Posts by doublelayer

Zero arrests, 2 correct matches, no criminals: London cops' facial recog tech slammed

Re: Numbers

Even if we go with a 98% FDR, the time wasted trying to keep up with the useless flags will be rather irritating. If, for every person we want to find, about 49 people are incorrectly flagged, and it takes five minutes to track that person, accost them, and question them enough to realize they aren't someone you're interested in, then over four person-hours are needed for each person you want to track just to deal with the mistakes from the system.

Now, we have to ask if five minutes is really going to give them enough time to track and exonerate each person. I'm assuming the tracking process can take a while, especially if they are running this in crowds, and that questioning them isn't as basic as saying "Aren't you that guy we're supposed to arrest?" I have no statistics on this, but I'm assuming that they have to verify documentation of identity and question the person as to their links to the person they want to find. I would also hope that paperwork is required to explain what the officer was doing at that time, which would take further time. I really think it will be longer than five minutes.

By the way, to those who say that many of the false detections can be removed just by having a police officer look at the pictures, that is almost certainly not the case. If the system matches the faces, the picture is either similar enough or of poor-enough quality as to make detection difficult. If we could just use humans to compare faces from indirect and low-resolution video with better accuracy than this system, we probably wouldn't even need this system.

You're in charge of change, and now you need to talk about DevOps hater Robin

Way to be completely unbiased

"Robin sits in the stand-up meeting, arms crossed, each morning mumbling "Well, I wrote some code" and then takes that long, loud sip of tea."

OK. You clearly went to some effort to always mention Robin in a nasty light. That's not going to help you convince those of us who think devops is pointless to like your argument. Maybe next time, you could grant that there are valid questions such that the most positive term for someone who didn't buy in immediately isn't "justifiably crotchety".

Here's the major problem with devops. We're not sure what it really is. We understand that you think it's a thing, but your articles seem to contain sections that were written by the "big buzzword generator". To me, a lot of them seem to say, under the jargon, that there are problems and we need to make them into not problems. It doesn't say how to do that, other than making some suggestions that we already know about. For example, having meetings and discussions seems to be a major part, with the various sections having contact with each other so that a problem noticed by one can be made known to the others so that one of them will make it into a not problem. That's not new. The reason a lot of software is crap isn't because people were thinking "I never thought that one group should note problems to the responsible group", but because the responsible group didn't care, the problem went to the bottom of a pile of papers, or management pushed the software through.

In addition, devops seems to be a project that, whatever else it does, will generate a lot of paperwork. I hate to inform you of this, but having systems requiring a lot of incidental work just to keep the system running makes things bog down. For example, the company I used to work for, where we all had to get up and have a meeting that often lasted at least four hours, during which everyone discussed their projects, was counterproductive. I had nothing to do with many of those projects, so while I listened to details that I couldn't use and problems I couldn't help to solve, I was not actually doing my job. That meeting system has, I'm told, been canceled.

Here's what you have to do to start to satisfy me with this devops system, if you're interested. First, stop using clearly biased language, perhaps to try to sound humorous. It's irritating me, at least. Second, be really clear about what each thing is that you're talking about. If a theory says do small unit tests, don't phrase it in several paragraphs of explaining what incidental systems are required, say "run many small unit tests". Finally, explain not just what the system is, but why it's going to work in practice in the real world. Explain how someone explains the philosophy to a manager who actually listens and wants to understand, and how to use it to avoid the problems that actually exist.

Your software hates you and your devices think you're stupid

Re: Older stuff WAS simplier...

On most microwaves I've seen, the door is attached to a switch. You still have a standard hinge and handle to pull open, but if you do that the microwave immediately turns off. It won't let you start it unless the door is closed. So as long as the door keeps itself shut, radiation shouldn't be a problem. That said, I don't really have anything against a specific control for opening the door, as it at least serves a purpose, unlike the many buttons I have on my microwave and have never actually used.

Re: Older stuff WAS simplier...

That's the control you don't need. Do what every microwave does, and I assume did, and just have a door. You pull it open. However, while I'll be the first to agree that I need very few microwave buttons, I'll take a number entry pad over logarithmic dial every time. I'll find how long I'm going to heat the things I typically do. Entering it at that point does not require me to look at the labels. Sure, there's an element of muscle memory in the dial which may make it more straightforward, but I find pressing 1, 5, start is pretty easy.

Re: There's an island somewhere...

I've produced audio tracks for cars before so that people can listen to audio books. Often the problem is that the car will forget your place and send you back, so my main task is chopping the audio up so that you can use the fast forward button to jump through sections as if they were separate tracks. I've written a piece of software that chops during long periods of silence which results in pretty good breaks and there isn't any weird popping noises and ... I'm off topic. Sorry. My original point was that these small portions were useful because pressing the fast forward button to jump from track to track was better than holding it down to try to move inside one. That never seems to get you anywhere. Either it moves forward at about 3x normal speed, so you have no chance of getting to the middle without having held your finger on the button for a minute or two, or they made it exponential. That seems logical, as usually you just want to jump a bit, but if it gets to the point where it's skipping five minutes at a go, the function is once again useless.

Of course the cars learned that I was about to defeat them, so they suddenly seem to have adopted the tactic of sometimes jumping to a completely random track when they finished with one. I've moved on to suggesting that people just get a car with an analog audio connection and use something else that will actually read the device and play the files in order.

Artificial intelligence is good for at least one thing – making hardware important again

Re: "Treat the AI as a team member,"

I can only hope that when these companies finally get the technology to create a sentient program, which will have a few tasks but will be rather useless, that they have learned enough not to make it. For now, I don't care that the process to improve the mathematical basis is called training--it takes in data of one format and outputs data of another format. That's not a colleague. If it starts giving me wrong answers, I'll fix it by deleting recent training data, not asking it questions. If it never works right, I'll fix it by throwing it away and starting over after an analysis of what went wrong. Maybe we want to do that to coworkers who aren't being helpful, but we don't actually get to try it.

IBM bans all removable storage, for all staff, everywhere

Re: Two use cases

That's fine, but you're likely to destroy a rather large quantity of USB disks. I might suggest using DVDs as much as possible. Not only would they be much cheaper to use even if you destroy them, but you would be rather certain that nobody has modified their contents, which you could encrypt rather easily. Of course, that doesn't help if you need more than 4.3 gb of space, but perhaps sometimes. I've considered using read-only USB devices under some of these circumstances, if only to prevent overuse of hammers.

Industry whispers: Qualcomm mulls Arm server processor exit

Re: Why should ARM Holdings help?

"[T]he WHOLE point of cloud is too pool customers at such scale that they can keep all cores utilized."

Is that really the case? Even cloud must have comparative downtimes. If some data center in Europe needs 100 units of performance for peak consumption by clients, which probably happens during working hours if these are business clients, then what would the requirement be at 3:30 in the morning. I don't doubt that there are people using those systems then, nor that there are tasks the servers perform off hours, nor that there may be businesses in other areas that are operating during the day and therefore using more power at European night than in European working hours, but even so I'd guess that there is a noticeable difference. If only 85 units are required, then using 85 or 90% of the cores and shutting the rest down for power savings can't hurt, right? Also, this could be a factor for companies that have systems not on the cloud but still need quite a bit of performance.

Re: RISC-V is the future

>Perhaps the downvoters could explain to us just why RISC-V is not the future?

I didn't vote either way, but I'll take a crack at it. I have nothing against that ISA. It's fine, and it shows some promise. That's not enough. We don't really need to keep inventing new ISAs for us to decide between; if we can get the ones we have running faster with less power, that gives us what we want. If this one will run really fast with low power and is relatively cheap to make, we will probably want it. However, it's at such an early stage that nobody can run on it now nor realistically plan on it for the near future. There have been enumerable new technologies that would have been great, if everything people said about them was true and we actually got it, but turned out never to become reality or was not that great compared to what we were told. For example, I've heard that basically every new file system for Linux/BSD servers would be the future. ButterFS was going to solve major problems. ZFS was nearly perfect. Even apple and microsoft did that. APFS will speed up the mac. REFS will do, actually I don't really know what it does but microsoft sent me a document about it, and I'm sure that it has some new features. Some of these will end up being very useful. Others won't be used in the longterm.

Thus, when you tell us that something is the future, without evidence of the future getting started now, a lot of us will assume that, whatever happens, it's not relevant now. In many cases, it won't be relevant at all. Everyone has at least one time where they said that something would be pointless and turned out to be wrong, but unfortunately everyone usually has many times where that turned out to be right.

Making calls? Ha, not what most peeps use phone for – Ofcom

Maybe, but

I'm sure a lot of smartphone users willing to and comfortable with installing apps, which is all that was measured, would like web browsing to work, but how much of that is for browsing data-heavy sites. Might it be that they wanted things like email and push notifications to be available all of the time? That would back up the findings from Monday's article on voice being considered more important and reliability being rated above speed, and it is definitely how I view my data connection.

Measure for measure: Why network surveys don't count what counts

I don't find the web to be most important for me. My primary web use case is e-mail, which doesn't take much data. Even those who upload a lot of photos won't need consistent bandwidth. In my case, if my data gets killed, then I have to wait to read my email until I'm in WiFi range. It could be annoying if I'm on the road for a long time, but that rarely happens and I can almost always wait. If I can count on being able to have a voice conversation that doesn't drop out and where people can hear what I'm saying, and that my text messages will arrive without being forgotten, then my main uses for my phone's wireless capabilities have been met. For those who use online apps more often, or for those who actually stream on data plans, the factors are undoubtedly different. For me, it's not that critical.

Microsoft vows to bridge phones to PCs, and this time it means it. Honest.

sorry microsoft, you had this already

I remember the days of Windows CE and windows mobile 6 devices. That's the closest they ever got. While those devices had very little connectivity, couldn't store much data, and had annoying restrictions on programs, they had a basis that could lead to what would be useful. At that point, you synced all your data over a cable, so that was annoying, but the theory was there: keep the same documents, e-mails, calendars, etc. on each device and make the programs to use them similar enough that you don't have to learn two, but not so similar that the interface for one of them is terrible. If they kept up with that theory, eventually uniting the two operating systems enough, they would have what people want if they want the two devices to be essentially the same. And people bought those devices, so they had an existing user base.

The hardware of today and of five years ago would be perfect for this. Extra storage would allow for keeping all the documents stored, at least for most people, on the mobile as well as the desktop or laptop. WiFi and bluetooth would ensure sync was fast and occurred in the background, so things wouldn't be missing from one. That'd also lead to a ton of cloud customers for a your data is completely backed up a la dropbox solution. It would have ensured customers for microsoft, as its control over the desktop and office areas would have helped to drive the sale of "completely connected" mobile devices.

They had the basis they could build on. So what did they do with it? They wrote windows phone 8. Well, I give up.

Risky business: You'd better have a plan for tech to go wrong

Re: @Doctor Syntax

I'd rather have the many different systems approach. If you have the one integrated system, you're benefits are that the manufacturer should know how it works, and if it breaks it's probably on them. This probably means that they have total control, and if you manage to break it in a way they dislike, they can put a lot of pressure on you. If I am running some piece of software and it turns out to be bad, I can more easily replace it. Depending on what it was, my reconfiguration time may be low or high, but I almost certainly won't have to reinstall my OS, let alone replace my hardware. As for the issue of loading java script from remote locations, I agree that that is crazy. I'm perfectly happy to use such libraries, but I'm going to check them out first and they will be hosted on my system. My issue is not with the fact that now I'm using someone's code, as I assume it's been tested by the other sites that use it, but that the approach to receive it is ill designed.

Google Pay heads for the desktop... and, we fear, an inevitable flop

Re: Google and android

That could be, but they have a lot of search and ads clients that don't have android, from their having chrome installed on everything, the default consideration of google over any other search engine, and google being default on firefox and safari (both IOS and OSX versions) and the google ad frames on many websites. I agree with you that it would be hard to undermine google's near complete victory in that realm, but if a business came up with a better ad system and people started installing that one on their sites, google could lose some without needing android to be weakened. Meanwhile, android, while it has large market share, has a distinct rival IOS and mini rivals in the mods for phones and the chance that eventually I will find the cheap feature phone I'm looking for. I think google could do a lot more to kill those than they have done, and the fact that they haven't suggests to me that they're pretty much fine with the smaller castle across the street because they know no cannons are going to take aim at their search and ads.

Google and android

"Of Google's dozen, or more, monopolies in online markets and services, Android is one of the most significant."

I'd argue that search and ads, and arguably gmail, is their major thing, and that google will live or die on it. Android's nice, but it's further down the list with maps, docs, and cloud. Stuff that's important, but not critical.

"Over 80 per cent of the world's phones run Android, and it is estimated more than two billion Android devices are in active use every month globally."

Can't we find a better system for cheap phones? Really. I've been trying to find a straightforward feature phone for my father that has good battery life and can make calls, and the ones I'm finding that aren't used or nailed into a contract cost more than $50. I can get a pretty crappy android phone for that that will do a lot more. Why can't we make cheaper phones?

"And on Android, 19 of the top 25 apps downloaded more than one billion times are Google's own apps."

Well then, stop putting them on my phone. Do you know how frequently I use street view? Zero times. That's how often. Why is there a street view icon in my apps list. That wasn't there when I bought this phone, and I certainly didn't ask for it.

Microsoft's latest Windows 10 update downs Chrome, Cortana

Re: Try Linux. - Or DON'T! (My love/hate Linux rant.)

The language indication in the task bar refers only to the language of the keyboard at the time. It is possible that, while enabling the Slovak keyboard, the Slovak language pack was downloaded too and set as default. I'm not sure why that could set only some of the labels, but it's the best theory I can think of for the result.

Re: It it ain't broke, you're not trying

High sierra installs have become very unreliable. Some people have no problems at all. Others have some difficulty getting the install back into working order but once that's accomplished, everything works perfectly. That's my sister, whose mac I had to reinstall and restore from backup but there has been no problem since. Others have noted frequent application crashes and operating system instability. And then there's me, where something that happened at the update has put my machine into a seemingly endless loop of bricking itself for about two weeks before restarting fine for about a week before starting again. This issue did not happen before the high sierra update and it persists through clean installs of high sierra, sierra, and el capitan, so I'm presuming some kind of firmware problem. In addition, various security bugs have been found in High Sierra which has led to questioning of the quality control on this release. I assume some corporates are dealing with these problems and have become perhaps excessively irritated due to a small sample set of machines that failed. Then again, I'm the same--I'm not sure I could buy another mac, given the fact that apple has failed to make any change or diagnose the problem that my machine has.

Re: No issues so far

Are you running one of the affected GPUs? It seems the bug will not affect many users, as these GPUs are only some of the many intel make. If indeed you are running one of the mentioned chipsets, then maybe it requires more specific stress.

Re: Try Linux. - Or DON'T! (My love/hate Linux rant.)

If you want to change the language back, you should be able to do the following steps:

1. Open the new windows 10 settings thing. If typing "settings" in the search box won't access it for you because of the language change, try using the quick tools menu. Press windows+x, which should pull up a menu of some more useful tools. Settings should be sixth from the bottom. Second from the bottom is a submenu with shutdown options, so settings is four above that.

If that doesn't work, try to use the command prompt. The system settings executable is well-hidden. First, cd into

c:\Windows\SoftwareDistribution\SIH\bin\cbs

Then, try any subdirectory of there. Hopefully, you're like me and you just have one subdirectory. Once you've entered it, you need to enter another subdirectory. It starts with "amd64_microsoft-windows-i..ntrolpanel." and includes some numbers, so enter that much and press tab for the rest. Then, try to run the command "start systemsettings.exe". Hopefully, you'll never need to do that.

2. Select "Time and Language" from the settings screen. On build 1709, it's the third element of five on the second row. On 1607, it's the one in the middle (second row, second column).

3. Select "Region and Language". That is the second of three.

4. This screen should have a list of languages, all of which are written in their own script, so your desired one should be available.

5. After clicking it, if it is not set as ddefault, there should be three buttons. The first one is the one to set as default, so click that.

6. At that point, you may want to click the other language and remove it from the system.

If these instructions are not helpful enough, you can describe what has happened and I'll try to correct them. Otherwise, I believe I have a friend who speaks Slovak, so she may be able to help me provide exact translations of UI elements you will need on your way.

Yes, this has been derailed by discussions of Linux. Yes, the discussions aren't exactly covering new ground. Yes, said discussions are relevant. The article is discussing a windows problem that breaks components, and people are commenting with a suggested alternative, albeit one that is a tad unoriginal. Is your preference that more people just comment on all the other microsoft problems of the past? Many posts here are doing just that.

Silicon can now reconfigure itself with just a jolt of electricity

Yeah, that'll work

How exactly is a chip made of this silicon dust going to fix security problems in chips? Unless you're considering that an update could just zap your processor into line, which will never happen. Just consider how this would work with the entirely fictional processor manufacturer Entil and the equally fictional design flaw called FreezeUp.

1. Entil looks into FreezeUp and figures out that it's their fault. They design a protocol for fixing their chips to avoid it. This requires changing the gates on the chip because the microcode's just not good enough.

2. They send a command to your chip, because they will have gone to the effort to repair and make available to you a patch requiring molecular reassembly, using thousands of valuable worker hours, keeping chip design specialists, electrical engineers, microcode programmers, and the many, many molecular physicists they have employed away from their profitable job making new chips, but that is just the kind of company that Entil is.

3. Your computer, running the old, but still serviceable, "Entil Care Ls" processor from the six-year-old "gravel-covered causeway" family receives the update file from Entil's system. Nothing can get in the way of that. Not the operating system writers, who might want you to update to the latest edition, portholes 11, who might conceivably use this as leverage to make you update. Not the device manufacturer, who might want you to buy another machine and could deprecate this one from support on the basis of FreezeUp. Not from Entil, because they made sure to include drivers that can handle their updates in every single operating system out there, with teams of people designed to make the interoperation of their processor and your system as worry-free as possible. And certainly not from your system that could interfere with the process silently or corrupt the file.

4. The update gets your permission to install. It came with an easy-to-manage control system, after all. One that would allow you to easily decide whether to try the update or take your chances. It also came with a system that would make it easy for nontechnical people to manage the update. No work required on their part.

5. The processor sets up its update environment, making sure that it doesn't have any problems at all, as the chip was built with special safety systems to make sure that critical security updates could be run without problems. It also prepares its automatic rollback system. Entil, knowing that this would happen, built the chip to recognize any time it was failing and roll back parts of the update. If the partial rollback didn't work, the chip is rescrambled and they try the update again. If it still fails, they return you to your original configuration, which they protect with very strong measures. This process takes only two seconds.

6. When activating the chip's components, the chip manufacturer has already thought of weird environmental things. The particles are limited to a zone. They can't possibly be affected by anything. Not even that big magnet Dad has stored under his work bench, on top of which is the computer. If they hadn't thought of that, imagine what could happen. Particles could fly away from the processor when they are reconfigured by that system. They could lose some. Those could go flying off at high speed and get into stuff. Hard drives, perhaps. They're small particles, so they can get inside those cases. You could only hope that they'd land somewhere that wouldn't kill the drive. But don't worry. Entil thought of that, and they are good at making sure nothing can go wrong, FreezeUp excepted.

7. While they're at it, your Care Ls was designed with extremely strong mechanisms for keeping all those particles together and in line. There is no way that Entil didn't think of conditions that could put stress on a chip that is essentially made of dust. You can subject your chip to extreme temperatures, high winds from real wind or airflow, magnetic fields, physical jolts, electrical imbalances, and your nephew repeatedly touching it as you assembled your motherboard, all without losing any of the chip's structure.

8. The chip, once it has installed the update, will run everything perfectly. How could it not? All the instructions are the same, after all. Never mind that the whole chip was disintegrated to fix the problem and the microcode wasn't good enough to solve it, meaning that that has almost certainly been changed, too, and that now your chip has not only a release version and a microcode version but now also a molecular design version.

9. Entil immediately informs you about security problems. After all, if they didn't, you'd find out anyway from the announcement when your chip updated itself. Therefore, you have a full history of all design problems in Entil's chips. There is no way that anyone would collect those and make exploits for major ones, because they would grab the molecular composition version number and run those exploits that haven't been patched yet. Entil is just that good; any attempt to do that would fail, because the updates have already been run successfully on every Entil chip on the planet and, by the way, those chips in orbit on the ISS or in satellites. Radiation and gravity being basically zero doesn't affect this chip design, either.

10. There is absolutely no way that the ability for Entil to change their processor's mechanism would be used to send out processors before everything was ready, with the company trusting that they will fix them in the field. The company doesn't have that type of culture.

11. On that topic, the company also doesn't have the type of culture wherein they'd "accidentally" break your chip when it was too old and you should buy a new one, or you had found a way of overclocking it or enabling disabled features (or, for that matter, disabling the wonderful Entil chip component known as the Execution Mechanism which by the way has no security problems whatsoever because it gets updates too in order to fix any possible security problem) to encourage you to stay away. That's not how companies work anywhere, but if a company were to act like that, it would definitely not be Entil.

12. There is no way for an attacker to forge a processor update from Entil. Not even if a big bunch of armed spies from a hostile nation that probably wouldn't want to do so anyway, but if we had to name them we might suggest the Republic for the Democratic Governance of the Rakeon People, were to break in and start copying disks left and right. Only Entil can send out those updates, and they will be tamperproof at all times. There is also no way for an attacker to trigger a problem in the processor, for example, part of it that would convince it that there was a problem during the last update and its intelligent problem resolution system needed to roll back, then using that opening to infect it with the security exploit that the most recent update was designed to prevent. That could never happen.

Actually, it all seems fine. I withdraw my previous objections. All steam ahead.

Nvidia quickly kills its AMD-screwing GeForce 'partner program' amid monopoly probe threat

Not exactly. It sounds like they were forcing manufacturers to decide whether their flagship brand would have nvidia chips only, or only non-nvidia chips. The companies that have gaming brands have to choose one, because their brands that have former knowledge will sell better than new ones just started. Therefore, they have to essentially choose to give nvidia a guarantee of placement in all their models, or settle for the likelihood that nvidia wouldn't let them sell any of their GPUs. That hurts the computer manufacturers a little, nvidia a little but they were thinking longterm, and customers a lot, both now and in the longterm. It's not about expecting one type of GPU and getting another, because you can easily determine what you're buying, it's about how much you paid for it and how much choice you will be offered to have whatever GPU you'd like to buy.

MacBook Pro petition begs Apple for total recall of krap keyboards

keyboards

I have noticed this as a result of moving back to my old macbook (the plastic version) owing to apple's bricking of my newer pro. I didn't have these low-travel keys, but I had the version before that. Now that I'm typing more often on a keyboard with more key travel owing to the deeper case, it feels so nice to do so. Of course, the machine is showing its age--the battery lasts about three hours maximum and the core 2 duo doesn't exactly cruise when running VMs, but at least the keyboard will be a nice factor while I prepare to obtain a newer machine.

Blame everything on 'computer error' – no one will contradict you

Re: Don't give your opinion!

"Our company noticed that many problems that were reported - didn't get followed up and resolved. So it was decided that anyone who reported a problem became its "owner"

No. Can something, even a company, be that stupid? They made "That's your problem" official policy? I need hope. I'm out of hope.

Whoa, Gartner drops a truth bomb: Blockchain is overhyped and top IT bods don't want it

Re: Your all missing the point.

If there comes a need for a relatively tamperproof system for auditing, my suggestion would be to buy a ton of blank CDs/DVDs. If you live near me, I'll give you some because somehow I ended up with like a hundred of them and I don't know where they came from. Then write the audit data, along with checksums for all preceding data stored on disks, and keep multiple copies. Label them well. If you're afraid that someone sneaky will overuse hashing power to find a way to hash incorrect data on a disk, store the original data plus encrypted versions of it with predefined keys so that you'll never get a hash collision while the universe is in existence.

Total network costs: 0.

Total data storage costs: Lots of blank CDs. Call my friends; they must be putting them in my house.

Total infrastructure costs: Four or five cardboard boxes on shelves in different buildings.

Total recovery costs: USB Optical drive ($10)

I've got way too much cash, thinks Jeff Bezos. Hmmm, pay more tax? Pay staff more? Nah, let's just go into space

Re: I disagree...

"It's only when you leave school and you're expected to move out, pay your own way and actually find yourself HAVING to do your own washing, cleaning and such, because there's nobody there do just do it for you, that you finally develop an appreciation for the need to do it."

The history of people going to far-off places, where either there weren't any people or they killed a lot of the people through enslavement, where technology was not advanced enough to connect these places, effectively making it similar to the planet scenario, is not on your side. Those places aren't doing dramatically different from other places. The problem you run into is that there are a lot of things that could cause you difficulty getting there in the first place. If, for example, some resources could be put into making life feasible for more people here on Earth, the likelihood of a major war with major weapons obliterating your launch sites before sufficiently advanced craft can launch you to a new planet will be decreased. Even if you consider travel to other planets being a major concern for right now, you might want to look at earth a bit before you go all in on the new tech.

AWS sends noise to Signal: You can't use our servers to beat censors

Re: Yet another reason

Good point. I can only hope that they come up with something else that'll work, as it seems that if domain fronting isn't shut down fully yet, it will be soon. I'm sure they have some tests ongoing.

Re: Yet another reason

I am skeptical about the cloud, too, but how would signal do something to circumvent censorship in this way without using it? The best I can come up with is that they start their own cloud and then allow this to function through their clients' sites, meaning that instead of signal being blocked, signal and all their cloud business clients that signal doesn't really want get blocked. You can't hide and be active if you aren't in a group with lots of other people, hence their cloud usage.

Scammers use Google Maps to skirt link-shortener crackdown

I use a different system

I know that short links are usually somewhat helpful, so I usually reserve a directory at root of the web server for such a system. For example, example.com/url/* is a shortened link, and I can make it clear what they'll see at that page and logical. People still know that it's my site they're contacting, and although the links may be longer than some of the shorteners out there, they can be quite short because there is no competition that drives up the key length and they will fit into tweets or short messages should someone want to send them.

Virtual desktops won’t save cash in clouds or on-prem. So why care?

Yeah...

I don't think that will work well. The benefits they cite seem to be mostly useful for locations where you want a lot of remote access, but RDP seems to serve that well. Otherwise, I don't see pretty much any of the benefits of the system. The sentence that most concerns me is the statement of less need for IT people to go to the desks and fix machines. What, pray tell, do they think the users are using to access their virtual desktops? Whatever it is, it looks a lot like a computer and it's on or under the desk. Users can mess that up just as well as they could a traditional machine. The main difference is that there is less access for the IT person if the user has managed some calamitous software problem. True, you can easily reboot it after you correctly plug in each cable, then reconnect to the virtual desktop, but if the issue is worse than that you will have to play around with whatever thin client it is to find it. Traditional desktops work fine most of the time, because they are straightforward to manage. If there is a location where repeatedly destroying or reimaging virtual desktops is a routine thing, there are probably problems that virtualization won't fix.

DIY device tinkerer iFixit weighs in on 15-month jail term for PC recycler

They are all wrong

I've read a bit too much about the case and I now think everyone and everything about it has several wrong elements.

Lundgren: What were you doing faking the disks to look like official microsoft or dell ones. I'm not saying leave them unlabeled, but it doesn't cost that much to change the label you put in the disk printer from "official dell/microsoft" to "windows restore disk for dell computers", and you could even get a free ad space for your business on it for people's support needs.

Microsoft: What were you thinking when you made up a number for lost money, when you know full well that the disks themselves were free and the license keys were already present meaning that you don't lose money.

Lundgren: If you're selling these at such a low price--essentially giving them away, why did you find it so necessary to have them have the labels on them? The recyclers are going to pay you the same amount, seeing as they're too lazy to burn their own copy. Surely you recognized that that would look sort of sketchy.

Microsoft: Why did you feel that saying the cliche malware line would work? You know full well that they were identical copies. I'd have a lot of sympathy if you found someone faking disks but including a set of preinstalled malware or for that matter any unnoted software of any kind, but that's not what these were. I'd have sympathy if these were windows disks with pirated license keys on them. None of that was the case.

Recyclers: Why did you buy these? Surely you know that they weren't official microsoft ones and also it doesn't cost much to write your own. I can't believe you didn't use the same suggestion I had earlier, to change the label to tell the users what was there and to refer them to you or someone who paid you, for support. Surely you could see that unofficial official install disks would be sort of sketchy?

Microsoft: Given you have announced an interest in getting recycled computers in use and having people on windows 10, don't you think that convincing customs that there wasn't a real case, owing to the basically zero value of the disks, and discussing with the recyclers to help them end their problem (especially if you could get them installing windows 10) could be useful for both of you? Why is the not-going-to-court solution so unpleasing to so many people?

IFixit: Can you see that this issue has nothing to do with the stuff you advocate? Microsoft and dell did not try to prevent people from using the machines, or even reinstalling windows on them. In fact, the data that they provided for free allowed people to erase and cleanly reinstall windows without buying anything or hopefully having to deal with any license management. Therefore, it was incredibly good of the companies given their usually terrible track record with repairs, as I would have expected such disks to be completely unavailable or to be well-hidden by the typical microsoft knowledge base web maze. No law or regulation or corporate policy prevented repairs in this case, and that's what we're going for. Go back to those laws that do exist and help us get rid of them.

Apple and The Notched One: It can't hide the X-sized iPhone let-down

Re: Commoditization is not going to be Apple's friend.

It's worth considering how many iPhones from old generations are still out there being used. The availability of updates results in some people keeping their iPhones longer than they did androids, and I know far more people using iPhones in the 6, 6s, and 7 ranges than I know people using the 8 or x. I can see that being a problem for apple if they keep it up with the expensive high-ends that don't really have a big selling point. Then again, I would expect the same to hold and even stronger with android phones that cost a similarly high amount, because there is now no lock-in reason to keep the people off the $200 phones with similar features. Yet, companies still make them and people must be buying, so what do I know anyway.

Firefox to feature sponsored content as of next week

Re: Kiss FF goodbye.

I'd gladly pay $2.50 a year for this purpose; I pay more for many other projects I use. However, the problems remain, and they are many:

There are those who would rather not identify themselves in order to make the payment at all, and even those who would not dislike this would probably not want to identify themselves to have the browser disable this for them, as they've just replaced one identifier with another. My payment in the year won't remove mozilla's problem, and they will keep going with this, so I have to ask whether they will actually care about my desire for privacy. If they are going to use my donation to create an ad tracking system, there are more deserving projects that respect me as a user, and the money would be better used paying them. Furthermore, there are users that should not be expected to pay for something that is, in fact, open source. I view the fact that firefox and related products can be used in less developed countries for free so that the people can use the internet to improve a situation I don't have to endure as a major advantage, and I would oppose any attempt to restrict that.

In short, I will donate to mozilla if they show they will respect me as a user and all other users by not inserting this advertisement crap, that they respect their open source roots and will keep the product free and open to alterations, and that this will continue and not be based on a short-term financial report. They haven't been doing that.

Let's be Frank: Bloke drags Google to the US Supreme Court over $8.5m privacy payout

Re: Wow

Uh...No.

"Yep, all those lawyers who work for the homeless or for absued children, or who help you make sure your scumbag neighbour isn't tossing dog doo into your garden -- all villains."

That's why we don't say "all". All stereotypes that claim to speak for all people are bad. But that doesn't change the point that it seems a bit bad that a settlement against google for messing up privacy for users results in $0 to users, a lot to lawyers, a lot to law schools that the users have little to do with, and a lot to things google was going to pay anyway. We might reasonably blame the lawyers for this.

"Or are lawyers not unlike 99.99% of IT people who are always skiving and taking bungs? Or so popular belief holds?"

Then popular belief is wrong. You'll see a lot of yelling about IT from everywhere, and a lot of it is true, but usually the administrator earns a lot less than the lawyer and doesn't break things because they hate the users. There are many jobs where things break and IT is blamed, but not all of those are because IT broke it. Also, it has almost nothing to do with the point being made, that one group of lawyers in this situation and perhaps too many lawyers in other positions are not serving the clients they claim to represent in favor of self-serving and adversary-serving decisions.

Windows 10 April 2018 Update lands today... ish

Re: "Keep clicking, Windows-lovers! It's bound to come along soon."

"several million voluntary beta testers that have signed up for Windows Insider have already tested it."

Microsoft has 1.5 million insiders. That is not several million. Also, many of them are like me--I'm in the insiders group but it has been a long time since I last really tested something rather than just firing up the system once in a while to see if I notice something. The updates are a bit annoying, and I don't spend enough time in windows to make it worth my time. I'm mostly in the group because I was four years ago and why bother withdrawing? I just checked my VM with the insider build on it--seems I haven't used it since February.

Exposing 145m Equifax customer deets: $240m. Legal fees: $28.9m. Insurance: Priceless

Re: Let's not lose our perspective here

I'm thinking. I'm thinking three things:

1. 145 million people

2. data not provided by choice, collected by company

3. data very difficult to change if leaked

Please let me know what I'm missing that makes this minor. Thanks.

Quick math + I like privacy and consequences = anger

So, this company has spent $242.7m dealing with their security problems. And those security problems caused the leak of data for 145 million people.

$242700000/145000000 people = $1.674 per person.

OK. That's nice. I suggest legislation that makes these companies liable on a per-user basis for say, $200. A small system mixup leaks two hundred users: $40k, enough to give the company a notice that that's not OK and to get in line, but not enough to hurt them. A large system mismanagement leaks ten thousand users: $2m, enough to indicate that you've messed up and you have responsibilities to your users. A complete lack of regard causes the leak of a hundred million users: $20b, hopefully enough to know that the company will be in really bad financial status at the end. The company should think that through before they decide to not care. That's the law I'd suggest if I ever ran for office. Now if I could actually ensure its passage, that formula would be edited somewhat, with the multiplication sign removed and the exponent sign added.

How do I get that passed without running for office and getting a ton of friends to do that too?

Leave it to Beaver: Unity is long gone and you're on your GNOME

Re: New Linux poweruser here ...

"If you can't search for information, critically evaluate what you find and come to your own conclusions, you are lacking essential skills, not just for this industry, but for your life in general."

That's true, but missing the point and I feel dropping a bit much on the original question. The original question surrounded opinion. I'm sure they could have looked up what systemd is or how it works. They seemingly didn't have to ask that. They could have looked up the very many discussions about systemd and started to read all ten thousand pages about it, but that would lead to a morass of random junk that wouldn't answer the question "Why do some people dislike systemd so much?". Here's what they would find:

1. People discussing implementation details of systemd that have changed and are no longer problematic

2. people discussing the theory behind systemd and why they like or dislike it

3. people talking about the systemd/bsd interaction without easily understanding whether these people are fairly representing either side

4. people getting way too far into the weeds of how to write systemd scripts

5. people discussing parts of the systemd source code that dramatically impairs understanding without reading the whole source

6. people discussing the problems they had migrating to systemd, without making it clear whether the problem was on systemd, on previous developers of code, or on sysadmins who weren't familiar with how Linux administration should work

This is all the stuff I found when doing a quick search. So why is it so bad that the comment said "There are a lot of people here who dislike systemd rather intensely, and I don't know why they do so. Why don't I just ask them?".

If you can't accept finding knowledgeable people to give you information and opinion, critically evaluate what they say, and come to your own conclusions, you are also lacking essential skills.

Penguins in a sandbox: Google nudges Linux apps toward Chrome OS

If you want to have an unmodifiable OS image, you could take any Linux or other OS that can be booted as a live environment, set it up in the way you like, and make a live CD style image of it. Then drop that as the boot image to a read-only partition. If necessary, lock down the BIOS as well so I can't possibly change it without your code. You could technically execute malicious software on it, but the software cannot persist across boots. You could also restrict which binaries can run so nobody can download code or bring it on a USB disk and run it. Just make sure the home directories are stored on a separate read/write partition so you can keep documents, browser cookies, and the like across boots. That shouldn't take that long to assemble--I have a similar setup on my fix-a-system USB disk so I neither have to re-download tools from the repositories nor find an alternate place to store notes I want to remain available after shutdown.

Chromebooks seem to exist because Google wants to train users to use only google apps, as not many others will work at all, and google's are the only ones with support baked directly into the OS. I always get nervous when I see these used for schools, because I know google wants people to use systems for data storage and word processing that only google controls, so they can later count on a stream of users of the "free" products.

Re: "does this read more like an ad-icle"

Couldn't you buy some cheap laptops and put Linux on them? Some of the chromebooks have the ability to boot externally (I think by people finding BIOS access somewhere, but I try to avoid those), and there are a lot of windows laptops with similar components and prices. Although you might not want to buy them to run windows on given the problems shoving windows and running programs into 2gb or less of memory, most Linux distros should work OK for that. Then you could have them use firefox or chrome, whichever you or they prefer, and use any Linux applications you want. Administration would be basically the same, and if the network dies the machines can still be used for a few things.

We wanted a camera, they gave us the eye of Gemini – and an eSIM

Re: "Inspired" by PSION

I never used the original, and I also don't really want to buy this in its current form, but how exactly did you intend them to create a system like the one most of us want and have it run forever. I can build you a system like this that runs for a long time on convenient batteries. With some design help, it will be nice to look at, as my hardware design knowledge is flaky. The device will do next to nothing. A bit of typing, some calendar perhaps.

That's not what we want. We mostly want to run our applications on the Linux side, with all the requirements those have. The most important one is connectivity. Some people are fine with just WiFi, while others want the convenience of cellular, as they will be taking this with them most of the time. That takes a lot of power. From what I've seen online, the original psion devices had connectivity in physical ports that were almost certainly disconnected most of the time. Also, I assume it was a bit larger to account for the RS-232 port, but I've never seen one, so I could be wrong. We probably also want the screen to be better, as there is a lot more video and images about these days. Sometimes we do need that. The wikipedia article also states that it ran about 20 hours use on the batteries. I admit that's quite a lot, and that it probably had great performance when it was off for a while, but you could just turn it off if you don't want background tasks.

I would like to see a keyboard that doesn't get at best indifferent reviews, as I'd be using the thing for typing. Either the typing must be great or I must have my punctuation for coding and terminal use in semi-normal positions on physical keys. Having neither makes this a no-go for me. However, I have hope for the next addition, and I don't see why this is such a letdown for the many psion-owners here.

Eurocrats double down on .eu Brexit boot-out

Re: What's the difference?

I just want pretty much all of those annoying new TLDs to go away. We need neither .accountant nor .accountants, and ICANN was clearly intending a truly massive joke when they put both of those up. Do they really expect that to end well? And what do they expect us to put in the TLD .airforce. I hate to break it to them, but none of us citizens own an airforce, even for those who have private jets and enjoy the humor. I'm going to state now that I will consider any website under .associates to be so dodgy as to be immediately blacklisted without a visit. .attorney is similarly weird, and of course there's also .law, .lawyer, and .legal. Thanks mates. Then they do the same singular-plural thing with .auto and .autos, when of course they have .car *and* .cars. And that's just the As. I haven't even checked out the rest of the alphabet for my own sanity. Kill them now!

Ozzie Ozzie Ozzie, oi oi oi! Tech zillionaire Ray's backdoor crypto for the Feds is Clipper chip v2

No good answer

There is not a good way for this to occur, but I recently heard a suggestion from some researchers as to a way that is significantly less terrible than all these ill-considered alternatives. That's not to say it is good, just that it is better because people tried to think a bit before just shouting "I've got an idea so pay me some money and I'll solve all your problems right now and by the time you find out that it doesn't produce infinite security and access to data for only the people you want I'll have a nice big house and a squadron of lawyers for my defense". In short, the system involves the use of weaker encryption where the keys are not known by any group and where multiple keys are used. There is one master key that is long, so breaking that takes a lot of effort. Each message also has at least one short key that is unique. The process to break it requires a certain amount of brute forcing, but can be done at a cost. Governments would be able to get this, whereas small criminal organizations probably wouldn't. Mass surveillance would be made difficult because it costs the same amount of power to decrypt each separate communication, so you have to do a cost-benefit analysis on each one you want to see. So I quite like the math they use. The problem remains that, firstly, it won't solve the problem of good encryption inside government encryption, and secondly, that bad actors inside or outside government will break it because they don't want a good solution to the problem.

Now that I've heard an idea that actually has a modicum of merit, all the other solutions that are essentially the same terrible one make me wonder why their designers are so stupid.

Good news: AI could solve the pension crisis – by triggering a nuclear apocalypse by 2040

Re: I'm really not that worried about this

One minor detail: South Africa got rid of their nukes, so they don't have a need for that anymore. In some cases, I could see a country like North Korea setting up an AI for autolaunch because nobody could blow things up like their supreme leader, so if he gets killed by a strike, they need to launch now. Also, they'd like to put that on their propaganda, with the usual lack of any clue of what it means or how to talk about it without sounding like someone randomized all their words "a system for the intelligent use of the nuclear weapons of the supreme leader, president of the DPRK and chairman of the Korean Workers' Party, an artificial control" Other than that, all the current nuclear powers are smart enough to realize that that doesn't make any sense.

Re: If I am to judge by existing "learning algos"...

President: Do we have confirmation? The nukes have been launched?

General: Yes. The satelites verified it; they will land in minutes. If we launch our response now, the inevitable land war will have some chance.

President: *pauses to think about it* All right. I regret that I have to do this. *to nuclear control system* Launch at targets 3, 5, 13, and 18.

Nuclear Control System: ...

President: What's it doing?

General: I don't know. The developer got vaporized just now, so...

Nuclear Control System: Thinking a moment, please wait...

President: What now?

Nuclear Control System: Here's what I found on the web for "launch on targets". Target Corporation is the second-largest discount store retailer in the United States. In 1995, the first SuperTarget hypermarket opened in Omaha, Nebraska and the Target Guest Card, the discount retail industry's first store credit card, was launched. Would you like to hear more about this topic?

Incoming missiles: Boom.

Other general: Well, that happened.

Prime Minister: I don't remember launching this! I just tried to order lunch on the intercom! Who set this up to listen on that?

General: I'll find out.

Technical director: Unfortunately, it seems the logs for this were kept on an AWS bucket in the U.S. so...

Microsoft Lean's in: Slimmed-down Windows 10 OS option spotted

I can see your point, but I disagree. It makes sense that a system needs a browser on it at the beginning, so you can use it to download another browser. If Microsoft likes edge over IE, I'm fine if that's the browser that they put on. Likewise, I think most OS should have a basic text file editor, so basically notepad. That's a bit more subjective, but it becomes useful when you're working on something that isn't your setup. However, there isn't a good case for the random applications that most people don't use and many of us, myself included, don't know what they are. There was a great list posted earlier about the bloat on the current windows 10. I'm sure that Groove Music does stuff that Windows Media Player doesn't do, but I haven't a clue what that stuff is, I'm not using either of them, and it's kind of irritating that Microsoft puts them both on. If you need a media player at all, take the better one, combine any code from the other that you need for some reason, and give me one. Or, you could just not put one on because I doubt it will take me long to find one if I need it. But fine, that's Microsoft's decision to write their own applications and put them on. There is even less of a case for them to put tiles up for applications that they didn't write, essentially as advertisements. I know that, if I want them, I can go get them. So does everyone else. Apple doesn't decide you'd like some games so they'll just put them onto your phone. If some android phone manufacturer does that, they are known for the peddlers of bloatware that they are, and their phones don't sell as well--I'm pretty sure candy crush isn't installed on any of the flagships by default.

New and inventive code is transforming your business – and bringing with it new and inventive ways for things to fail

What?

"Businesses are becoming increasingly digitalised, with operations and customer experiences relying on data and devices being online all the time."

As opposed to...what? Last year, when everything was the same? Five years ago when the viruses were a bit different but they still existed so everything was the same? Ten years ago where the viruses were different and the windows wasn't wobbly but basically everything was the same? Fifteen years ago where the computers were slow and noisy, but viruses still happened and things would break from user error or security problems, so everything was the same?

"If you’re starting to think: “Hang about, that sounds like DevOps” – a philosophy that’s making inroads in software development - then you should be, because that’s where we are."

Oh. That's what this is. Carry on then.

I'd like to see the DevOps people define it. But not like they have been doing, with the weird words, the seeming lack of knowing how the IT department works and/or interacts with the company, or the part where they tell you that you have a problem but the solution is essentially "think some more about it". I'd like to here the one-sentence, maximum-twenty-words, maximum-length-of-words-ten-letters, definition of DevOps. And don't try the recursive definition, either.

Twenty years ago today: Windows 98 crashed live on stage with Bill Gates. Let's watch it again...

Re: Bill Gates

To clarify on the autobiography comments, the book was not an autobiography--it was a standard biography, not written by a ghost writer helping Jobs, but by a writer who wrote about him. The writer in question is Walter Isaacson. He got approval to interview people, including long interviews with Jobs and his family, as well as many people who worked with, lived with, knew in some capacity, or talked about Jobs at some point. The number of times the word "jerk" and less complementary synonyms appeared should at least assuage the comments that the book will tell only the story from Jobs's perspective.

Facebook privacy audit by auditors finds everything is awesome!

A leaked message between PWC staff

Please note: part of this message has been redacted by ________ in the interest of protecting confidential ________.

We have completed our 2015-2017 audit of facebook, and things have gone remarkably well. We found facebook employees to be incredibly helpful; every time we arrived to audit their data center without announcing it first, some nice staffer would take us to lunch for two hours or so. Once we got back, they were very accommodating and allowed us to investigate all the parts of their data center, which is remarkably small for a company that big. It seems facebook has been storing next-to-nothing. Really, you can go to your facebook page and download their archive and that's all they have. Honest.

Meanwhile, we'd like to inform the independence unit division about the answers to their many questions about the integrity of this audit. Facebook has taken no actions to try to sway us at all ________, and they would like to meet all the members of the independence unit at a time of your choosing so they can ________. They told us that if you don't have time, they could ________. However, if you would rather not, you should probably know that ________ information that you probably don't ________. For example, in the case of independence unit researcher ________, it turns out that they ________, ________, ________, and ________. I'm sure you will be available for facebook to inform you of any relevant information, which I'm sure you would like to read before you sign off on our document. We remain extremely interested in the integrity of our reporting. ________. Please let us know if you have any questions.

With the warmest regards,

Committee for the Analysis of Facebook's Privacy Endeavor

Price Waterhouse Coopers (PWC)

1 Hacker Way,

Menlo Park, CA, 94025

Facebook previews GDPR privacy tools and, yep, it's the same old BS

Re: Wow

>I would have thought that moving data dominion from Ireland (GDPR) to the USA (using some diluted GDPR analogue like Privacy Shield) would not make a significant difference to our rights as EU-located individuals.

Sorry. I might have been unclear. You are correct--no matter where the data lies, facebook is obliged to adhere to GDPR for its European users. However, it is moving data that it might previously have been stored in Europe out of it so that that data is not affected. For example, a user in Algeria whose data might previously have been stored in Europe for the locality now has their data stored outside the area so that it won't be covered by GDPR.

Here's the description of who has to comply with GDPR:

The regulation applies if the data controller (an organisation that collects data from EU residents), or processor (an organisation that processes data on behalf of a data controller like cloud service providers), or the data subject (person) is based in the EU. The regulation also applies to organisations based outside the EU if they collect or process personal data of individuals located inside the EU.