avatar image

Welcome, Guest. Please login or register.
Did you miss your activation email?

Login with username, password and session length
  • January 18, 2018, 03:58 PM
  • Proudly celebrating 10 years online.
  • Donate now to become a lifetime supporting member of the site and get a non-expiring license key for all of our programs.
  • donate

Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.

Messages - SeraphimLabs [ switch to compact view ]

Pages: [1] 2 3 4 5 6 ... 13next

IOW, to magnetically 'polarize' solid rock by temporarily transforming it into a molten state in the presence of a magnetic field, which is neither practical nor feasible for Mars.

I wouldn't be too sure of this. A sufficient quantity of nuclear devices detonated at once would result in glassing of the planet's surface, creating the necessary semimolten state to allow the iron particles to align to a suitable magnetic field.

The question then becomes how did earth gets its magnetic field in the first place, since that would probably lead to clues in how you would generate a planetary magnetic field in order to initially charge up the martian crust.

Without the charged particle deflection offered by a properly aligned magnetic field, the atmosphere just gets blown away. Using machinery to just replace it nonstop would result in the planet slowly but surely losing mass.

It is also possible to magnetize iron by way of impact. Vibrations will disrupt the crystalline structure of the metal enough to allow for magnetic polarity alignment, and this can happen through almost any method of making a piece of steel vibrate.

What if a comet crashed to earth at some point in the past that happened to be mostly iron with a pretty strong field? The impact energy would have made earth's surface hot, and the seismic waves of the impact would have vibrated the earth's crust. If that comet was magnetic, the whole planet would have become magnetic basically overnight.

Living Room / Re: Does anyone here use Bitcoins?
« on: August 13, 2015, 08:33 PM »
And that's game.

New York State is requiring sites and services that handle digital currencies on behalf of other people to apply for a "BitLicense" intended to protect end users from the many vicious scams out there and to prevent criminal activity like money laundering and terrorism.

In response though, the majority of the bitcoin-related sites I've been using have opted to simply discontinue service to NYS, listing it alongside Iran and Syria as an area that is not allowed to use their service.

This is the end for bitcoin if other states follow suit. They'll little by little regulate it away until everyone has forgotten what it could have been, then ban it outright and mop up the stubborn people who wouldn't let it go peacefully.

Living Room / Re: New vulnerability found in older Intel processors
« on: August 09, 2015, 01:12 PM »
"Willing but maybe not able" is an interesting category I might fall into.

"Willing" is an hour's worth of convincing. But if it's harder than "upgrade PaleMoon/Floash/Java" then we might be stuck.

It is, and a dangerous process to boot that is best left to people who have experience with it and know how to recover from a failure.

Because a botched BIOS update can brick your system, and a lot of OEM boards do not have functional recovery systems because how dare you try to update the bios outside of the supervision of a factory trained technician.

Naturally this kind of flaw is something the NSA would have known about and been using for the past decade, and it only got exposed because they found a better one to abuse.

It would be worth looking at if Windows 10 wasn't so keen on stealing your personal data to sell to the highest bidder.

Unfortunately there have been multiple levels of privacy intrusions confirmed with Windows 10, and no easy way to disable them.

Not only that, but Microsoft decided to help themselves to your upload bandwidth using their p2p update feature, the details of which I am not clear on but it sounds suspiciously like they intend to make Windows share updates with nearby systems both in your LAN and in your neighborhood WAN at the cost of your network capacity.

Called this.

Sooner or later if it isn't already, sharing a picture of a kitten leaping into a bucket with your mom will be a high profile crime worthy of half a million dollars in fines and fees and 20-life prison time.

I've been telling people to just practice civil disobedience. This is a law that obviously cannot be enforced because collectively the citizenship of the country simply ignores it and has no interest in obeying it.

If the courts actually worked as intended they would recognize this law is of corrupt intentions and does not benefit the people of the country, so they would order it striken from the books without any further negotiation.

Justice is blind, but she can clearly smell money.

Very sad to hear. Can't have anything nice in America.

Even though he made it from coast to coast in Canada, within two weeks of trying to do same in America someone had totally wrecked him.

Living Room / Re: Be prepared against ransomware viruses..
« on: July 04, 2015, 05:45 PM »
However we are talking about ransomware and I fear ransomware can't be stopped by limited privileges. Encrypting data is not a system operation, so I think ransomware are allowed to do it even if privileges are low.
I think limited privileges are useful against other kinds of malware only.

It can be stopped by limited privileges from accessing backups on the network and other machines.  Which was the most tragic part of the incident in the OP.

Not necessarily.

Mapped network drives can be created and accessed by users without administrative access unless a group policy exists saying otherwise.

And Windows also allows users to access removable devices regardless of administrative access. Including any remote network filesystem that it has read-write access to.

Messing with user privilege would not have any impact at all on the speed of ransomware encrypting files unless that user privelage change also had associated restrictions on CPU and IPOS resource consumption.

Living Room / Re: Save Domain Privacy
« on: June 30, 2015, 05:09 PM »
I formerly used a PO box, but I was paying for a PO box just for this.  Then I used privacy, but GoDaddy are greedy.  So now I use Tucows, and they give it for free.  And as far as the pre-paid, you still have to top it up, or they'll cut it off.

PO Box costs me $50 per year, I use it as my main mailbox anyway since I don't trust my neighbors to not snoop.

The prepaid phone I have has two expirations. The first is based on minutes used- it counts down till its gone. It also keeps track of service days remaining, again counting down till its gone.

And its all too easy for one of these phones to end up with like 300 minutes and several years of service time on it, making it ideal for a low cost point of contact to comply with ICANN regulations.

My domains are registered with dyn, and they want like $20 a year on top of the registration for the privacy service. Multiplied by the list of domains I have.

Using the PO box and prepaid phone instead, my info is compliant with ICANN regulations, but at the same time I am not inviting unwanted guests or getting a lot of telemarketters like I would be with a normal phone line and mailbox arrangement.

Living Room / Re: Save Domain Privacy
« on: June 30, 2015, 04:40 PM »
They haven't yet made it against policy to use a PO Box as your contact info.

And the phone number is allowed to be a crappy prepaid phone that is never actually used- it just has to actually ring and be answered after a couple of attempts.

As usual the only people to lose out to policies like these are the masses. The savvy can still evade them as easily as they please.

I've been rather annoyed lately with the ICANN policy changes, and the general attitude of 'we must regulate and make a profit from everything that exists on the internet  and how dare you try to do otherwise'

namecoin is looking rather attractive right now as an alternative technology to the existing ICANN regulated DNS infrastructure.

I was actually thinking about what it would take to create a namecoin-like decentralized SSL certificate system, along the lines of your sitename and the public fingerprint of your site's SSL certificate are published in a blockchain so that anyone at any time can check to see if they have the correct certificate for the site in question. This would eliminate the need for a central authority for SSL certificates much the same way Namecoin eliminates the need for a central authority to serve DNS.

Living Room / Re: Be prepared against ransomware viruses..
« on: June 27, 2015, 09:40 PM »
Are there any fast global checks? Like are the ransomed files renamed to some bizarre file extension, or just ".zip" (that happens not to be unzippable)? So then you could put a list of all sane file extensions somewhere, and then some kind of deep background process that says "hey, if you find yourself creating anything evil, stop all activity and holler"?

The one I encountered turned every image, office document, email, and html file into a .EXX added on to its normal extension.

I literally had dodged a bullet with it- the night before I had noticed it acting funny and kicked it off the network suspecting malware. Next morning it had the cryptolocker ransome notice up and while it still ran all of the documents on it had been encrypted.

If I had left it alone it would have tried to encrypt everything it could reach on the fileserver.

Living Room / Re: Be prepared against ransomware viruses..
« on: June 27, 2015, 07:36 AM »
My only encounter with it thus far it came in as an email attachment. A fake PDF that was actually executable and would install malware.

Though in my case I would have survived it just fine. It would have of course encrypted the main network shares on the server, but backups of those shares are taken daily via rsync to another box and then only offered up as read-only so if I need to retrieve something I can.

Still hard to believe anyone would actually send a payment, but a lot of people would have no clue what to do about it and wouldn't want to lose their stuff.

...whoever wrote malware like this should be executed by firing squad.

Living Room / Re: The end of the hard disk
« on: June 25, 2015, 12:08 PM »
Next time I will be more precise and call it "storage device"...that should make everyone happy. :)

I was further confused by the "SSDs are already old hat" because the PCI-e mounted SSDs are SSDs. That, combined with "hard disks that work according to SSD principles" made me think this was some new technology I hadn't heard about before.

I'm not trying to be pedantic. I'm just trying to explain why what you said confused me. :)

It's all good though. The miscommunication has been cleared up.

It could have been this technology.

Surely you guys have heard of the 'core memory' from the 1960s that was literally a woven grid of wires and cores, and ended up setting the standard geometry and behavior for main ram for decades to follow.

Ferroelectric RAM is technology going full circle like it has done in a lot of markets lately. It reimplements effectively core memory on silicon, creating a non-volatile memory that is fast enough to serve as a main system ram while also being sufficiently stable to operate as a mass storage device.

Though there are currently density limitations that have kept it from becoming used in the mainstream, once those limitations are overcome this is a technology that could blow SSDs and the now commonplace DRAM out of the water.

I'm still not too sure on SSDs for long-term reliability. Though they do offer great improvements in performance and the price tag is now comparable for applications not requiring large capcacities. I've been deploying workstations based on 120GB SSDs in place of 320GB HDDs for a couple of years now with good results, and am considering the possibility of using one or more SSDs in my next server build as either a filesystem cache or as the actual primary storage device.

Pricing for larger capacity devices is still a sore spot, but it has gone down considerably since they were first introduced.

Oh and Renegade, you DID mismatch the two devices right? RAID1 of SSDs is not safe if you ordered both devices around the same time and they have been together in the raid the whole time. The result is that they will die by way of media wearout within a few days of one another usually.

To counter this, order each device separately from either a different vendor or a few weeks apart. That way you aren't as likely to run into SSDs from the same manufacturing batch or design revision. The manufacturing variation between batches makes enough of a difference in the actual usage life to expand what could be only a few days window between failures to be several weeks- enough to install a new device and rebuild the array.

Living Room / Re: Tindie
« on: May 20, 2015, 07:51 AM »
Poking through it, the concept looks rather interesting.

Maybe instead of making only one ignition coil driver I'll make like 10 of them and sell the extras for other hobbyists who want one. I would just have to open source my design for it, which isn't too bad considering its largely derived from well known designs.

Living Room / Re: Http vs Https Universally
« on: May 16, 2015, 07:14 PM »
There's actually two even bigger problems with https than just the cost of getting certs. Also I use self-signed certificates for most of my stuff, which provide the same encryption bonus free of charge. Tradeoff is you then no longer can be sure of what server you are talking to unless you've made your own certificate authority and have traceability to your own root certificates.

The first is IPv4 depletion. SSL only allows one site per IP, and sites with it have always had an additional overhead cost in provisioning the dedicated IPv4 required to make it work. IPv6 would help mitigate this, but all too many ISPs are behind the times and haven't even looked at IPv6 rollout on their networks. After all IPv4 is still working, why should they spend their precious profits installing new IPv6 capable infrastructure when its not broken yet.

The second is caching, which really helps keep the internet bandwidth-efficient especially in the Americas where people are still using Dialup here in 2015. By definition, https cannot be cached because that would require the proxy to be able to decrypt the content in order to make the decision of if it should keep it or not. And a properly functioning encryption the data will change each time the page loads, completely defeating any possibility of caching it without having to trust the proxy with unencrypted data. Browsers will do some caching though, but a lot less of it is possible on https.

Living Room / Re: New Virus or ??
« on: May 05, 2015, 08:57 PM »
Quarantine the offending system- disconnect it from any and all networks. Do not put any writeable media in it, any incoming tools must be brought in using finalized CDRs so that whatever it is cannot spread.

Is there anything worth noting in the Windows event logs?

Does it still run the malware when started in safe mode?

Also have you tried booting from a Linux LiveCD and looking at the filesystem to verify it the data is actually gone. At this point I would be hesitant to copy data off of the machine until you know what you are dealing with, but important info can at least be retyped into another system.

It does sound like some type of virus, quite possibly a ransomware that then retaliates like this if not paid off.

Living Room / Re: Kickstarter Highlight: Onion Omega
« on: May 05, 2015, 04:25 PM »
Using a .io domain name instead of one of the major TLDs. -3 right off the top
-SeraphimLabs (May 05, 2015, 11:49 AM)

The only .io sites I've heard of have been reputable. Two off the top of my head:

It may very well be legit.

But lately I've seen a lot of startups and pump & dump scams with sites based on the .io tld.

Most of them have been bitcoin related though.

The general plan though is to pitch an idea and collect funding for it, then go quiet for a few months during which time little progress is made, followed by a surge of drama as a web of lies comes crashing in.

Its not at all for certain, especially because there are a lot of scams on any TLD out there, but seeing a new startup using .io is making me a little hesitant. Beyond that everything else looks amazing, I want to try these devices out.

Living Room / Re: Kickstarter Highlight: Onion Omega
« on: May 05, 2015, 11:49 AM »
Using a .io domain name instead of one of the major TLDs. -3 right off the top

I have had a lot of problems with sites on .io domains being either ponzi schemes or outright scams.

The concept is solid though, and if they have a working prototype then it looks to me like this could be a very interesting thing.

It even uses a Linux based OS that I am familiar with, using OpenWRT-equipped routers as wifi hotspots in my workplace in favor of actual business class access points.

I'll keep tabs on it as it progresses.

^^ One of the biggest downers to me as I became competent fixing cars was the "Classic Movement" or whatever one wants to call it.  The main advantage of being willing to crawl around under cars in nearly freezing weather, enduring icy water dripping down your neck when doing front end alignments, and putting up with too hot engines(lean burn) when doing tune ups, was the fact you could pick up a car with some years and miles on it and make it run well on the cheap.

That's still quite possible, and experiences I know all too well. Clunked the barrier on ice in January a few years back and spent several days under a tarp in a blizzard with the front end of the car apart stitching the ~300 wire PCM harness back into a sane configuration so that the engine will fire up again. Or one of my more recent ones, on my way to a christmas dinner with the wife's family and I burst a transmission cooler hose. Ended up getting her and her brother helping me push it more than a mile to the parking lot where I work, all my tools were there to piece it back together.

Its just harder to do. As cars become increasingly automated, they become less tolerant of components wearing out and systems not quite performing as they used to. It makes the PCM liable to just freak out and shut the car down, then throw tantrums at you when you try to troubleshoot and get it going again. Even then though, I have no problems picking up a car that is one step away from the scrap heap for $1000, putting new CV joints and brakes on it, and driving it for 3-5 more years and 50k miles further before it finally passes beyond my ability to keep it clunking and goes to the scrap anyway.

In that regard the cash for clunkers program really screwed me over. Thousands of perfectly good candidates that were the right age and while quite travelled still had plenty of useful life were collected and intentionally destroyed in the name of government subsidy. Since then the price of old cars has gone way up because an entire decade's worth of easily repaired used cars was lost.

Living Room / Re: Ethernet cables: UTP vs. STP
« on: April 27, 2015, 04:41 PM »
I personally hate STP. The shield is annoying to work with when preparing the cable, and is of no benefit at all unless connected to a good earth ground at one end only. UTP cable is cheaper and easier to install.

Most installations don't take enough care in preparing the cable and properly grounding it to actually achieve any benefit from having it shielded, and accidentally grounding the cable at both ends can make the interference worse by way of ground loop.

Gigabit ethernet will work over even Cat3 at short distances despite being out of spec, but the further you go the higher quality cable you need for reliable transmission. These days I usually use Cat6 for all new installs, even though I have 5e cables over 100 feet long still in service and achieving reliable 1Gbps speeds.

The quality of the switchgear and NIC matters as much as the cable itself does. Cheap NICs and switches do a poor job of filtering noise and will be unstable at speed over a long or old cable, while the good ones can meet and even exceed the standards for network performance and noise immunity.

Its not hyperbole if people are constantly replacing their phones because they have to have the latest fashionable new model even though their current one is still fully functional.
-SeraphimLabs (April 26, 2015, 06:12 PM)

The problem with that statement is that it isn't the devices.  It's the people.  Which isn't what this article was about.  So... hyperbole.

Not if after a generation or two the device was redesigned with a 1-2 year upgrade program in mind. If people aren't going to keep the same device for longer than that, why design the device to last longer? Make it last only as long as the average consumer will use it, and never mind the outliers that keep the same device for years on end they are obviously not the people you should be designing for.
-SeraphimLabs (April 26, 2015, 10:56 PM)

But there is no proof that they are.  As shown by the fact that my iPad 1 is still in excellent condition, and sells for 1/5 the price I bought it for 5 years ago.  There is the consumer use case, but the outliers are very much still there, especially with the upgrade policies as they are and breakage/loss.

That isn't how design life works though. And the first generation of a product line is often overbuilt compared to those that follow because the typical use case has not been as well established. Later generations incorporate wear and failure analysis of previous generations, correcting weak spots while at the same time weakening strong points to cut costs.

According to the equations in my mechanical engineering books, automotive mechanisms should have a wear allowance sufficient for approximately 160,000 miles. This is pretty close to the factory warranty on most vehicles interestingly enough, the warranty expires around the time the vehicle is expected to have used up its designed-in wear tolerance.

Even though my car is currently at 202,000 with no major mechanical problems that I am aware of. It has gone well beyond its design life on most of its components, and other than the components I have replaced is a device which has exceeded its design specifications. On the other hand most cars of the same age have already been crushed for scrap, most of them due to wearing out or being damaged beyond where it is economical to repair.

Design life is not an exact science. You are designing to where the majority of a product will operate for the calculated time period without major issues. It is possible to exceed that lifetime if you take good care of your belongings or it was built with quality, but in the field most of what was produced is expected to be replaced failure or not.

Its not hyperbole if people are constantly replacing their phones because they have to have the latest fashionable new model even though their current one is still fully functional.
-SeraphimLabs (April 26, 2015, 06:12 PM)

The problem with that statement is that it isn't the devices.  It's the people.  Which isn't what this article was about.  So... hyperbole.

Not if after a generation or two the device was redesigned with a 1-2 year upgrade program in mind. If people aren't going to keep the same device for longer than that, why design the device to last longer? Make it last only as long as the average consumer will use it, and never mind the outliers that keep the same device for years on end they are obviously not the people you should be designing for.

Perhaps usage statistics have been kept showing how long on average they last.

I've said this for a long time as well. Most of your 'environmentally friendly' measures implemented on American cars are in fact a burden that reduces your fuel economy. And it is already a federal crime to remove or bypass an emissions control device, even though basically every performance equipped custom vehicle started out by removing most if not all of the emissions devices in the name of performance gains.

And I like so many other technically inclined people have a healthy enough distrust of computing devices to not ever allow a system to tell me what I cannot do. It either does what I tell it to, or it finds itself being reprogrammed to a system that I can be in control of. Its a shame cars are too varied to really allow a GPL drop-in replacement for the typical PCM, but perhaps a platform that could be coupled with DIY hardware using a modular software structure akin to most Linuxes would make it fairly easy for someone with a working knowledge of computers to assemble and program an open source PCM to fit their vehicle. I wouldn't mind trying it myself.

But a lot of cars you can completely redefine their operating behavior as easily as changing a memory cassette in the PCM, with more modern vehicles being dealer-reprogrammable by simply re-flashing the software and tables inside it. I see this as a play by the automakers to guarantee their stranglehold on that capability, so that nobody outside of their authorized dealerships can legally have the tools or software to do modifications on that level, modifications often required by high end performance modding to allow the PCM to work with performance parts.

Far as I know though, the ultimate you can get in fuel economy for a gasoline engine with modern technology is the following. Anything else just burdens the motor or makes you use more fuel than you need. :

Feedback Carburetor- this does have an oxygen sensor and computer, and uses feedback from the oxygen sensor to adjust bypass air. Like so you set the base mixture to suit the engine, and then the bypass trims it richer or leaner as needed to compensate for variations in driving conditions and engine behavior.

"Straight" Gasoline - NONE of that Ethanol crap! I've tested this myself, using '87 octane' gasoline purchased in bulk for a farm tested against '87 octane' pump gasoline from the station downtown that is marked as containing up to 10% ethanol. Well, on a 1.6L carbureted engine, the straight gas was giving me almost 40 MPG, while the E10 was barely making 30 MPG. Now think about this a minute- gas consumed vs mileage travelled, and you suddenly realize that E10 gasoline is causing me to burn more actual gasoline than i would be using on straight gas.

Electronic spark advance - this and the feedback carburetor are among the very few places where the electronics do beat their mechanical counterparts. Simply because the ignition timing can be made far more precise and can respond to changes in operating conditions far more easily than can be done mechanically. Plus the breaker points are a high maintenance area that is often neglected, letting the electronics win in this area eliminates an often neglected maintenance item that has fuel economy penalties for ignoring.

Its not hyperbole if people are constantly replacing their phones because they have to have the latest fashionable new model even though their current one is still fully functional.

So even though it probably could last longer, in practice it will not and it is designed with that in mind.

Its like how cars are litearlly DESIGNED to last 160,000 miles. I have mechanical engineering books with equations that literally let you calculate how long the components will last so that you can shave off costs as much as possible by shortening its service life and then setting a warranty that expires when the designed service life is also used up.

Disposable Society is finally showing itself to be the problem it truely is. Certainly took long enough.

Valve and Bethesda have introduced a program where game mods for Skyrim can be purchased from the Steam Workshop, where traditionally game mods for any game have been available for free.

Needless to say this has kicked over the beehive of a game modding community that has been thriving for the past 10 years without any corporate oversight or financial incentives, and the past few days have been rather drama-packed.

Many mod makers have come right out and said they will not use the new service out of principle, and a lot of the mod makers who have joined the dark side and listed their mods with Steam as buyable packages are facing community blacklisting over how quickly they choose to be greedy and demand money for code that has traditionally been free and is part of a complicated spiderweb of dependancies.

The greatest achievement of PC gaming- a platform that can be easily customized to suit a user's taste has been undone overnight, as suddenly the community spirit that has kept it going all these years is shattered by greed and the cold reality of the for profit mindset.

And seeing all this happening, it is so incredibly tempting to instruct paypal to chargeback every purchase I have made on Steam. Steam would of course ban my account for this action, but I would get my money back and Steam would be one returning customer less.

My opinion of the situation is that game mods are a hobby- something you do because you enjoy it and are having fun with it. The minute you start making a profit from it, its a business interest not a hobby, and you have exposed yourself to all kinds of legal problems between taxes, IP rights, and backlash from people who enjoyed what you made that do not want to pay for what they were getting for free.

Its funny how this has gone almost completely unnoticed, probably because of how rediculous it is.

Meanwhile the internet is in an uproar over what Steam is up to.

Well the joke's on Valve. I always use paypal when buying from Steam, so they can go kill each other over my purchases I'll just be here playing Age of Empires Gold edition, which I bought nearly 15 years ago and being free from DRM and callhomes still works almost as good as when I bought it. Just gotta load up a VM of Windows 98 to make it run right.

Pages: [1] 2 3 4 5 6 ... 13next