Preach, brother. Vmware's "vCloud" is a huge, complicated load of crap. I would like to say it's easier to install and use than OpenStack, but I've dealt with both... I'd almost rather quit my 6-figure job to flip burgers than deal with either of them. (and yes, there are companies that sell virtual infrastructure using vCloud. and that speaks *volumes* about how horrible openstack was to setup at the time.)
The strength of OpenStack is in it's "free". (and thanks to a number of distros, it's getting much easier to install) The strength of VMware is a long list of powerful features (vmotion, HA, Fault Tolerance, etc) and near brainless simplicity to install and setup -- esxi: 5min, vcenter appliance: under 30min.
I don't know what part of Raleigh you live in, but in my parts, the power is quite stable. You may be thinking of Apex... where the city bills you for CP&L services. (they have something like 2 bucket trucks to actually fix anything.)
(Decades back at NCSU, I was in Bragaw, on the phone, when a squirrel was carbonized in the substation that feeds the entire main campus. Very loud boom and no power for most of the day.)
That might have been the first 30s of his work, but he went a a lot further than that in his tinkering. ('tho he claims to have never had or ran the actual BitMover client, he sure as hell interacted with them. 90% of the beef here is that he continued to refine his client after agreeing to stop.)
Actually, it appears to be rolling along at the same pace it always has. The only thing that has changed (*AGAIN*) is Linus's idiotic idea of release numbering. He should stop sipping from the teacup of Microsoft and look instead at Cisco -- while it may have some oddity with special branching, it's had relatively the same release vs. development process for decades. I'm not saying we should return to the insanity that was 0.99pl5j1e, but 3.x isn't any better.
... and nothing to validate the subject cn against.
Sure there is... the SSID, the BSSID (AP MAC), the enterprise domain to which you're trying to join, etc. But, yes, it's all moot if the OS presents the user what a "whatever, just do it already" button.
And what you fail to realize is their estimates where just that... estimates. They were way wrong then, and even more so now. Such a project today would run into the trillions, over-run every budget imaginable, take a century to "build", and yet never be finished or fully functional. There are simply too many underhanded people out to make a buck to engineer and build anything on this scale.
The "military grade" systems (which have been around since the 90s -- 'tho I'm unaware of them being anywhere but the test ranges at Aberdeen) don't use batteries; they use massive (gold) capacitors. I don't care what sort of battery you have, it will not dump as fast as a capacitor.
Actually, NO. Semi-auto firearms are slightly complicated mechanical designs. They are inherently "fully automatic", but their design incorporates elements to halt the cycle.
Making a full-auto firearm requires paperwork and licenses. It is a felony for anyone without such to fabricate a full-auto firearm (or alter/convert to full-auto) -- even discussing it is illegal.
I was thinking the same thing. While there are metal printers, they, and their output, are super-expensive. And there's ZERO reason to go to all that expense.
For the record, I built the same thing (a "rail gun") in high school, in 1988. Obviously, they have access to much better batteries today -- we used car batteries -- and fancy electronics -- despite doing this in an electronics class, ours was far more low-tech (blob of mercury(!), garden hose, and a series of pins... whip the hose and a nail goes flying through a 2x4.) Their system appears to have limited range, no stability (the nails are tumbling 99% of the time) and thus low penetration. [I still wouldn't advise standing in front of it.]
It's a good start to the latest chapter in their FBI file.
The mom-n-pop places likely still do it by hand -- and the keys they make are far more likely to work. But the mega-marts just don't want to have to pay people to do it -- and the people they have (had) doing it are complete crap at it.
The controller on a lot of HID systems is an XP box
The actual controller of these systems is a small, almost brainless, microcontroller. The thing humans use to interact with it (the "ui") is a windows application. When the microcontroller loses it's cookies, and the windows system doesn't have a backup, well, that'll take a while to sort out.
Your post is full of so many illegal things... a fire alarm that can be disabled by fire before going off: I'm pretty sure that's illegal everywhere in the US.
If you're talking about mag-locks -- electromagnets that hold the door closed from ingress and egress, then that is not "low-security"; it is, in fact, the definition of high security... you don't get through that door, in any direction, without the appropriate access. Again, by code, they must release when the fire alarm is activated so people can evacuate however necessary.
When was the last time you had a human cut a key at Walmart? (or Lowes, Home Depot, etc.) It's been about 2 years, here. It's a kiosk now. (for the record, the keys maybe by the human... none, 0%, of them worked.)
You've not looked at many modern web "applications", have you? The amount of javascript, style sheets, and html markup is ENORMOUS. It's common for sites to save 50-75% of bandwidth enabling compression. (for sites that aren't primarily images, etc.)
It works because it's not ethernet. PPP, even "oE", is still point-to-point. Cablemodems and Passive-Optical-Networks aren't ethernet either -- at best they're "ether-like", i.e. capable of carrying an ethernet frame (what's called MAC encaps today.) (See also: Frame-Relay hub-and-spoke vs. fully-meshed) By your definition, VLANs are a perversion of ethernet.
None of this is "new". Cablemodems have been around for over a decade, assigning a single IP per connection. And DSL was already doing it years before that. And UNIX(tm) hosts were doing something very similar long before "the internet" was a word. (see also: proxy arp)
(Also, IPv4 and Ethernet are isolated, independent standards.)
In the IPv4 world... granting a static subnet to a consumer wastes a minimum of 3 addresses (network, broadcast, and router) and often several more as the entire block doesn't get used.
The common ways IPv4 has been provisioned for, well, ever has been PPP (both ends are a/32), or various "split horizon" multi-access setups such as cablemodems, and PONs where it's a large subnet but each node is isolated from each other. (That used to cause problems with two local nodes wanting to communicate, but modern technology has addresses that)
It's a well established fact that "users" (in the collective sense) are morons who cannot secure their network or the device(s) they connect to it. This is precisely why we have port blocking. (for the record, since the days of dialup!) These blocks are the rudimentary ISP-side "firewalls" -- and that's pretty much as fancy as it's ever going to get. The ability to remove them is questionable (see above re: morons.)
(the slashdot crowd is really the wrong group to use as a metric)
Ok, troll. Unless I'm selling (or giving away) VPN access ("service") to others, no, it's not. By your, extremely lame, definition, any inbound listening port is a "server".
The problem is defining "server" in a manner that doesn't read like the "knives in school" statute. (which in NC, is MANY pages) Thanks to we-don't-want-to-be-pinned-to-a-corner legalese, there's room to drive a truck through FCC's 10-201 (all 194 pages of it.) Section 72 gives the ISP broad power to define whatever "tiers" they wish. And section 77 allows those tiers to include a "no servers" clause "to effectively and reasonably manage their networks".
No. This is about misuse of a service that's intended for consumer use. What you're talking about -- and the relatively little traffic it entails -- would fit within that definition. Throwing up a website for Bob's Boats -- you being Bob, who makes model boats -- would not; because the primary purpose of such a site would not be for Bob, but for the internet at large to access, with potentially high bandwidth results.
Well, you *DO* live in such a divided world. And you bought a (much cheaper) "consumer" internet connection to use as a "producer". There really no issue with running services on your home connection for your use. The issue is when those services are not primarily for the owner's use. Hosting Exchange for a dozen companies, hundreds of blogs, e-commerce stores, etc. are undeniably "business" uses of one's "consumer" internet account.
(Note: One can run any number of services without exposing a single one to the internet. It's spelled vee-pee-en, children. I've done so for well over a decade.)
Slashdot Mirror
Preach, brother. Vmware's "vCloud" is a huge, complicated load of crap. I would like to say it's easier to install and use than OpenStack, but I've dealt with both... I'd almost rather quit my 6-figure job to flip burgers than deal with either of them. (and yes, there are companies that sell virtual infrastructure using vCloud. and that speaks *volumes* about how horrible openstack was to setup at the time.)
The strength of OpenStack is in it's "free". (and thanks to a number of distros, it's getting much easier to install) The strength of VMware is a long list of powerful features (vmotion, HA, Fault Tolerance, etc) and near brainless simplicity to install and setup -- esxi: 5min, vcenter appliance: under 30min.
I don't know what part of Raleigh you live in, but in my parts, the power is quite stable. You may be thinking of Apex... where the city bills you for CP&L services. (they have something like 2 bucket trucks to actually fix anything.)
Or squirrels! Don't forget squirrels.
(Decades back at NCSU, I was in Bragaw, on the phone, when a squirrel was carbonized in the substation that feeds the entire main campus. Very loud boom and no power for most of the day.)
That might have been the first 30s of his work, but he went a a lot further than that in his tinkering. ('tho he claims to have never had or ran the actual BitMover client, he sure as hell interacted with them. 90% of the beef here is that he continued to refine his client after agreeing to stop.)
Actually, it appears to be rolling along at the same pace it always has. The only thing that has changed (*AGAIN*) is Linus's idiotic idea of release numbering. He should stop sipping from the teacup of Microsoft and look instead at Cisco -- while it may have some oddity with special branching, it's had relatively the same release vs. development process for decades. I'm not saying we should return to the insanity that was 0.99pl5j1e, but 3.x isn't any better.
Sure there is... the SSID, the BSSID (AP MAC), the enterprise domain to which you're trying to join, etc. But, yes, it's all moot if the OS presents the user what a "whatever, just do it already" button.
And what you fail to realize is their estimates where just that... estimates. They were way wrong then, and even more so now. Such a project today would run into the trillions, over-run every budget imaginable, take a century to "build", and yet never be finished or fully functional. There are simply too many underhanded people out to make a buck to engineer and build anything on this scale.
The "military grade" systems (which have been around since the 90s -- 'tho I'm unaware of them being anywhere but the test ranges at Aberdeen) don't use batteries; they use massive (gold) capacitors. I don't care what sort of battery you have, it will not dump as fast as a capacitor.
Actually, NO. Semi-auto firearms are slightly complicated mechanical designs. They are inherently "fully automatic", but their design incorporates elements to halt the cycle.
Making a full-auto firearm requires paperwork and licenses. It is a felony for anyone without such to fabricate a full-auto firearm (or alter/convert to full-auto) -- even discussing it is illegal.
I was thinking the same thing. While there are metal printers, they, and their output, are super-expensive. And there's ZERO reason to go to all that expense.
For the record, I built the same thing (a "rail gun") in high school, in 1988. Obviously, they have access to much better batteries today -- we used car batteries -- and fancy electronics -- despite doing this in an electronics class, ours was far more low-tech (blob of mercury(!), garden hose, and a series of pins... whip the hose and a nail goes flying through a 2x4.) Their system appears to have limited range, no stability (the nails are tumbling 99% of the time) and thus low penetration. [I still wouldn't advise standing in front of it.]
It's a good start to the latest chapter in their FBI file.
The mom-n-pop places likely still do it by hand -- and the keys they make are far more likely to work. But the mega-marts just don't want to have to pay people to do it -- and the people they have (had) doing it are complete crap at it.
The actual controller of these systems is a small, almost brainless, microcontroller. The thing humans use to interact with it (the "ui") is a windows application. When the microcontroller loses it's cookies, and the windows system doesn't have a backup, well, that'll take a while to sort out.
Go Alien 3 and put a breathalyzer on every door. *grin*
Or better... the nuts who installed the motion sensor put it on a ceiling tile that goes all the way over the door. Thump tile, door unlocks. :-)
Your post is full of so many illegal things... a fire alarm that can be disabled by fire before going off: I'm pretty sure that's illegal everywhere in the US.
If you're talking about mag-locks -- electromagnets that hold the door closed from ingress and egress, then that is not "low-security"; it is, in fact, the definition of high security... you don't get through that door, in any direction, without the appropriate access. Again, by code, they must release when the fire alarm is activated so people can evacuate however necessary.
One can get the same audit trail with dry-contact or magnetic sensors. If done well, no one will ever notice they're there.
When was the last time you had a human cut a key at Walmart? (or Lowes, Home Depot, etc.) It's been about 2 years, here. It's a kiosk now. (for the record, the keys maybe by the human... none, 0%, of them worked.)
You've not looked at many modern web "applications", have you? The amount of javascript, style sheets, and html markup is ENORMOUS. It's common for sites to save 50-75% of bandwidth enabling compression. (for sites that aren't primarily images, etc.)
It works because it's not ethernet. PPP, even "oE", is still point-to-point. Cablemodems and Passive-Optical-Networks aren't ethernet either -- at best they're "ether-like", i.e. capable of carrying an ethernet frame (what's called MAC encaps today.) (See also: Frame-Relay hub-and-spoke vs. fully-meshed) By your definition, VLANs are a perversion of ethernet.
None of this is "new". Cablemodems have been around for over a decade, assigning a single IP per connection. And DSL was already doing it years before that. And UNIX(tm) hosts were doing something very similar long before "the internet" was a word. (see also: proxy arp)
(Also, IPv4 and Ethernet are isolated, independent standards.)
In the IPv4 world... granting a static subnet to a consumer wastes a minimum of 3 addresses (network, broadcast, and router) and often several more as the entire block doesn't get used.
The common ways IPv4 has been provisioned for, well, ever has been PPP (both ends are a /32), or various "split horizon" multi-access setups such as cablemodems, and PONs where it's a large subnet but each node is isolated from each other. (That used to cause problems with two local nodes wanting to communicate, but modern technology has addresses that)
It's a well established fact that "users" (in the collective sense) are morons who cannot secure their network or the device(s) they connect to it. This is precisely why we have port blocking. (for the record, since the days of dialup!) These blocks are the rudimentary ISP-side "firewalls" -- and that's pretty much as fancy as it's ever going to get. The ability to remove them is questionable (see above re: morons.)
(the slashdot crowd is really the wrong group to use as a metric)
Ok, troll. Unless I'm selling (or giving away) VPN access ("service") to others, no, it's not. By your, extremely lame, definition, any inbound listening port is a "server".
The problem is defining "server" in a manner that doesn't read like the "knives in school" statute. (which in NC, is MANY pages) Thanks to we-don't-want-to-be-pinned-to-a-corner legalese, there's room to drive a truck through FCC's 10-201 (all 194 pages of it.) Section 72 gives the ISP broad power to define whatever "tiers" they wish. And section 77 allows those tiers to include a "no servers" clause "to effectively and reasonably manage their networks".
No. This is about misuse of a service that's intended for consumer use. What you're talking about -- and the relatively little traffic it entails -- would fit within that definition. Throwing up a website for Bob's Boats -- you being Bob, who makes model boats -- would not; because the primary purpose of such a site would not be for Bob, but for the internet at large to access, with potentially high bandwidth results.
Well, you *DO* live in such a divided world. And you bought a (much cheaper) "consumer" internet connection to use as a "producer". There really no issue with running services on your home connection for your use. The issue is when those services are not primarily for the owner's use. Hosting Exchange for a dozen companies, hundreds of blogs, e-commerce stores, etc. are undeniably "business" uses of one's "consumer" internet account.
(Note: One can run any number of services without exposing a single one to the internet. It's spelled vee-pee-en, children. I've done so for well over a decade.)