Microsoft: We’ve taken down the botnets. Europol: Would Sir like a kill switch, too? Last December, Microsoft intercepted traffic on users’ PCs and helped break up a botnet. And nobody complained. So the company very tentatively asked at a session on ethics and policy in Brussels this week whether it should do more. John Frank, Microsoft's VP of European Government Affairs, explained how Microsoft had helped …
This post has been deleted by its author
C'mon 'coder, you know that is not an option for everyone at this time.
Many industries, their software suites and workflows don't have Linux/BSD/OSX/Whatnot alternatives. One uses the software that one's clients and partners use.
People who are thus tied to Windows are right to express their concerns and criticisms.
>People who are thus tied to Windows deserve to suffer ever increasing pain.
Eh?
The whole fucking point is that they, the users, have no fucking say-so in the OS they use; they are tied to their industry standard applications. What part of you fails to grok that, you pathetic misanthrope?
""Windows 7 vs. Linux: the Desktop Comparison"
https://www.youtube.com/watch?v=QHCDU-CUoaQ"
I've been playing with it on and off since March 2014. The latest MS imbroglio makes my departure from their products more likely. I've been using their stuff since Win 2x, and a slightly earlier DOS variant. What has happened this year disgusts me even more than previous slipperiness.
By virtue of their latest tricks they are sounding their own death knell.
I have to say I'm pleasantly surprised with Zorin, I put 32 bit Zorinos 9 on an old Compaq Presario C500 and it's at least as fast as 64bit Win7 Asus A53u laptop which is years younger & has vastly superior specs, plus the office suite is included.
I'm looking very hard at Zorin as an upgrade path to avoid Win10.
This post has been deleted by its author
Icon selected for reactions I expect to receive -->
Actually I like the idea, with caveats. But first I'm quite certain that Windows 10 EULA allows this, and much more. So nothing new actually, just a new use of extremely invasive, but apparently legal, technology. Now, here are the caveats:
1) "permitted on the internet" this implies the computer is not being entirely disabled, just some parts of its functionality which depend on connection to Internet won't work anymore. Yes I know we all need Internet to do things we need to do, like fill taxes, chat to family etc. but if your bloody Windows is spewing viruses or DDoS-ing important infrastructure I do not give a sh*t about what you need it for. You are running Windows and by agreeing to EULA you allowed Microsoft to do whatever it wants to do with your machine, now go crying elsewhere (or go ask your friends for Mint Live CD)
2) backdoors. That's a big one, but how much of a backdoor would be actually needed? We are talking about installation or remote activation of a component rendering network component of a computer inoperative, and I argue that what Microsoft has done with Windows 10 users is by far much more intrusive than this. I do not think anything more intrusive would be actually needed and Microsoft already has PKI infrastructure in place for strong enough verification of a signature of any action they might want to take.
Now whether Microsoft abuses or not such a kill-switch is entirely separate discussion - even if they demonstrate the capability to abuse it, it would open them to whole lot of work ("cooperation requests") from law enforcement they might not like. I guess this is actually the question they are asking now - how much of a finger to give, without risking the integrity of the whole arm.
Good side-effect this might have, after 10 or 20 years of such technology no-one will even dream of attaching SCADA and other critical systems to Internet, or anywhere close where they might self-update or do anything else triggered by the vendor and not by the user. Or ideally, no-one will use Windows for anything critical.
"I'm quite certain that Windows 10 EULA allows this,"
Depending on where in the world you are, sections of the EULA are not worth the bytes they take up. From a comment on another story I gather that in the USA "contract is king" and MS may well be able to do this, but in the UK and the rest of the EU and probably most other non-USA jurisdictions, the EULA is subservient to law and consumer rights and can't sign away those rights by agreeing to a EULA.
It might be a different matter if MS were offering to help ISPs detect network abuse or outgoing signals to known C&C servers as the result of infections and give them the ability to easily sandbox those users computers by limiting their access to an ISP website containing advice and downloads to help clean up the offending PC. This way it's only the service providers network access being restricted without any remote control of a users PC and very unlikely to fall foul of existing legislation or require new legislation.
At what point does Windows 10 present you with a EULA and give you the choice to decline? Windows 10 forces itself on people with choices like 'install now' or 'install later'. They cannot stand behind an un-inforcable EULA and say you gave them permission. Can I use that defence if I rob someone? I asked them for their wallet and 'they' gave it to me. (I was holding a gun at the time). It is illegal to frig with someone's computer without their consent.
The law is more important, and the law says that any clause that a consumer who has no easy ability to change EULA clauses would not expect, is invalid an unenforceable.
To pick a daft example:
If the EULA said that you owe Microsoft the blood of your first-born child, would that be valid?
As long as it's all documented in the EULA, it seems reasonable. If you don't like it, use an alternative os.
Whilst we can argue about the need or not for security software that is separate to the OS/platform, it is clear this is another attempt by MS to push AV vendors aside. Remember MS aren't the only one's operating cloud-based security services; and were a rather late entrant to the market...
Interestingly, one of the reasons why Windows has done so well is because of the ecosystem it spawned. So whilst it might seem like a good idea for MS to do more, in fact it will only serve to destroy another part of the Windows ecosystem...
>Interestingly, one of the reasons why Windows has done so well is because of the ecosystem it spawned. So whilst it might seem like a good idea for MS to do more, in fact it will only serve to destroy another part of the Windows ecosystem...
Yes, Windows has done well because it has spawned an ecosystem of productivity applications.
However, the idea that people choose Windows because it has a wider selection of anti-virus software than other OSs seems a bit of a stretch.
It's the servicifation servicisation financialisation of the economy. Soon you won't be able to buy a PC - you'll only be able to lease one, with capital and maintenance costs rolled up into a single monthly payment. We're already there with cars, with mobile phones, even with our homes. Why buy when you can rent instead? The mind boggles.
Eh?
We're living in an age where a £30 PC can run a GUI desktop and traditional office applications, as well as playing HD video.
Phones, likewise. I'm seeing more 'bought outright' phones amongst Joe Public, like the One+ and Cubot, and also people keeping their 2-4 year-old phone and switching to a 'SIM only tariff'. At the moment, to tide me over 'til I can afford a new screen for my 'proper' phone, I'm using an Android 4 phone purchased unlocked from Sainsbury's for £25*.
For sure, there are parts of the world where £25 is still a hefty investment, but not in the markets that have bought IT gear to date.
*It makes phones calls clearly. Android instates my contacts without fuss. Battery is pretty good. It lets me text and email. All in all, a handy spare to keep around. Alacatel Pixi 3.
“We detect when your PC is infected and 'phones home' as much as four times an hour. We then redirect that back to our sink hole and identify that with our national computers, and work to get those machines cleaned up,”
Oh so that call from "support from Microsoft partner" about the virus I had on my computer was genuine after all... no I feel bad having played idiot for nearly 45 minutes with the poor engineer trying to get a remote session on my PC. I'm sure he'll be calling back, one of his colleague already called me a few months ago :-)
"
Just waiting for the day when cops can execute a "search and seizure" of your home PC if it's identified as having a malware infection.
"
Yes, that would be great. And even better, the same law would almost certainly also allow search, seizure & arrest if your computer is detected as connecting to an ISIS website. Or a communist website. Or an anti-Christian website. Etc.
Wouldn't that make the World a far better place?
A kill switch would be a very bad idea - mission or otherwise critical boxes can get infected too (they shouldn't, but it's not a perfect world) and simply switching them off would lead to all kinds of problems. However, re-routing traffic plus a notification seems like a good idea and would help a lot in sorting things out, and faster.
It's actually a reasonable question to ask. If a PC is part of a botnet then it's likely being used for something illegal. Why should ISPs and the like let such machines continue to access their network?
The answer is they shouldn't but they will because profit. So it falls to companies like MS to do something since they're the only ones with both the ability and inclination to do something.
Personally I'm a fan of redirecting all traffic from infected machines to a page with instructions on how to fix the problem.
"how is it shooting themselves in the foot"
Microsoft have raised a possible future tactic that relies entirely on the public trusting Microsoft, and by extension all the governments to which they must answer.
None of those entities are trustworthy. Not Microsoft, not the governments. The fact that Microsoft does not understand this - does not seem to be capable of comprehending the importance of trust - is the footbullet.
Microsoft just thunder around like a monopoly: in their minds there is no need for trust because noone has a choice but to use them. I hope they are proven wrong and driven out of business. With extreme prejudice.
>None of those entities are trustworthy. Not Microsoft, not the governments.
I don't have the coding and mathematical skills to audit all the software I use, so I would rely on trusting others, be them FOSS advocates or otherwise. Aware of my limitations, i would have to work out a chain of trust to someone who knows their onions. That is tricky.
Maybe I would be trusting people on the same continent as me. Chances are, I would have to trust a white, English-speaking man (true of Linux, OSX, Windows, Plan 9 etc). This approach takes us down an unpleasant line of logic, and one remembers that mathematicians and chess players often become nutters, but artists seldom do.
Anyway, software back-doors are not the chief threat to our privacy. In a few years time, sensors will be so cheap and ubiquitous, and the real-time processing so powerful, that we can just be observed at source.
I don't want software back doors, but I do want those who profess to have the skills to stop bashing last decade's bogeyman and instead concentrate their attention on worthier battles.
A kill switch is less of an issue than a backdoor. You won't know if you have been backdoored, unless you are technically sophisticated, or unlucky. If you suddenly find you can only access websites that inform you you have been killswitched, it is very much in the open and up-front, and you'll know you need to do something about it.
In motoring analogy terms: if killswitched, your PC has just failed it's DOT (MOT for old-timers), and until you have the necessary remedial work done, you are not getting back on the road.
People don;t object to having their car inspected for roadworthiness at regular intervals: perhaps the same approach should apply to Internet-connected devices?
It depends on the criteria they use for killing access. From the article they suggest using it if they detect an infection or a crime. I don't mind about the infection part as almost by definition that will be doing something I don't want. But a crime might be something I choose to do in which case I don't want my computer colluding against me with the authorities.
And there's also the question of the degree of infection. MS include Defender with all their OS's now and it's adequate. But what happens when something does get past it? Would you be helplessly booted off the Internet? Do they kick you off at the slightest sign of infection or only if you're wreaking absolute havoc? Who makes that decision? There's a lot to unpack in the details here.
"But a crime might be something I choose to do"
Or quite possibly is something legal in your own country (or a civil case, not criminal) and not in the USA where no doubt they would decide on such action.
Either way, it is a slippery slope to go down. For example, can we then blame MS if they fail to stop botnets DDoS'ing a web site, etc, on the grounds those machines are "under the control of MS"?
So, you object to LoJack on the basis that you reserve the right to use your car in a bank robbery and regard removing it to prevent vehicle tracking helping to apprehend you as an excessive imposition?
That's bonkers. There are arguments to be made here, but keeping your criminal career as effort free (you can always install Linux) as possible isn't one of them.
"
People don;t object to having their car inspected for roadworthiness at regular intervals: perhaps the same approach should apply to Internet-connected devices?
"
OK - so all operating systems must have government approval before they are legal to use on the Internet (Microsoft and Apple will be able to afford to get approval, but it will probably be illegal for Linux users to connect to the Internet). All users must take every Internet device they own to a dealer and pay $$$ for an inspection and certificate every year. After any major change to the system (e.g. installing a new application) the user may not connect to the Internet before taking the device to a dealer for approval.
Also 3rd party insurance will be mandatory in case your computer gets infected and causes damage to someone else's system.
Still sounding like a good idea?
"People don;t object to having their car inspected for roadworthiness at regular intervals: perhaps the same approach should apply to Internet-connected devices?"
My car is inspected by any of thousands of licensed mechanics in my city all of whom must meet regulatory criteria that is regularly reviewed by my government and subject to the input of industry experts. There is a vibrant industry of competition in the provisioning of the vehicles, the maintenance of these vehicles and the inspection thereof.
My car is not subject to the whims of a monopolist who has proven repeatedly that they absolutely cannot be trusted. Microsoft is such a monopolist and they absolutely have proven themselves untrustworthy.
There is no universe in which I will hand over control of my desktop to Microsoft. Not to them, and sure as all hell not to their government.
Microsoft cannot be trusted.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
