Rapid prototyping

Given five or six dimensional constraints, compare the performance of the two software , To speak with data . These five or six dimensions , Not limited to packet loss rate , Inherent delay , Shake, etc .

If two dimensions , A table can show , If there are four dimensions , A four-dimensional cube can show , Storing another two-dimensional table in a two-dimensional cell is equivalent , But it's already ugly . What about the six dimensions ？ No one can imagine the appearance of a six dimensional cube , Can no longer tolerate dolls two-dimensional tables .

Raw data , Obsessed with display , This typically makes the mistake of idealism , Fell into the abyss of metaphysics . Struggling with how to present , In fact, I already have hope in my heart for what it will look like , Just don't know how to do , This may be the manager's request , Maybe it's just that everyone shows it like this , Just follow suit .

If there is no need to show , How to put raw data ？ It can't be scattered at will .

MVC To solve this problem .

I came into contact with this concept more than ten years ago , But never deliberately used some recited principles to design anything , More than ten years of career has made my practice conform to MVC normal form , This may be the power of vocational training , Or call it 1 The law of ten thousand hours , Muscle memory , No matter what word .

I think so . Before considering how to present , I want to save the raw data in a structured container in a convenient way , This container is always placed somewhere , When presenting data , Detect the container according to the fields that need to be displayed . In this way , Storage , retrieval , Show complete decoupling , Mutual dependence .

The exhibition finally becomes a natural result , Not limited to tables , It can also be all kinds of pictures . Presentation and data are completely separated .

In order not to rely on database design ( The main thing is that I can't , Relying on others to cooperate may cause communication delays ), I need to program myself to realize the design of data structure and data storage , retrieval . A master programmer may act immediately , But I can't program , And want to quickly produce a prototype .

That's what I do . Using file system tree structure , For quick retrieval , May adopt ramfs.

• stay ramfs Build a directory tree organized by dimensions , Each dimension has a layer .
• The leaf of the multi-layer directory tree is a text file , Save the raw data falling on the coordinates of this dimension .
• adopt ls, cat, echo, grep/sed/awk, find, basename Wait for the command to add, delete, modify and check the directory tree .
• take shell The result of command retrieval is fed to the front-end display program , Realize the display .

Here's the picture ：

ramfs The effect of placing a directory tree in the is achieved by writing a program tree The effect of adding, deleting, modifying and checking the structure is completely the same . however , If you don't understand what rapid prototyping means , This practice will definitely be fiercely rejected due to poor performance , I want to ask , Performance all day long , How many good things should we miss before admitting that performance is the last resort .

ramfs The effect of placing a directory tree in the is similar to SQL The effect of retrieving the designed database is completely consistent .anyway,ramfs Put a tree inside ,SQL Retrieve database , Programming to realize tree Add, delete, modify and check the structure , Are all means to achieve the same goal , To prototype , Of course, which one to use quickly . The goal is not to write a high-performance , Robust programs , It's storage , Retrieve and display a pile of multi-dimensional raw data .

The essence of programming is combinatorial logic , Not in any language , What techniques are used . Quickly make a prototype , These details need to be discarded , At the same time, we must remove the strong correlation between goals and means , It's not easy .

Since more than a decade ago , I even studied a lot about design patterns in College ,MVC The book of , But after work, I almost never involved in related . Although not relevant , But imperceptibly, it has been doing layering , Decoupling , Modular work , Finally formed “ Muscle memory ”. Distinguish the levels , Clear the theme , Instead of saying anything , To solve problems efficiently . such as , I posted a circle of friends in the morning “ The lack of security is behind the pursuit of greatness and completeness , It is just fine , No more and no less is the most noble . Unfortunately, very few people can understand this . Including many people who are materially rich , Hundreds of square meters of large houses are also piled full . There is no need to own something that is used once more than ten days . Always want to store everything , For a rainy day , Pile a pile of things instead of losing space , This is insecurity . So is the enterprise , What do , We all strive to be bigger and stronger , It's too big to fall .buffbloat,aimd It's also caused by byte hoarding , Even if it's bbr This idea , People still want to send one more byte if they want to send one more byte . Can't , Small and beautiful , It is just fine , Is not understood , unfortunately , There is no sense of security because there is an idea driving , Must compete , Or we'll fall behind , This is the necessity of the environment with more wolves and less meat , In short , The result of insecurity , It's just a word , volume .” My original intention is “ In comparison 160 Flat large apartment ,330 Flat floor ,410 Feeling after the flat villa , According to your own needs , Enough is enough. , The bigger the better ？” As a result, there is a bbr, Let's focus on bbr Discuss , But ignore the noumenon . Have to say , This is a common problem of programmers , It can even be said to be a problem , Programmers only stare at the words they can see directly , It's a bit like Chaplin screwing the buttons on a girl's skirt with a wrench . Therefore, it is inevitable to lose a lot of really useful information , Thus missing the essence of the prototype , Not to mention making it quickly . Enough is enough. ？ The bigger the better ？ I believe there is no reason not to choose the latter . Write some ideas .

Zhejiang Wenzhou leather shoes wet , It's not fat when it's raining .