Not "for free". Settlement Free Peering is based on a mostly balanced flow of traffic. The instant that ratio moves from 1:1 to 100:1 .... "free" isn't in the room anymore.

And THIS is what makes them not common carriers; ISPs can do this. In the Telco world, interconnect fees are required to be symmetric, for example: if the agreement is that charges carrier A $0.05 per call record to terminate onto carrier B's network, then it must charge carrier B $0.05 a call to terminate onto carrier A's network, it's not allowed to charge carrier A $0.05 per call and give carrier B free service into carrier A's network. An interconnect agreement cannot be terminated or repriced to favor specific networks, just based on the ratio of calls in or out.

That would have zero impact. This is like the telephone company in city A have 96 channels to the telephone company in city B, but then 100 people try to make calls. Only some of them will go through, and that's a capacity issue, not regulated by Common Carrier status.

That's not the scenario. It's not capacity between cities, in this case, it is capacity between networks. The problem is they are discriminating against some networks and refusing to build capacity at the same time as they are building capacity to other networks for free; that's not a common carrier.

A telco expands capacity based on utilization, and in building more capacity to other networks in the same area: it's not a case of some networks get capacity built to them for free and some have to pay, it's..... each telco pays their own costs to build that capacity needed by their customers AND asymmetric usage is settled through the USAGE fees associated with LD termination on each call.

P.S. Alternatively, the information can be uploaded in encrypted format + Base64 to places such as Pastebin, or Freenet, or other massively distributed publication platform.

The card can then contain just a few 40-character URLs followed by 512-bits worth of cryptocurrency wallet addresses.

Then a couple of 256-bit decryption keys for the coded messages and the rest of the card can be used for a list of randomly generated initialization vectors that will be used for further encrypted messages.

So the website can contain an arbitrarily large amount of information which can then be decrypted using the data on the card.

Also, additional information can be added later by creating a spend transaction to one of the cryptocurrency addresses listed on the card, and publishing the information in the public blockchain, but on the public blockchain the text can be encrypted with the key and one of the initialization vectors on the card can be used. More random initialization vectors and additional addresses and crypto keys required to be provided inside each Nth encrypted message uploaded to the blockchain.

As long as they're going to build and provide plugins for the common use cases, who cares? It could be pretty cool. Sounds like a more flexible architecture if they can support multiple package systems.

On the other hand, it might just be unneeded complexity. Frankly, in a lot of places I like things that are SIMPLE, Reliable, and Fast. I prefer programs that do one thing well and work with other programs using standard text-based communications to meet the rest of needs instead of attempting to accomadate every possible use case within one piece of software ----- a plugin architecture with proprietary interactions is not necessarily "working well" with other programs; it's kind of the opposite idea, expanding one program with addons.

It makes sense for a PACKAGE system to support multiple kinds of storage repositories, like Yum or APT does.... you can have your CD-based Repos, FTP based Repos, and private mirrors.

I'm not sure it makes sense for a package management system to support arbitrary backend plugins.

Or just anonymizing the records as soon as no longer required.

However, 5 years is not very much time if they may need to investigate the possibility of fraudulent applications in the future.

They're late comers to this party: We already have community repos. Chocolatey and BoxStarter. Why would we need OneGet?

It's Internet Explorer vs Netscape all over again :)

Let's change that up slightly, to use 3715 bits out of the 7104 available, approximately 50%:

• E-mail address = 40 bytes
• Social Security Number binary encoded - 9 digits = 29 bits.
• Health Insurance Provider Name - 16 alphanumeric characters = 12 bytes
• Health Plan ID - Encoded 6 bits per symbol 8 symbols = 48 bits.
• ZIP CODE of City of birth = 15 bits
• GPS Latitude and Longitude of current primary workplace (two 32-bit floats) = 64 bits
• Employer company name - 16 alphanumeric characters (encoded 6 bits per character) = 12 bytes
• Driver's License Number - 10 digits = 32 bits.
• Driver's license State (number from 00 to 49)= 6 bits
• Driver's license Expiration date (Number of days Since Jan 1, 1970) = 15 bits
• Current vehicle license plate 9 alphanumeric characters (encoded 6 bits per character) = 54 bits
• Current vehicle VIN number 17 alphanumeric characters (encoded 6 bits per character) = 102 bits
• Job Title - 16 alphanumeric characters = 12 bytes
• Annual Income in US Dollars - 1 to 14 digits = 47 bits
• Mother's maiden name (max: 20 characters) = 15 bytes
• Date of birth = 15 bits
• Telephone number with area code - 10 digits = 34 bits
• Full name - Encoded using 6 bits per character, Uppercase alphabetic characters, digits, spaces, field separator, and NULs only 50 characters = 37 bytes
• ZIP CODE of Previous residence = 15 bits
• Date moved into current residence = 15 bits
• ZIP CODE of Current residence = 15 bits
• GPS Latitude and Longitude of current residence (two 32-bit floats) = 64 bits
• Street name and house number of current resident Address (6 bits per character ) = max 20 bytes
• Apartment number or suite number = max 20 bytes
• Bank1 - Account number = 29 bits
• Bank1 - Routing number 12 digits = 37 bits
• Bank2 - Account number = 29 bits
• Bank2 - Routing number 12 digits = 37 bits
• Credit card 1 - primary account number - 12 digits = 37 bits
• Credit card 1 - CVV number - 3 digits = 10 bits
• Credit card 1 - Track 1 data 79 alphanumeric characters = 60 bytes
• Credit card 1 - Track 2 data 40 digits = 17 bytes
• Credit card 2 - primary account number - 12 digits = 37 bits
• Credit card 2 - CVV number - 3 digits = 10 bits
• Credit card 2 - Track 1 data 79 alphanumeric characters = 60 bytes
• Credit card 2 - Track 2 data 40 digits = 17 bytes
• Credit card 3 - primary account number - 12 digits = 37 bits
• Credit card 3 - CVV number - 3 digits = 10 bits
• Credit card 3 - Track 1 data 79 alphanumeric characters = 60 bytes
• Credit card 3 - Track 2 data 40 digits = 17 bytes

No... that was just Theo's excuse. He called OpenSSL's memory allocation strategy an "exploit mitigation countermeasure."

Actually, that was just a side effect, and what OpenSSL does that "counteracted" the defense is extremely common in software and software libraries.

It's also generally a good idea as far as performance is concerned ---- and with a library such as SSL which needs to process network traffic (HTTPS, for example); adequate performance is pretty darned important.

They run the plate and get the vehicle description. If it matches, odds are its not stolen or switched plates. If it doesn't match, they boot it or tow it.

Towing requires time and dispatch of the proper equipment. They'll probably just find the VIN somewhere else or break in in some manner.

If it's switched plates and they boot it, the boot can likely be removed in about 2 minutes by picking the lock then using a standard ratchet + spark plug socket. Clamps could also be taken off pretty quickly with a hand drill and a $1.00 grinding wheel or a hammer and chisel to cut through some spot welds in the design of these things, but the city authorities might try and bring up some bullshit about "damage" to city property that had to be done by the owner in order to expeditiously recover the essential use of their vehicle.

I'd have a hard time finding it. It'd require me to read the car manual to check on the location of it.

The police know where the manufacturers put the number, which include some additional locations outside and beneath the vehicle which can be read with a flashlight or optically scanned.

But they are all less convenient for the officer than looking down at the dash, or forcing entry to read it off the door frame or pop the hood to read one of the plates off the engine block, major vehicle components, or one of the other dozens of locations where extra hidden VIN plates are placed.

Perhaps because, in the USA, don't you physically change the licence plate every year?

No.... the plate is easily removable, and someone could steal it or swap an incorrect or forged plate there, but every year or 2 years you get a new special little sticker to attach to a corner your plate to show the new registration expiration date.

They'll just break out the drill if you make it too hard to pick quickly. Or the screwdriver. It's amazing what a long-handled flat-bladed screwdriver will do to your average pin/wafer tumbler lock...

I suggest the use of a lever tumbler lock design with some defense of the lock surface using hard plate steel containing tungsten-carbide chips, randomization of placement, and false drill points where an entering drill will pierce a pressurized bladder triggering separate re-locking mechanisms to prevent the car from being started or the door from being opened.

Keep your VIN number covered up.

Obstructing VIN = Violation of the law, possible Ticket.

Sufficient probable cause for police to force entry into the vehicle to investigate.

Suspicion of car theft, may result in you being detained.

"Any person who, individually or in association with one or more others, knowingly removes, changes, alters, or conceals any motor number, serial, or other identification number, decal or device affixed to a motor vehicle, trailer, semitrailer or motor vehicle part as required by federal law without the consent of the Department, shall be guilty of a Class 6 felony."