realizing it. Something (I don't know what) now makes it perfectly safe to open attachments from people you know. It's only attachments from people you don't know that are unsafe.
Now I can open that mail from my mate Tom, the one titled "Look at these pics of what I got up to when I was drunk last Saturday." It's perfectly safe. It's not just El Reg telling me this, it's almost every news site on teh interwebz. New technology, released without announcement, now means it's perfectly safe to open attachments as long as they're from people you know.
Before everyone downthumbs me, I did see the bit about "or if you were not expecting them." The thing is, I often get unexpected mail from friends because they're eager to tell me about something I didn't even know happened. Such mail may contain attachments. And if the text is brief enough and generic enough, it will seem genuine to a large proportion of people.
So the real advice is "Don't open attachments. Ever." Oh, but there was that bug in some mail readers where merely opening the mail passed attachments through to rendering s/w so you could see a thumbnail of it, and the rendering s/w was flawed. So the real advice is "Don't read mail. Ever." Oh, but wasn't there a mail reader that would pre-generate the thumbnail before you opened the mail, and that had a flaw? So the real advice is "Don't fire up your mail reader. Ever."
Toshiba draws back from fab foundry lock-out foolishness
Flying robots are great... until they meet flying humans, anyway
Management's agile, digital (insert buzzword here) strategy ossifying? Blame the Red Queen
China staggering under WannaCrypt outbreak
Tuesday 16th May 2017 12:51 GMT
Re: Nervous criminals
To summarize:
1) China and Russia got hit hard.
2) China and Russia play hard ball.
3) Fingerprints point at the Norks.
Which got me thinking. What if somebody was upset with the Norks but couldn't deal with them as he'd like to because it would upset the Chinese? Releasing something like WannaCry with Nork fingerprints on it would solve his problems. The Chinese wouldn't object to fatty-boy getting assassinated because they'd be the ones doing the assassinating.
The Trump just isn't smart enough to come up with a plan like that. But the CIA are.
Behold, auto-completing Android bug reports – because you're not very thorough
Monday 15th May 2017 10:11 GMT 
Mixed blessings
Google, autocomplete, bug reports, and mixed blessings. Hmmm, that reminds me of something I discovered a week ago, reported to the android team and was told it was a browser problem.
On an android phone...
1) Fire up chrome.
2) Go to mail.google.com (if you don't have a gmail account this will have to remain a thought experiment for you).
3) Sign in. Sign out. Shake it all about. Repeat a few times.
4) Go to mail.google.com. Note that it helpfully knows who you are (not so helpful for you if somebody has nicked your phone, because now they know your e-mail address) but don't sign in (yet). Well, there is a "sign out completely" option, but how many people bother with that? Maybe a few more now I've posted this.
5) Click on the "show password" icon (the eye with a line through it).
6) Now enter your password. Note how the autocomplete shows up. Watch, as you enter more characters of your password, how the autocomplete homes in on your password. If you have non-alpha characters near the start of your password the full password shows up in the autocomplete very quickly.
Whoops! The sign-in not only gives away your account's username (if you can't be arsed to sign out completely each time), it drastically reduces the entropy of your password. Paradoxically, the stronger your password (with several non-alpha characters) the more entropy it discards.
Very secure.
I wonder if there's an API for apps to interrogate the autocomplete dictionary. If there is, a malicious app (which could pass all playstore security tests) could drastically reduce the search space for your password. Especially if you've used a good password. How often does "k*hr}39rq" occur in ordinary text? Damn, now I have to change my gmail password.
Firefox is worse. Click on the "show password" icon and it shows your full password in a browser-generated autocomplete box (as well as popping up the soft keyboard autocomplete) without you having to type anything. Assuming you've previously logged into gmail using firefox, of course.
The android team tell me this is a browser issue, not an android one. I've repeated this test with IE on Linux and Windows, Edge on Windows, Chrome on Linux and Windows, and Konqueror on Linux and none of them exhibit this flaw. None of them exhibit this flaw because the Linux and Windows systems I tried it on have real keyboards, not a soft keyboard with autocomplete.
Since the android team tells me it's not an android problem, there's no reason why I shouldn't post the details here. At least that way some people get to learn there's a security hole in the thing. If any of you can be arsed to contact every browser app for android (there are a lot of them) then be my guest.
I expect responses telling me I'm an idiot and it really is a browser flaw. Or a gmail sign-in page flaw. Maybe you're right. After all, it's not like android, gmail and chrome are all produced by the same company whose development teams ought to have lines of internal communication unavailable to the rest of us. It's unreasonable of me to expect that informing one member of the triad of such a problem would result in them communicating internally to figure out what they could do to resolve a problem which seems to arise because of the interaction of all three of their products.
It's a good job I use the lock screen and have a decent password on that. The short time-out I set can make it a real pain to use sometimes, but it's necessary.
Warm, wet, mysterious... sound familiar? Ah, yes, you've heard of this second Neptune, too
Sunday 14th May 2017 13:32 GMT
Re: To a first approximation
To a first approximation, if you think that first approximations should lead to a true statement then your thinking is flawed.
All scientific understanding is an approximation. Sometimes a very good one. Sometimes one that does not apply in rare circumstances. Newtonian gravity and mechanics is a very good approximation that fails in some circumstances. If you're designing particle accelerators or navigation satellites then you need Einsteinian relativity, otherwise Newton does just fine. Even with theories which give very good results and have no known exceptions we cannot prove (or even know) that they're true because new evidence or thinking could cause them to be supplanted.
In the case of first-order approximations all we ask is that they provide answers that are close enough to be useful. Often all they have to do is let you decide if something might be feasible (so try a more accurate approximation) or clearly impossible (so no need to do all the extra calculations).
You're right that a lot of the elements important to us are not metals. That doesn't alter the fact that most of the universe is hydrogen and helium. From periodictable.com's abundances we get hydrogen 75%, helium 23%. All the others add up to 2%. So, to a first approximation, the universe is hydrogen and helium.
Excluding hydrogen and helium, there are 15 non-metals in the periodic table and 7 metalloids. If you class metalloids as non-metals that means 70 out of 92 elements are metals. Class metalloids as metals and that means 77 out of 92 elements are metals. English doesn't have a word meaning "every element but hydrogen and helium" so when you're looking at stellar spectra to see what's in stars, after you exclude hydrogen and helium lines then most of the lines you see are due to metals. This is expecially true because elements with high atomic numbers are metals and have large numbers of transition levels. What you're going to see in a star's spectrum is mostly hydrogen, helium and metals.
It's only when you look at the second-order approximation that things change. Oxygen, carbon and neon are most abundant after helium. Iron after them. Then silicon (metalloid) and magnesium. Then sulfur and argon. Etc.
And then you get to the third-order approximation. During planet formation, a lot of the lighter gases were driven away from the inner accretion disc by the solar wind, so our planet is has a massive iron (relative abundance in universe 0.11%) and nickel (0.006%) core.
To a first approximation, the universe is hydrogen, helium and metals. Especially if you spend a lot of time studying stellar spectra. A geologist would give you a different answer for planet earth.
Saturday 13th May 2017 10:46 GMT
To a first approximation
To a first approximation, every every element in the periodic table is a metal (the vast majority of elements are metals).
To a first approximation, everything in the universe is either hydrogen or helium (the vast majority of the universe is one or other of those).
Therefore, to a first approximation, everything in the universe that isn't hydrogen or helium is a metal.
You have to use a second approximation to start considering things like carbon, oxygen, nitrogen and chlorine.
Saturday 13th May 2017 10:41 GMT 
I'm stupid
We think that the process of planet formation puts more of the heavy elements into smaller planets
Which just goes to show how stupid I am. It's all so obvious now. There's a cosmic cake-frosting device that squirts heavy elements onto small planets. Which are lighter than heavy planets because they are composed mainly of heavy elements whilst the heavy planets are composed mainly of light elements.
I'm still a little confused, but I expect it will make even more sense as I approach a state of alternative sobriety later.
For now, GNU GPL is an enforceable contract, says US federal judge
Saturday 13th May 2017 12:27 GMT
Re: Technical Point
American law is, if anything, strong as regards "consideration." The US has a long tradition of making what are effectively gifts into contracts by stating that "X received Y for $1 and other valuable considerations."
The real point can be found in your acknowledgement that the consideration (you called it a payment) need not be of money. So you receive the GPL s/w and your consideration in return is the conditional promise that if you modify it and distribute the modified s/w you will submit the modifications back to the project. You get to use it for free (no contract). You get to modify it for your own use for free (no contract). You distribute your modifications then you have to pay by submitting your modifications back (contract).
Actually, it's a shade more complicated than that, and IANAL. But you get the idea.
Saturday 13th May 2017 12:21 GMT
Re: That's different:
That's different: That's a plain and simple copyright violation.
Not different.
Under the Berne Convention, any unique software is automatically copyright.
Therefore, downloading any software, be it Windows or be it some GPL'd s/w, is copying copyrighted material.
The holder of the copyright gets to dictate who may make copies of the work and under what circumstances. Microsoft says you can copy their stuff if you hand over some wonga. Other s/w may invoke different terms, such as the GPL.
The GPL says you are entitled to make a copy provided you comply with certain conditions. If you make a copy but do not comply with those conditions then you have violated copyright. You have made a copy under conditions which the copyright holder explicitly states to be impermissible.
Which bit of this is hard to understand?
Amazon's Alexa is worst receptionist ever: Crazy exes, stalkers' calls put through automatically
Saturday 13th May 2017 10:07 GMT 
Fail... ail... ail...
Do I hear an echo of "fail"? I surely do.
I have installed and then deleted quite a few apps on my android phone. They look good, but when I try them I discover they don't do what I thought, or are flawed in some way. The point is that I can delete them. If I can't then they are malicious apps.
So once you've installed this Alexa Calling, you can't uninstall it except by calling support? That is totally unfuckingacceptable (forgive my tmesis). Are they channelling Microsoft?
I'd already decided, for several reasons, that echo was not something I would ever wish to use. I would have said yesterday that it was possible (but unlikely) that Amazon might modify echo to the point where I would reconsider that decision. I have now concluded that there is no fucking way I will ever use echo, no matter how much they improve it.
Oracle crushed in defeat as Java world votes 'No' to modular overhaul
Friday 12th May 2017 20:34 GMT
Re: Rules of thumb
Ooops, meant to add the qualifying term "software" in my rant. As you said, the hardware was quite good.
And you're right about Solaris too - not too bad, although it was considered to be slow enough that it was often called Slowlaris. But, like all the proprietary *nixen, it suffered from "we're going to do this our way, to differentiate us from the competition," which actually meant "we're going to fragment the market so Microsoft can dominate it." Short-term gain, long-term loss for all those proprietary *nixen.
Friday 12th May 2017 19:19 GMT
Rules of thumb
My rules of thumb are as follows. I expect many will disagree and (semi-ironically) downthumb me. Nevertheless, I continue to follow these rules of thumb because they work (for some values of work).
1) If Sun invented it, it is buggy and full of security holes. See NFS (all major versions, despite each major version being slightly less full of security holes than the one it replaced), Sun's variant of RPC, Java.
2) If Sun bought it, they broke it, at least a little. See MySQL, Star Office.
3) Oracle took whatever Sun did badly and made it far worse. See everything.
And yes, I know, the downthumbs will mainly come from Java lovers. Because Java is different (they think) from all the rest of Sun's children in that it isn't a buggy, bloated, slow pile of shit. And then they'll ooh and aah at how Java SE 10 is going to add all the things that will finally allow it to deliver all the programming advantages of OOP that the very first version was supposed to (but didn't) deliver. Oh yes, SE 10 is actually going to be useful and usable, just like the prophets foretold back in the dim history of time (back when they were talking about JDK 1.0).
O2 continues to splash out on 4G ahead of rumoured IPO
Friday 12th May 2017 19:49 GMT
O2 "bargain" deal
Coincidentally, my paper today carried an advert for a "fantastic" data package from O2.
30GB/month for the price of 10GB/month. On phones supplied by them in the first place. Which is probably why they don't bother to tell you what you'd pay for 10GB/month (from the ad it appears this varies according to unspecified factors, possibly including phase of the moon). Oh, and it ties you in to a 24-month contract. After which it reverts to their standard 30G price. This special offer is available until 28th June.
Their standard 30G price is £38/month. Really.
I have a SIM-only package from Three which gives me 30G/month. It allows tethering (the O2 package doesn't say it permits tethering, so it probably doesn't). The price was £17.50/month (less than half the O2 standard price). Except there was some sort of special offer when I got it, so I have 12 months at only £15/month. Except there was another special offer on top of that, such that after I'd paid for 3 months I got the next 3 months free. So over a year that works out at £11.25/month.
So, prior to this special offer starting and before it ends on 28th June, O2 works out more than three times what I'm paying Three. And to qualify for it I'd need a handset from O2, which would add to the cost.
I'm persuaded! I can't see understand why anybody would think O2 are cunts when they can offer fantastic deals like that.
Uber is a taxi company, not internet, European Court of Justice advised
Thursday 11th May 2017 15:01 GMT
Re: Demarchy
I had to look this one up. First result included a Wizard of ID cartoon.
Lackey: "I've just uncovered a plot to underthrow the government."
King: "You mean overthrow."
Lackey: "No, they say they can't throw it far enough."
Yes, demarchy solves some problems. Introduces others.
1) It might give better results for the country as a whole, but if it happened to pick you then would you be happy? If you ran a business, a four-year gap might screw it thoroughly. If you were in a good position in IT, the four-year gap and consequent lack of current knowledge/experience would hamper your chances of getting an equivalent position afterwards. People try to evade a few weeks of jury duty for similar reasons...
2) Foreseeing that it might be hard to get a job as good as the one you had before serving in government, would you be tempted to fuck things up so badly that there had to be a new lottery after a couple of weeks?
Well this is awkward. As Microsoft was bragging about Office at Build, Office 365 went down
Another IoT botnet has been found feasting on vulnerable IP cameras
Take a sneak peek at Google's Android replacement, Fuchsia
Wednesday 10th May 2017 12:07 GMT
Re: Now, which syllable has the emphasis in macadamia?
I must confess that is a question which has occasionally sprung to mind ever since I read of the biological naming conventions that gave rise to it. Not just which syllable has the emphasis but whether the third "a" is long or short.
This would all be so much better in ancient Hebrew or Aramaic. No characters for vowels so reading any unfamiliar written word could give a very large number of possibilities. You might think that makes matters worse, but in that situation as long as you get the consonants right then however you say it must be correct.
Wednesday 10th May 2017 12:01 GMT
Re: A wonderful name
"That 20% includes a lot of the non-English-Speaking world, the internet is global, you know."
Ummm, that 80% also includes a lot of the non-English-speaking world, too. And there is good reason to suppose that some of the non-English-speaking world (particularly those who speak German) are more likely to pronounce it correctly because "fucks ya" may not sound rude to them. Pronouncing it correctly makes it less likely for them to think it must be spelled as if it were pronounced "few sha".
"I maintain your "almost nobody in the English-speaking world" comment was a bit of a stretch."
I though it sufficiently obvious that the post was humorous that intelligent people would tolerate exaggeration for comedic effect. Obviously I was wrong about some part of that assumption.
"FWIW I did not downvote you."
FWIW, that analysis of voting motives was intended to be humorous. Or do you really also think I believe there's a commentard busily digging up his fuchsias because he doesn't want to pronounce their name correctly and cannot stand mispronouncing anything? Again, I was obviously wrong about some of my assumptions.
Tuesday 9th May 2017 21:05 GMT
Re: A wonderful name
"I don't recall ever actually seeing fuchsia misspelt."
You haven't spent enough time in usenet newsgroups then. Or tried using google's advanced search: "fuschia" 24,900,000 results; "fuchsia" 94,300,000 results. So ignoring various other mis-spellings, it's mis-spelled around 20% of the time.
"But spelling it is no different than any other word. You don't learn to spell by learning a word phonetically and then trying to figure out how it's written; you see it written, and you learn to write it the same way."
Sure, intelligent people might do it that way. Not everyone is intelligent. They've heard the word but forgotten how to spell it, if they ever knew. If they were intelligent they might check in a dictionary, but they're not intelligent. Hence "fuschia."
"So I imagine the downvotes are not for pedantry, but for fundamental rejection of the premise that almost nobody in the English-speaking world can spell "fuchsia","
I exaggerated for comedic effect. 20% is still a fuck of a lot. Really. Enough that no sane marketing person would ever consider using that as a product name. Oh, hang on, "sane" and "marketing person" - my error.
My guess is the seven (at the time of writing) downvotes came from:
1) The guy who hates it when people use the word "fuck" in a post.
2) The guy who got upset because I disparaged Microsoft and Internet Exploder.
3) The guy who has to dig up all the fuchsias in his garden because he doesn't like that he now has to call them "fucks yas" or deliberately mispronounce their name.
4) The guy who thinks everybody can spell "fuchsia" correctly. Mentioning no anonymous cowards.
5) The guy who has mis-spelled "fuchsia" his entire life and is now feeling deeply embarrassed.
6) Two people who are deeply upset at realizing google has announced a product that, when correctly pronounced, states that "Google fucks ya." They blame me for that even though I had nothing to do with naming it.
Tuesday 9th May 2017 12:19 GMT
A wonderful name
Fuchsia. What a wonderful name. So easy to spell.
Almost nobody in the English-speaking world can spell "fuchsia" because almost everybody in the English-speaking world mispronounces it. To understand why requires a little knowledge of biological nomenclature.
Biologists have the habit of naming new species after other biologists (naming a new species after oneself would be naff) by latinizing the name. Often the latinization consists of appending "ia". So there's a shrub named after Mr Banks which is called a Banksia. There's a tree named after Mr Macadam which is called a Macadamia. You get the idea.
The Fuchsia is named after the 16th-century German botanist Leonhart Fuchs. "Fuchs" is the German cognate for the English "fox." So if that botanist had instead been Leonard Fox the plant would have been called a foxia, and pronounced "focksia," not "fyosha."
But to English-speaking ears, the German "ch" (similar to the Scots and Welsh "ch") sounds all too much like "ck." So if you pronounce "fuchsia" correctly (the way the rest of the world does) it sounds (to English-speaking ears) like "fucks ya." Hence "fyoosha."
Which amused the fuck out of me when Microsoft invented "friendly color names" for Internet Exploder. They could have chosen "magenta" for the mix of saturated red + saturated blue, like the printing industry and colour TV industry had long done. And made the browser also accept "majenta" for those who were bad at spelling. But they chose "fuchsia," which almost nobody can spell. And, if you look at the flowers of a standard shrub fuchsia, you see only the red outer petal (and perhaps pink and yellow tips of stamens) but not the inner petal. The inner petal is not magenta, anyway, it's closer to indigo. So a complete fuck-up all around with that name. Why does that not surprise me?
As Kojak never said, in the X-rated episode that was never made, "who fuchsia, baby?"
Crooks can nick Brits' identities just by picking up the phone and lying
Wednesday 10th May 2017 10:55 GMT
Re: 'Security' questions?
"One of my banks allows the customer to choose their own web banking username."
I'm channelling XKCD here...
You chose "password" as your username and set "username" as your password.
The last time I was forced to choose a telephone password for something I was never going to use again, I picked "none."
FCC blames DDoS for weekend web lockout
Tuesday 9th May 2017 12:28 GMT
Obvious ploy
If we enforced net neutrality then ISPs would commit an offence if they blocked DDoS attacks (since net neutrality would require that all packets must be treated equally). Therefore net neutrality is a silly idea, as that DDoS attack on our very own site just happened, fortuitously, to demonstrate. QED.
BTW, Ivan, the money has been paid into your Swiss bank account. Thanks for the DDoS attack.
Facebook fake news: Sort it out yourself, readers
Sorry, Dave, I can't code that: AI's prejudice problem
Monday 8th May 2017 10:45 GMT
This is a known problem with a known solution
Let me summarize the problem:
1) AIs can have hidden bias caused by poor datasets and/or algorithms.
2) With certain types of algorithm, particularly neural nets, it can be impossible to figure out what rules the AI is using to reach its decision, and therefore impossible to know whether or not the decision is biased other than by statistical analysis over many trials.
Now let me summarize a parallel problem:
1) Humans can have hidden bias caused by poor teaching.
2) Humans use neural nets, so it can be impossible to figure out what rules the human is using to reach its decision, and therefore impossible to know whether or not the decision is biased other than by statistical analysis over many trials.
What's the difference between a neural-net based AI and a neural-net based human? Scale. But that only makes it harder to know what the larger neural net in a human is really doing (as opposed to analysing results).
The solution that was applied to the human problem? Procedures and rules designed to stamp out individuality (and creativity, and intelligence, and adaptability). I.e., what the civil service uses to ensure you get a consistent result no matter which individual deals with you. It may be consistently bad, with little hope of correction because overseers are bound by the same rules the underlings are, but it is (relatively) free from bias.
If we ever achieve strong AI, that is artificial sapience, it's going to be as biased and stupid as we are. But it may be a lot faster at being so. Forget the singularity with a god-like AI that is compassionate, caring, loving, and wise (as the Xtian God is meant to be), think ancient Greek and Roman gods (and the Judaic JHVH). Those god were essentially humans with all the standard human failings (stupidity, greed, petulance, laziness, anger, jealousy, etc.) with some added magical powers. If the singularity ever happens, it's not going to end well for humans.
Hackers emit 9GB of stolen Macron 'emails' two days before French presidential election
Today's bonkers bug report: Microsoft Edge can't print numbers
Windows 10 S forces Bing, Edge on your kids. If you don't like it, get Win10 Pro – Microsoft
Thursday 4th May 2017 11:41 GMT 
I remember when...
It wasn't that long ago (four or five years) when any El Reg commentard criticising Microsoft or praising Linux would get near-infinite down-thumbage and many scathing responses. Even something really mild like:
I love Windows. I hate Linux. But the other day I hit a use case, one that will probably only ever come up one in a hundred million times, where Linux did a *slightly* better job than Windows.
Even that would get massive scorn. The possibility that there could ever be a use case where Linux might be slightly better than Windows could not be countenanced.
Now? Just about every post attacks Microsoft. Those posts get a few down-thumbs but many more up-thumbs. It's not that Microsoft's "lock in users and screw them for all they can pay" policy has ever changed, it's just that they've become a lot more blatant about it. So blatant that even former Microsoft lovers have had enough and come to recognise what some of us have long known.
I remember when it was all fields around here...
ISPs must ensure half of punters get advertised max speeds
Thursday 4th May 2017 10:46 GMT
Bugger maximum speeds
Make the bastards guarantee a minimum speed.
Oh, I must be one of the unlucky 50% because my ultrafast super-hyper-mega broadband connection is slower than using avian carriers. But 50% of their customers are getting the advertised speed so I have no cause for complaint.
Fuck that for a game of soldiers.
Make them guarantee a minimum speed. Or, as an earlier poster suggested, pro-rata billing.
Sure, they can give a max speed in their advertising if they want, but it must be in smaller type. One of the footnotes most people never read. The speed figure in the biggest, boldest type must be the minimum speed they guarantee you'll get at all times of the day.
OK, I'll accept a 99% of the time type of qualification to allow for rare events. Even 90% of the time. That figure, too, has to be in big fucking type.
Male escort forgot pregnancy protection, scores data protection instead
It's Russian hackers, FBI and Wikileaks wot won it – Hillary Clinton on her devastating election loss
Wednesday 3rd May 2017 11:28 GMT
Re: Pedantic note
True, we are all subject to the OSA. But only a handful of people have ever signed the OSA. Some, perhaps many, of the commentards here will have signed a piece of paper acknowledging that they have been acquainted with the OSA, its purposes and the penalties for breaching it. Only monarchs get to sign the act itself, when it is passed into law.
How would you pronounce 'Cyxtera'?
SpaceX spin-out plans to put virtual machines in orbit
Tuesday 2nd May 2017 19:39 GMT
Re: one hot use case coming up
Hmmmm.
I just went through your entire post and substituted "vm" with "operating system." Seems equally plausible that way. No need for a vm.
If you wanted to argue that it's safer to have some sort of vm so that in case of fuck-ups you could rely upon the host continuing to work and you could re-init the guest then I'd see your point. But that, unless I misinterpreted it, was not what the article was on about. The article seemed to be more about some sort of magic cloud satelliting idea.
I still don't see the point, if I understood the original article correctly. I once watched somebody run freeswitch (better than asterisk) on a linux guest on top of a windows host. Before he started I pointed out that neither are real-time OSs and that unsynched interrupt handling on the two would likely screw things up in a "dalek gargling down a drainpipe" sort of way. He countered that it was a shit-hot, latest h/w windows box which had yet to be put to any other use and was perfectly adequate. I turned out to be right. He then tried it on an ageing linux box handling a couple of hundred web sites and their associated mail and it worked flawlessly.
I'm still somewhat skeptical about the idea. You've written nothing to persuade me to your side.
Biting the hand that feeds IT
