Protocol stack implant chip is imperative, IPSec is a mistake

This article is a bit extreme, but it may also cater to the meaning of certain individuals or organizations. This article is short, I will continue to write this topic later. The heart is too painful, the person is too bad.
1. protocol stack implant chip
If you think that the stack of chips is consuming material and cost, then you are wrong. The Tcp/ip/ethernet protocol stack has now become a de facto standard, and many teenagers have not changed. Cured things in view of the circuit design, open mode, silicon wafer technology, such as overhead, is generally the more constant things to consider curing, otherwise you have to pay a huge price, and is not conducive to the protection of customer investment. In fact, the protocol stack has long been stable, has not changed much, but why is it worth curing now? Because the CPU of the previous PC can bear the complexity of software, it is now micro-terminal, its arm CPU can't bear the huge software computational cost, and the hardware calculation will be better. I'm talking about this from the consumer's point of view. Although in the beginning, may be more expensive, but the hardware of things is so, to recover the upfront investment costs, follow-up tariff will continue to decline, as for the chip heating problem, this technical problem is absolutely can be solved, difficult to solve is the problem of consumer inertia.
Why many people or organizations do not want to do this, the first is to consider the cost. I don't know much about economics, but I don't understand it at all, I know a little bit. What I understand is that the current big brand is not in the hands of consumers, this is a very bad omen. In fact, technology in the lead consumers, if there is no 1nm process (exaggerated, the Department of the People do not seriously) did not come out, the consumers do not use, so simple, not to say that consumers need, Intel, TSMC will certainly be able to do it, the big card is also not held in Huawei HiSilicon Hand, design chip is on the one hand, Doing it is another matter. This is not the end of the Middle Ages until the 19th century, the time of the scuffle, that time is, I need what, you will give me to do out, do not come out, I find someone else! That era made the Sun King Louis 14, the achievement of Napoleon ...
Something that is worth curing is a universal thing. With the popularity of the home terminal, the rich Internet business, the family exports will encounter bottlenecks, this bottleneck is not hardware, but can use hardware breakthrough software to endure. Wait a while.
2. About IPSec
IPSec is reversing the history of the drive. As I mentioned earlier, it's foolish to encrypt the video stream with an IP layer VPN, and the network layering model is dead. I am standing on the premise of my own long-term research in this field, and not about any company or individual who is currently concerned about this area of business. If you want to solve a specific problem or be coerced, do it. The end of technology is metaphysical, if only scripted, it is wrong. Like me, I can say and make any detail of a field such as a, but I still say a is a doesn't excrement, behind it, half is metaphysical understanding, the other half is pious.
The topic of security, which is inherently application-related, is not just data security, as some traditionally think, because the security operations themselves affect the behavior of the data. It's foolish to have an ECB encryption on a video stream, and encryption should be done immediately after the video is encoded, rather than being forced to do it in an unsuspecting, yet sensitive, middle box, and I always see that you're trying. Video is just an example of a wide range of examples, full blown. How foolish it is to use the legacy IPSec to encrypt the IP layer uniformly! The IP layer is simple, simple, and does not connect to the best effort service. IPSec added two one-way SAS, which made an IP stream a two-way flow in a secure sense .... How sad. Admittedly, IPSec was successful in the file transfer era, but now it is doomed to fail. Security is application-aware and is automatically resolved by the application itself. There is no unified IP layer Security solution, if you want to do, you need to reluctantly guess, guess the contents of the packet, depth analysis, how tired.
Why do you want to cater to TCP because it's a lot of scenes?

Enters August 2, lasts, lasts.

Protocol stack implant chip is imperative, IPSec is a mistake