Another point that needs to be emphasized is that most journals and most "good" conferences in the field of computer science are very limited in scale. Many good journals only publish a dozen or even three or four papers in one issue, and some are quarterly or bimonthly. Many good conferences only employ thirty or forty or even twenty papers a year. So, don't be surprised when you find that there are several top publications and conferences in every field of computer.
The best magazine in the whole computer science field is JACM Journal of Computer Society. This publication is the official publication of ACM and is most widely respected. However, because the magazine claims to publish only those papers that have a long-term impact on computer science, there is inevitably a theoretical deviation. In fact, it is true: although JACM's essays cover most areas of computers, most of the papers published by JACM are purely theoretical things such as algorithms, complexity, graph theory and combinatorial mathematics. Papers in other fields are difficult to enter.
Another publication with great influence in the field of computer science is the communication of ACM (CACM). In a sense, CACM is more like nature/science than JACM. JACM is full of long speeches, definitions, theorems and proofs, not to mention that ordinary readers can't understand it, even experts in very similar fields may not understand it. CACM is a magazine, which contains not only high-level academic papers and abstracts, but also various popular science articles and news. Even for papers, CACM requires articles to be easy to understand, not to pursue strict mathematical proof, but to pursue easy-to-understand intuitive description. Ten or twenty years ago, almost all articles of CACM were classics. However, in recent years, due to the further popularization of CACM, its academic quality has declined slightly.
IEEE Computer Transactions is the best IEEE computer publication. But due to the characteristics of IEEE, it pays more attention to computer engineering than computer science. In other words, IEEE Transaction on Computers mainly publishes things in the fields of system, architecture and hardware, although its scope is wider than most publications.
As far as the quality of publications is concerned, ACM Transactions series is generally higher than IEEE Transactions series, but it cannot be generalized. Most ACM transactions are the best publications or one of the best publications in this field. Most IEEE journals are good publications in this field, but some are the best or average. Non-ACM/IEEE also has good or even the best publications. For example, SIAM Computing Magazine is considered as one of the best magazines in theory.
In fact, conference papers in the field of computer science play a greater role than journal papers. Most meetings are held once a year, occasionally every other year. Formal conference papers need anonymous review by 2-4 or more reviewers, and the received papers will be formally collected and published.
Most ACM conferences are top or excellent conferences in this field. Most IEEE conferences are excellent conferences in this field, but there are also top-level or general conferences.
The grade of a conference can usually be expressed by the employment rate of papers. Top-level meetings are usually around 20% or lower, and sometimes they can reach around 10%. The lowest employment rate I know is 7%. Good meetings are usually around 30%. When it reaches more than 40%, the reputation of the conference will be very general. More than 60% of meetings are usually difficult to get respect.
But there are exceptions. The famous STOC (ACM Computing Theory Seminar) has an employment rate of over 30%, but it is undoubtedly the best conference in theory. This situation is mainly caused by the fact that there are not many theoretical workers, and most people have a psychology of respecting and fearing STOC.
"Although we sometimes claim to be happy, our initial enthusiasm for life has disappeared." This is a sentence in the prefa