goq/telegram-list: a curated Telegram directory maintained as a Markdown file

The problem: Telegram discovery for developers is search-hostile

Telegram search is built for finding a chat you already know about, not for browsing a category. There is no tag system, no public index of groups by topic, and no reliable way to tell a dormant channel from an active one before you join. For a developer looking for, say, a Russian-language QA chat or a Data Science jobs channel, the practical options are word of mouth, a link posted in another chat, or a directory someone else maintains. goq/telegram-list is that directory. The README states its scope plainly: a list of interesting Telegram groups, channels and bots, aimed at programmers, developers, geeks, sysadmins and DevOps people. The audience is Russian-speaking, and the entries are mostly Russian-language resources. It is a discovery aid for humans, not an API or a dataset. If you are looking for a structured feed of Telegram metadata, this repository is the wrong shape for that job.

What is actually in the repository

The README carries the bulk of the content. It opens with two meta-entries: a news channel for the list itself and a companion chat for discussion and suggestions. Then it names a channel of the week, Data Science Jobs, described as covering work and vacancies in Data Science, Machine Learning, Deep Learning, neural networks, natural language processing and data mining. After that comes a menu that splits the material into channels and chats. The channel side is divided into news, programming and software development, frontend, information security, machine learning and data science, DevOps, business and startups, libraries and knowledge bases, jobs, miscellaneous, and personal blogs. The chat side is finer-grained: programming languages, frameworks and libraries, developer communities, game development and graphics, system and network administration plus telephony, DevOps and configuration management, QA, databases, big data and machine learning, information security and cryptography, operating systems, mathematics and natural sciences, mobile development, jobs, project management and Agile, Telegram and VK bot development, geographic information systems, general IT communities, Russian regional IT communities, electronics and embedded systems, and books and philosophy. Bots live in a separate file, bots.md, linked from the menu. Each entry is a Markdown link to a t.me address followed by a short description after a dash.

The staleness marker is the most interesting design decision

The README defines a convention: a marker next to a channel or chat name means the resource has not been updated for a long time, and the note adds that the information may still be useful but outdated. This is a small thing that most link lists skip. A directory of Telegram links decays silently, because a dead channel still resolves and still shows old posts. Marking decay inside the list itself is a manual, imperfect substitute for a health check, but it is honest about the problem rather than pretending the list is evergreen. The catch is that the marker depends on a maintainer noticing. Nothing in the repository appears to automate that check, so treat the absence of a marker as unknown status rather than proof of activity. The README does not state how often the marker is re-evaluated, and I could not confirm any cadence from the material.

Contributions arrive as pull requests against Markdown

The README invites readers to send suggestions and pull requests for channels, chats or bots relevant to programming and IT topics, and adds a caution that only relevant resources are added. That sentence is doing real work: it tells you the maintainers apply an editorial filter, so the list is curated rather than exhaustive. The contribution path is the standard GitHub one, editing the README or bots.md on the default branch, which is master. There is no schema, no front matter and no validation step described. A pull request that adds a line in the wrong section, or in a slightly different format, has nothing to fail against automatically. For a contributor that is low friction. For a consumer it means the list is only as consistent as the reviewers are. The repository shows no releases, so there is no versioned snapshot to pin against. If you build anything on top of this list, you are tracking a moving Markdown file.

Getting it running means cloning or forking, not installing

There is nothing to install. The README documents no package, no CLI and no configuration keys, because the project is a document. The practical commands are the ordinary Git ones. To get a local copy, run git clone https://github.com/goq/telegram-list.git and then read README.md, which is the main list, and bots.md, which holds the bot entries. If you want to contribute, fork the repository, edit the Markdown, and open a pull request against master. If you want to reuse the content programmatically, you would parse the Markdown yourself: the entries follow a link-then-dash-then-description shape, which is regular enough to extract with a small script, but the repository does not ship one. There is no build step, no test suite and no generated output. Anyone expecting an npm package, a Python module or a JSON export will not find one here.

Where the list breaks down

The first limitation is language. The README is in Russian and the described resources are predominantly Russian-language, so an English-speaking developer gets less value from most entries. The second is that a link list cannot tell you whether a chat is worth your time. Membership size, moderation quality, signal-to-noise ratio and whether a channel is now a paid promotion feed are all invisible in a one-line description. Third, the staleness marker is manual, as noted above, so the list can carry entries that look current and are not. Fourth, there is no licence information in the material I have, which matters if you intend to republish the list or fold it into a product. Without a stated licence, the safe assumption is that default copyright applies and redistribution needs the maintainer's permission, but that is a question for the maintainers or a lawyer, not something I can settle here. Finally, the repository has no releases and no automation, so there is no mechanism to notify you when an entry changes or dies.

How it differs from awesome-telegram and from Telegram's own search

The closest comparison is the awesome-list pattern, for example the various awesome-telegram repositories that collect Telegram clients, libraries and bots. Those lists are aimed at developers building on Telegram: they index SDKs, bot frameworks and API wrappers. goq/telegram-list is aimed at developers using Telegram: it indexes places to read and talk. The two overlap only at the edges, such as the Telegram bot development chat category. A second comparison is Telegram's in-app search. Search is query-driven and returns whatever matches your string, with no editorial layer and no category browsing. goq/telegram-list is browse-driven and pre-filtered by maintainers, which is better for exploration and worse for recall. If you need to find one specific channel by name, use search. If you want to see what exists in a category, use the list. Neither replaces the other.

Maintenance cost and who should pick it up

The maintenance model is human review of pull requests plus periodic manual marking of dormant entries. That is cheap to run and cheap to fork: the whole artifact is text, so a team can copy the sections it cares about, delete the rest, and keep its own version under internal review. The cost that does not go away is link rot. Every entry is a t.me URL pointing at a resource the maintainers do not control, so entries can go private, change topic or stop posting without any change to the repository. If you fork it, budget for someone to re-check the subset you keep. As for adoption: this list suits a Russian-speaking developer who wants a broad starting set of channels and chats, and it suits anyone who wants to fork the Markdown and curate a narrower slice. It does not suit anyone building a product that needs structured, current Telegram directory data, and it does not suit readers who need English-language resources. Verify the last push date on master, confirm the licence position with the maintainers before republishing, and check the marker convention in the README before trusting any single entry.

Editorial conclusion

Adopt goq/telegram-list if you want a ready-made, human-reviewed starting point for Russian-language developer channels and chats, or if you want to fork the Markdown and keep your own trimmed copy. Do not adopt it if you need a machine-checkable directory, an English-language list, or live membership data, because the repository supplies none of those. Before relying on it, open the file on the default master branch, check the last push date, and scan for the warning marker the README describes next to channel names.

Official sources

1. goq/telegram-list on GitHub
2. Issues
3. README