What exactly is this “weekly assessment” about
For the first time, the title “the weekly review of the building” was seen, and many would think that it was another personal diary in the form of a running book. But if you break down the brackets carefully — regularly updated, dynamic, structured, cloud technology, algorithms, back end, front end, listen/see, english, books, videos, good songs, novelty — you find this to be a very mature personal knowledge management framework. It has a structured archive of all high-value information that a person has been exposed to within a week according to technology, input, output and four dimensions of life。
I've been doing this for years, and i've stepped more than i've written. At first, i had a mustache, and i put something interesting in my notes, and then three months later i looked back, and those notes were nothing like a dump -- no way to find, no use, no memory. Then i began to figure out the logic of this “weekly review”: regular updating is the rhythm, dynamic is the source of information, architecture is the skeleton, cloud technology and algorithm is the hard core content, back and back is the landing, listening and english is the input channel, book viewing is the spiritual supply, and novelty explores the boundary。
The core question that this framework addresses is: how to make the input of fragmentation a reusable knowledge asset in an information overload. It is for all those who need continuous learning — whether you are a new developer or a technocrat who has worked for many years, or even a content creator who has no knowledge of technology but wants to build an individual knowledge system. Because it's not technology, it's technologyInformation processingThe mode of thinking。
I've seen too many people equate "learning" with "receiving." see a good article, collection; see a good project, collection; see a good experience, collection. It's been three years since the collection was filled with ash. The value of a weekly evaluation is that it forces you to do information metabolisms once a week -- digest the input, re-formulate it in your own words, and then output it. The process itself is the best learning。
How does a critical bone fit: from information flow to knowledge base
The daily reviews are too fragmented, and you may spend more time each day organizing than learning, and much of the information needs to sink for a day or two to see value. The monthly assessment was too rough, and the inspiration of a month earlier had been cold by the end of the month, and the details had been forgotten. Week is a perfect rhythm -- short enough to keep memories fresh; long enough to see the correlations and trends of information。
My own habit is to spend an hour and a half every sunday night doing this. All input sources for this week are quickly passed in the first half hour: browser bookmarks, later reading lists, podcast notes, code submission logs, wiring. In the middle of 40 minutes, filtering and categorizing, and picking out something really valuable. In the last 20 minutes, the weekly review body was written to make the core harvest clear in its own words。
Note: don't try to put everything in a weekly review. I wanted to remember everything from the beginning, and the result was 8,000 words a week, and i didn't want to read it again. Then i set a hard rule: each one has no more than three words, and the entire section has no more than ten entries. Instead of forcing yourself to make a deduction, it makes what remains more valuable。
2. 2 classification logic of four dimensions
The list of keywords in brackets appears to be confusing and can actually be grouped into four categories:
The dimensions, including keywords, central roles, handling
Technical hard core
Structure, cloud technology, algorithms, back end, front end
Professional empowerment
Depth read + manual validation
Information input
Dynamics, listen/read, english
Widening the information view
Quick filter + summary record
Knowledge deposition
Books, movies, songs
Long-term literacy accumulation
Full consumption + personal awareness
Exploration of borders
New
Stay curious
Subsurface understanding + judgement depth
This classification isn't coming from a headshot. I have tried using the “work/learning/life” score, which is too general; the “important/emergency” score, which is too utilitarian; and the “input/output” score, which is not specific enough. And finally, it turns out that the best thing to do with the use of information -- the technical hard core stuff -- is to use it, the information stuff is to be judged, the knowledge stuff is to feed, and the exploration stuff is to play with. The scenes are different, and the coping strategies are naturally different。
2. 3 regularly updated rhythm
The words “regularly updated” are simple and the hardest to do. I've seen too many people write 5,000 words in the first week, drag the second week until wednesday, and cut the third week directly. The problem is to consider weekly evaluation as a task, not a habit。
My approach is to embed it in the established rhythm of life. Every sunday night is my fixed “cleaning time”, and it's not working. If there's a problem on sunday, it's until saturday night. The point is, don't go for perfection -- there's really nothing to write about one week, just three words: "this week is mostly a busy project, with zero technology input, one book, two albums." that's a hundred times stronger than a break. The continuity of custom is much more important than the quality of a single event。
3. Technical hard core component: architecture, cloud technology, how algorithms are assessed
The structure is the most “heavy” part of the weekly review. A good architecture article, with a very high level of information density, is not enough to read it. My filter is simple: can you draw a picture? If you read an article, you can draw the core of it on the whiteboard, which means you really understand; if you don't, either the article is bad or you don't read it。
I usually have three levels of structure:
Good thinking: don't be greedy about structure. It's good to get through a structure in a week. I've tried to read five architecture articles a week, and the result is only the title. It was changed to read only one item a week, but asked for 10 minutes to make it clear to others, which was more than 10 times better。
3. 2 framework for cloud technology assessment
The content of cloud technology is being updated very quickly, and today's best practice may become obsolete tomorrow. So i'm evaluating cloud technology in weekly reviews, not new, just steady. Specifically, i would ask three questions:
What did this technology solve before it solved? If it is simply “more convenient” or “better performance”, the priority is lower. What is its core abstraction? For example, the core abstraction of packagings is “declaration api” and the core abstraction of serverless is “incident driven”. Seizing abstraction captures the essence. What are the costs of moving? If a programme worked well but was costly to move, care would have to be taken。
I usually score two points for each cloud technology entry in the weekly evaluation: maturity (1-5 points) and learning priority (1-5 points). If maturity is less than three points, wait and see; if learning priorities are higher than four points, the next week is scheduled to begin。
3. 3 method of digestion of algorithmic content
Algorithm is the most "hard" part of the weekly review and the easiest part to give up. My strategy is to take down the shock -- not to try to get all the details at once, but to build a hunch, then to go deeper。
The operation is in three steps:
N. B.: algorithms should not only be collected but not practiced. I've seen too many people collect hundreds of algorithms, and even a two-point search is the result of interviews. Each algorithm recorded in the weekly evaluation must be accompanied by at least one code that it can run itself。
4. Front-end and dynamic: the principle of trade-off for front-end content in land-fall instruments and information sources 4. 1
The speed of upgrading the front-end technology warehouse is the fastest in all technical areas. A framework was learned today and a new version may be available tomorrow. So i comment on the front-end in my weekly reviews, looking at only two types: one is a paradigm shift, from command ui to declaration ui; the other is a technique that can be used immediately, such as a new use of a css attribute。
I don't even write a weekly review. Since the update is normal, it is not worth the weekly review. Unless the version brings a destructive change or introduces a completely new programming model。
Here's my own front-end weekly review template:
At the end of the week:
- changes in paradigm: [yes/no], specifically:
- practical techniques: [listed 1-3]
- portal recording: [listed 1-2]
- plan for next week: [specific learning]This template forces me to make a deduction, to record only what really matters。
4. 2 depth requirements for backend content
The back end has a longer intellectual half-life than the front end. A ten-year-old back-end design principle may still be applicable today. So i'm evaluating the backend in my weekly review, and i'm asking for more。
Specifically, the back-end content will be asked about the principle layer. See, for example, an article on database indexing that i would not just write about "b+ tree fits to range" and i asked, "why is b+ tree fits to range?" what's the advantage of a b-plus tree? At what level of data is the advantage obvious
It takes time, but it's worth it. Because back-end knowledge is transferable -- you understand the b+ tree today, and the lsm tree tomorrow can be quickly understood, because they solve the same kind of problem。
4. 3 dynamic and listening information source management
“voice” and “hearing/watching” are the most easily neglected parts of the weekly reviews. Many people feel that these are “soft content” that is not worth the time to sort out. But my experience is quite the opposite -- these soft elements often have the most valuable signals。
For example, a technocratic community that you've been interested in, the recent week has seen a dramatic shift from “micro-services” to “single-return”, which reflects changes in the direction of the industry. Such signals are invisible in hard-tech articles and can only be captured through continuous tracking of developments。
My approach is: it takes 20 minutes a week to brush the source, but not to read in depth. See interesting titles, write down; see recurring themes, mark them; see views of subversive consciousness, highlight. This week's information trends are then summarized in three or five sentences in the weekly review。
It's the same thing to listen to. Podcasts, videos, live broadcasts, the media have less information density than text, but it's better to be associated with -- you can spend it in commuters, gymnastics, housework. I usually record in my weekly review what i heard this week, what i saw, what i should recommend。
5. English, books, video, good songs: how to organize spiritual supplies 5. 1 input strategy for english content
English is a technocrat information arbitrage tool. The same technical topics are often discussed in the english-speaking world half to a year before the chinese-speaking world. So i reserved a place for english in the weekly review。
But there is a trap for english input: it is prone to the error of learning english. Many people use english as a learning material in english, checking words, analyzing syntax, resulting in no learning of technology and no improvement in english. My approach is to learn technology in english, not technology. If you encounter an unknown word, skip it directly, as long as it does not affect the understanding of the core. You read too much, you know the words。
I usually record in my weekly review what english content has been read this week, what the core views are, and what the difference is between discussions in the chinese world. This “differentiation” is often the most valuable component。
5. 2 methods of deposition of books and visuals
Books and videos are the slowest part of the weekly review. They are not as readily available as technical articles, but they have accumulated over time and have shaped the way of thinking。
I have a principle: do not write a summary, only touch. There's no need to repeat it. What's really worth it is that the book touches me at what point and changes my mind. For example, reading a book about system thinking, i would not sum up its chapter structure, and i would write: “this book reminds me that the way i dealt with it was `dismantling', but the system thought told me that there are problems that can't be broken down, and they can't be broken.”
Video is the same. I don't write about the play, i write about what i didn't see before this video. It may be the choice of a role, it may be the metaphor of a scene, it may be the resonance of some emotion。
5. 3 good song and novel recording
Good songs and novelty are the "sortants" in the weekly reviews. It's exhausting to look at the technical content, so it's easy to do something at a time when it's sustainable。
The way i write a good song is simple: the name of the song plus the feeling of a word. Like, "a song -- a prelude goes back to a summer." there's no need for music. There's no need for analysis. There's no need for analysis。
The novelty is very broad: a new tool, a new concept, a new play. I usually record: what is it, why is it new to me, how am i going to know more? These three questions will help me judge whether this novelty is worth going deep or just new。
6. Practical exercise: writing a weekly review from scratch 6. 1 preparation: establishing your information-gathering conduit
Writing weekly reviews presupposes something to write about. If you don't normally collect information, open the file on sunday night, and have a blank head, it won't work。
Here's what i've been collecting:
It's true: the simpler the collection pipe, the better. I tried to use sophisticated labelling systems and automation tools, resulting in the maintenance tools themselves taking more time than collecting information. It was then returned to the original mode — bookmarks + memos + plain text — which was the most efficient。
6. 2 screening: what is worth writing in the weekly review
After collecting a bunch of material, the next step is screening. My screening criteria are three questions:
This thing changed what i knew before? If it's just "know," it's not worth writing. It has to be “as it was, i was wrong”. Can i use this next week? If you can use it, write first. If not, but important, it is also written, but labelled as “long-term concern”. Can i explain this thing in my own words? If it is not clear, the statement is not fully understood, either read it or not。
These three questions filter out 80% of the material. The remaining 20 per cent is really worth writing。
6. 3 writing: improved efficiency through templates
The weekly reviews are written, and naturally they form their own templates. Here's my template:
# week x
# technical hard core #
- [structure] summarizing one sentence + core harvest
- [loud technology] summarizing one sentence + core harvest
- [ algorithm] summarizing one sentence + core harvest
# back and forth
- [furrent end] paradigm change/putting techniques/put-down recording
- [back-end] principle-level questioning + core harvest
# information input
- [ dynamics] information trends this week
- [english] what's the difference with the chinese world?
- [listening/watching] recommended content + a feeling
# knowledge's settled #
- [book] touch point + cognitive change
- [video] see anything new?
- [good song] song name + a word feeling
new
- what is it? - why is it newThe advantage of this template is to be structured and flexible. The weekly entries are not the same, but are in the same format to facilitate subsequent review and retrieval。
6. 4 retrospect: making weekly assessments yield dividends
After the weekly review, don't just throw it there. I'm going to spend half an hour a month reviewing the weekly reviews of the past four weeks to see if there's a recurring theme, if there's a missing signal, if there's any direction to adjust。
This retrospect moves the weekly review from "record" to "repeated disk." the records are linear and the double disks are networked. Linear records only let you know “what i did” and web-based doubles show you “why i did it” and “what to do next”。
Note: read not only your weekly review but also the target set at the beginning of the month. If weeks of weekly evaluation deviate from the target, either the target is adjusted or action is adjusted. Weekly evaluation is a means, not an end。
7. Record of common problems and screening techniques 7. 1
This is the most common problem. My line of thought is:
Symptoms, possible causes, solutions
Open the document blank
No quality input this week
Lower the standard and write three sentences
I can't do it anymore
The material is not digested
Go back and read the material
It's a lot of bullshit
No filter
Refilter with three questions
I don't want to write for weeks
It's not built
Embedding weekly assessments into established rhythms and lowering expectations
The most fundamental solution is to collect it properly. If you are consciously entering and recording every day, there's naturally something to write by the end of the week. If you don't normally collect it and hold it on weekends, it won't work。
What if it's too loose
It's a lot easier to turn into a mess. Everything's in there. My solution is to add a "label" to each entry, indicating which dimension it belongs to. Technological hard cores, information input, knowledge deposition, border exploration, with four labels, are immediately structured。
It's a good thing if there's a particular dimension in it, which means you've invested a lot in that direction this week. If a dimension is empty for weeks, it means your input structure is unbalanced and needs adjustment。
What if we can't hold on
The underlying reason for not staying the course is usually that expectations are too high. You want to write high-quality content every week, but the reality is that there are weeks that are flat. Accept this reality and allow yourself to write “hydrography”。
My bottom line is: write at least three words a week. Even if the three sentences were “to work too much this week, to learn nothing, to sleep well”, it would be better than breaking. Because once it's broken, the cost of picking it up is ten times higher。
Also, don't write alone. Find one or two of your friends who are also doing weekly reviews, look at each other's weekly reviews, and comment on each other. Social pressure is the best incentive to insist。
What's the difference between weekly reviews and journals
A lot of people ask me that. My answer is: the diary is written for yourself, the weekly review is written for the future. The journals record emotions and events, and weekly reviews record awareness and growth. Journals can be very private, and they better be public — public means you have to be responsible for what you write, to make sure that what you write is valuable to others。
My own weekly reviews are sent in a small circle that only a dozen people can see. But it's these dozen people that gave me the momentum to keep writing. Because i know if i write too much this week, i'll be embarrassed to see you next week。
8. Some through pits and lessons learned
I made a big mistake at the beginning of my weekly review: consider it a technical note. Each article is written in detail, the code, the configuration, the rationale, and the whole article must be repeated. As a result, one of the weekly reviews wrote more than 10,000 words and didn't want to read it again。
It became clear to me that the core value of weekly evaluation is not how much information is recorded, but how much information is digested. A good weekly review should be index, not full text. It tells you what happened this week, what deserves attention and what to do next, rather than putting all the details in it。
The other pit is seeking perfect format. I tried to use tools, templates, layouts, and the result was more time in format than content. Then i went back to marktown, the most simple, the title, the list, the thicker, and three things. The format is for content and does not reverse the present。
There is also a pit where only writing technology does not write life. There was a time when my weekly review was all technical, read like a working paper. Then i forced myself to join books, videos, good songs, and the weekly reviews immediately became hot. Technology is work, life is the background. Technology makes sense only when the bottom is rich。
Finally share a small technique: add a "premonition next week" to the weekly assessment. At the end of the week, write a sentence or two about what we're going to do next week. This forecast does not need to be formal, but it will help you build continuity between weeks. When we write the weekly review next week, we'll take a look at last week's preview, check it out and analyze the reasons before it is finished. It's a simple move that can make your weekly evaluation move from an "insolved event" to a "continuous growth record."。
This could be further expanded by consolidating the monthly weekly assessments into monthly reports, quarterly monthly reports into quarterly reports and an annual review at the end of the year. The layer is moving forward, and you'll find your own growth trajectory visible。
